Detection and localization of 14-3-2 protein in primary cultures of embryonic rat brain

Brain Research, 184 (1980) 455--466 © Elsevier/North-Holland Biomedical Press

455

D E T E C T I O N A N D L O C A L I Z A T I O N OF 14-3-2 P R O T E I N IN P R I M A R Y C U L T U R E S OF E M B R Y O N I C RAT BRAIN

JEAN SECCHI, D O M I N I Q U E LECAQUE, M A R I E - A N N E COUSIN, DANIELLE LANDO, LUC1ENNE L E G A U L T - D E M A R E and JEAN-PIERRE R A Y N A U D

Centre de Recherches Roussel-Uclaf, 93230 Romainville and (L.L.-D.) Laboratoire de Biochlmie Cellulaire, Colldge de France, 75231 Paris Cedex O1 (France) (Accepted July 19th, 1979)

Key words: neuron specific enolase - - synapses - - cell culture - - immunocytochemistry

SUMMARY

The development of embryonic rat brain m cell cultures was studied by an immunocytochemical method based on the detection of 14-3-2 protein (neuronspecific enolase or NSE), a neuron-specific protein. This protein was already present in undifferentiated neurons (less than 5 days in culture), being dispersed throughout the cytoplasm, though seemingly concentrated in the vicinity of polyribosomal structures. It was not found in nuclei, in mitochondria or in the Golgi apparatus. During neuron differentiation, the location of 14-3-2 protein was related to neurite development insofar as it was detected along the axon and even in what could be taken to be the presynaptic region of numerous interneuron contacts. In contact areas, a thickening of the junction membrane was observed but the presence of 14-3-2 protein was always unilateral demonstrating the absence of a true synapse and reflecting the halt in neurite development observed after 15 days in culture. The presence of 14-3-2 protein in the cell cultures was confirmed by a microcomplement fixation assay. The protein detected in cell cultures had the same immunological properties as that found in the 17-day-old embryo, but was slightly different from that found in adult rat brain. This observation can be confronted with the lack of neuron maturation in the immunocytochemical studies.

INTRODUCTION

Since its discovery, its isolation and the preparation of antisera from bovine nervous tissue 15, the 14-3-2 protein (neuron-specific enolase or NSE protein) has been extensively investigated. It is found in many animal species with a high degree of immunological cross-reactivity and is only located in neurons4, t7. It is an acidic

456 protein, with enolase activity ~ and the monomer has a molecular weight of about 4 0 , 0 0 0 1 1 . It appears early during development. It is detected in the chick embryo as soon as the cell-types are morphologically recognizable 6, m mouse nervous t~ssue its level increases from the day after birth until approximately 6 months of age 5 and in the superior cervical ganglion of the rat its level increases during the first 5 weeks of life8. Immunocytochemistry and hght microscopy reveal its presence in the cytoplasm of neurons and in the neurltes of brain tissue17; the combined use of electron microscopy and peroxidase-conjugated antiserum has enabled the detection of immune complexes on plasmic and perinuclear membranes and on endoplasmic ret~culum m the pre- and postsynaptic regions, but not in ghal cells and in neighboring non-nervous tissue7, T M . These observations imply that 14-3-2 protein has a neuronal origin. It is transported from the site of synthesis in the neuronal perlkaryon to the nerve-endings 12, 17. In vitro studies using the mlcrocomplement fixation technique on neuroblastoma have shown that 14-3-2 protein is present throughout all stages of cellular growth and that its activity is markedly increased during the stationary phase1,1°, 13. In view of these results, it seemed worthwhile applying the ~mmunocytochem~cal detection method of the 14-3-2 protein to primary cultures of embryonic rat brain on the premise that the detection of in vitro 14-3-2 protein is a valuable means of neuron identification. Antiserum activity was controlled on neuroblastoma, on other cell-lines and on sections of brain tissue from adult animals. Until now, the identification of different cellular patterns, and in particular of neurons, in primary cultures, has been difficult, unreliable and, more often than not, has depended upon correlation of results from multiple procedures. The subcellular localization and ontogeny of the 14-3-2 protein in primary cell cultures should provide useful information on neuronal differentiation and a comparative study w~th adult brain tissue could enable the evaluation of neuron maturity m vitro. MATERIALS AND METHODS

Animals Embryonic brains used for the cultures came from pregnant Sprague-Dawley rats (Iffa-Credo, France). Brain sections were obtained from adult male rats weighing 150-200 g (from the same origin). The animals were fed standard pellets and water ad libitum. Production of antiserum to 14-3-2protein Antisera to 14-3-2 protein (serum 61 and serum 62) were prepared at the Pasteur Institute. The 14-3-2 protein, extracted from bovine brain, was purified according to a published method 9. Rabbits were injected intravenously with 1 mg (10 × 0.2 ml) of the protein. A booster injection was given 3 weeks later and again one week after that. The sera were tested by an immunodiffusion method (Fig. 1). Primary cultures Seventeen-day pregnant rats were decapitated. The embryos in their sacs were

457

Fig. 1. Double diffusion in Agar of antisera prepared against purified 14-3-2 protein, a: central well: purified bovine 14-3-2 protein, b: central well:crude soluble extracts of mouse brain. Anti-bovine 14-3-2protein sera: serum 61 (wells 1,5),serum 62 (well 2), serum9(well 3); anU-rat-NSEserum (well4); well 6 was empty. removed and placed in a Ham F 10 (Gibco) culture medium (pH 7.2) containing 10 fetal calf serum (Eurobio), glucose (6 g/l), penicillin (100 units/ml), streptomycin (100 #g/ml) and fungizone (25 #g/ml). The upper part of the encephalon (diencephalon and telencephalon), following removal of the meninges, was suspended in a Hepes buffer (25 mM Hepes, 137 mM NaC1, 5 mM KC1, 10 mM glucose, 0.7 mM NazHPO4 adjusted to pH 7.2 with 1 N NaOH) under a binocular lens. The tissue was minced and dissociated, using a Pasteur pipette to take up and gently expel the suspension 5-10 times. The dispersed cells were centrifuged (7 min at 800 rpm), resuspended in the culture medium and plated on 35 mm diameter polystyrene Petri dishes (Corning) (2.5 × 105 cells/sq, cm in 1.5 ml of medium). The fleshly plated cultures were incubated at 37 °C in 95 ~ air-5 ~o COg. The medium was replaced on the fifth day of culture and then changed 3 times a week. Cell-line cultures Experiments were performed with the following cell-lines: mouse neuroblastoma (clones NIE-115) kindly provided by Dr. M. Nirenberg; rat glial cell-line C-6, primary mouse culture A/J embryonic fibroblasts and mouse glial cell-line TRM6 obtained from Dr. Y. Netter. Immunocytochemistry assays were carried out on these different cell-lines taken at day 3 of the stationary phase. Brain tissue The rats were anesthetized with ether and decapitated. The brains were immediately removed, cross-sectioned at the infundibulum" level to reveal the hippo-

458 campal area and then frozen in isopentane chilled in liquid nitrogen. Eight/~m cryostat sections were made at --15 °C and placed on glass slides.

Immunocytochemistry with light and electron microscopy 19 Having carefully removed the incubation medium, the cultures were immediately fixed for 45 rain at 4 °C in a solution containing 4 % paraformaldehyde and 0.1% picric acid in PBS (phosphate-saline buffer, pH 7.4). The hippocampal sections were fixed in the same way. After washing the cultures and sections in PBS at 4 °C fol 1 h, they were incubated as follows: 30 mln in rabbit 14-3-2 antiserum (antiserum 61) diluted to 1/200 with Tris.NaC1 buffer (pH 7.6) containing 0.1% gelatine; 30 min in anti-rabbit sheep serum (lgG) also diluted to 1/200 in Tris.NaC1; 30 rain in PAP (peroxydase-antiperoxydase, Bionetics) diluted to 1/20 in Tris.NaCl. After each step the preparations were washed twice in Tris.NaC1 buffer for 5 min. The preparations were then stained for 10 min with freshly prepared Tris buffer (pH 7.6) containing 0.05 ~ DAB (3,3 diamino-benzidin-tetrahydrochlorlde, Merck) and 0.01 ~ hydrogen peroxide. After washing in distilled water (15 min) the hippocampal sections were dehydrated in ethanol, cleaned in xylene and mounted for light microscopy. The washed cultures were postfixed in PBS buffer containing 1 ~o osmium tetroxlde at room temperature, dehydrated in a graded series of ethanol and embedded in a 1-2 mm thick layer of Epon. After polymerization (at 60 °C for 4 h) the Epon discs with the cell monolayer and substrate were separated from the bottom of the culture dishes. Cell areas, selected by light microscopy, were cut out, reembedded in Epon, and then sectioned parallel to the plane of the cell monolayer using a Ultrotome IlI LKB with a diamond knife. Thin, unstained sections from selected areas were examined under a Siemens Elmlskop 101 electron microscope. The specificity of the immunocytochemical reaction was checked by: (a) replacing the specific serum by a non-immune rabbit serum; and (b)by omitting one of the fundamental steps of the reaction. Quantitative complement fixation assay for 14-3-2 protein The assay was performed according to a modification of the microcomplement fixation technique 20. RESULTS

Observations: light microscopy Examination of primary cultures of embryonic rat brain at different stages in their development has shown that several cells react with 14-3-2 antiserum as reflected by the intensity of staining. Up until 5 days in culture (Fig. 2a) the reactive cells remained in a cluster and appeared rounded or ovoid in shape; diffuse dark brown staining appeared in the cytoplasm whilst the nucleus remained virtually unstained. The intensity of staining varied according to the cell, the proliferating glial and fibroblast cells being easily distinguished by the presence of background coloration only. After 5 days in culture, fine stained cytoplasmic processes gave the cell a bi- or

459

t

,J

Fig. 2. Immunocytochemmal reaction with anti-14-3-2 antibody (light microscopy) (× 650 for a-g, × 300 for h and 1). Primary cultures, a: after 5 days in culture' clustered cells with stained cytoplasm (see arrows), b and c: after 12 days in culture: differentiated neurons with stained perinuclear cytoplasm (cy) and processes; one neurite (ne) is highly stained up to the growth cone (gc). f: after 12 days in culture with non-immune rabbit serum:no staining, g: after 12 days m culture with 40 mM KCI. well-differentiated neurons with highly-stained cytoplasm and neurltes. Neuroblastoma NIE 115 : d and e: granular staining round the nuclei and along the processes up to the growth cones (gc). Adult rat hippocampus: h: cryostat section showing high staining of the neurons of the pyramidal striatum (SP) only. SO ~trmtum orient" SR ~triatum radiatHm i" control with non-immune serum: no stainint,.

460

b Fig. 3. Immunocytochemical reaction with anti-14-3-2 antibody (electron microscopy) after 12 days in culture ( × 40,000). a: part of the cell body of a neuron with numerous PAP complexes dispersed throughout the cytoplasm. The mitochondria (m), the cisternae of the reticular endoplasmlc reticulum (rer) and the nucleus (N) are unstained. Inset" Golgi apparatus (G). b: control experiment with nonimmune rabbit serum showing no staining.

461 tripolar appearance. The stained cells were more dispersed and were located abo~,e the bottom-layer cells. After 12 days in culture, many bi- or tripolar stained cells with short cytoplasmic processes were visible but few had processes at least twice the length of the cell body (Fig. 2b and c). In all cases cytoplasmic and neurite staining paralleled cell maturation and intense, discontinuous, chaplet-like staining was usually observed in the longest processes. The growth cones were lightly stained (Fig. 2b). These observations are comparable to those on differentiated neuroblastoma NIE 115 celllines (Fig. 2d and e) in which, howevel, most cells were strongly stained whereas, in primary cultures, the glial cells and fibrob!asts, which had proliferated to form a bottom layer, had a pale non-specific background coloration. Controls without antiserum were unstained (Fig. 2f). Between 5 and 30 days in culture there seemed to be a halt in neurite development and neuron differentiation remained incomplete (unshown). An increase in the potassium ion concentration (3.8-40 mM) of the culture medium (Fig. 2g), however, led to the appearance of differentiated cells with long and ramified processes and to a reduction in the network of glial cells, fibroblasts and other cells. The 14-3-2 antiserum was inactive in other cultures (Schwannoma, Glioma, Fibroblastoma). In hippocampal sections (Fig. 2h) only the cell bodies and neurites of the Amon's horn pyramidal cells were strongly stained; the astrocytes, oligodendrocytes and endothelial cells of the blood vessels in the adjacent layers remained unstained. Observations: electron microscopy Between 11 and 15 days in culture marked anti-14-3-2 peroxydase activity was observed in the cytoplasm and in cell processes with a positive reaction in light microscopy. The PAP complexes were dispersed throughout the hyaloplasm and were located in the vicinity of the ribosomes (Fig. 3a), either in contact with the numerous polysomes present at this stage of development, or in contact with membrane ribosomes. No precipitates were visible in the lumen of the reticulum, in the Golgi apparatus, in the perinuclear space, in the nuclear matrix or on the inner nuclear membrane. The plasma membrane was unevenly and probably unspecifically stained around the edge. Little or no reaction was observed with non-immune serum in the cytoplasm and in the nucleus (Fig. 3b). Some cell processes also had numerous dispersed PAP complexes (Fig. 4) and were in contact with unreactive processes with unspecifically stained membranes (Fig. 4a, b, c and f). These contacts, which were either contacts 'en passant' or contacts between an axon-like ending and a neurite or cell body, were usually desmosome-like. The fibres in contact were swollen but only one of them was highly stained and contained mitochondria and sometimes vesicles scattered in the hyaloplasm (Fig. 4e). Similar contacts with thickening of the membranes and presence of synaptic vesicles have been seen in sections stained with uranyl acetate and lead (Fig. 5). In the immunocytochemical studies, membranes in contact were symmetrically thickened and more intensely stained than the other parts of the plasma membrane. Contact was generally between stained and unstained fibres except in the case of stained neurites

462

Fig. 4. Immunocytochem~cal reaction with anti-14-3-2 antibody (electron microscopy) after 12 days m culture. Neunte-neurite contacts (a, b, c and f) ( × 40,000). In each case dispersed PAP complexes were present in one neunte only; thickening of the membrane (see arrows) was observed in the contact area. e: ( ~. 30,000) presence of several synaptlc-hke vesicles (sv) In the reactive neurite, d: ( × 40,000) desmosome-hke contact (see arrow) between neurlte (n) and cell-body (cb) with PAP complexes present m the hyaloplasm, m, mitochondria.

463

Fig. 5. Developing synapse after 12 days in culture. Synaptic vesicles (sv) are present m the axon-termmal and polysomes (see arrows) are present in the postsynaptic zone. m, mitochondrla; mt, microtubtdes. (Uranyl q- lead, ~ 40,000). and neuronal cell soma (Fig. 4d). The cells neighboring the reactive cells remained unstained; only in the macrophages did we find numerous opaque, dense, electron inclusions. The immune complexes in the neuroblastoma N I E 115 cells were located as m the neurons o f primary cultures, but the absence o f synapses prevented comparison with primary cultures. I n all cases, the control preparations revealed no P A P precipitates in the cytoplasm and in the cell processes o f the neurons.

100C

n .r;OI

°//"

71~

o'.~'\:e ~.:.A~

1~o

looo

"'A.°" ".A

lo(~oo

Protein (ng) Fig. 6. Microcomplement fixation with antiserum to 14-3-2protein in primary cell cultures from embt yomc rat brain (A .... A) (29 days m culture), 17-day-old embryonic rat brain (© .... O), and adult rat brain (~-------~). Antiserum (no. 61) ddution = 1 : 10,000.

464

Assay of the 14-3-2protein by microcomplementfixation Assays of the 14-3-2 protein were performed by the microcomplement fixation technique to confirm the presence of this protein m primary cultures (Fig. 6). The concentration recorded in cell cultures was about the same as that found in a 17-dayold embryo but much lower ( ~ 1/10th) than that in adult brain. The protein detected in cell cultures was similar to that of the embryo, but some immunologic dissimilality existed between the immature (embryo and cell cultures) and mature (adult) protein in so far as only 50 ~ of the maximum fixation of the mature rat was recorded for the immature rat. DISCUSSION According to available data, the 14-3-2 protein is exclusively located in neurons6,7,16. Our results on hippocampal sections, in which only the large neurons of the pyramidal striatum were stained, go alongwith these data and are comparable to the results obtained by Pickel et alA v on brain stem sections using the Sternberger ~9 method (unlabeled antibody enzyme method using PAP complex). The staining of our sections (short duration of fixation without embedding) was, however, more marked since they were made with a cryostat at low temperature, as described for the immunocytochemical detection of proteins S100 and GFA in glial cells ~4, thus leading to greater sensitivity. Similarly, all the cells of neuroblastoma cultures (known to contain 14-3-2 protein) were stained, but to different extents, whereas no cellular staining was observed in non-neuronal culture lines, in which no 14-3-2 protein has been detected by the microfixation complement technique. Our experiments have furthermore revealed the presence of the 14-3-2 protein in primary cultures of embryonic rat brain tissue in cells possessing the morphological characteristics of neurons. The protein was detected as early as 5 days in culture, and remained present throughout culture. However, as regards the specific localization of the immune complex in electron microscopy studies, our results in primary cultures do not concord in all points with observations by Grasso et al. 7 and Persson et al. 16 on nervous tissue sections from adult rats. No reaction was observed in the internal nuclear membrane of the cell body or the internal layer of the nuclear envelope and immune complexes were detected not only in the membranes of the granular endoplasmic reticulum but also throughout the hyaloplasm. The fact that staining was more intense in the vicinity of free polysomes (which are widespread in the cytoplasm of differentiating cells, whilst the granular and endoplasmic retlculum is only slightly developed) reflects active protein synthesis, as demonstrated by Zomzely-Neurath et al. 21 with free polysomes from rat cerebral cortex. The detection of 14-3-2 protein along and at the extremities of the cell processes, on the other hand, supports the results of Plckel et al. 17 on NSE protein in brain stem sections (NSE protein was prepared from the rat but has all the characteristics o f o x 143-2 protein) The protein, being dispersed throughout the hyaloplasm, can migrate towards the neurlte endings, as observed by Marangos et al. lz for the axoplasmic transport of 14-3-2 protein in the chick optic nerve According to these authors, the

465 protein is synthesized on free polysomes and its migration is related to the slow component of the flux. In the case of differentiating neuroblasts, as studied here, the protein appeared before neurite formation and its migration along the neurites, in particular in the axon, closely followed neurite development. Its presence in the growth cone, as observed in light microscopy, constituted even further evidence for such migration and might indicate a possible role in synaptogenesis or in synaptic function. According to our observations by electron microscopy, the long stained processes, probably axon-like, originating from immunoreactive cells set up contacts with numerous unstained processes of undefined nature and origin. Most of these contacts were desmosome-like with symmetrical thickening of the membranes, but occasionally the morphological appearance of the contact suggested a developing synapse. Indeed, it is known that in neuron cultures desmosome-like contacts are formed and thickening of the membranes occurs before synaptogenesisa. However, the concentration of opaque material on each side of a synaptic membrane characterizing a true synapse was not seen in our cultures, though it has been noted for 14-3-2 protein in sections of frontal cortex from adult rats. In these sections, the 14-3-2 concentration was greater in the postsynaptic web and membrane, though also high in the presynaptic membranes and densities16,la. Thus it would appear that, under our culture conditions, the junction complexes were not completely formed, maybe because neuron differentiation and consequently 14-3-2 migration to the postsynaptic zone were incomplete. This lack of maturation observed in immunocytochemical studies was confirmed by a microcomplement fixation assay. The protein detected in cell cultures had the same immunological properties as that found in the 17-day-old embryo, but was slightly different from that found in adult rat brain. This difference could be explained by the fact that, in the embryo, only the hybrid enolase form (1 NSE subunit-1 NNE subunit) is probably present2L Furthermore, we have recently noted (unpublished data) by the microcomplement fixation assay that the hybrid enolase form is qualitatively different from the enolase 3 form (2 NSE subunits). Further experiments are necessary to establish the presence of 14-3-2 protein in the two compartments of the synapse in primary cell cultures as already observed in adult neurons 16 and to determine whether this bilateral localization can be related to synapse function. ACKNOWLEDGEMENTS We thank Dr. Claire Zomzely-Neurath (Roche Institute for Molecular Biology) for the kind gift of anti-NSE rat serum, Dr. Blake W. Moore (Washington University School of Medicine), Dr. Alfonso Grasso (Instituto di Biologia Cellulare, Rome) and Professor A. E. Bussard (Institut Pasteur) for anti-14-3-2 protein bovine serum. The technical assistance of A. Hannecart (experimental work) and of A. Holbach (photographic work) is gratefully acknowledged. REFERENCES 1

Augusti-Tocco,G., Casola, L. and Grasso, A., Neuroblastomacells and 14-3-2, a brain specific protein, Cell Diff., 2 (1973) 157-161.

466 2 Bock, E., Fletcher, L., Rider, C. C. and Taylor, C. B., The nature of the two proteins of brain specific antigen 14-3-2, J. Neurochem., 30 (1978) 181-185. 3 Bunge, M. B., Bunge, R. P. and Peterson, E. R., The onset of synapse formatmn rn spinal cord cultures as studied by electron microscopy, Brain Research, 6 (1967) 728-749. 4 Cicero, T. J., Cowan, W. M., Moore, B. W. and Suntzeff, V., The cellular locahzatlon of the two brain specific proteins, S-100 and 14-3-2, Brain Research, 18 (1970) 25-34. 5 Cicero, T. J., Ferrendelh, J. A., Suntzeff, V. and Moore, B. W., Regional changes m CNS levels of the S-100 and 14-3-2 proteins during development and aging of the mouse, J. Neurochem., 19 (1972) 2119-2125. 6 Cicero, T. J. and Provme, R. R., The levels of the brain-specific protein, S-100 and 14-3-2, m the developing chick spinal cord, Brain Research, 44 (1972) 294-298. 7 Grasso, A., Haghd, K. G., Hansson, H. A., Persson, L. and Ronnb~ck, L., Locahzatlon of 14-3-2 protein m the rat brain by immunoelectron microscopy, Brain Research, 122 (1977) 582-585. 8 Grasso, A. and Plrazzl, R., Changes m the concentration of the brain specific protein 14-3-2, during the development of the superior cervical ganghon of the rat and effects of surgical decentralization, Brain Research, 90 (1975) 324-328. 9 Grasso, A., Roda, G., Hogue-AngeletU, R A., Moore, B. W. and Perez, V. J., Preparation and properties of the brain specific protein 14-3-2, Brain Research, 150 (1978) 117-133. 10 Herschman, H. R. and Lerner, M. P., Production of a nervous-system-specific protein (14-3-2) by human neuroblastoma ceils in culture, Nature New Biol., 241 (1973) 242-244. 11 Marangos, P. J., Zls, A. P., Clark, R. L and Goodwin, F. K., Neuronal, non-neuronal and hybrid forms of enolase in brain: structural, immunological and functional comparisons, Brain Research, 150 (1978) 117-133. 12 Marangos, P., Zomzely-Neurath, C., York, C. and Bondy, S. C., Axoplasmlc transport of a brainspecific soluble protein, Bwchim. biophys. Acta (Amst.), 392 (1975) 75-81. 13 McMorris, F. A., Kolber, A. R., Moore, B. W. and Perumal, A. S., Expression of the neuron-speofic protein, 14-3-2, and steroid sulfatase m neuroblastoma cell hybrids, J. cell. Physiol., 84 (1974) 473--480. 14 Moller, M., Ingdd, A. and Bock, E., Immunohistochemlcal demonstration of S-100 protein and GFA protein m interstitial cells of rat pineal gland, Brain Research, 140 (1978) 1-13 15 Moore, B. W. and McGregor, D., Chromatographic and electrophoretic fracttonatmn of soluble proteins of brain and hver, J. biol. Chem., 240 (1965) 1647-1653. 16 Persson, L., R6nnb~ck, L., Grasso, A., Haghd, K. G., Hansson, H.-A., Dolonms, L., Mohn, S.-O. and Nygren, H., 14-3-2 Protein m rat brain, J neurol Sci., 35 (1978) 381-390. 17 Plckel,V. M.,Re~s,D.J.,Marangos,P. J. andZomzely-Neurath, C.,Immunocytochemicallocahzation of nervous system specific protein (NSP-R) m rat brain, Brain Research, 105 (1976) 184-187 18 Ronnb~ck, L., Persson, L., Hansson, H. A., Haghd, K. G. and Grasso, A., 14-3-2 Protein m rat brain synapses, Experlentia (Basel), 33 (1977) 1094-1095. 19 Sternberger, L. A., Hardy, P. H., Jr., Cucuhs, J. J. and Meyer, H. G., The unlabeled anubody enzyme method of lmmunohistochemlstry. Preparation and properties of soluble antJgen/antzbody complex (horseradish peroxidase-antihorseradish peroxldase) and its use m identification of spirochetes, J. Histochem. Cytochem., 18 (1970) 315-333. 20 Wasserman, E. and Levine, L , Quantitative mlcrocomplement fixation and its use in the study of antigenic structure by specific antlgen antibody mhlbltmn, J. Immunol., 87 (l 961) 290-295. 21 Zomzely-Neurath, C., York, C. and Moore, B. W , In wtro synthesis of two brain-specific proteins (S100 and 14-3-2) by polynbosomes from rat brain. I. Site of synthesis and programming by polysome-derived messenger RNA, Arch. Biochem., 155 (1973) 58-69. 22 Zomzely-Neurath, C. and Keller, A , The different forms of brain enolase Isolation, characterization, cell specificity and physmlogical significance. In S. Roberts, A. Lajtha and W. H. Glspen (Eds.), Mechanisms, Regulation and Special Functions of Protein Synthesis in the Brain, Vol 2, Elsevier, Amsterdam, 1977, pp. 279-298.