Brain Research, 509 (1990) 161-164 Elsevier

161

BRES 23963

A monoclonal antibody 5E5 recognizes an intranuclear antigen selectively present in a subpopulation of the neurons Kazunori Yoshimura 1, Kimio Akagawa 2, Yozo Nishimura 3 and Keiichi Uyemura 1 1Department of Physiology, Saitama Medical School, Saitama (Japan), 2Department of Neurochemistry, Institute for Physiological Science, Aichi (Japan) and 3Department of Anatomy, Jichi Medical School, Tochigi (Japan) (Accepted 27 October 1989)

Key words: Nuclear antigen; Monoclonal antibody 5E5; 5E5 antigen; Central nervous system; Single-stranded DNA-protein complex; Neuron

Monoclonal antibody 5E5 labeled the nuclear antigen of the neurons in the guinea pig and rat central nervous systems including the cerebrum, cerebellum, spinal cord and retina. This antibody could discriminate neurons even among the same cell class. In in vitro study, only 10% of dividing PC12 ceils was labeled with this antibody. An electron microscopic immunohistochemical study also revealed that this antibody selectively labeled heterochromatins in the neurons. Although we could not obtain any positive result by an immunoblot study, the antigenecity was remarkably diminished by the DNase I or S1 nuclease treatment on the tissue sections whereas RNase and trypsin was ineffective. These results suggested that this antigen might be a single-stranded DNA-protein complex resistant to proteolytic procedures, and possibly related to cell function or state of differentiation. The nervous system is highly differentiated and consisted of morphologically and physiologically heterogeneous cell populations. This heterogeneity represents molecular diversity of which expression was possibly controlled by specific factors in cell nuclei. Therefore identification of specific nuclear factors is prerequisite to understand the mechanism underlying neuronal differentiations, but it has been difficult to study such factors because of their diversity and extremely small quantities. While recent advances in monoclonal antibody method provided an approach to overcome this difficulty since they could identify unknown molecules of small amounts even in crude CNS preparations 1'6. In this context, we tried to raise monoclonal antibodies recognizing such possible factors characteristic of the nuclei in the mammalian nervous system. We here report that a monoclonal antibody, 5E5, raised against a crude non-histone protein fraction of guinea pig brains selectively recognizes a nuclear heterochromatin antigen of a distinct neuronal subpopulation. The nuclear fraction of adult guinea pig brains were prepared according to the method of Uyemura et al, H. The nuclear fraction was suspended with EDTA buffer (0.2 mM EDTA, 75 mM Tris-HC1, pH 7.4) and centrifuged at 1200 g for 10 min to obtain a pellet (P3). P3 was further resuspended in a salt buffer (0.35 M NaC1, 1 mM phenylmethysuifonyl fluoride, 1% dimethylsulfoxide, 10

mM Tris-HCl, pH 7.4), stirred overnight at 4 °C and centrifuged for 10 min at 12,000 g. The resulting pellet (P4) was resuspended into a urea buffer (6 M urea, 10 mM Tris-HCl, pH 7.4) and centrifuged for 1 h at 65,000 g to obtain a supernatant (BS5), which was expected to contain much non-histone proteins of the neuronal nuclei 5. BS5 (100 /~g protein) in Freund's complete adjuvant were subcutaneously injected into Balb/C mice twice with 3 weeks interval and the final booster (50 ~g protein in PBS) was given via the tail vein. Four days later, the spleen cells were obtained and fused with the myeloma cells (P3-NSI/1-Ag-4) using 50% polyethyleneglycol (mol. wt. 1500, Boehringer Mannheim). The specificity of the monoclonal antibody was immunohistochemically screened on cryostat sections of the tissues. The detail procedures of antibody production and indirect immunofluorescent (Rhodamine) studies were described elsewhere 2,3'6. For the enzyme treatments, sections were incubated with various concentration of either trypsin (17-100 U/ml), DNase I (0.2-5 mg/ml), RNase (0.2-5 mg/ml), S1 nuclease (2-20 U//A) or pronase 5-10 /~g/ml) in the nuclease buffer (10 mM Tris-HCl, pH 7.5, 50 mM MgC12, 0.1 mM PMSF, 50 mM NaCi) or PBS (for proteolytic digestion) containing 5% normal goat serum (NGS) at room temperature for 30 min. Then, the sections were washed extensively in 5% NGS and processed immunohistochemically. The electron micro-

Correspondence: K. Yoshimura, Department of Physiology, Saitama Medical School, Saitama, 350-04 Japan. 0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

162 scopic immunohistochemistry was carried out as described previously4"1°. PC12 cells were cultured as described by Senba et al. s. Of about 1200 hybridoma colonies obtained, mono-

clonai antibodies from a few cell lines selectively labeled nuclear antigens of the guinea pig nervous system. One of them, designated 5E5 antibody, preferentially labeled intranuclear structures of the n e u r o n s by a speckled

Fig. 1. Immunohistochemical labeling with 5E5 antibody in neuronal tissues and PC12 cells. A: immunofluorescent staining of the motor area (layers 2-3) of guinea pig cerebral cortex, x 160. B: layer 5 of labeled pyramidal cells. Arrows indicate unlabeled cells, x655. C: guinea pig cerebellum. Arrows indicate Purkinje cells. ×325. D: part of the PC12 cells was also stained with 5E5 antibody (arrowheads). Arrows indicate negative cells, x655. E: electron microscopic immunohistochemistry on a stellate cells of a rat cerebellum. Arrows and arrowheads indicate heterochromatins and nuclear membrane, respectively. × 10,200.

163 TABLE I

TABLE II

Localization of 5E5 antigens in various tissues of a guinea pig

The effects of enzyme treatment on adult rat cerebral cortices for 5E5 antigens

G, gray matter; W, white matter; 1, subpopulation of neurons; -, exclusively negative with higher concentration (>100) of antibody; +, weakly positive with higher concentration (>100) of antibody; + +, positive; + + +, strongly positive. Tissue

5E5 antigens

Cerebrum

+++

G W Cerebellum G W Spinal cord G W Retina Heart Liver Small intestine Colon Spleen Adrenal medulla Kidney Ovary Skeletal muscle

+++ +++ ++ _+ + -+_+ + + + _+ +

pattern (Fig. 1). In the guinea pig cerebral cortex (1A), positive cells were found in all layers except layer 1. These positive cells were neurons, since a double labeling study with G F A P (glial fibrillary acidic protein) (Labsystems) showed no double-labeled cells (data not shown) and also the white matter remained unstained. As shown in Fig. 1B, however, not all the cortical neurons were labeled. In the cerebellum (Fig. 1C), 5E5 stained the nuclei of all Purkinje cells (arrows), cells in the molecular layer (stellate or basket cells) and granule cells. Interestingly, immunohistochemically positive or negative cell population were observed simultaneously among granule cells, demonstrating that this antibody could discriminate distinct neurons among an identical cell class. Cells in the white matter was not labeled. These preferential localization and staining pattern was almost the same as those in the rat nervous system. Also in the in vitro study, 10% of undifferentiated, dividing PC12 cells was 5E5 positive (Fig. 1D) despite the adrenal medulla in vivo was not labeled. This result also suggested that only a part of the cells among the same cell class expressed 5E5 antigen. A n electron microscopic immunohistochemistry on the rat cerebellum demonstrated that 5E5 reacted positively with heterochromatins in the nuclei of a neuronal subpopulation (Fig. 1E). Presence of immunohistochemically positive cells was confined to the nervous system including cerebral and cerebellar cortices, spinal cord and retina of guinea pigs (Table I) and rats. In rat retina, most ganglion cells, if not all, were labeled with 5E5, while only a few photoreceptors were positively stained. All Purkinje cells in rat cerebellum were also labeled. No

+ + +, antigenicity had disappeared; +, antigenicity had somewhat decreased;-, no effect. Enzyme

Sensitivity to enzyme treatment

DNase I RNase S1 nuclease Trypsin Pronase

+++ +++ +

other tissue than the nervous system was not prominently stained with a conventional antibody concentration. Moreover, cultured retinal neurons and human neuroblastoma obtained by the surgical operation were labeled with 5E5 while C6 glioma cells were not (manuscripts, in preparation), suggesting that the 5E5 antigen-containing cells were involved mostly in neuronal or neuron-related cel~ types. In order to identify the antigenic molecule of 5E5, we first tried immunoblot studies. But no positive result was obtained. Therefore, we further attempted to treat the sections with several digestive enzymes to study if the antigenicity was affected. As shown in Table II, it was revealed that treatment with either DNase I or S1 nuclease, which cleaves single chain D N A , remarkably diminished the antigenicity whereas RNase and trypsin were ineffective. Pronase treatment slightly weakened the antigenicity. However, purified D N A from the rat CNS did not react with 5E5 on dot blot analysis. Chloroform/methanol could not extract the antigen either (data not shown), demonstrating that this antigen could not be lipid in nature. Taken together, these results implied that the 5E5 antigen might be a single-stranded DNA-protein complex peculiar to subpopulation of neurons. As for the staining pattern, 5E5 labeling resembled that of the proliferating cell nuclear protein in dividing cells 7 or the nuclear matrix antigen reported by Lehner et al. 9. However, it seemed unlikely that 5E5 corresponded to either of these antigens since neurons retaining 5E5 do not proliferate and tissue distribution are quite different from those of these antigens. On the other hand, the characteristic localization within the nervous system suggested that this nuclear antigen might be related to some unknown neuronal function or differentiational state of the neurons. However, no physiological meaning nor characterization of this antigen is clear at present and further studies will be required to know the possibilities discussed above. This work was supported by Special Coordination Funds of the Science and Technology Agency of the Japanese government.

64 1 Akagawa, K. and Barnstable, C.J., Identification and characterization of cell types in monolayer cultures of rat retina using monoclonal antibodies, Brain Research, 383 (1986) 110-120. 2 Akagawa, K. and Barnstable, C.J., Identification and characterization of cell types accumulating GABA in rat retinal cultures using cell type specific monoclonal antibodies, Brain Research, 408 (1986) 154-162. 3 Akagawa, K., Hicks, D. and Barnstable, C.J., Histotypic organization and cell differentiation in rat retinal reaggregate cultures, Brain Research, 437 (1987) 298-308. 4 Akagawa, K., Nishimura, Y., Sakamoto, Y., Uyemura, K., Matoh, M. and Barnstable, C.J., A monoclonal antibody 2.6A recognizes a minor component protein of 38K in the plain synaptic vesicles, Biomed. Res., 9 (1988) 161-168. 5 Altieri, E, Allegra, P., Lonigro, I.R., Scarpa, S. and Caiafa, P., Distribution of tissue-specific tightly bound non-histone proteins in the first level of repeating chromatin structures, Eur+ J. Biochem., 154 (1983) 147-152. 6 Barnstable, C.J., Akagawa, K., Holstein, R. and Horn, J.P., Monoclonal antibodies that label discrete cell types in the

7 8

9

10 11

mammalian nervous system, Cold Spring ttarbor ,syrup. Quant+ Biol., 48 (1983) 863-876. Bravo, R., Synthesis of the nuclear protein cyclin (PCNA) and its relationship with DNA replication, Exp+ Cell Res., 163 (1986) 287-293. Katoh-Senba, R., Kitajima, S., Yamazaki, Y. and Sano, M., Neuritic growth from a new subline of PC12 pheochromocytoma cells; cyclic AMP mimics the action of nerve growth factor, J. Neurosci. Res., 17 (1987) 36-44. Lehner, C.E, Eppenberger, H.M., Fakan, S. and Nigg, E.A., Nuclear substructure antigens monoclonal antibodies against components of nuclear matrix preparations, Exp. Cell Resr. 162 (1986) 205-219. Nishimura, Y., Schwarz, M.L. and Rakic, P., Localization of 7-aminobutyric acid and glutamic acid decarboxylase in rhesus monkey retina, Brain Research, 359 (1985) 351-355. Uyemura, K., Nakayama, T., Kitamura, K , Yamanaka, T. and Hirano, S+, Nuclear proteins of the guinea pig brain, J~ Neurochem., 23 (1974) 65-70.

A monoclonal antibody 5E5 recognizes an intranuclear antigen selectively present in a subpopulation of the neurons.

Monoclonal antibody 5E5 labeled the nuclear antigen of the neurons in the guinea pig and rat central nervous systems including the cerebrum, cerebellu...
1MB Sizes 0 Downloads 0 Views