a combined immunohistochemical and electron microscopic study of the second cell type in the...

Cop>righI 0 MunkJguurd. 1997

Journal of Pineal Research ISSN 0742-3098

A combined immunohstochemical and electron microscoDic studv of the second cell type in the I rl

developing sheep pineal gland

Franco A, Regodon S, Masot AJ, Redondo E. A combined immunohistochemical and electron microscopic study of the second cell type in the developing sheep pineal gland. J. Pineal Res. 1997; 22:130- 1 3 6 . 0 Munksgaard, Copenhagen

Abstract: Ultrastructural and immunohistochemical techniques were used to study the second cell type in sheep embryo pineal glands. Thirty-two embryos were studied from day 54 of development through birth. Specimens were arranged in four age groups, defined in terms of the most relevant histological features: Group 1 (54-67 days of prenatal development), Group 2 (71-92 days), Group 3 (98-1 13 days), and Group 4 (1 18-1 50 days). At 98 days, a second cell type was observed which differed from pinealoblasts and showed uniform ultrastructural characteristics similar to those of astrocytes in the central nervous system. Ultrastructural homogeneity was not matched by the results of histochemical and immunohistochemical analysis: while all Type I1 cells stained positive to phosphotungstic acid hematoxylin, only 50% expressed glial fibrillary acidic protein. In the course of ovine intrauterine development, the vascular affinity of this second cell population, composed of glial-like or astrocytic cells at varying stages of maturity, leads to the formation of a limiting pineal barrier. This barrier may constitute the morphological expression of a hypothetical functional involvement in the exchange of substances between blood and pineal parenchyma.

Antonio Franco, Sergio Regodon, Antonio J. Masot, and Eloy Redondo Department of Anatomy and Histology, Faculty of Veterinary Medicine, University of Extremadura, 10071 Caceres, Spain

Key words: immunohistochemical- ultrastructural - second cell type - pineal gland - sheep

Address reprint requests to Dr. Antonio Franco Rubio, Anatomia y Embriologia, Facultad de Veterinaria, Universidad de Extremadura, E-10071 Caceres, Espana.

Received October 2, 1996; accepted February 18,1997.

introduction golden hamsters [Huang et al., 19841, carnivores

Three cell types have been defined in mammal pineal gland parenchyma: pinealocytes or principal cells, supportive or interstitial cells, and pigmented cells [Pevet, 19771. Numerous terms have been used to designate the second type, including interstitial cells [Wolfe, 1965; Moller et al., 1978; Huang et al., 1984; Cozzi, 19861, Type I1 pinealocytes [Calvo and Boya, 1983, 1984 a,b; Boya and Calvo, 19841, glial cells [Anderson, 1965; Calvo et al., 1988a1, and astrocytes [Xu Zang et al., 1985; Lbpez-Mufioz et al., 1992a; Boya and Calvo, 19931.

The morphology of this second cell type has been studied using silver impregnation methods with gold chloride sublimate [Del Rio-Hortega, 19221 or silver carbonate [Scharenberg and Liss, 19651. Elec- tron microscopic studies of the pineal gland in mice and rats [Luo et al., 1984; Schachner et al., 1984; Calvo and Boya, 1984b, 1985; Calvo et al., 1988b],

130

[Calvo et al., 1988a, 1990; Boya et al., 19951, horses [Cozzi, 19861, and ruminants [Anderson, 19651 have all pointed to the abundance of microfilaments in the perikaryon and cell processes as the most strik- ing ultrastructural characteristic of these cells.

Immunohistochemical demonstration of glial antigens, including glial fibrillary acidic protein (GFAP), vimentin, and S-100, in the second pineal cell type of various mammals has underlined the glial nature of this population Moller et al., 1978; Sozos Papasozomenos, 1983; Calvo et al., 1988b; L6pez-Muiioz et al., 1992a,b; Boya and Calvo, 19931.

Expression of antigens has largely been demonstrated by means of light microscopy. The scale of the immunohistochemical reaction of glial cells and cell processes often renders impossible the correct identification of cells [Schachner et al., 19771; hence, the importance of electron microscopy in this type of research. Even so, and except for a study in

Second cell type in sheep pineal gland

made manually using projected images. Mor- phometrical analysis was based on random sam- pling, the sample being considered satisfactory when the standard deviation was less than 5% of the mean; the formula proposed by Aherne and Dunnill [ 19821 was employed: n = (d0.05 x ) ~ where n = number of samples, s = standard deviation of sample, x = sample mean, 0.05 = desired error.

rats by Calvo et al. [1988b], little work has been done in this area. There has been no previous research in sheep, particularly during prenatal development. Thus, the purpose of the present study was to ascertain the nature of the second cell type in developing sheep pineal gland.

Materials and methods

Animals

Thirty-two ovine embryos at different stages of development, ranging in age from 54-150 days, were used for this study. Specimens were arranged in four age groups, each containing eight embryos (four males, four females): Group 1 (54-67 days of prenatal development), Group 2 (7 1-92 days), Group 3 (98-113 days), and Group 4 (118-150 days). To obtain embryos at different stages of development, cesarean section was performed after syn- chronization of estrus, using hormonal techniques and uterine flushing. Fluorogesterone acetate was administered 14 days before introduction of males. Six hundred IU of pregnant mare serum globulin (PMSG) (Sigma Aldrich Quimica, Madrid, Spain) was then inoculated. Cesarean section was performed from day 54 following introduction of males. Animals were tranquilized with an intramuscular injection of 0.5 mg/100 kg body weight of propionyl phenothiazine, and anesthesia was induced by intravenous injection of sodium thiopental (4 g in a 20% aqueous solution). Once separated from maternal linking, embryos were killed by umbilical vein administration of 4 g sodium thiopental in a 20% aqueous solution.

Light microscopy

Embryo pineal glands were sliced parasagitally after 1 hr in Carnoy’s fluid. One of the two portions thus obtained was fixed in 10% neutral formalin saline and processed by paraffin-embedding methods. Sections 3 pm thick were cut and stained with hematoxylin and eosin (HE) for routine morphological examination, and with phosphotungstic acid hematoxylin (PTAH) for detection of glial-type cells.

Morphometrical analysis

Numerical cell density was determined by calculat- ing the number of PTAH- and GFAP-positive cells. Five sections, separated from each other by a dis- tance of roughly 50 pm, were taken for each gland. Ten fields measuring 10,000 pm2 were randomly se- lected per section. Fields and sections were identi- cal for both staining techniques. Calculations were

Electron microscopy (EM)

The other half of each pineal gland was used for ultrastructural analysis. Tissue blocks (1 mm3) were immersed in ice-cold 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.2), with or without the addition of 3% sucrose. After washing in buffer, tissue blocks were postfixed in 1% OsO4 in 0.1 M cacodylate buffer (pH 7.2), dehydrated through al- cohol series and embedded in epoxy resin. Ultrathin sections were cut, stained with lead citrate, and uranyl acetate and examined with a Jeol Jem 100 S-X electron microscope.

Light immunohistochemical analysis

An avidin-biotin-peroxidase complex (ABPC) was carried out on deparaffinized pineal samples for detection of glial fibrillary acidic protein (GFAP), the main protein component of intermediate astrocyte filaments. Sections were incubated in diluted (150) normal swine serum (Dako, Madrid, Spain) for 15 min to reduce background and then incubated in diluted (1: 1,000) rabbit anti-bovine GFAP (Dako, Madrid, Spain) for 3 hr at 20°C. Diluted (1:1,500) biotinylated swine anti-rabbit IgG (Dako, Madrid, Spain) was placed on the sections for 30 min, and sections were then incubated with diluted (1 5 0 ) ABPC reagent (Dako, Madrid, Spain) for 1 hr. Af- ter diaminobenzidine reaction, a nuclear counter- staining with hematoxylin was applied. Negative controls consisted of the substitution of primary an- tisera for nonimmune rabbit serum. Adjacent brain tissue sections served as positive controls.

lmmunohistochemical electron microscopic analysis (IEM)

Ultrathin sections cut from blocks obtained for electron microscopy were stained with colloidal gold for detection of GFAP-positive cells. Sections were treated with a saturated aqueous solution of sodium metaperiodate and then incubated in a moist chamber at 37°C with TRIS-buffered saline (TBS) 0.05 M (pH 7.6) and 1% bovine serum albumin (Sigma Aldrich Quimica, Madrid, Spain). Sections were subsequently incubated with primary antibody (rabbit anti-bovine GFAP, diluted 1 : 1,000) at 37°C for

131

Franco et al.

3 hr. After incubation, sample sections were washed three times in TBS 0.05 M, pH 7.6. Finally, they were incubated with diluted (1 50) anti-rabbit IgG, 10 nm gold conjugate (Biocell, UK) for 60 min in a moist chamber at room temperature and contrasted with lead citrate and uranyl acetate.

Results

Structural examination

The pineal parenchyma in Groups 1 and 2 displayed a uniform morphology; cells were all of one type, identified as pinealoblasts. They are elongated cells, with round nuclei in which the chromatin are not homogeneously distributed and small nucleoli. In Groups 3 and 4, in addition to pinealoblasts a second cell type with different characteristics was detected. Cell morphology was similar in both these groups, and cells were generally located in the perivascular space (Fig. 1). Nuclei were small, oval, or rounded, and electron-dense (Figs. 2, 3) , and thus readily distinguishable from the vesicular nuclei of pinealoblasts.

Morphometrical examination

Numerical cell density using the PTAH stain was 18 * 2 celldfield in Group 3 and around 28 * 4 celldfield in Group 4. The number of embryonic pineal gland cells expressing GFAP was greater in Group 4 (14 * 2) than in Group 3 (9 * 1).

Among all glial cells (PTAH-positive), only a small amount express GFAP positivity. The numerical density of GFAP-positive cells was lower than that of PTAH-positive cells.

Ultrastructural examination

Ultrastructural analysis confirmed light microscopy findings: a second cell type was evident in Groups 3 and 4, in addition to the pinealoblasts present in all groups. Type I1 cells were less numerous, more electron-dense, and showed a clear preference for perivascular locations (Fig. 4). The most characteristic feature of these cells was the presence of very long processes with numerous microfilaments (Fig. 5). The endoplasmic reticulum was mostly granular; cis- ternae had fairly narrow lumina. Lysosomes with clearly defined limiting membrane; diplosomes and ribosomes were observed in perinuclear cytoplasm.

Ultrastructurally, Type I1 cells in Group 4 closely resembled those observed in Group 3, except that the former exhibited a greater number of filaments (Fig. 6). The processes of Type I1 cells are closely associated with pinealoblast processes.

132

lmmunohistochemical examination

GFAP-positive cells were observed in the embryonic pineal parenchyma in Groups 3 and 4. In Group 3 , these were distributed uniformly throughout the gland, preferentially located close to blood vessels (Fig. 7). Immunopositive cells (whose appearance resembled that of the CNS astrocytes used as positive control) displayed small, dense, ovoid nuclei, and an intensely-staining rim of cytoplasm border- ing negative nuclei (Fig. 8). GFAP-positive cells displayed a small number of processes, with varying diameters and arranged both longitudinally and transversely (Fig. 8). Cell processes were interwo- ven among pinealoblasts, and around blood vessels to form a limiting barrier (Fig. 9).

In Group 4 embryos, GFAP-positive cells were either oval or elongated in shape (Fig. 10). Cyto- plasmic processes, which were more numerous than in Group 3, varied in both diameter and orientation (Fig. 10).

In both groups, colloidal gold labeling revealed expression of GFAP by cells whose morphology closely resembled that described ultrastructurally for the second cell type. These cells were observed in perivascular locations (Fig. 11). exhibited ovoid a n d or elongated non-staining nuclei, and strong cytoplasmic positivity, with clear affinity for the microfilaments of cytoplasmic processes (Figs. l l, 12). Immunostaining was more intense in Group 4 (Fig. 12) than in Group 3 embryos (Fig. 11). Adjacent pinealoblasts remained immunonegative.

Discussion

The results obtained indicate the existence of a second cell type, in addition to pinealoblasts, during the prenatal development of sheep pineal gland. Light microscopy highlighted the resemblance between these cells and the second cell type described in the pineal gland of other species, including mice and rats [Calvo and Boya, 1983; Luo et al., 1984; Schachner et al., 1984; Calvo et al., 1988b], golden hamsters [Huang et al., 19841, rabbits [Garcia- Mauriiio and Boya, 19921, carnivores [Calvo et al., 1990; Boya and Calvo, 1993; Boya et al., 199.51, horses [Cozzi, 19861, and ruminants [Anderson, 19651. However, ultrastructural homogeneity rendered impossible the detection of possible subtypes. Histochemical and immunological tests were thus performed with a view to overcoming this limita- tion and at the same time enhancing our poor knowl- edge of this second cell type. The first step was to determine the presence of glial-like cells by PTAH immunostaining .

GFAP is widely considered a valid label for the de-


Fig. 1. Embryo, 98 days. Pineal parenchyma (detail) showing pinealoblasts and second cell type with perivascular affinity (arrows). H-E x 500. Fig. 2 . Embryo, 113 days. Type I1 cell nuclei (arrows); nuclei are small, dense and ovoid. PTAH x 350. Fig. 3. Embryo, 118 days. Type I1 cell nuclei (arrows). PTAH x 350. Fig. 4 . Embryo, 102 days. Perivascular space; Type I1 cell

cytoplasmic process with abundant microfilaments. EM x 10,000. Fig. 5. Embryo, 113 days. Type I1 cell cytoplasmic processes (asterisks) containing abundant microfilaments. EM x 15,000. Fig. 6. Embryo, 150 days. Moderately electron-dense Type I1 cell. Perinuclear cytoplasm contains abundant ribosomes and microfilaments. EM x 10,000.

tection of astrocyte development [Valentino et al., 19831, and more particularly as evidence of the presence of astrocytic cells at a certain stage of maturity [Huang et al., 1984; L6pez-Muiioz et al., 1992a,b]. The second step was thus to determine GFAP expression.

Comparison of the results of these two tests

showed that of all glial (PTAH-positive cells), only a proportion were positive to GFAP. The numerical density of GFAP-positive cells was lower than that of PTAH-positive cells, suggesting that a certain proportion of glial (PTAH-positive) cells may be immature astrocytes not expressing GFAP, a find-

133

Franco et al.

Fig. 7. Embryo, 98 days. Positive-staining cells (arrows) scattered throughout the pineal gland. ABPC x 350. Fig. 8. Embryo, 106 days. Intense cytoplasmic staining re- sponse along the border of negative nuclei (arrows); longitudinally arranged cell process. ABPC x 500. Fig. 9. Embryo, 11 3 days. Immunopositive cells close to blood vessels and scattered throughout the pineal parenchyma (arrows). ABPC x 500.

Fig. I f f . Embryo, 123 days. Abundant elongated immunopositive cells. ABPC x 250. Fig. 11. Embryo, 98 days. Immune electronmicroscopic demonstration of GFAP in the cytoplasm of Type 11 cells (arrows). IEM x 10,000. Fig. 12. Embryo, 118 days. Immune electronmicroscopic demonstration of GFAP, nonstained nuclei and positive cytoplasm in Type I1 cells. IEM x 20,000.

ing already reported in rats [Schachner et al., 19841 and in carnivores [Lopez-Mufioz et al., 1992a,b; Boya and Calvo, 19931.

The results obtained here indicate that the second cell population in developing ovine pineal gland is in fact a combination of glial-astrocyte cells at

134

varying stages of maturity. This hypothesis has already been advanced in the case of adult rat pineal gland [Lopez-Mufioz et al., 1992al.

Because the second cell type was detected in the present study at 98 days gestation, sheep can be classified as a species showing early astrocyte matu-


One finding reported for a number of mammalian species is that the GFAP-positive cell processes were located close to perivascular spaces [Moller et al., 1978; Huang et al., 1984; Schachner et al., 1984; Cozzi, 1986; Jeffrey et al., 19901. Although Lopez- Muiioz et al. [ 1992 a,b] report no close relation between GFAP-positive cells and blood vessels, the present results confirming the close relationship suggested by a number of other authors [Wolfe, 1965; Moller et al., 1978; Calvo and Boya, 1983, 1985; Sozos Papasozomenos, 1983; Calvo et al., 1988b; Boya and Calvo, 19931, may indicate a hypothetical functional significance of Type I1 cells as sub- strate for the exchange of substances between the pineal parenchyma and the bloodstream [Schachner et al., 1984; Xu Zang et al., 1985; Suarez et al., 19871. During embryonic development of ovine pineal gland, the perivascular-located Type 11 cells might constitute some kind of barrier function, but we cannot a f E i that it is a blood-pineal barrier similar to the one detected in human fetal pineal gland [Sozos Papasozomenos, 1983; Xu Zang et al., 19851, in cows [Anderson, 19651, and in cats [Xu Zang et al., 19851. This hypothetical functional role would complement the support function (similar to that of astrocytes in the CNS) traditionally attributed to these cells. The irregular disposition of Type I1 cells interspersed throughout the pineal parenchyma may represent morphological evidence of this support function [L6pez-Muiioz et al., 1992a,b].

ration. As Boya and Calvo [1993] have suggested in the case of dogs and cats, this early maturation, together with the implicit exclusion of certain fac- tors present in the CNS (direct contact with neurons) make the ovine pineal gland an excellent model for the study of astrocyte development.

Comparison of the present results with those reported elsewhere reveals considerable discrepancies with regard to the timing of the appearance of Type 11 cells in the pineal gland. Our findings agree with those of Sozos Papasozomenos [1983], who reports their appearance in human fetal pineal gland at 24 weeks gestation (final term). Other authors, however, disagree. Boya and Calvo [I9931 make no reference to a second cell type in carnivores until the second week post- partum. A comparison is in any case rendered difficult because of the paucity of studies on embryonic development of mammalian pineal gland.

Both ultrastructurally and in terms of antigen expression, the second cell population, consisting of PTAH- and GFAP-positive cells, exhibited characteristics similar to those reported for astrocytes [Cozzi, 1986; Calvo et al., 1988b; L6pez-Muiioz et al., 1992a,b; Boya and Calvo, 19931. This similar- ity, together with the lack of material available for more specific comparison, led to the use of astrocyte development in the CNS of various mammalian species as a point of reference for comparative purposes. A considerable discrepancy was immedi- ately evident: Some authors have failed to detect astrocytes during embryonic development of the CNS [Schnitzer et al., 19811; others report their appearance during the first term [Levitt and Rakic, 1980; Choi, 19811 or the second term of pregnancy [Hewicker-Trautwein et al., 19941. In the present study, the number of Type I1 cells increased as gestation progressed, a finding also reported for human fetal pineal gland [Sozos Papasozomenos, 19831. A direct relationship was detected, as fetal age advanced, between the increase in cytoplasmic filaments and increased intensity of the cytoplasmic reaction to GFAP. Similar observations are reported by Cozzi [ 19861 in equine pineal gland.

Comparison of the location of Type I1 cells in various mammalian species suggests that cell topography is closely linked to the anatomical location of the gland. In sheep, where the gland is deeply located within nerve structures, GFAP-positive cells are detected across the whole gland surface. Simi- lar results have been reported in hamsters [Sheridan and Reiter, 19701, and in cats and dogs [Boya and Calvo, 19931. In rats, however, where the pineal gland is more superficially located, these cells are found only in the stalk and the most superficial por- tion of the gland [Luo et al., 1984; Mikkelson and Moller, 1990; L6pez-Muiioz et al., 1992a,b].

Acknowledgments

The authors are grateful to Miguel Angel Gdmez and BelCn Garcia Abadias of the Department of Pathological Anatomy, Faculty of Veterinary Medicine, Murcia, and German Fernandez of the Faculty of Veterinary Medicine, Chceres, for technical assistance.

Literature cited

AHERNE, W.A., M.S. DUNNILL, eds.( 1982) Morphometry. Lon- don: Arnold.

ANDERSON, E. (1965) The anatomy of bovine and ovine pineals. Light and electron microscopic studies. J. Ultrastruct. Res. 12: 1-80.

BOYA, J., J.L. CALVO (1984) Structure and ultrastructure of the aging rat pineal gland. J. Pineal Res. 1:83-89.

BOYA, J., J.L. CALVO (1993) Immunohistochemical study of the pineal astrocytes in the postnatal development of the cat and dog pineal gland. J. Pineal Res. 15:13-20.

BOYA, J., J.L. CALVO, D. RANCANO (1995) Structure of pineal gland in the adult cat. J. Pineal Res. 18:112-118.

CALVO, J.L., J. BOYA (1983) Postnatal development of cell types in the rat pineal gland. J. Anat. 186:185-195.

CALVO, J.L., J. BOYA (1984a) Postnatal evolution of the rat pineal gland: Light microscopy. J. Anat. 138:45-53.

CALVO, J.L., J. BOYA (1984b) Ultrastructure of the pineal gland in the adult rat. J. Anat. 138:405409.

CALVO, J.L., J. BOYA (1985) Ultrastructure of the rat pineal stalk. Acta Anat. 123:172-177.

135

Franco et al.

CALVO, J.L., J . BOYA, J.E. G A R C I A - M A U R I ~ O (1988a) Ultra- structure of the pineal gland in the adult dog. J. Pineal Res. 5:479487.

(1988b) Presence of glial cells in the rat pineal gland: A light and electron microscopic immunohistochemical study. Anat. Rec. 220:424428.

CALVO, J.L., J. BOYA, J.E. GARCIA-MAURINO, A. LOPEZ CARBONEL (1990) Postnatal development of the dog pineal gland: Elec- tron microscopy. J. Pineal Res. 8345-254.

CHOI, B.H. (1981) Radial glial of developing human fetal spi- nal cord: Golgi, immunohistochemical and electron microscopic study. Devel. Brain Res. 1:249-267.

Cozzi, B. (1986) Cell types in the pineal gland of the horse: An ultrastructural and immunocytochemical study. Anat. Rec.

DEL RIO-HORTEGA, P. ( I 922) Constitucidn histol6gica de la glandula pineal. Arch. Neurob. 3:359-389.

GARCIA-MAURINO, J.E., J. BOYA (1992) Postnatal maturation of the parenchymal cell types in the rabbit pineal gland. Histol. Histopathol. 7:75-81.

HEWICKER-TRAUTWEIN, M., N. KRUEGER, K. URBAN, G. TRACIT- WEIN (1994) Immunohistochemical localization of glial and neuronal cell markers in the developing bovine brain. Anat. Histol. Embryol. 23: 154-165.

HUANG, S.K., M. NOBILING, M. SCHACHNER, R. TAUCNER (1984) Interstitial and parenchymal cells in the pineal gland of the golden hamster. A combined thin-section, freeze-fracture and immunofluorescence study. Cell Tissue Res. 235:327-337.

JEFFREY, M., G.A.H. WELLS, A.W. BRIDGES (1990) An immunohistochemical study of the topography and cellular localization of three neural proteins in the sheep nervous system. J. Comp. Pathol. 103:23-35.

LEVITT, P., P. RAKIC (1980) Immunoperoxidase localization of glial fibrillary acidic protein in the radial cells and astrocytes of the developing rhesus monkey brain. J. Comp. Neurol.

LOPEZ-MUNOZ, F., J . BOYA, J.L. CALVO, F. MARIN (1992a) Im- munohistochemical localization of glial fibrillary acidic protein (GFAP) in rat pineal stalk astrocytes. Histol. Hispathol.

.OPEZ-MU~OZ, F., J.L. CALVO, J. BOYA, A. LOPEZ CARBONEL (1992b) Coexpresion of vimentin and glial fibrillary acidic protein (GFAP) in glial cells of the adult rat pineal gland. J. Pineal Res. 12:145-148. ,LIO, Z.R., R.L. SCHULTZ, E.F. WHITTER, L. VOLLRATH (1984) Ul- trastructural characterization of glial cells in the rat pineal

CALVO, J.L., J . BOYA, A. BORREGON, J.E. GARCIA-MAI!RINO

21 6: 165-174.

193:815-840.

7:643-646.

gland with special reference to the pineal stalk. Anat. Rec.

MIKKELSEN, J.D., M. MOLLER (1990) A direct neural connec- tion from the intergeniculate leaflet on the lateral geniculate nucleus to the deep pineal gland demonstrated with PHA-L in the rat. Brain Res. 520:342-347.

MOLLER, M., A. INCILD, E. BOCK (1978) Immunohistochemical demonstration of S-100 protein and GFA protein in interstitial cells of the rat pineal gland. Brain Res. 14O:l-13.

PEVET, P. (1977) On the presence of different populations of pinealocytes in the mammalian pineal gland. J. Neural Transm. 40:289-304.

SCHACHNER, M., T. HEDDLEY-WHYTE, D. Hsu, G. SCHOONMAKER, A. BIGNAMI (1977) Ultrastructural localization of glial fibrillary acidic protein in mouse cerebellum by immunoperoxidase labelling. J . Cell Biol. 75:67-73.

SCHACHNER, M., S.K. HUANG, P. ZIEGELMULLER, B. BIZZINI, R. TAUGNER (1984) Glial cells in the pineal gland of mice and rats. A combined immunofluorescence and electron-microscopic study. Cell Tissue Res. 237:245-252.

SCHARENBERG, K., L. Liss (1965) The histologic structure of the human pineal body. Prog. Brain Res. 10:193-217.

SCHNITZER, J., W. FRANKE, M. SCHACHNER (1981) Immunocy- tochemical demonstration of vimentin in astrocytes and ependymal cells of developing and adult mouse nervous system. J . Cell Biol. 90:435447.

SHERIDAN, M.N., R.J. REITER (1970) Observations on the pineal system in the hamster. 11: Fine structure of the deep pineal.

SOZOS PASOZOMENOS, CH. (1983) Glial fibrillary acidic (GFA) protein-containing cells in the human pineal gland. J . Neuropathol. Exp. Neurol. 42:391408.

SUAREZ, I., B. FERNANDEZ, G. BODEGA, P. TRANQUE, G. OLMOS, L.M. GARCIA-SEGURA (1987) Postnatal development of glial fibrillary acidic protein immunoreactivity in the hamster ar- cuate nucleus. Devel. Brain Res. 37:89-95.

VALENTINO, K.L.. E.G. JONES, S.A. KANE (1983) Expresion of GFAP immnureactivity during development of long fiber tracts in the rat CNS. Devel. Brain Res. 9:317-336.

WOLF^, D.E. (1965) The epiphyseal cells: An electron-microscopic study of its intercellular relationship and intracellular morphology in the pineal body of the albino rat. Prog. Brain Res. 10:322-376.

Xu ZANC, X., G. NILAVER, B.M. STEIN, M.R. FETELL, P.E. DUFFY (1 985) Immunocytochemistry of pineal astrocytes: Species differences and functional implications. J. Neuropathol. Exp. Neurol. 44:486495.

210:663-674.

J . Morphol. 131:163-177.

136

a combined immunohistochemical and electron microscopic study of the second cell type in the...

Documents