Cell Tissue Res (1991) 266:325-330

Cell and Tissue Research 9 Springer-Verlag1991

Transient expression of a calcium-binding protein (spot 35-calbindin) and its mRNA in the immature pituicytes of embryonic rats Hiroshi Abe 1, Osamu Amano 1, Thoru Yamakuni 2, Ryozo Kuwano 2, Yasuo Takahashi 2, and Hisatake Kondo 1 1 Department of Anatomy,Tohoku UniversitySchool of Medicine, Seiryo-machi2-1, Sendai, 980 Japan 2 Department of Neuropharmacology,Brain Research Institute,Niigata University,Niigata, Japan Accepted June 26, 1991

Summary. Spot 35 protein is a Ca-binding protein originating from the rat cerebellum; it is now referred to spot 35-calbindin. This protein is expressed in immature pituicytes of the neurohypophyseal anlage in the E1 IE18 rat embryo. The gene expression of spot 35-calbindin was detected by in-situ hybridization analysis only at stage E11-El 2. Profiles of spot 35-positive nerve fibers of a neurosecretory nature were found in anlage at stage El6. At this stage, some immature pituicytes are partially immunopositive for spot 35-calbindin only in their peripheral cytoplasm; others are immunonegative. At birth and thereafter through adulthood, abundant nerve fibers are the sole structures immunoreactive for spot 35-calbindin; all the pituicytes are immunonegative, resulting in a light-microscopic appearance of numerous immunonegative round profiles, corresponding to pituicytes, and capillaries embedded in the granularly immunostained neurohypophysis. The present findings suggest that, during specific embryonic stages, immature pituicytes exert some as yet unidentified roles related to Ca-mediated functions involving the expression of spot 35-calbindin. Key words: Calbindin - Neurohypophysis - Development, ontogenetic - Immunohistochemistry - In-situ hybridization - Electron microscopy - Rat (Wistar)

Substantial evidence exists for a class of soluble proteins that bind intracellular calcium ions and that may be involved in controlling intracellular calcium concentrations (Carafoli 1987). Among various calcium-binding proteins, spot 35 protein was originally isolated from the rat cerebellum; it is characterized as having a molecular weight of 27 kDa and a pI of 5.3 (Yoshida and Takahashi 1980). Immunohistochemical analysis has localized spot 35-immunoreactivity specifically in the cerebellar Purkinje neurons throughout entire neuronal doOffprint requests to: H. Abe

mains including dendritic trees, spines, and axon terminals (Yamakuni et al. 1984). In a recent molecular cloning and nucleotide sequence analysis (Yamakuni et al. 1986, 1987), a high degree of homology has been revealed in the amino acid sequence (79%) and nucleotide sequence (77%) of spot 35 protein and 28 kDa vitamin D-dependent intestinal Ca-binding protein of chicks (CaBP-28 kDa, chick calbindin) (Wasserman and Taylor ]966; Wilson etal. 1985; Hunziker 1986). After our complete analysis of the nucleotide sequence of spot 35 protein, cDNA clones have subsequently been isolated from a rat brain cDNA library using cDNA to chick intestinal calbindin. As a consequence, the isolated cDNA from the rat brain has been revealed to be identical to that of spot 35 protein (Hunziker and Schrickel 1988). Because spot 35 protein was originally isolated from the rat brain, and not from the chick intestine, and because the nucleotide sequence analysis of s p o t 35 protein was completed prior to that of the chickoriginating brain-type calbindin, we consider that the name 'spot 35' should be retained for the protein of rat cerebellar origin; it is therefore appropriate to rename spot 35 protein as spot 35-calbindin (Abe et al. 1990). It should be noted here that the occurrence of all of the calcium-binding proteins, including the spot 35calbindin, has so far been observed only in neurons; they are not found in typical glial cells of the mature nervous system, except for some ependymal cells (Garcia-Segura et al. 1984). However, a recent study of the pineal organ has revealed the occurrence of spot 35immunoreactivity in S-100 immunoreactive interstitial cells, which are of an essentially glial nature, and which enclose the pinealocytes of neuronal origin (Yamamoto et al. 1990). This finding suggests that the pineal interstitial cells exert as yet undetermined calcium-mediated functions that are not shared with other glial cells. This also leads us to question whether spot 35-imnmnoreactivity is expressed in the neurohypophysis, which is derived from another outgrowth of the diencephalon during development, and which functions in a way different from typical neural tissues. In this regard, it should be

326 noted that previous studies have d e m o n s t r a t e d the occurrence o f i m m u n o r e a c t i v i t y for chick intestine-originating calbindin in the n e u r o h y p o p h y s e a l anlage o f rat e m b r y o s (Enderlin et al. 1987), b u t only in the p a r a v e n tricular and supraoptic nuclei, which send neurosecretory axons into the n e u r o h y p o p h y s i s o f adult rats (Garcia-Segra et al. 1984; Jande et al. 1981). The present study was therefore u n d e r t a k e n to examine the chronological change in localization o f spot 35i m m u n o r e a c t i v i t y in the n e u r o h y p o p h y s i s o f rats f r o m e m b r y o n i c stages t h r o u g h a d u l t h o o d , at the light- and electron-microscopic levels. Furthermore, in-situ hybridization analysis was also p e r f o r m e d to examine the expression status o f the gene for spot 35-calbindin in the n e u r o h y p o p h y s e a l anlage. Materials and methods Adult female rats with regular estrous cycles (Wistar Imamichi) were mated with males. Vaginal smears were examined every morning and the day of sperm positivity was designated day 0 of pregnancy (EO). The pregnant females usually gave birth in the afternoon of day 22 of pregnancy, which was designated day 0 of postnatal life (PO). Under anesthesia by Nembutal (40 mg/kg b.w., intraperitoneally), fetuses at stages Ell, E12, El4, El6, El8, E20 were taken out by Cesarean section. The fetuses and postnatal rats at different ages were perfused through the heart with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.3, at room temperature. Some fetuses were irradiated in the fixative by microwaves (2450 MHz, 500 W, Panasonic NE-M340, Japan) for 15-20 s with the final temperature of the fixative below 37~ C. Thereafter, younger fetuses or excised hypophyses from older fetuses and postnatal specimens were placed in the same fixative for 4 h at 4~ C, and subsequently transferred into 30% sucrose in 0.1 M phosphate buffer, pH 7.3, at 4~ C. Specimens were frozen and cut at a thickness of 10 20 gm on a cryostat. The tissue sections were incubated with a rabbit antispot 35-calbindin antiserum at a dilution of 1:1000 for 12 h at room temperature. Control serum was obtained by the addition of 10 gg rat spot 35-calbindin to 1 ml corresponding antiserum diluted 1:1000. Detailed procedures for the isolation of spot 35calbindin and the preparation and specificity of the antiserum have been described elsewhere (Yamakuni etal. 1984; Kondo et al. 1988). The tissue sections were placed in biotinylated second antibody (anti-rabbit IgG, Vector Lab., USA) at a dilution of 1 : 200 for I h, and then in peroxidase-conjugated streptavidin (Dacopatts, Denmark) at a dilution of 1:600, or in avidin/biotin-peroxidase complex (Vectastain ABC kit, Vector Lab., USA) for 1 h. The sites of antigen-antibody reaction were made visible by the DAB (3,3'diaminobenzidine tetrahydrochloride) reaction (0.01% DAB in 0.05 M TRIS HC1, pH 7.6, with 0.002% H202). For immunoelectron microscopy, tissue sections were post-osmicated and embedded in Epon 812. cDNA for spot 35-calbindin was isolated from a cDNA library prepared from poly(A) RNA of rat cerebellum as previously described (Yamakuni et al. 1987). A BglII-PvuII fragment, about 400 bp in length, of the Y-non-coding region of spot 35-calbindincDNA was selected as a cDNA probe (probe A) specific to spot 35-calbindin. In addition, another BglII-PvuIIfragment, about 700 bp in length, was cut into two fragments by RsaI, and a second cDNA probe (probe B), also about 400 bp in length, was obtained. Cryostat sections mounted on gelatin-coated glass slides were subject to incubation for 2 h at room temperature in prehybridization mixture containing 50% deionized formamide, 4 • SSC (standard saline citrate), Denhardt's solution, 2% Sarkosyl, 1 mM EDTA and 250 gg/ml heat-denatured salmon sperm DNA in 0.1 M sodi-

um phosphate buffer (pH 7.2). The prehybridization mixture on each slide was replaced with 50 Ixl hybridization mixture, consisting of the prehybridization mixture with the addition of 10% dextran sulfate, 100 mM dithiothreitol and 5 x 105 dpm per slide of the cDNA probe (mixture of probe A and probe B) radiolabeled with [35S]dCTP by a random primed DNA labeling kit (Boehringer Mannheim, FRG). After incubation at 37~ overnight, the sections were washed 3 times in 0.1 x SSC, 0.1% Sarkosyl at 37~ for 40 min. After dehydration, the glass slides were dipped in NTB2 nuclear track emulsion (Kodak) and exposed for 14 days, developed and analyzed with bright and dark field microscopy. Results At stage E l l , which is the earliest stage examined, only a few cells exhibiting m o d e r a t e i m m u n o r e a c t i v i t y for spot 35-calbindin were identified in the n e u r o h y p o p h y seal anlage, which a p p e a r e d as the infundibular process o f the diencephalon a p p o s e d to R a t h k e ' s p o u c h . T h e spot 35-immunoreactive cells increased in n u m b e r r e m a r k a b l y by stage E l 2 , and almost all the cells in the n e u r o h y p o p h y s e a l anlage were intensely i m m u n o r e active for spot 35-calbindin (Fig. 1). The i m m u n o r e a c tive cells were elongated and aligned with their long axis perpendicular to the plane o f the neuroepithelium. In the in-situ hybridization analysis using the 35S-labeled probe, the hybridization signals were c o n c e n t r a t e d in the n e u r o h y p o p h y s e a l anlage region at stages E l l 12 (Fig. 2a, b). The silver grains were evenly distributed at the c o n c e n t r a t i o n used and no discrete aggregations c o r r e s p o n d i n g to individual cells were seen in the region. I m m u n o - e l e c t r o n m i c r o s c o p y o f the E12-13 anlage revealed that the i m m u n o r e a c t i v e cells were oval or polygonal, and h a d a large o b l o n g nucleus with a p r o m i n e n t nucleolus in the center. M a n y m i t o c h o n d r i a were observed but Golgi a p p a r a t u s and endoplasmic reticulum were n o t well developed. Junctions o f the intermediate type were f o r m e d between adjacent i m m u n o r e a c t i v e cells. The i m m u n o r e a c t i v e material was distributed diffusely t h r o u g h o u t the c y t o p l a s m o f the cells, except for the interior o f the nucleus and certain organelles, including m i t o c h o n d r i a and endoplasmic reticulum. N o nerve fibers with or w i t h o u t i m m u n o r e a c t i o n were f o u n d a m o n g the i m m u n o r e a c t i v e cells (Fig. 3). The whole n e u r o h y p o p h y s e a l anlage o f stage E l 4 was still distinctly i m m u n o r e a c t i v e as seen by light microscopy (Fig. 4). However, in the in-situ hybridization analysis, a significant a c c u m u l a t i o n o f the silver grains was no longer detected in the n e u r o h y p o p h y s e a l anlage by E14 a n d thereafter (Fig. 5). By E16 a n d 18, the n e u r o h y p o p h y s e a l anlage generally appeared immunoreactive, but m a n y small r o u n d profiles were seen t h r o u g h o u t the anlage. These profiles were i m m u n o n e g a t i v e and seemed to c o r r e s p o n d to cells and capillaries (Fig. 6). By i m m u n o - e l e c t r o n microscopy, a m a r k e d variety in the intensity and distribution o f the i m m u n o r e a c t i o n was noted in individual cells o f the anlage. W h e r e a s some cells were diffusely i m m u n o r e a c tive t h r o u g h o u t the entire cytoplasm, but n o t in the nucleus, m a n y others were partially i m m u n o r e a c t i v e in the peripheral c y t o p l a s m (Fig. 7). Substantial n u m b e r s o f cells i m m u n o n e g a t i v e for spot 35-calbindin were also

327

Fig. 1. A sagittal section of the diencephalon of an E12 rat embryo showing the neurohypophyseal anlage intensely immunoreactive for spot 35-calbindin. D Diocoel; R Rathke's pouch, x 150. Bar: 100 gm Fig. 2a, b. Dark-field (a) and bright-field (b) micrographs of the diencephalon of an E12 embryo processed for in-situ hybridization with cDNA specific to spot 35-calbindin. Note the dense accumulation of autoradiographic silver grains on the neurohypophyseal anlage (arrowheads) apposing Rathke's pouch (R). D Diocoel. • 150. Bar: 100 gm

ture pituicytes. Immunoreactive material is diffusely distributed throughout the cytoplasm, but not in the nucleus (iV). No nerve fibers are present among the cells, x 5100. Bar: I gm Fig. 4. Sagittal section of the diencephalon of an E 14 embryo showing the neurohypophyseal anlage (asterisk) still intensely but diffusely immunoreactive for spot 35-calbindin. D Diocoel; R Rathke's pouch, x 150. Bar: 100 gm

Fig. 3. Immuno-electron micrograph of the neurohypophyseal anlage of an El2 rat embryo showing spot 35-immunoreactive imma-

Fig. 5. Dark-field micrograph of the diencephalon from the same embryo as in Fig. 4, processed for in-situ hybridization with cDNA specific to spot 35-calbindin. No significant hybridization signals are detected in the neurohypophyseal anlage (asterisk). D Diocoel; R Rathke's pouch. • 160. Bar: 100 gm

f o u n d to be r a n d o m l y distributed in the anlage. Thin nerve fibers, 0.3-0.6 g m in diameter and characterized by n u m e r o u s vesicles, were first f o u n d a m o n g the immunoreactive and i m m u n o n e g a t i v e cells at stage E16, as previously reported by conventional electron m i c r o s c o p y (Fink and Smith 1971) (Fig. 8). T h e y were intensely imm u n o r e a c t i v e for spot 35-calbindin and were in direct apposition to or partially enclosed by i m m u n o r e a c t i v e or i m m u n o n e g a t i v e cells. At birth and thereafter, the n e u r o h y p o p h y s i s was moderately i m m u n o r e a c t i v e (Fig. 9). I m m u n o - e l e c t r o n

m i c r o s c o p y d e m o n s t r a t e d that all cells c o m p o s i n g the n e u r o h y p o p h y s i s were immunonegative, and nerve fibers were the sole structures i m m u n o r e a c t i v e for spot 35-calbindin (Fig. 10). The i m m u n o r e a c t i v e nerve fibers were 0.3-3.0 g m in diameter and contained granular vesicles o f varying sizes, 90-160 n m in diameter, together with microvesicles, 50-60 n m in diameter. The i m m u n o reaction was localized diffusely in the axoplasm, except for the interior o f the vesicles. The organization o f these cells and nerve fibers h a d essentially the same appearance as in the adult n e u r o h y p o p h y s i s . F r o m these find-

Fig. 6. Immuno-light micrograph of the neurohypophyseal anlage at stage E18. Note the many small round profiles with no distinct immunoreaction, intermingled with immunopositive cells; see also Fig. 9. x 290. Bar: 50 gm

Fig. 7. Immuno-electron micrograph of the neurohypophyseal anlage at stage E18. Note the marked variety of immunoreaction for spot 35-calbindin in individual pituicytes (P): some cells are extensively immunoreactive, whereas others exhibit the immunoreaction in their peripheral cytoplasm or no immunoreaction at all. x 8600. Bar: 1 g m Fig. 8. Immuno-electron micrograph showing a spot 35-immunoreactive nerve fiber (n), containing various vesicles and apposing

an immunonegative pituicyte (P) at stage El8. 0.1 gm

x 38900. Bar:

Fig. 9. Immuno-light micrograph of the neurohypophysis of a P7 rat. Spot 35-immunoreactive structures with a granular appearance are distributed around many clear round profiles of varying sizes corresponding to immunonegative pituicytes (arrows) and capillaries (asterisk). x 290. Bar: 50 gm Fig. 10. Immuno-electron micrograph of the neurohypophysis of a P7 rat showing an immunonegative pituicyte (P) adjacent to neurosecretory axons (n) immunoreactive for spot 35-calbindin. Li Lipid droplet, x 24900. Bar: 1 g m

329 ings, the cells under discussion were regarded as being immature pituicytes and m o s t of the nerve fibers as being neurosecretory axons. After incubation of sections with the control serum in place of the anti-spot 35-calbindin antiserum, no cells or fibers were found to exhibit positive immunoreactivity in the neurohypophysis at any stage of development.

Discussion The present immunohistochemical study demonstrates, in accord with a previous one (Enderlin et al. 1987), that the immature pituicytes of rats display an intense immunoreactivity for Ca-binding spot 35-calbindin, albeit only transiently during a limited time in embryonic development. Although according to Enderlin et al. (1987) calbindin immunoreactivity in the neurohypophysis appears not earlier than at stage E14, the present study shows that this type of immunoreactivity is evident at stage E l l - E l 2 in the neurohypophyseal anlage. The discrepancy m a y be due to the different strains of experimental animals or to the fixatives used in the studies. The in-situ hybridization analysis confirms that the immunoreactivity is a result of the presence of authentic spot 35-calbindin because the m R N A for this protein has been detected in the neurohypophyseal anlage. It should be noted that the hybridization signals are detected significantly in the neurohypophyseal anlage region only at stages E11-12, but are no longer detectable in the region at stage E14, when immature pituicytes still remain immunoreactive for this protein. This finding indicates that the expression of the gene for spot 35calbindin is down-regulated at the transcriptional level at stage E14, and that the remaining immunoreactivity is attributable to the longer turnover time of this protein than that of its m R N A . The occurrence of immature pituicytes partially immunoreactive for this protein together with the immunonegative pituicytes in the E l 6 - 1 8 anlage adds weight to this suggestion. Such a gradual disappearance of the immunoreactivity in the immature pituicytes and the concomitant invasion of spot 35-immunoreactive neurosecretory nerve fibers at stage E16, and thereafter, result in the superficial immunoreactivity of the entire neurohypophyseal anlage throughout the late embryonic and postnatal stages. A time lag exists between the down-regulation of the gene expression and the invasion of the immunoreactive neurosecretory fibers. This suggests that the down-regulation of the gene expression for spot 35-calbindin in the pituicytes is intrinsically p r o g r a m m e d , and it excludes the possibility that contact of the immature pituicytes by the invading nerves m a y be involved in the down-regulation; however, several reports indicate that the gene expression o f certain proteins in cells of a glial nature is regulated by apposing neurons (Taniuchi et al. 1988; Mirsky et al. 1980; Trapp et al. 1988). It is well known that intracellular calcium ions are involved significantly in various cellular functions, such as the stimulation ofcalcium-calmodulin-dependent protein kinases and protein kinase C, the operation of calci-

urn-dependent potassium channels and the secretion of various substances, including neurotransmitters (Carafoli 1987). Moreover, previous studies have shown that calcium ions m a y play a crucial role in neural tissue development, such as neural fold elevation, neurulation and neurite elongation (Moore and Stanisstreet 1986; Smedley and Stanisstreet 1985; Kater et al. 1988). The above evidence and the transient occurrence of the calcium-binding spot 35-calbindin in immature pituicytes imply that, during a limited embryonic period, these cells exert some as yet unidentified roles related to the calcium-mediated functions described above, but not shared with other glial cells.

Acknowledgements.This work was supported in part by a grant from Takefu Psychiatry Hospital Foundation, Fukui Prefecture, Japan. References Abe H, Amano O, Yamakuni T, Takahashi Y, Kondo H (1990) Localization of spot 35-calbindin (rat cerebellar calbindin) in the anterior pituitary of the rat: development and sexual differences. Arch Histol Cytol 53:585-591 Carafoli E (1987) Intracellular calcium homeostasis. Annu Rev Biochem 56:395-433 Enderlin S, Norman AW, Celio MR (1987) Ontogeny of the calcium binding protein calbindin D-28k in the rat nervous system. Anat Embryol 177:15-28 Fink G, Smith GC (1971) Ultrastructural features of the developing hypothalamo-hypophysial axis in the rat. A correlative study. Z Zellforsch 119:208-226 Garcia-Segura LM, Baetens D, Roth J, Norman AW, Orci L (1984) Immunohistochemical mapping of calcium-binding protein immunoreactivity in the rat central nervous system. Brain Res 296:75-86 Hunziker W (1986) The 28-kDa vitamin D-dependent calciumbinding protein has a six-domain structure. Proc Natl Acad Sci USA 83:7578-7582 Hunziker W, Schrickel S (1988) Rat brain calbindin D28: six domain structure and extensive amino acid homology with chicken calbindin D28. Mol Endocrinol 2:465-473 Jande SS, Maler L, Lawson DEM (1981) Immunohistochemical mapping of vitamin D-dependent calcium-binding protein in brain. Nature 294:765-767 Kater SB, Mattson MP, Cohan C, Connor J (1988) Calcium regulation of the neuronal growth one. Trends Neurosci 11:315 321 Kondo H, Yamamoto M, Yamakuni T, Takahashi Y (1988) An immunohistochemical study of the ontogeny of the horizontal cell in the rat retina using an antiserum against spot 35 protein, a novel Purkinje cell-specific protein, as a marker. Anat Rec 222:103-109 Mirsky R, Winter J, Abney ER, Pruss RM, Gavrilovic J, Raft MC (1980) Myelin-specific proteins and glycolipids in rat Schwann cells and oligodendrocytes in culture. J Cell Biol 84: 483-494 Moore DCP, Stanisstreet M (1986) Calcium requirement for neural fold elevation in rat embryos. Cytobios 47:167-177 Smedley M J, Stanisstreet M (1985) Calcium and neurulation in mammalian embryos. J Embryol Exp Morphol 89 : 1 14 Taniuchi M, Clark HB, Schweitzer JB, Johnson EM Jr (1988) Expression of nerve growth facter receptors by Schwann cells of axotomized peripheral nerves: ultrastructural location, suppression by axonal contact, and binding properties. J Neurosci 8 : 664-681 Trapp BD, Hauer P, Lemke G (1988) Axonal regulation of myelin

330 protein mRNA levels in actively myelinating Schwann cells. J Neurosci 8 : 3515-3521 Wasserman RH, Taylor AN (1966) Vitamin D3-induced calciumbinding protein in chick intestinal mucosa. Science 152 : 791-793 Wilson PW, Harding M, Lawson DEM (1985) Putative amino acid sequence of chick calcium-binding protein deduced from a complementary DNA sequence. Nucleic Acids Res 13:88678881 Yamakuni T, Usui H, Iwanaga T, Kondo H, Odani S, Takahashi Y (1984) Isolation and immunohistochemical localization of a cerebellar protein. Neurosci Lett 45: 235-240 Yamakuni T, Kuwano R, Odani S, Miki N, Yamaguchi Y, Takahashi Y (/986) Nucleotide sequence of cDNA to mRNA for

a cerebellar Ca-binding protein, spot 35 protein. Nucleic Acids Res 14:6768 Yamakuni T, Kuwano R, Odani S, Miki N, Yamaguchi K, Takahashi Y (1987) Molecular cloning of cDNA to mRNA for a cerebellar spot 35 protein. J Neurochem 48 : 1590-1596 Yamamoto M, Kondo H, Yamamkuni T, Takahashi Y (1990) Expression of immunoreactivity for Ca-binding protein, spot 35 in the interstitial cell of the rat pineal organ. Histochem J 22 :4~ 10 Yoshida Y, Takahashi Y (1980) Compositional changes in soluble proteins of cerebral mantle, cerebellum, and brain stem of rat brain during development. A two-dimensional gel electrophoretic analysis. Neurochem Res 5:81-96