Journal of Neuroscience Research 29:107-113 (1991)
Monoclonal Antibody Br4 Recognizes Specific Neuronal Cell Types M. Saito and M. Saito Division of Neurobiology , Nathan S . Kline Institute for Psychiatric Research, Orangeburg, New York A monoclonal antibody termed as Br4 was prepared by a fusion of the myeloma P3X63-AG8-653 with spleen cells from BALB/c mice immunized with chick embryonic brain cells. Immunocytochemically, it reacts strongly with certain neurons in cerebral cortex, hippocampus, substantia nigra, and granular layer of cerebellum from rat brain but does not react with the white matter. A monoclonal antibody Br4 also reacts with primary cultured chick neurons, but not with cultured astrocytes. Western blots show that Br4 recognizes three proteins of 145,000, 108,000, and 97,000 molecular weight from the rat brain. The protein with molecular weight 145,000 (p145) and the protein with 97,000 molecular weight (p97) are essentially soluble; p145 is especially enriched in the synaptosomal soluble fraction. The protein with 108,000 molecular weight (plO8) is found in both membranebound and soluble forms. Western blots show that among the tissues examined the three proteins are found most abundant in brain and especially enriched in the cerebral cortex and hippocampus. The Br4 antigens may be useful in identifying neuronal cell subpopulations in the central nervous system.
Key words: immunocytochemistry, immunoblotting, neuronal proteins, synaptic soluble fraction
INTRODUCTION When considering the diversity of neuronal cell function, it seems important to investigate specific components in certain specialized neuronal subgroups. These components may be enzymes or receptors related to neurotransmitters, or the molecules may not be related to neuronal transmission but may identify unique neuronal subpopulations. The technique of monoclonal antibody production has been successfully applied to detect specific molecules for certain subsets of neurons. Antibodies specific for peripheral and central neurons have been identified (Vulliamy et al., 1981; Cohen and Selvendran, 1983). Antibodies that react with different classes of hippocampal neurons (Miller and Benzer, 1983; Hinton et al., 1988; Moskal and Schaffner, 1986; Woodhams et 0 1991 Wiley-Liss, Inc.
al., 1989) or specific for the granule cells in cerebellum (Langley et al., 1985; Hirsh et al., 1983) have been reported. A monoclonal antibody Cat-30 1 recognizes specific subpopulation of cortical and thalamic neurons (Hendry et al., 1988) and B30 identifies small population of cells in the developing CNS (Stainier and Gilbert, 1989). Antibodies that recognize cell subpopulation in the olfactory system have been reported (Allen and Akeson, 1985; Hempstead and Morgan, 1985), and a photoreceptor antigen was identifed using Drosophila nervous tissue as an immunogen (Fujita et al., 1982; Zipursky et al., 1984). In our laboratory, chick embryonic brains have been used as the immunogen for preparing monoclonal antibodies that are specific for neurons or that define molecules specific for certain subsets of neurons, since chick embryo is very accessible and has the best studied developing nervous system and the pure cultured neurons can be prepared from embryonic telencephalon (Pettman et al., 1979). We report here the antibody (Br4) that reacts with certain neurons in hippocampus, substantia nigra, cerebral cortex, and granular layer of cerebellum from rat brain and describe partial characterization of the unique Br4 antigens.
MATERIALS AND METHODS Monoclonal Antibody Production BALB/c mice were immunized 6 times at intervals of 2 weeks with single cell suspension. [lo7 cells in 200 p1 of Dulbecco’s modified Eagle’s medium (DMEM)] prepared from 11 day-old chick embryonic total brain. Three days after the final injection, the single cell suspension (1.76 X lo8 cells) prepared from the spleen of the immunized mouse was mixed with myeloma (P3 X 63-AG 8.653) cell suspension (2.5 X lo7 cells).
Received August 29, 1990; accepted September 17, 1990 Address reprint requests to Dr. Mitsuo Saito, Division of Neurobiology, Nathan S. Kline Institute for Psychiatric Research, Orangeburg. NY 10962.
108
Saito and Saito
The mixed cell suspension was centrifuged and washed with 50 ml of DMEM without fetal calf serum (FCS). The packed cells were loosened by tapping after removing supernatant. One ml of 50% polyethylene glycol solution (PEG) (50% PEG 4000 in DMEM, GIBCO), kept at 37"C, was added to the loosened cells slowly (in 1 min) with continuous shaking. The cell suspension was shaken for one more min, and 3 ml of DMEM without FCS was added to the cell suspension during next 3 min. Subsequently, 7 ml of DMEM was added during next 3 min. The final cell suspension was centrifuged to remove PEG solution. The packed cells were resuspended in 30 ml of HAT medium (GIBCO) supplemented with 20% FCS and 50% of the conditioned medium obtained from mouse thymus cell culture (2 X 10' cells in 20 ml DMEM with 20% FCS). The cell suspension was plated on 300microtiter wells. Hybridoma culture medium was screened on cultured chick neurons prepared as described below. Positive cell lines were cloned by limiting dilution. The culture supernatant harvested from one of such positive clones (termed Br4) was used for these experiments. In some cases, ascites fluid produced by injecting the cloned cells was also used. The Ig subclass of Br4 was tested using a monoclonal antibody typing kit (Sigma).
Immunocytochemical Staining of Cultured Neural Cells and Rat Brain The primary cultured neurons or astrocytes were prepared from chick embryonic telencephalon by the method of Pettman et al. (1979). The immunostaining was performed according to DeJong et al. (1985) with modification. The cultured cells were fixed by 4% formaldehyde in phosphate-buffered saline (PBS) overnight, washed three times with PBS, incubated for 3 hr in PBS containing 1% bovine serum albumin (BSA) and incubated with Br4 hybridoma culture supernatant at 4°C overnight. The cells were then washed and incubated with alkaline phosphatase conjugated antimouse IgM and IgG antibody (Boehringer Mannheim) for 2 hr at 23"C, washed and stained by incubating in 0.2 M Tris-HC1 buffer (pH 9.1) containing 10 mM MgCl,, 0.017% of 5-bromo-4-chloro-3-indolyl phoshate, and 0.034% of nitro blue tetrazolium at 23°C for 15 min. The sections (10 pm) of frozen rat cerebrum and cerebellum by a cryotome were attached on slide glasses by heating at 60°C for 5 min, fixed by 4% formaldehyde in PBS overnight and stained as described above for the neural cells.
Teflon glass homogenizer at 4°C. The homogenate was centrifuged at 100,OOOg for 1 hr to obtain soluble and particulate fractions. The subcellular fractionation was performed according to Gray and Whittaker (1962), with slight modification. The rat forebrains were homogenized in 10 vol of 10% sucrose using a Potter-Elvehjem glass homogenizer (5 strokes) and a Teflon glass homogenizer (10 strokes) at 4°C. (All the following procedures were done at 4°C.) The homogenate was centrifuged at 700g for 10 min and the pellet obtained was rehomogenized in 10 vol of 10% sucrose and centrifuged. The supernatant obtained from each spin was combined and centrifuged at 10,OOOg for 30 min to obtain P, (crude mitochondrial) fraction. The supernatant was centrifuged at 100,OOOg for 1 hr to obtain P, (microsomal) fraction and S, (cytosol) fraction. The fraction was resuspended in 10% sucrose and was layered on 27%-41% sucrose layers in a centrifuge tube and centrifuged at 100,OOOg for 2 hr. The material floating at the 10%-27% interface was separated and centrifuged at 100,OOOg for 1 hr to obtain the myelin fraction. The material at the 27%-41% interface was separated and centrifuged at 100,OOOg for 1 hr to obtain the synaptosomal fraction. The pellet under the 41 % sucrose layer was separated as the mitochondriallysosomal fraction. The microsomal, myelin, synaptosomal, and mitochondrial-lysosomal fractions were washed once by resuspending in 10% sucrose and repelleting by centrifugation. The part of the synaptosomal fraction was lysed by incubating in 5 mM TrisiHCl0.1mM EDTA (pH 8.1) for 30 min at 0°C and centrifuged to obtain synaptic soluble and synaptic particulate fractions. In some experiments, the synaptosomes were lysed in 5 mM Tris/HCl-10 mM CaC1, (pH 8.1).
Gel Electrophoresis and Immunoblot Analysis Tissue samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) after protein determination using Bio-Rad protein assay kit. SDS-PAGE was carried out in 6.5% polyacrylamide gels according to the procedure of Laemmli (1970). M, values were estimated by use of Sigma-prestained molecular-weight standard. The gels were stained with Coomassie blue or electroblotted onto nitrocellulose sheets (Bio-Rad) using a Transblot apparatus (Hoefer Scientitic). The blots were blocked with 5% BSA-0.5% Tween in PBS for 2 hr and incubated for 3 hr at 25°C with monoclonal antibody Br4 in PBS containing 0.5% Tween 20-1% BSA-1 M glucose-10% glycerol Preparation of Tissue Samples and Subcellular according to the method of Birk and Koepsell (1987) Fractionation of Rat Brain with modification. The nitrocellulose sheets were then Tissue samples were prepared from adult Sprague- rinsed in 0.5% Tween 20 in PBS and incubated with Dawley rats by homogenization in 10 vol of 5 mM Tris- antimouse IgG and IgM conjugated with horseradish perHCl (pH 7.5)-1 mM EDTA-0.2 mM PMSF using a oxidase (HRP) (Boehringer Mannheim) and followed by
Neuron-Specific Monoclonal Antibody development in diaminobenzidine and H,O, (De Jong et al., 1985).
RESULTS Immunocytochemical Staining of Cultured Chick Neurons and Astrocytes A hybridoma clone secreting a IgM monoclonal antibody termed Br4 that reacts strongly with cultured chick neurons, but not with the astrocytes, has been raised. Cultured neurons were prepared from 8-day-old chick embryonic telencephalon and incubated for 3, 6, or 10 days in DMEM containing 15% FCS and immunostained with the monoclonal antibody Br4 as described under Materials and Methods. The cultured neurons prepared are known to be highly pure and to show the neuronal differentiation during the culture period (Pettman et al., 1979). On day 3 after starting the culture, the cell bodies and neurites of virtually all neurons react with the antibody (Fig. 1A,D). On day 6, polar staining of the cell bodies and punctate staining of the neurites is often observed, and some of the neurons become unstained (Fig. 1B,E). After 10 days of culture in the presence of FCS, flat cells that are reported to be astrocytes (Pettman et al., 1979) appear and gradually increase in number, but these cells do not react with Br4 (Fig. 1C,F). Also, when preparing pure astrocytes from the 14-day-old chick embryonic telencephalon, the cells do not react with the antibody (data not shown). The astrocytes prepared from rat were also not recognized by the antibody. Cells under any conditions described above are not stained when the medium from a nonproducing hybridoma line is used instead of Br4. Immunocytochemical Staining of Adult Rat Brain The sections (10 pm thick) of frozen rat cerebrum and cerebellum were stained with antibody Br4 by alkaline phosphatase method. In cerebrum, there is clear immunoreactivity in the cortex, hippocampus, and substantia nigra, but the white matter does not react with antibody (Fig. 2A). In cerebellum, the strongest staining is observed in the granular layer; some fine dot structures are stained in the molecular layer (Fig. 2B). Purkinje cells do not react with the antibody. The background staining of these sections (tested by using the medium from a nonproducing hybridoma line) is very low. Br4 immunoreactivity is weakly expressed in rat embryonic brain compared with that of the adult brain (data not shown). Distribution of Antigenic Proteins in Subcellular Fractions of Rat Brain Immunoblot analysis was performed to identify antigenic proteins recognized by Br4. After subcellular
109
fractionation of adult rat forebrains, proteins of each fraction were separated on SDS-PAGE (6.5%) and immunoblotted by the antibody. As shown in Figure 3 , four proteins of apparent molecular weight, 145,000, 108,000, 97,000, and 33,000 are recognized by this antibody. The protein with 145,000 M , (p145) and the protein with 97,000 M , (p97) exist essentially as soluble forms; p145 is especially enriched in the synaptosomal soluble fraction. The protein with 108,000 M , (p108) is found as both membrane-bound and soluble forms. The protein with 33,000 M , detected in the myelin fraction is probably nonspecific because this band is also stained by several other monoclonal antibodies. The mobilities of these proteins on SDS-PAGE are not affected in the presence or absence of P-mercaptoethanol in the sample solution.
Regional Distribution of the Antigenic Proteins Figure 4 shows the distribution of the antigenic proteins among several rat organs examined by immunoblotting. p145, p108, and p97 seem to be essentially brain specific, although in adrenal, two protein bands that have molecular weight similar to p97 are detected. Figure 5 shows regional distribution of the antigenic proteins in rat brain. The levels of all three proteins are found less in corpora quadrigemina, pons, medulla, and spinal cord compared with the other regions. Especially the difference in the amount of p97 was noticeable among these regions. Effects of Calcium Ions on Solubility of the Antigens Although p145 and p97 are essentially found as soluble proteins, their solubility decreases in the presence of calcium ions. When the lysis of synaptosomes is performed in the presence of calcium ions, p 145 and p97 are found in the synaptosomal particulate fraction and not found in the soluble fraction (Fig. 6a). Also, p145 can be precipitated from the synaptic soluble fraction by incubation at 30°C for 1 hr with 5 mM calcium ions (Fig. 6b). In this case, p108 an p97 are not found in either the soluble or particulate fraction, indicating the presence of proteases activated by calcium ions. DISCUSSION We have raised a monoclonal antibody termed Br4 that specifically recognizes some neuronal cells. The immunostaining pattern of rat brain sections indicates that the antibody reacts with some neuronal cells, especially in the area of cerebral cortex, hippocampus, substantia nigra, and the granular layer of cerebellum, and does not react with the white matter. However, not all neurons are stained; for example, Purkinje cells are not recognized
110
Saito and Saito
Fig. 1. Immunocytochemical staining of cultured chick neurons with a monoclonal antibody Br4. Cultured chick neurons prepared from 8-day-old chick embryonic telencephalon as described by Pettman et al. (1979) were incubated in Dulbecco’s modified Eagle’s medium containing 15% FCS. After 3 days (A), 6 days (B), or 10 days (C), these cells were stained with
the monoclonal antibody Br4 by the alkaline phosphatase method. D-F: phase contrast microscopic pictures of A-C, respectively. In A(D), virtually all neurons are stained, while some neurons lose their antigenicity in B(E) and C(F) (arrowheads). Astrocytes observed in F (arrows) are not stained. Scale bar = 50 km.
by this antibody. The antibody reacts strongly with cultured chick neurons. In the early stage of culture, cell bodies and neurites of almost all neurons are reacted by
the antibody, while in the later stage, some neurons lose their antigenicity, suggesting that the antigenic materials localize in certain subsets of neurons with in vitro devel-
Neuron-Specific Monoclonal Antibody
A
B
C
D
E
F
111
Fig. 2 . Immunocytochemical staining of rat brains. Sections (10 k m thick) of frozen rat adult cerebrum (A) and rat adult cerebellum (B) were stained with the monoclonal antibody Br4 by the alkaline phosphatase method. There is clear immunoreactivity in the cortex (c), hippocampus (hpc), and substantia nigra (sn) (A) and the granular (8) and molecular (m) layers of the cerebellum (B). No staining is observed in Purkinje cells (p). White matter (w) of cerebrum and cerebellum is not stained. Ventricle is filled with blood (bld).
G
kD P145-P108.P97-
P33--
-1 80 -1 16
A
B
C
D
E
F
G
H
- 84
- 58 - 49 - 37 - 27
Fig. 3. Subcellular distribution of antigenic proteins recognized by the monoclonal antibody Br4. Subcellular fractionation of rat adult forebrains was performed and the aliquots of the fractions (10 pg protein) were separated on 6.5% SDSPAGE, blotted, and developed with Br4, using the horseradish peroxidase method. A: Total homogenate. B: Cytosol. C: Microsomes. D: Synaptic soluble. E: Synaptic particulate. F: Myelin. G: Mytochondrial-lysosomal fraction. The major proteins recognized by the antibody are p145, p108, p97, and p33. p145 is especially enriched in synaptic soluble fraction.
-- P145 -- P108
-- P 9 7
Fig. 4. Tissue specificity of antigenic proteins. The aliquots (20 pg protein) of the soluble fractions obtained from a range of tissues from adult rats were separated on 6.5% SDS-PAGE and blotted and developed by Br4. A: Liver. B: Kidney. C: Adrenal. D: Muscle. E: Heart. F: Spleen. G: Cerebellum. H: Forebrain.
112
Saito and Saito
A
B
C
D
E
F
G
H
I
1
-- P l 4 -- P l d -- P W
Fig. 5. Regional distribution of antigenic proteins in rat brain. Several regions of rat adult brain were dissected. Each soluble fraction of their regions (40 pg protein) was separated on 6.5% SDS-PAGE and analyzed by Western blotting. A: Cortex. B:
Olfactory lobe. C: Striatum. D Hippocampus. E: Massa intermedia. F: Hypothalamus. G: Corpora quadrigemina. H: Pons. I: Medulla. J: Spinal cord.
445 .-P108 .-P97
+Ca% +CC?+ u u Sol. Pt.
sup. Ppt.
Fig. 6. Effects of calcium ions on the solubility of the antigens. a: The synaptosomal fraction from rat forebrain was lysed in 5 mM Tris-HC1 (pH 8.1) containing 0.1 mM EDTA or 10 mM CaCI, at 4°C for 30 min and centrifuged to obtain soluble and particulate fractions. The synaptic soluble fraction (Sol.) obtained in the presence of 1 mM EDTA (A) or 10 mM CaCI, (B) or the synaptic particulate fractions (Pt.) obtained in the presence of 1 mM EDTA (C) or 10 mM CaCI, (D) were
separated by 6.5% SDS-PAGE, blotted and stained with Br4 using the horseradish peroxidase method. b: The synaptic soluble fraction obtained in the presence of 1 mh4 EDTA was incubated for 1 hr at 30°C in the presence of 10 mM CaCl,, followed by centrifugation at 100,OOOg for 1 hr. The supernatant (Sup.) (A) and the precipitate (Ppt.) (B) were analyzed as described in a. The synaptic soluble fraction which was incubated at 30°C without calcium ions was also analyzed (C).
opment. The antibody does not react with cultured astrocytes from chick o r rat brains. Although the early stages of cultured chick neurons are highly reactive with the antibody, the rat embryonic brain shows less antigenicity than the adult brain. It may reflect the difference between the species or the differentiation stages.
The staining pattern of the cultured neurons indicates the intracellular localization of the antigens. Immunoblots show that three proteins from rat brains, p145, p108 and p97, are recognized by this antibody. So far, it is unknown which proteins are recognized immunocytochemically . However, subcellular localization of p145 indicates its primary localization in synaptosomes.
Neuron-Specific Monoclonal Antibody
p145 and p97 exist both in cytosol and in synaptic soluble fractions, which matches the immunocytochemical staining. Also the regional distribution of the proteins in rat brain correlates with the immunocytochemical staining pattern. Especially, the amount of p97 seems very low in pons, medulla, and spinal cord. It seems that p108 and p97 are not breakdown products of p145, since the subcellular distribution of these three proteins are different. The proteins are found more abundant in brains than in other organs, except that p97 exists quite significantly in adrenal. It is observed that calcium ions decrease the solubility of these three proteins. It may relate to the phenomenon reported by Garner (1990) that lysis of synaptosomes in the presence of calcium ions causes a decrease in solubility of the presynaptic slow component b proteins. The three Br4 antigens exist not only in rat brain but also in chick and bovine brain (data not shown). We are now attempting the purification of these proteins from bovine brain. In summary, we have raised a monoclonal antibody Br4 that reacts a subset of neurons and recognizes three antigenic molecules. Further studies are necessary to determine the nature of neurons and the proteins recognized by this antibody.
REFERENCES Allen WK, Akeson R (1985): Identification of a cell surface glycoprotein family of olfactory receptor neurons with a monoclonal antibody. J Neurosci 5:284-296. Birk H-W, Koepsell H (1987): Reaction of monoclonal antibodies with plasma membrane proteins after binding on nitrocellulose: renaturation of antigen sites and reduction of nonspecific antibody binding. Anal Biochem 164:12-22. Cohen J, Selvendran SY (1983): A neuronal surface antigen is found in the CNS but not in peripheral neurons. Nature 305:424-427. DeJong ASH, Van Kessel-Ban Vark M, Raap AK (1985): Sensitivity of various visualization methods for peroxidase and alkaline phosphatase activity in immunoenzyme histochemistry . Histochem J 17:1119-1130. Fujita SC, Zipursky SL, Benzer S , Ferrus A, Shotwell SL (1982): Monoclonal antibodies against the Drosophila nervous system. Proc Natl Acad Sci USA 79:7929-7933. Garner JA (1990): Selective alterations in presynaptic cytomatrix pro-
113
tein organization induced by calcium and other divalent cations that modulate exocytosis. J Neurochem 54: 1770-1708. Gray EG, Whittaker VP (1962): The isolation of nerve endings from brain: An electron microscopic study: Cell fragments derived by homogenization and centrifugation. J Anat (Lond) 96:7987. Hempstead JL, Morgan JI (1985): A panel of monoclonal antibodies to the rat olfactory epithelium. J Neurosci 5:438-449. Hendry SHC, Jones EG, Hockfield S, McKay RDG (1988): Neuronal populations stained with monoclonal antibody Cat-301 in the mammalian cerebral cortex and thalamus. J Neurosci 8:5 18542. Hinton DR, Henderson VW, Blanks JC, Rudnicka M, Miller CA (1988): Monoclonal antibodies react with neuronal subpopulations in the human nervous system. J Comp Neurol 267:398408. Hirsh M-R, Langley OK, Ghandour MS, Hirn M, Gombos G, Goridis G (1983): A monoclonal antibody recognizing subpopulations of neurons in mouse brain, J Neuroimmunol 4:175-186. Hishinuma A, Hockfield S , McKay R, Hildebrand JG (1988): Monoclonal antibodies reveal cell-type-specific antigens in the sexually dimorphic olfactory system of Munducu sextu. I. Generation of monoclonal antibodies and partial characterization of the antigens. J Neurosci 8:296-307. Laemmli UK (1970): Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227680-685. Langley OK, Foucaud B, Ghandour MS, Barry JDe, Schladenhaufen Y, Gombos G (1985): A developmentally modified neuronspecific marker in rodent cerebellum. Neuroscience 14:147157.
Miller CA, Benzer S (1983): Monoclonal antibody cross-reactions between Drosophilu and human brain. Proc Natl Acad Sci USA 80:764 1-7645. Moskal JR, Schaffner AE (1986): Monoclonal antibodies to the dentate gyms: Immunocytochemical characterization and flow cytometric analysis of hippocampal neurons bearing a unique cell surface antigen. J Neurosci 6:2045-2053. Pettman B, Louis JC, Sensenbrenner M (1979): Morphological and biochemical maturation of neurons cultured in the absence of glial cells. Nature 281:378-380. Stainier DY, Gilbert W (1989): The monoclonal antibody 1330 recognizes a specific neuronal cell surface antigen in the developing mesencephalic trigeminal nucleus of the mouse. J Neurosci 9:2466-2485. Vulliamy T, Rattray S, Mirsky R (1981): Cell-surface antigen distinguishes sensory and autonomic peripheral neurons from central neurons. Nature 291:418-420. Woodhams PL, Webb M, Atkinson DJ, Seeley PJ (1989): A monoclonal antibody, Py, distinguishes different classes of hippocampal neurons. J Neurosci 9:2170-2181. Zipursky SL, Venkatesh TR, Teplow DB, Benzer S (1984): Neuronal development in the Drosophilu retina: Monoclonal antibodies as molecular probes. Cell 36:15-26.
Monoclonal Antibody Br4 Recognizes Specific Neuronal Cell Types M. Saito and M. Saito Division of Neurobiology , Nathan S . Kline Institute for Psychiatric Research, Orangeburg, New York A monoclonal antibody termed as Br4 was prepared by a fusion of the myeloma P3X63-AG8-653 with spleen cells from BALB/c mice immunized with chick embryonic brain cells. Immunocytochemically, it reacts strongly with certain neurons in cerebral cortex, hippocampus, substantia nigra, and granular layer of cerebellum from rat brain but does not react with the white matter. A monoclonal antibody Br4 also reacts with primary cultured chick neurons, but not with cultured astrocytes. Western blots show that Br4 recognizes three proteins of 145,000, 108,000, and 97,000 molecular weight from the rat brain. The protein with molecular weight 145,000 (p145) and the protein with 97,000 molecular weight (p97) are essentially soluble; p145 is especially enriched in the synaptosomal soluble fraction. The protein with 108,000 molecular weight (plO8) is found in both membranebound and soluble forms. Western blots show that among the tissues examined the three proteins are found most abundant in brain and especially enriched in the cerebral cortex and hippocampus. The Br4 antigens may be useful in identifying neuronal cell subpopulations in the central nervous system.
Key words: immunocytochemistry, immunoblotting, neuronal proteins, synaptic soluble fraction
INTRODUCTION When considering the diversity of neuronal cell function, it seems important to investigate specific components in certain specialized neuronal subgroups. These components may be enzymes or receptors related to neurotransmitters, or the molecules may not be related to neuronal transmission but may identify unique neuronal subpopulations. The technique of monoclonal antibody production has been successfully applied to detect specific molecules for certain subsets of neurons. Antibodies specific for peripheral and central neurons have been identified (Vulliamy et al., 1981; Cohen and Selvendran, 1983). Antibodies that react with different classes of hippocampal neurons (Miller and Benzer, 1983; Hinton et al., 1988; Moskal and Schaffner, 1986; Woodhams et 0 1991 Wiley-Liss, Inc.
al., 1989) or specific for the granule cells in cerebellum (Langley et al., 1985; Hirsh et al., 1983) have been reported. A monoclonal antibody Cat-30 1 recognizes specific subpopulation of cortical and thalamic neurons (Hendry et al., 1988) and B30 identifies small population of cells in the developing CNS (Stainier and Gilbert, 1989). Antibodies that recognize cell subpopulation in the olfactory system have been reported (Allen and Akeson, 1985; Hempstead and Morgan, 1985), and a photoreceptor antigen was identifed using Drosophila nervous tissue as an immunogen (Fujita et al., 1982; Zipursky et al., 1984). In our laboratory, chick embryonic brains have been used as the immunogen for preparing monoclonal antibodies that are specific for neurons or that define molecules specific for certain subsets of neurons, since chick embryo is very accessible and has the best studied developing nervous system and the pure cultured neurons can be prepared from embryonic telencephalon (Pettman et al., 1979). We report here the antibody (Br4) that reacts with certain neurons in hippocampus, substantia nigra, cerebral cortex, and granular layer of cerebellum from rat brain and describe partial characterization of the unique Br4 antigens.
MATERIALS AND METHODS Monoclonal Antibody Production BALB/c mice were immunized 6 times at intervals of 2 weeks with single cell suspension. [lo7 cells in 200 p1 of Dulbecco’s modified Eagle’s medium (DMEM)] prepared from 11 day-old chick embryonic total brain. Three days after the final injection, the single cell suspension (1.76 X lo8 cells) prepared from the spleen of the immunized mouse was mixed with myeloma (P3 X 63-AG 8.653) cell suspension (2.5 X lo7 cells).
Received August 29, 1990; accepted September 17, 1990 Address reprint requests to Dr. Mitsuo Saito, Division of Neurobiology, Nathan S. Kline Institute for Psychiatric Research, Orangeburg. NY 10962.
108
Saito and Saito
The mixed cell suspension was centrifuged and washed with 50 ml of DMEM without fetal calf serum (FCS). The packed cells were loosened by tapping after removing supernatant. One ml of 50% polyethylene glycol solution (PEG) (50% PEG 4000 in DMEM, GIBCO), kept at 37"C, was added to the loosened cells slowly (in 1 min) with continuous shaking. The cell suspension was shaken for one more min, and 3 ml of DMEM without FCS was added to the cell suspension during next 3 min. Subsequently, 7 ml of DMEM was added during next 3 min. The final cell suspension was centrifuged to remove PEG solution. The packed cells were resuspended in 30 ml of HAT medium (GIBCO) supplemented with 20% FCS and 50% of the conditioned medium obtained from mouse thymus cell culture (2 X 10' cells in 20 ml DMEM with 20% FCS). The cell suspension was plated on 300microtiter wells. Hybridoma culture medium was screened on cultured chick neurons prepared as described below. Positive cell lines were cloned by limiting dilution. The culture supernatant harvested from one of such positive clones (termed Br4) was used for these experiments. In some cases, ascites fluid produced by injecting the cloned cells was also used. The Ig subclass of Br4 was tested using a monoclonal antibody typing kit (Sigma).
Immunocytochemical Staining of Cultured Neural Cells and Rat Brain The primary cultured neurons or astrocytes were prepared from chick embryonic telencephalon by the method of Pettman et al. (1979). The immunostaining was performed according to DeJong et al. (1985) with modification. The cultured cells were fixed by 4% formaldehyde in phosphate-buffered saline (PBS) overnight, washed three times with PBS, incubated for 3 hr in PBS containing 1% bovine serum albumin (BSA) and incubated with Br4 hybridoma culture supernatant at 4°C overnight. The cells were then washed and incubated with alkaline phosphatase conjugated antimouse IgM and IgG antibody (Boehringer Mannheim) for 2 hr at 23"C, washed and stained by incubating in 0.2 M Tris-HC1 buffer (pH 9.1) containing 10 mM MgCl,, 0.017% of 5-bromo-4-chloro-3-indolyl phoshate, and 0.034% of nitro blue tetrazolium at 23°C for 15 min. The sections (10 pm) of frozen rat cerebrum and cerebellum by a cryotome were attached on slide glasses by heating at 60°C for 5 min, fixed by 4% formaldehyde in PBS overnight and stained as described above for the neural cells.
Teflon glass homogenizer at 4°C. The homogenate was centrifuged at 100,OOOg for 1 hr to obtain soluble and particulate fractions. The subcellular fractionation was performed according to Gray and Whittaker (1962), with slight modification. The rat forebrains were homogenized in 10 vol of 10% sucrose using a Potter-Elvehjem glass homogenizer (5 strokes) and a Teflon glass homogenizer (10 strokes) at 4°C. (All the following procedures were done at 4°C.) The homogenate was centrifuged at 700g for 10 min and the pellet obtained was rehomogenized in 10 vol of 10% sucrose and centrifuged. The supernatant obtained from each spin was combined and centrifuged at 10,OOOg for 30 min to obtain P, (crude mitochondrial) fraction. The supernatant was centrifuged at 100,OOOg for 1 hr to obtain P, (microsomal) fraction and S, (cytosol) fraction. The fraction was resuspended in 10% sucrose and was layered on 27%-41% sucrose layers in a centrifuge tube and centrifuged at 100,OOOg for 2 hr. The material floating at the 10%-27% interface was separated and centrifuged at 100,OOOg for 1 hr to obtain the myelin fraction. The material at the 27%-41% interface was separated and centrifuged at 100,OOOg for 1 hr to obtain the synaptosomal fraction. The pellet under the 41 % sucrose layer was separated as the mitochondriallysosomal fraction. The microsomal, myelin, synaptosomal, and mitochondrial-lysosomal fractions were washed once by resuspending in 10% sucrose and repelleting by centrifugation. The part of the synaptosomal fraction was lysed by incubating in 5 mM TrisiHCl0.1mM EDTA (pH 8.1) for 30 min at 0°C and centrifuged to obtain synaptic soluble and synaptic particulate fractions. In some experiments, the synaptosomes were lysed in 5 mM Tris/HCl-10 mM CaC1, (pH 8.1).
Gel Electrophoresis and Immunoblot Analysis Tissue samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) after protein determination using Bio-Rad protein assay kit. SDS-PAGE was carried out in 6.5% polyacrylamide gels according to the procedure of Laemmli (1970). M, values were estimated by use of Sigma-prestained molecular-weight standard. The gels were stained with Coomassie blue or electroblotted onto nitrocellulose sheets (Bio-Rad) using a Transblot apparatus (Hoefer Scientitic). The blots were blocked with 5% BSA-0.5% Tween in PBS for 2 hr and incubated for 3 hr at 25°C with monoclonal antibody Br4 in PBS containing 0.5% Tween 20-1% BSA-1 M glucose-10% glycerol Preparation of Tissue Samples and Subcellular according to the method of Birk and Koepsell (1987) Fractionation of Rat Brain with modification. The nitrocellulose sheets were then Tissue samples were prepared from adult Sprague- rinsed in 0.5% Tween 20 in PBS and incubated with Dawley rats by homogenization in 10 vol of 5 mM Tris- antimouse IgG and IgM conjugated with horseradish perHCl (pH 7.5)-1 mM EDTA-0.2 mM PMSF using a oxidase (HRP) (Boehringer Mannheim) and followed by
Neuron-Specific Monoclonal Antibody development in diaminobenzidine and H,O, (De Jong et al., 1985).
RESULTS Immunocytochemical Staining of Cultured Chick Neurons and Astrocytes A hybridoma clone secreting a IgM monoclonal antibody termed Br4 that reacts strongly with cultured chick neurons, but not with the astrocytes, has been raised. Cultured neurons were prepared from 8-day-old chick embryonic telencephalon and incubated for 3, 6, or 10 days in DMEM containing 15% FCS and immunostained with the monoclonal antibody Br4 as described under Materials and Methods. The cultured neurons prepared are known to be highly pure and to show the neuronal differentiation during the culture period (Pettman et al., 1979). On day 3 after starting the culture, the cell bodies and neurites of virtually all neurons react with the antibody (Fig. 1A,D). On day 6, polar staining of the cell bodies and punctate staining of the neurites is often observed, and some of the neurons become unstained (Fig. 1B,E). After 10 days of culture in the presence of FCS, flat cells that are reported to be astrocytes (Pettman et al., 1979) appear and gradually increase in number, but these cells do not react with Br4 (Fig. 1C,F). Also, when preparing pure astrocytes from the 14-day-old chick embryonic telencephalon, the cells do not react with the antibody (data not shown). The astrocytes prepared from rat were also not recognized by the antibody. Cells under any conditions described above are not stained when the medium from a nonproducing hybridoma line is used instead of Br4. Immunocytochemical Staining of Adult Rat Brain The sections (10 pm thick) of frozen rat cerebrum and cerebellum were stained with antibody Br4 by alkaline phosphatase method. In cerebrum, there is clear immunoreactivity in the cortex, hippocampus, and substantia nigra, but the white matter does not react with antibody (Fig. 2A). In cerebellum, the strongest staining is observed in the granular layer; some fine dot structures are stained in the molecular layer (Fig. 2B). Purkinje cells do not react with the antibody. The background staining of these sections (tested by using the medium from a nonproducing hybridoma line) is very low. Br4 immunoreactivity is weakly expressed in rat embryonic brain compared with that of the adult brain (data not shown). Distribution of Antigenic Proteins in Subcellular Fractions of Rat Brain Immunoblot analysis was performed to identify antigenic proteins recognized by Br4. After subcellular
109
fractionation of adult rat forebrains, proteins of each fraction were separated on SDS-PAGE (6.5%) and immunoblotted by the antibody. As shown in Figure 3 , four proteins of apparent molecular weight, 145,000, 108,000, 97,000, and 33,000 are recognized by this antibody. The protein with 145,000 M , (p145) and the protein with 97,000 M , (p97) exist essentially as soluble forms; p145 is especially enriched in the synaptosomal soluble fraction. The protein with 108,000 M , (p108) is found as both membrane-bound and soluble forms. The protein with 33,000 M , detected in the myelin fraction is probably nonspecific because this band is also stained by several other monoclonal antibodies. The mobilities of these proteins on SDS-PAGE are not affected in the presence or absence of P-mercaptoethanol in the sample solution.
Regional Distribution of the Antigenic Proteins Figure 4 shows the distribution of the antigenic proteins among several rat organs examined by immunoblotting. p145, p108, and p97 seem to be essentially brain specific, although in adrenal, two protein bands that have molecular weight similar to p97 are detected. Figure 5 shows regional distribution of the antigenic proteins in rat brain. The levels of all three proteins are found less in corpora quadrigemina, pons, medulla, and spinal cord compared with the other regions. Especially the difference in the amount of p97 was noticeable among these regions. Effects of Calcium Ions on Solubility of the Antigens Although p145 and p97 are essentially found as soluble proteins, their solubility decreases in the presence of calcium ions. When the lysis of synaptosomes is performed in the presence of calcium ions, p 145 and p97 are found in the synaptosomal particulate fraction and not found in the soluble fraction (Fig. 6a). Also, p145 can be precipitated from the synaptic soluble fraction by incubation at 30°C for 1 hr with 5 mM calcium ions (Fig. 6b). In this case, p108 an p97 are not found in either the soluble or particulate fraction, indicating the presence of proteases activated by calcium ions. DISCUSSION We have raised a monoclonal antibody termed Br4 that specifically recognizes some neuronal cells. The immunostaining pattern of rat brain sections indicates that the antibody reacts with some neuronal cells, especially in the area of cerebral cortex, hippocampus, substantia nigra, and the granular layer of cerebellum, and does not react with the white matter. However, not all neurons are stained; for example, Purkinje cells are not recognized
110
Saito and Saito
Fig. 1. Immunocytochemical staining of cultured chick neurons with a monoclonal antibody Br4. Cultured chick neurons prepared from 8-day-old chick embryonic telencephalon as described by Pettman et al. (1979) were incubated in Dulbecco’s modified Eagle’s medium containing 15% FCS. After 3 days (A), 6 days (B), or 10 days (C), these cells were stained with
the monoclonal antibody Br4 by the alkaline phosphatase method. D-F: phase contrast microscopic pictures of A-C, respectively. In A(D), virtually all neurons are stained, while some neurons lose their antigenicity in B(E) and C(F) (arrowheads). Astrocytes observed in F (arrows) are not stained. Scale bar = 50 km.
by this antibody. The antibody reacts strongly with cultured chick neurons. In the early stage of culture, cell bodies and neurites of almost all neurons are reacted by
the antibody, while in the later stage, some neurons lose their antigenicity, suggesting that the antigenic materials localize in certain subsets of neurons with in vitro devel-
Neuron-Specific Monoclonal Antibody
A
B
C
D
E
F
111
Fig. 2 . Immunocytochemical staining of rat brains. Sections (10 k m thick) of frozen rat adult cerebrum (A) and rat adult cerebellum (B) were stained with the monoclonal antibody Br4 by the alkaline phosphatase method. There is clear immunoreactivity in the cortex (c), hippocampus (hpc), and substantia nigra (sn) (A) and the granular (8) and molecular (m) layers of the cerebellum (B). No staining is observed in Purkinje cells (p). White matter (w) of cerebrum and cerebellum is not stained. Ventricle is filled with blood (bld).
G
kD P145-P108.P97-
P33--
-1 80 -1 16
A
B
C
D
E
F
G
H
- 84
- 58 - 49 - 37 - 27
Fig. 3. Subcellular distribution of antigenic proteins recognized by the monoclonal antibody Br4. Subcellular fractionation of rat adult forebrains was performed and the aliquots of the fractions (10 pg protein) were separated on 6.5% SDSPAGE, blotted, and developed with Br4, using the horseradish peroxidase method. A: Total homogenate. B: Cytosol. C: Microsomes. D: Synaptic soluble. E: Synaptic particulate. F: Myelin. G: Mytochondrial-lysosomal fraction. The major proteins recognized by the antibody are p145, p108, p97, and p33. p145 is especially enriched in synaptic soluble fraction.
-- P145 -- P108
-- P 9 7
Fig. 4. Tissue specificity of antigenic proteins. The aliquots (20 pg protein) of the soluble fractions obtained from a range of tissues from adult rats were separated on 6.5% SDS-PAGE and blotted and developed by Br4. A: Liver. B: Kidney. C: Adrenal. D: Muscle. E: Heart. F: Spleen. G: Cerebellum. H: Forebrain.
112
Saito and Saito
A
B
C
D
E
F
G
H
I
1
-- P l 4 -- P l d -- P W
Fig. 5. Regional distribution of antigenic proteins in rat brain. Several regions of rat adult brain were dissected. Each soluble fraction of their regions (40 pg protein) was separated on 6.5% SDS-PAGE and analyzed by Western blotting. A: Cortex. B:
Olfactory lobe. C: Striatum. D Hippocampus. E: Massa intermedia. F: Hypothalamus. G: Corpora quadrigemina. H: Pons. I: Medulla. J: Spinal cord.
445 .-P108 .-P97
+Ca% +CC?+ u u Sol. Pt.
sup. Ppt.
Fig. 6. Effects of calcium ions on the solubility of the antigens. a: The synaptosomal fraction from rat forebrain was lysed in 5 mM Tris-HC1 (pH 8.1) containing 0.1 mM EDTA or 10 mM CaCI, at 4°C for 30 min and centrifuged to obtain soluble and particulate fractions. The synaptic soluble fraction (Sol.) obtained in the presence of 1 mM EDTA (A) or 10 mM CaCI, (B) or the synaptic particulate fractions (Pt.) obtained in the presence of 1 mM EDTA (C) or 10 mM CaCI, (D) were
separated by 6.5% SDS-PAGE, blotted and stained with Br4 using the horseradish peroxidase method. b: The synaptic soluble fraction obtained in the presence of 1 mh4 EDTA was incubated for 1 hr at 30°C in the presence of 10 mM CaCl,, followed by centrifugation at 100,OOOg for 1 hr. The supernatant (Sup.) (A) and the precipitate (Ppt.) (B) were analyzed as described in a. The synaptic soluble fraction which was incubated at 30°C without calcium ions was also analyzed (C).
opment. The antibody does not react with cultured astrocytes from chick o r rat brains. Although the early stages of cultured chick neurons are highly reactive with the antibody, the rat embryonic brain shows less antigenicity than the adult brain. It may reflect the difference between the species or the differentiation stages.
The staining pattern of the cultured neurons indicates the intracellular localization of the antigens. Immunoblots show that three proteins from rat brains, p145, p108 and p97, are recognized by this antibody. So far, it is unknown which proteins are recognized immunocytochemically . However, subcellular localization of p145 indicates its primary localization in synaptosomes.
Neuron-Specific Monoclonal Antibody
p145 and p97 exist both in cytosol and in synaptic soluble fractions, which matches the immunocytochemical staining. Also the regional distribution of the proteins in rat brain correlates with the immunocytochemical staining pattern. Especially, the amount of p97 seems very low in pons, medulla, and spinal cord. It seems that p108 and p97 are not breakdown products of p145, since the subcellular distribution of these three proteins are different. The proteins are found more abundant in brains than in other organs, except that p97 exists quite significantly in adrenal. It is observed that calcium ions decrease the solubility of these three proteins. It may relate to the phenomenon reported by Garner (1990) that lysis of synaptosomes in the presence of calcium ions causes a decrease in solubility of the presynaptic slow component b proteins. The three Br4 antigens exist not only in rat brain but also in chick and bovine brain (data not shown). We are now attempting the purification of these proteins from bovine brain. In summary, we have raised a monoclonal antibody Br4 that reacts a subset of neurons and recognizes three antigenic molecules. Further studies are necessary to determine the nature of neurons and the proteins recognized by this antibody.
REFERENCES Allen WK, Akeson R (1985): Identification of a cell surface glycoprotein family of olfactory receptor neurons with a monoclonal antibody. J Neurosci 5:284-296. Birk H-W, Koepsell H (1987): Reaction of monoclonal antibodies with plasma membrane proteins after binding on nitrocellulose: renaturation of antigen sites and reduction of nonspecific antibody binding. Anal Biochem 164:12-22. Cohen J, Selvendran SY (1983): A neuronal surface antigen is found in the CNS but not in peripheral neurons. Nature 305:424-427. DeJong ASH, Van Kessel-Ban Vark M, Raap AK (1985): Sensitivity of various visualization methods for peroxidase and alkaline phosphatase activity in immunoenzyme histochemistry . Histochem J 17:1119-1130. Fujita SC, Zipursky SL, Benzer S , Ferrus A, Shotwell SL (1982): Monoclonal antibodies against the Drosophila nervous system. Proc Natl Acad Sci USA 79:7929-7933. Garner JA (1990): Selective alterations in presynaptic cytomatrix pro-
113
tein organization induced by calcium and other divalent cations that modulate exocytosis. J Neurochem 54: 1770-1708. Gray EG, Whittaker VP (1962): The isolation of nerve endings from brain: An electron microscopic study: Cell fragments derived by homogenization and centrifugation. J Anat (Lond) 96:7987. Hempstead JL, Morgan JI (1985): A panel of monoclonal antibodies to the rat olfactory epithelium. J Neurosci 5:438-449. Hendry SHC, Jones EG, Hockfield S, McKay RDG (1988): Neuronal populations stained with monoclonal antibody Cat-301 in the mammalian cerebral cortex and thalamus. J Neurosci 8:5 18542. Hinton DR, Henderson VW, Blanks JC, Rudnicka M, Miller CA (1988): Monoclonal antibodies react with neuronal subpopulations in the human nervous system. J Comp Neurol 267:398408. Hirsh M-R, Langley OK, Ghandour MS, Hirn M, Gombos G, Goridis G (1983): A monoclonal antibody recognizing subpopulations of neurons in mouse brain, J Neuroimmunol 4:175-186. Hishinuma A, Hockfield S , McKay R, Hildebrand JG (1988): Monoclonal antibodies reveal cell-type-specific antigens in the sexually dimorphic olfactory system of Munducu sextu. I. Generation of monoclonal antibodies and partial characterization of the antigens. J Neurosci 8:296-307. Laemmli UK (1970): Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227680-685. Langley OK, Foucaud B, Ghandour MS, Barry JDe, Schladenhaufen Y, Gombos G (1985): A developmentally modified neuronspecific marker in rodent cerebellum. Neuroscience 14:147157.
Miller CA, Benzer S (1983): Monoclonal antibody cross-reactions between Drosophilu and human brain. Proc Natl Acad Sci USA 80:764 1-7645. Moskal JR, Schaffner AE (1986): Monoclonal antibodies to the dentate gyms: Immunocytochemical characterization and flow cytometric analysis of hippocampal neurons bearing a unique cell surface antigen. J Neurosci 6:2045-2053. Pettman B, Louis JC, Sensenbrenner M (1979): Morphological and biochemical maturation of neurons cultured in the absence of glial cells. Nature 281:378-380. Stainier DY, Gilbert W (1989): The monoclonal antibody 1330 recognizes a specific neuronal cell surface antigen in the developing mesencephalic trigeminal nucleus of the mouse. J Neurosci 9:2466-2485. Vulliamy T, Rattray S, Mirsky R (1981): Cell-surface antigen distinguishes sensory and autonomic peripheral neurons from central neurons. Nature 291:418-420. Woodhams PL, Webb M, Atkinson DJ, Seeley PJ (1989): A monoclonal antibody, Py, distinguishes different classes of hippocampal neurons. J Neurosci 9:2170-2181. Zipursky SL, Venkatesh TR, Teplow DB, Benzer S (1984): Neuronal development in the Drosophilu retina: Monoclonal antibodies as molecular probes. Cell 36:15-26.
