- Email: [email protected]
Glial fibrillary acidic protein (GFAP) in rat brain tumors transplacentally induced by ethylnitrosourea (ENU)
Journal of the Neurological Sciences, 1983, 61:349-355 Elsevier
349
GLIAL FIBRILLARY ACIDIC PROTEIN (GFAP) IN RAT BRAIN TUMORS TRANSPLACENTALLY INDUCED BY ETHYLNITROSOUREA (ENU)
A. MAURO, M.T. GIORDANA, A. MIGHELI and D. SCHIFFER
2nd Neurological Clinic, University of Turin, Turin (Italy) (Received 8 March, 1983) (Revised, received 9 May, 1983) (Accepted 8 June, 1983)
SUMMARY
The immunohistochemical distribution ofglial fibrillary acidic protein (GFAP) in neoplastic lesions induced in the rat by ENU is reported. GFAP was present in hypertrophic reactive astrocytes, which were numerous in early neoplastic proliferations, in microtumors of the white matter, and in those collected at the periphery of large tumors. They were absent in cortical oligodendroglial foci and microtumors. No GFAP-positive cells were observed in hyperplasias of the white matter: astrocyte-like cells of large tumors were GFAP-negative. The significance of reactive astrocytes and the problem of the astrocytic component in transplacental ENU tumors are discussed.
Key words: A s t r o c y t e s - B r a i n t u m o u r - Ethylnitrosourea ( E N U ) - Glial fibrillary acidic protein ( G F A P ) - R a t
INTRODUCTION
Oligodendrogliomas are the most common tumors induced in the rat by transplacental administration of ethylnitrosourea (ENU) (Grossi-Paoletti etal. 1970; Koestner et al. 1971). Astrocytic participation, as reactive or tumoral astrocytes, has been repeatedly emphasized; a few microtumors and tumors have been diagnosed as
This work was supported by CNR, Special Project "Control of Neoplastic Growth", Grant No. 81.01462.96. Requests for reprints to: A. Mauro, 2nd Neurological Clinic, University of Turin, Via Cherasco, 15, 1-10126 Turin, Italy. 0022-510X/83/$03.00 © 1983 Elsevier Science Publishers B.V.
350 astrocytomas and terms such as "polymorphous" or "mixed gliomas" have been frequently used to indicate the occurrence of astrocytic and oligodendrocytic and/or ependymoma-like cells in the same tumor (Janisch and Schreiber 1969; Wechsler et al. 1969; Schiffer et al. 1970, 1971, 1978; Koestner et al. 1971; Druckrey et al. 1972; Lantos 1972; Pilkington and Lantos 1979). As for the so-called ependymomas, it has been recently demonstrated that they are not true ependymal tumors (Mandybur and Alvira 1982). By means of the GFA-protein demonstration we attempted to investigate and define the astrocytic component in gliomas induced by transplacental ENU in the rat. MATERIAL AND METHODS
From our collection of CNS tumors induced by transplacental ENU in Fisher-344 rats (20 mg/kg ENU i.v. on day 17 of gestation), we selected 6 cell hyperplasias of the white matter, 10 early neoplastic proliferations (ENPs), 6 cortical oligodendroglial foci, 8 microtumors, 11 isomorphic and 10 polymorphic oligodendrogliomas. These lesions represent the different stages of tumoral development, as described by Koestner et al. (1971) and by us (Schiffer et al. 1980). Ten normal rats, killed from the 30th day of extrauterine life to adult life, were used as controls. Brains fixed in Carnoy at 0-4 °C and paraffin-embedded were cut in 5 #m thick serial sections. Different histological methods were used. The GFAP was evidenced by the immunohistochemical method of peroxidase-antiperoxidase (PAP), according to Sternberger et al. (1970). Rabbit GFAPantiserum was that of Dako Corporation, Santa Barbara, CA (PAP Kit K 507 for GFA). Sections were counterstained with hematoxylin, dehydrated and mounted. RESULTS In control rats a positive reaction was observed in fibrous astrocytes found in the molecular layer of the cerebral cortex, the paraventricular white matter, the "limitans gliae" membrane, the hippocampus, the subependymal layer, the thalamus, the white matter, in the granular layer of the cerebellum and in Bergmann's glia. Occasionally, the perivascular framework of astrocytic processes was clearly evident. In ENU-treated rats the appearance of cell hyperplasias in the paraventricular white matter, between the 30th and the 60th day of extrauterine life, did not modify the distribution and frequency of GFAP-positive astrocytes. ENPs, appearing after the 60th day of extrauterine life, were characterized by abundant hypertrophic astrocytes with thick, long processes, which were intensely positive for GFAP (Figs. la and b); they often showed bizarre or multiple nuclei. These astrocytes were distributed over an area wider than the tumoral lesion, affecting also the adjacent cortex. They developed processes abutting on blood vessels. The same picture was observed in the microtumors arising from ENPs: GFAPpositive hypertrophic astrocytes were found inside and around the lesion (Fig. lc). In these microtumors, circumscribed clusters of closely packed tumoral cells, with a
351
Fig. 1. a: Reactive astrocytes in a ENP of the paraventricular white matter. GFAP-PAP, x 150. b: The same as in a at higher magnification. c: Microtumor with hypertrophic astrocytes and a cluster of tumoral cells. GFAP-PAP, x 200. d: Cortical oligodendroglial focus without hypertrophic astrocytes. GFAP-PAP, x 150. perinuclear rim o f cytoplasm, were completely G F A P - n e g a t i v e . They were often s u r r o u n d e d by G F A P - p o s i t i v e processes o f hypertrophic astrocytes. Neither the cortical oligodendroglial foci (Fig. l d ) , microtumors, nor isomorphic oligodendrogliomas c o n t a i n e d G F A P - p o s i t i v e astrocytes: but if the t u m o r outgrew the
352
Fig, 2. a: Isomorphic oligodendroglioma. GFAP-positive astrocytes at the periphery and in the surrounding normal tissue. GFAP-PAP, x 150. b: Hypertrophic astrocytes at the periphery of an isomorphic oligodendroglioma. GFAP-PAP, x 400.
cortex, invading the hippocampus and the white matter, GFAP-positive astrocytes appeared at the periphery of the tumor and in the surrounding normal tissue (Figs. 2 and b). In the polymorphic gliomas tumoral cells, even those with abundant cytoplasm, resembling astroglial cells, were GFAP-negative. Hypertrophic GFAP-positive astrocytes were again present in the peripheral areas of the tumor and in the surrounding normal tissue. DISCUSSION
The distribution of GFAP-positive astrocytes found by us in normal rats corresponds to that described by different authors (Bignami et al. 1980). Stellate astrocytes were particularly evident in hemispheric white matter and hippocampus. In our series of over 400 tumors transplacentally induced by ENU we observed neither pure ependymomas nor isomorphic astrocytomas (Schiffer et al. 1970, 1971, 1978, 1980), in accordance with Janisch and Schreiber (1977). However, an astrocytic component was clearly evident in the so-called "polymorphic" microtumors and gliomas, on the basis of light-microscopic similarities of some tumoral cells with astrocytes. Inconsistent with these observations and with those of Conley (1979), our present results fail to demonstrate the exsistence of GFAP-positive tumoral astrocytes. We believe that the large, intensely GFAP-positive astrocytes observed within and around
353 the tumoral lesions, were actually reactive astrocytes. As a matter of fact, not only their spider aspect, with thick, long processes, but also their relationship with vessels and their distribution support this interpretation: they increased from the center to the periphery of the tumor and progressively decreased in the surrounding brain. The occurrence of reactive astrocytes in ENU tumors is generally acknowledged. However, it is not easy to distinguish them from neoplastic astrocytes even from the ultrastructural point of view (Lantos 1974). In this regard it must be noted that the occurrence of bizarre or multiple nuclei in astrocytes is not definitive evidence of their neoplastic nature. Mandybur and Alvira (1982) also consider the astrocytes encountered in ependymoma-like tumors induced by ENU as reactive cells. The lack of GFAP-positive tumoral cells in ENU tumors is surprising and its interpretation is not easy. It must be noted that the identification of the astrocytic nature of cells by means of GFAP presents some problems, even though there is no doubt that GFAP is a specific marker of astrocytes (Biguami et al. 1972; Uyeda et al. 1972). A positive immunochemical reaction is due to the occurrence of glial filaments or to their subunits (Eng and Bigbee 1978; Rueger et al. 1979; Lucas et al. 1980). It could be negative in cells of astroglial origin, either too primitive or too anaplastic to produce GFAP (De Armond et al. 1980). On the other hand, GFAP could be positive in astrocytic cells without filaments, due to the existence of a soluble form of the GFA protein (De Armond et al. 1980). The possibility has also been demonstrated that glial cells, other than astrocytes, show positive GFAP reaction (Deck et al. 1978; Van der Meulen et al. 1978; Eng and Rubinstein 1978; Duffy et al. 1979; Velasco et al. 1980). In human oligodendrogliomas, for example, GFAP-positive tumoral cells, morphologically indistinguishable from tumoral oligodendrocytes, have been demonstrated (Eng and Rubinstein 1978; Van der Meulen et al. 1978; Velasco et al. 1980; Giordana et al. 1982; Schiffer et al. 1983). We did not observe this in ENU oligodendrogliomas. In conclusion, due to the lack of other morphological signs of immaturity and of regression in most ENU tumors, we are inclined to consider the GFAP-negative response of tumor cells as indicative of a real absence of tumoral astrocytes. Of paramount importance is the observation that the cell hyperplasias of the white matter, representing the earliest morphological alterations produced by transplacental ENU, did not contain GFAP-positive cells. This means that astrocytes do not participate in the earliest neoplastic lesions; since reactive astrocytes are also absent, we may say that these lesions do not represent an adequate stimulus for glial reaction. The pathogenesis of glial reaction in ENPs is unknown; damage to the myelin sheaths (Lantos and Pilkington 1979), might be responsible. The intensity of the astrocytic response may be related to the location of the tumor: the sites normally rich in stellate astrocytes such as the hippocampus and the paraventricular white matter, displayed the strongest reaction. We formerly interpreted the clusters of packed cells in microtumors as foci of astrocytic differentiation, because of the light-microscopic features (Schiffer and Giordana 1974; Schiffer et al. 1978). The negative response of GFAP of these cells
354 c a s t s d o u b t o n t h e i r a s t r o c y t i c n a t u r e . P o s s i b l y t h e y a r e c e n t r e s o f cell a n a p l a s i a , as Engelhardt and Bannasch
( 1 9 7 8 ) m a i n t a i n , o r e v e n c e n t r e s o f cell p r o l i f e r a t i o n in
oligodendroglial tumors.
REFERENCES Bignami, A., L.F. Eng, D. Dahl and C.T. Uyeda (1972) Localization of the glial fibrillary acidic protein in astrocytes by immunofluorescence, Brain Res., 43: 429-435. Bignami, A., D. Dahl and D.C. Rueger (1980) Glial fibrillary acidic protein (GFA) in normal neural cells and in pathological conditions. In: S. Fedoroff and L. Hertz (Eds.) Advances in Cellular Neurobiology, Academic Press, New York, pp. 285-310. Conley, F.K. (1979) The immunocytochemical localization of GFA protein in experimental murine CNS tumors, Acta Neuropath. (Berl.), 45: 9-16. De Armond, S.J., L.F. Eng and L.J. Rubinstein (1980) The application of glial fibrillary acidic (GFA) protein immunohistochemistry in neurooncology, Path. Res. Pract., 168: 374-379. Deck, J. H. N., L.F. Eng, J. Bigbee and S.M. Woodcock (1978) The role of glial fibrillary acidic protein in the diagnosis of central nervous system tumors, Acta Neuropath. (Berl.), 42: 183-190. Druckrey, H., S. Ivankovic, R. Preussmann, J. Z~ilch and H.D. Mennel (1972) Selective induction of malignant tumors of the nervous system by resorptive carcinogens. In: W. M. Kirsch, E. Grossi-Paoletti and P. Paoletti (Eds.), The Experimental Biology of Brain Tumors, Charles C. Thomas, Springfield, IL, pp. 85-147. Duffy, P.E., L. Graf, Y. Huang and M.M. Rapport (1979) Glial fibrillary acidic protein in ependymomas and other brain tumors, J. Neurol. Sci., 40: 133-146. Eng, L.F. and J.W. Bigbee (1978) Immunohistochemistry of nervous system specific antigens, Adv. Neurochem., 3: 43-98. Eng, L.F. and L.J. Rubinstein (1978) Contribution of immunohistochemistry to diagnostic problems of human cerebral tumors, J. Histochem. Cytochem., 26: 513-522. Engelhardt, A. and P. Bannasch (1978) Histochemie saurer Mucopolysaccharide w~ihrend der Genese Methylnitrosoharnstoff-induzierter Hirntumoren der Ratte, Acta Neuropath. (Berl.), 32: 197-204. Giordana, M. T., A. Mauro, A. Migheli and D. Schiffer (1982) GFAP usefullness in resolving open problems of human and experimental neuro-oncology, IXth Intern. Congr. Neuropath., Vienna, Abstract 1-53. Grossi-Paoletti, E., P. Paoletti, D. Schiffer and A. Fabiani (1970) Experimental brain tumours induced in rats by nitrosourea derivatives, Part 2 (Morphological aspects of nitrosoethylurea tumours obtained by transplacental induction), J. Neurol. Sci., 11: 573-581. J~nisch, W. and D. Schreiber (1969) Experimentelle Geschwiilste des Zentralnervensystems, Fischer Verlag, Jena. J~inisch, W. and D. Schreiber (1977) In: D.D. Bigner and S.A. Swenberg (Eds.), Experimental Tumors of the Central Nervous System, 1st english edition, Upjohn Co. Pub., Kalamazoo, MI. Koestner, A., J.A. Swenberg and W. Wechsler (1971) Transplacental production with ethylnitrosourea of neoplasms of the nervous system in Sprague-Dawley rats, Amer. J. Path., 63: 37-56. Lantos, P.L. (1972) The fine structure of periventricular pleomorphic gliomas induced transplacentally by N-ethyl-N-nitrosourea in BD-IX rats, with a note on their origin, J. Neurol. Sci., 17: 443-460. Lantos, P.L. (1974) An electron microscopic study of reacting astrocytes in gliomas induced by N-ethylN-nitrosourea in rats, Acta Neuropath. (Berl.), 30: 175-181. Lantos, P.L. and G.J. Pilkington (1979) The development of experimental brain tumours - - A sequential light and electron microscope study of the subependymal plate, Part 1 (Early lesions (abnormal cell clusters)), Acta Neuropath. (Berl.), 45: 167-175. Lucas, C.V., K.G. Bensch and L.F. Eng (1980) In vitro polymerization of the glial fibrillary acidic protein extracted from multiple sclerosis brain, Neurochem. Res., 5: 247-255. M andybur, T. I. and M. M. Alvira (1982) Ultrastructural findings in so-called ependymal rat tumors induced by transplacental administration of ethylnitrosourea (ENU), Acta Neuropath. (Berl.); 57: 51-58. Pilkington, G.J. and P.L. Lantos (1979) The development of experimental brain tumours - - A sequential light and electron microscope study of the subependymal plate, Part 2 (Microtumours), Acta Neuropath. (Berl.), 45: 177-185. Rueger, D. C., J. S. Huston, D. Dahl and A. Bignami (1979) Formation of 100 A filaments from purified glial fibrillary acidic protein in vitro, J. Mol. Biol., 135: 53-68.
355 Schiffer, D. and M.T. Giordana (1974) On the occurrence and significance of acid mucopolysaccharides in oligodendrogliomas experimentally induced in the rat by nitrosourea derivatives. In: D. Schreiber and W. J~inisch (Eds.), Experimentelle Neuroonkologie, Wiss. Ber. Martin-Luther Universit~it, HalleWittenberg, pp. 101-108. Schiffer, D., A. Fabiani, E. Grossi-Paoletti and P. Paoletti (1970) Experimental brain tumours induced in rats by nitrosourea derivatives, Part 1 (Morphological aspects of methylnitrosourea tumours), J. Neurol. Sci., 11: 559-572. Schiffer, D., A. Fabiani, E. Grossi-Paoletti and P. Paoletti (1971) Tumori cerebrali sperimentali del ratto da derivati della nitrosourea - - Considerazioni patogenetiche sui gliomi misti e polimorfi, Tumori, 57: 333-341. Schiffer, D., M.T. Giordana, S. Pezzotta, C. Lechner and P. Paoletti (1978) Cerebral tumors induced by transplacental ENU - - Study of the different tumoral stages, particularly of early proliferations, Acta Neuropath. (Berl.), 41: 27-31. Schiffer, D., M.T. Giordana, A. Mauro, G. Racagni, F. Bruno, S. Pezzotta and P. Paoletti (1980) Experimental brain tumors by transplacental ENU - - Multifactorial study of the latency period, Acta Neuropath. (Berl.), 49:117-122. Schiffer, D., M.T. Giordana, A. Mauro and A. Migheli (1983) Glial fibrillary acidic protein (GFAP) in human cerebral tumors - - An immunohistochemical study, Tumori, 69: 95-104. Sternberger, L.A., P.H. Hardy, Jr., F.F. Cuculis and H.G. Meyer (1970) The unlabeled antibody enzyme method of immunohistochemistry ~ Preparation and properties of soluble antigen-antibody complex (horse-radish peroxidase-antiperoxidase) and its use in identification of spirochetes, J. Histochem. Cytochem., 18: 315-333. Uyeda, C.T., Eng, L.E. and A. Bignami (1972) Immunological study of the glial fibrillary acidic protein, Brain Res., 37: 81-89. Van der Meulen, J.D.M., H.J. Houthoff and E.J. Ebels (1978) Glial fibrillary acidic protein in human gliomas, Neuropath. Appl. Neurobiol., 4: 177-190. Velasco, M. E., D. Dahl, U. Roessmann and P. L. Gambetti (1980) Immunohistochemical localization ofglial fibrillary acidic protein in human glial neoplasms, Cancer, 45: 484-494. Wechsler, W., P. Kleihues, S. Matsumoto, K. J. Ztilch, S. Ivankovic, R. Preussmann and H. Druckrey (1969) Pathology of experimental neurogenic tumors chemically induced during prenatal and postnatal life, Ann. N.Y. Acad. Sci., 159: 360-408.
349
GLIAL FIBRILLARY ACIDIC PROTEIN (GFAP) IN RAT BRAIN TUMORS TRANSPLACENTALLY INDUCED BY ETHYLNITROSOUREA (ENU)
A. MAURO, M.T. GIORDANA, A. MIGHELI and D. SCHIFFER
2nd Neurological Clinic, University of Turin, Turin (Italy) (Received 8 March, 1983) (Revised, received 9 May, 1983) (Accepted 8 June, 1983)
SUMMARY
The immunohistochemical distribution ofglial fibrillary acidic protein (GFAP) in neoplastic lesions induced in the rat by ENU is reported. GFAP was present in hypertrophic reactive astrocytes, which were numerous in early neoplastic proliferations, in microtumors of the white matter, and in those collected at the periphery of large tumors. They were absent in cortical oligodendroglial foci and microtumors. No GFAP-positive cells were observed in hyperplasias of the white matter: astrocyte-like cells of large tumors were GFAP-negative. The significance of reactive astrocytes and the problem of the astrocytic component in transplacental ENU tumors are discussed.
Key words: A s t r o c y t e s - B r a i n t u m o u r - Ethylnitrosourea ( E N U ) - Glial fibrillary acidic protein ( G F A P ) - R a t
INTRODUCTION
Oligodendrogliomas are the most common tumors induced in the rat by transplacental administration of ethylnitrosourea (ENU) (Grossi-Paoletti etal. 1970; Koestner et al. 1971). Astrocytic participation, as reactive or tumoral astrocytes, has been repeatedly emphasized; a few microtumors and tumors have been diagnosed as
This work was supported by CNR, Special Project "Control of Neoplastic Growth", Grant No. 81.01462.96. Requests for reprints to: A. Mauro, 2nd Neurological Clinic, University of Turin, Via Cherasco, 15, 1-10126 Turin, Italy. 0022-510X/83/$03.00 © 1983 Elsevier Science Publishers B.V.
350 astrocytomas and terms such as "polymorphous" or "mixed gliomas" have been frequently used to indicate the occurrence of astrocytic and oligodendrocytic and/or ependymoma-like cells in the same tumor (Janisch and Schreiber 1969; Wechsler et al. 1969; Schiffer et al. 1970, 1971, 1978; Koestner et al. 1971; Druckrey et al. 1972; Lantos 1972; Pilkington and Lantos 1979). As for the so-called ependymomas, it has been recently demonstrated that they are not true ependymal tumors (Mandybur and Alvira 1982). By means of the GFA-protein demonstration we attempted to investigate and define the astrocytic component in gliomas induced by transplacental ENU in the rat. MATERIAL AND METHODS
From our collection of CNS tumors induced by transplacental ENU in Fisher-344 rats (20 mg/kg ENU i.v. on day 17 of gestation), we selected 6 cell hyperplasias of the white matter, 10 early neoplastic proliferations (ENPs), 6 cortical oligodendroglial foci, 8 microtumors, 11 isomorphic and 10 polymorphic oligodendrogliomas. These lesions represent the different stages of tumoral development, as described by Koestner et al. (1971) and by us (Schiffer et al. 1980). Ten normal rats, killed from the 30th day of extrauterine life to adult life, were used as controls. Brains fixed in Carnoy at 0-4 °C and paraffin-embedded were cut in 5 #m thick serial sections. Different histological methods were used. The GFAP was evidenced by the immunohistochemical method of peroxidase-antiperoxidase (PAP), according to Sternberger et al. (1970). Rabbit GFAPantiserum was that of Dako Corporation, Santa Barbara, CA (PAP Kit K 507 for GFA). Sections were counterstained with hematoxylin, dehydrated and mounted. RESULTS In control rats a positive reaction was observed in fibrous astrocytes found in the molecular layer of the cerebral cortex, the paraventricular white matter, the "limitans gliae" membrane, the hippocampus, the subependymal layer, the thalamus, the white matter, in the granular layer of the cerebellum and in Bergmann's glia. Occasionally, the perivascular framework of astrocytic processes was clearly evident. In ENU-treated rats the appearance of cell hyperplasias in the paraventricular white matter, between the 30th and the 60th day of extrauterine life, did not modify the distribution and frequency of GFAP-positive astrocytes. ENPs, appearing after the 60th day of extrauterine life, were characterized by abundant hypertrophic astrocytes with thick, long processes, which were intensely positive for GFAP (Figs. la and b); they often showed bizarre or multiple nuclei. These astrocytes were distributed over an area wider than the tumoral lesion, affecting also the adjacent cortex. They developed processes abutting on blood vessels. The same picture was observed in the microtumors arising from ENPs: GFAPpositive hypertrophic astrocytes were found inside and around the lesion (Fig. lc). In these microtumors, circumscribed clusters of closely packed tumoral cells, with a
351
Fig. 1. a: Reactive astrocytes in a ENP of the paraventricular white matter. GFAP-PAP, x 150. b: The same as in a at higher magnification. c: Microtumor with hypertrophic astrocytes and a cluster of tumoral cells. GFAP-PAP, x 200. d: Cortical oligodendroglial focus without hypertrophic astrocytes. GFAP-PAP, x 150. perinuclear rim o f cytoplasm, were completely G F A P - n e g a t i v e . They were often s u r r o u n d e d by G F A P - p o s i t i v e processes o f hypertrophic astrocytes. Neither the cortical oligodendroglial foci (Fig. l d ) , microtumors, nor isomorphic oligodendrogliomas c o n t a i n e d G F A P - p o s i t i v e astrocytes: but if the t u m o r outgrew the
352
Fig, 2. a: Isomorphic oligodendroglioma. GFAP-positive astrocytes at the periphery and in the surrounding normal tissue. GFAP-PAP, x 150. b: Hypertrophic astrocytes at the periphery of an isomorphic oligodendroglioma. GFAP-PAP, x 400.
cortex, invading the hippocampus and the white matter, GFAP-positive astrocytes appeared at the periphery of the tumor and in the surrounding normal tissue (Figs. 2 and b). In the polymorphic gliomas tumoral cells, even those with abundant cytoplasm, resembling astroglial cells, were GFAP-negative. Hypertrophic GFAP-positive astrocytes were again present in the peripheral areas of the tumor and in the surrounding normal tissue. DISCUSSION
The distribution of GFAP-positive astrocytes found by us in normal rats corresponds to that described by different authors (Bignami et al. 1980). Stellate astrocytes were particularly evident in hemispheric white matter and hippocampus. In our series of over 400 tumors transplacentally induced by ENU we observed neither pure ependymomas nor isomorphic astrocytomas (Schiffer et al. 1970, 1971, 1978, 1980), in accordance with Janisch and Schreiber (1977). However, an astrocytic component was clearly evident in the so-called "polymorphic" microtumors and gliomas, on the basis of light-microscopic similarities of some tumoral cells with astrocytes. Inconsistent with these observations and with those of Conley (1979), our present results fail to demonstrate the exsistence of GFAP-positive tumoral astrocytes. We believe that the large, intensely GFAP-positive astrocytes observed within and around
353 the tumoral lesions, were actually reactive astrocytes. As a matter of fact, not only their spider aspect, with thick, long processes, but also their relationship with vessels and their distribution support this interpretation: they increased from the center to the periphery of the tumor and progressively decreased in the surrounding brain. The occurrence of reactive astrocytes in ENU tumors is generally acknowledged. However, it is not easy to distinguish them from neoplastic astrocytes even from the ultrastructural point of view (Lantos 1974). In this regard it must be noted that the occurrence of bizarre or multiple nuclei in astrocytes is not definitive evidence of their neoplastic nature. Mandybur and Alvira (1982) also consider the astrocytes encountered in ependymoma-like tumors induced by ENU as reactive cells. The lack of GFAP-positive tumoral cells in ENU tumors is surprising and its interpretation is not easy. It must be noted that the identification of the astrocytic nature of cells by means of GFAP presents some problems, even though there is no doubt that GFAP is a specific marker of astrocytes (Biguami et al. 1972; Uyeda et al. 1972). A positive immunochemical reaction is due to the occurrence of glial filaments or to their subunits (Eng and Bigbee 1978; Rueger et al. 1979; Lucas et al. 1980). It could be negative in cells of astroglial origin, either too primitive or too anaplastic to produce GFAP (De Armond et al. 1980). On the other hand, GFAP could be positive in astrocytic cells without filaments, due to the existence of a soluble form of the GFA protein (De Armond et al. 1980). The possibility has also been demonstrated that glial cells, other than astrocytes, show positive GFAP reaction (Deck et al. 1978; Van der Meulen et al. 1978; Eng and Rubinstein 1978; Duffy et al. 1979; Velasco et al. 1980). In human oligodendrogliomas, for example, GFAP-positive tumoral cells, morphologically indistinguishable from tumoral oligodendrocytes, have been demonstrated (Eng and Rubinstein 1978; Van der Meulen et al. 1978; Velasco et al. 1980; Giordana et al. 1982; Schiffer et al. 1983). We did not observe this in ENU oligodendrogliomas. In conclusion, due to the lack of other morphological signs of immaturity and of regression in most ENU tumors, we are inclined to consider the GFAP-negative response of tumor cells as indicative of a real absence of tumoral astrocytes. Of paramount importance is the observation that the cell hyperplasias of the white matter, representing the earliest morphological alterations produced by transplacental ENU, did not contain GFAP-positive cells. This means that astrocytes do not participate in the earliest neoplastic lesions; since reactive astrocytes are also absent, we may say that these lesions do not represent an adequate stimulus for glial reaction. The pathogenesis of glial reaction in ENPs is unknown; damage to the myelin sheaths (Lantos and Pilkington 1979), might be responsible. The intensity of the astrocytic response may be related to the location of the tumor: the sites normally rich in stellate astrocytes such as the hippocampus and the paraventricular white matter, displayed the strongest reaction. We formerly interpreted the clusters of packed cells in microtumors as foci of astrocytic differentiation, because of the light-microscopic features (Schiffer and Giordana 1974; Schiffer et al. 1978). The negative response of GFAP of these cells
354 c a s t s d o u b t o n t h e i r a s t r o c y t i c n a t u r e . P o s s i b l y t h e y a r e c e n t r e s o f cell a n a p l a s i a , as Engelhardt and Bannasch
( 1 9 7 8 ) m a i n t a i n , o r e v e n c e n t r e s o f cell p r o l i f e r a t i o n in
oligodendroglial tumors.
REFERENCES Bignami, A., L.F. Eng, D. Dahl and C.T. Uyeda (1972) Localization of the glial fibrillary acidic protein in astrocytes by immunofluorescence, Brain Res., 43: 429-435. Bignami, A., D. Dahl and D.C. Rueger (1980) Glial fibrillary acidic protein (GFA) in normal neural cells and in pathological conditions. In: S. Fedoroff and L. Hertz (Eds.) Advances in Cellular Neurobiology, Academic Press, New York, pp. 285-310. Conley, F.K. (1979) The immunocytochemical localization of GFA protein in experimental murine CNS tumors, Acta Neuropath. (Berl.), 45: 9-16. De Armond, S.J., L.F. Eng and L.J. Rubinstein (1980) The application of glial fibrillary acidic (GFA) protein immunohistochemistry in neurooncology, Path. Res. Pract., 168: 374-379. Deck, J. H. N., L.F. Eng, J. Bigbee and S.M. Woodcock (1978) The role of glial fibrillary acidic protein in the diagnosis of central nervous system tumors, Acta Neuropath. (Berl.), 42: 183-190. Druckrey, H., S. Ivankovic, R. Preussmann, J. Z~ilch and H.D. Mennel (1972) Selective induction of malignant tumors of the nervous system by resorptive carcinogens. In: W. M. Kirsch, E. Grossi-Paoletti and P. Paoletti (Eds.), The Experimental Biology of Brain Tumors, Charles C. Thomas, Springfield, IL, pp. 85-147. Duffy, P.E., L. Graf, Y. Huang and M.M. Rapport (1979) Glial fibrillary acidic protein in ependymomas and other brain tumors, J. Neurol. Sci., 40: 133-146. Eng, L.F. and J.W. Bigbee (1978) Immunohistochemistry of nervous system specific antigens, Adv. Neurochem., 3: 43-98. Eng, L.F. and L.J. Rubinstein (1978) Contribution of immunohistochemistry to diagnostic problems of human cerebral tumors, J. Histochem. Cytochem., 26: 513-522. Engelhardt, A. and P. Bannasch (1978) Histochemie saurer Mucopolysaccharide w~ihrend der Genese Methylnitrosoharnstoff-induzierter Hirntumoren der Ratte, Acta Neuropath. (Berl.), 32: 197-204. Giordana, M. T., A. Mauro, A. Migheli and D. Schiffer (1982) GFAP usefullness in resolving open problems of human and experimental neuro-oncology, IXth Intern. Congr. Neuropath., Vienna, Abstract 1-53. Grossi-Paoletti, E., P. Paoletti, D. Schiffer and A. Fabiani (1970) Experimental brain tumours induced in rats by nitrosourea derivatives, Part 2 (Morphological aspects of nitrosoethylurea tumours obtained by transplacental induction), J. Neurol. Sci., 11: 573-581. J~nisch, W. and D. Schreiber (1969) Experimentelle Geschwiilste des Zentralnervensystems, Fischer Verlag, Jena. J~inisch, W. and D. Schreiber (1977) In: D.D. Bigner and S.A. Swenberg (Eds.), Experimental Tumors of the Central Nervous System, 1st english edition, Upjohn Co. Pub., Kalamazoo, MI. Koestner, A., J.A. Swenberg and W. Wechsler (1971) Transplacental production with ethylnitrosourea of neoplasms of the nervous system in Sprague-Dawley rats, Amer. J. Path., 63: 37-56. Lantos, P.L. (1972) The fine structure of periventricular pleomorphic gliomas induced transplacentally by N-ethyl-N-nitrosourea in BD-IX rats, with a note on their origin, J. Neurol. Sci., 17: 443-460. Lantos, P.L. (1974) An electron microscopic study of reacting astrocytes in gliomas induced by N-ethylN-nitrosourea in rats, Acta Neuropath. (Berl.), 30: 175-181. Lantos, P.L. and G.J. Pilkington (1979) The development of experimental brain tumours - - A sequential light and electron microscope study of the subependymal plate, Part 1 (Early lesions (abnormal cell clusters)), Acta Neuropath. (Berl.), 45: 167-175. Lucas, C.V., K.G. Bensch and L.F. Eng (1980) In vitro polymerization of the glial fibrillary acidic protein extracted from multiple sclerosis brain, Neurochem. Res., 5: 247-255. M andybur, T. I. and M. M. Alvira (1982) Ultrastructural findings in so-called ependymal rat tumors induced by transplacental administration of ethylnitrosourea (ENU), Acta Neuropath. (Berl.); 57: 51-58. Pilkington, G.J. and P.L. Lantos (1979) The development of experimental brain tumours - - A sequential light and electron microscope study of the subependymal plate, Part 2 (Microtumours), Acta Neuropath. (Berl.), 45: 177-185. Rueger, D. C., J. S. Huston, D. Dahl and A. Bignami (1979) Formation of 100 A filaments from purified glial fibrillary acidic protein in vitro, J. Mol. Biol., 135: 53-68.
355 Schiffer, D. and M.T. Giordana (1974) On the occurrence and significance of acid mucopolysaccharides in oligodendrogliomas experimentally induced in the rat by nitrosourea derivatives. In: D. Schreiber and W. J~inisch (Eds.), Experimentelle Neuroonkologie, Wiss. Ber. Martin-Luther Universit~it, HalleWittenberg, pp. 101-108. Schiffer, D., A. Fabiani, E. Grossi-Paoletti and P. Paoletti (1970) Experimental brain tumours induced in rats by nitrosourea derivatives, Part 1 (Morphological aspects of methylnitrosourea tumours), J. Neurol. Sci., 11: 559-572. Schiffer, D., A. Fabiani, E. Grossi-Paoletti and P. Paoletti (1971) Tumori cerebrali sperimentali del ratto da derivati della nitrosourea - - Considerazioni patogenetiche sui gliomi misti e polimorfi, Tumori, 57: 333-341. Schiffer, D., M.T. Giordana, S. Pezzotta, C. Lechner and P. Paoletti (1978) Cerebral tumors induced by transplacental ENU - - Study of the different tumoral stages, particularly of early proliferations, Acta Neuropath. (Berl.), 41: 27-31. Schiffer, D., M.T. Giordana, A. Mauro, G. Racagni, F. Bruno, S. Pezzotta and P. Paoletti (1980) Experimental brain tumors by transplacental ENU - - Multifactorial study of the latency period, Acta Neuropath. (Berl.), 49:117-122. Schiffer, D., M.T. Giordana, A. Mauro and A. Migheli (1983) Glial fibrillary acidic protein (GFAP) in human cerebral tumors - - An immunohistochemical study, Tumori, 69: 95-104. Sternberger, L.A., P.H. Hardy, Jr., F.F. Cuculis and H.G. Meyer (1970) The unlabeled antibody enzyme method of immunohistochemistry ~ Preparation and properties of soluble antigen-antibody complex (horse-radish peroxidase-antiperoxidase) and its use in identification of spirochetes, J. Histochem. Cytochem., 18: 315-333. Uyeda, C.T., Eng, L.E. and A. Bignami (1972) Immunological study of the glial fibrillary acidic protein, Brain Res., 37: 81-89. Van der Meulen, J.D.M., H.J. Houthoff and E.J. Ebels (1978) Glial fibrillary acidic protein in human gliomas, Neuropath. Appl. Neurobiol., 4: 177-190. Velasco, M. E., D. Dahl, U. Roessmann and P. L. Gambetti (1980) Immunohistochemical localization ofglial fibrillary acidic protein in human glial neoplasms, Cancer, 45: 484-494. Wechsler, W., P. Kleihues, S. Matsumoto, K. J. Ztilch, S. Ivankovic, R. Preussmann and H. Druckrey (1969) Pathology of experimental neurogenic tumors chemically induced during prenatal and postnatal life, Ann. N.Y. Acad. Sci., 159: 360-408.