GLIA 3:502-509 (1990)

Reactive Astrogliosis After Basic Fibroblast Growth Factor(bFGF) Injection in Injured Neonatal Rat-Brain FRANCOISE ECLANCHER,l FREDERIC PERRAUD? JAN FALTIN? GERARD LABOURDETTE,Z AND MONIQUE SENSENBRENNER' 'Departement de Neurophysiolo ie et de Biologie des Comportements and 'Departement de hreurochimie du Centre de Neurochimie du 8N.R.S. and INSERM U44, 67084 Strasbourg, France and 3.1nstitute of Physiology, Department of Molecular Neurobiology, 14220 Prague, Czechoslovakia

KEY WORDS

Brain injury, Astrocytes, Glial fibrillary acidic protein

ABSTRACT

Reactive gliosis was revealed by immunocytochemistry using antibodies against the glial fibrillary acidic protein (GFAP) after a stab or an electrolytic lesion administered to the cerebral cortex, corpus callosum, striatum, or hippocampus of a 6-day-old rat. The intensity of the gliosis was about the same in the various structures injured and did not change with the delay of 3,7, or 20 days between the injury and the sacrifice of the animals. When basic fibroblast growth factor (bFGF) was injected in the lesion locus just after the lesion was performed, it resulted (as soon as 3 days after injury) in a strong astrogliosis that was enhanced after a delay of 7 days, the astrocytes in the lesion area exhibiting enlarged cell processes and intense GFAP-positiveimmunoreactivity. After a delay of 20 days, the astrocytes were not dispersed any more but packed in three or four layers along the borders of the lesion, thus reducing its extension. This suggests a possible role for bFGF in promoting scar formation following brain injury.

INTRODUCTION Brain injury is followed by intensive hyperplasia of astrocytes that acquire a fibrous appearance. The thick, long processes contain numerous glial intermediate filaments, as revealed by immunocytochemistry using an antiserum to glial fibrillary acidic protein (GFAP),a specific marker for mature and reactive astrocytes. These reactive astrocytes lead to the formation of a structurally organized post-traumatic scar, at least in the adult animal. In the neonatal rat, some authors did not observe any reaction of the astrocytes following brain injury (Berry et al., 1983; Osterberg and Wattenberg, 1963; Sumi and Hager, 1968). On the contrary, other authors found after cerebral lesions in the newborn rat, local proliferation of reactive astrocytes but only after long delays of 1 to 3 months (Barrett et al., 1984; Bignami and Dahl, 1976; Janeczko, 1988). For Berry et al. (1983) normal mature scarring with astrocyte end-feet alignment over a glia limitans does not occur in wounds before 8-12 days post-partum. They suppose that the acquisition of the mature response to 0 1 9 9 0 Wiley-Liss, Inc.

injury from 8-12 days post-partum may be correlated with the presence of increasing titers of a fibroblast growth factor (FGF), derived from autolytic digestion of injured brain tissue. Acidic and basic FGFs (aFGF; bFGF) have been isolated from the brain (Gospodarowicz et al., 1984),localized in neurons (Pettmann et al.,,1986), and found to be released after brain injury (Finklestein et al., 1988). In vitro studies have shown that both FGFs stimulate the proliferation and maturation of astroblasts (Pettmann et al., 1985; Sensenbrenner et al., 1985; Weibel et al., 1985)as well as ofneuroblasts (Gensburger et al., 1987). They also promote the survival of cortical and hippocampal neurons, and enhance the outgrowth of their neurites (Morrison et al., 1986; Walicke et al., 1986). Basic FGF, but not aFGF, affects survival of ciliary ganglionic neurons and enhances choline acetyltransReceived October 16,1989; accepted May 30,1990. Address reprint requests to Dr. Francoise Eclancher, DNBC, Centre de Neurochimie du C.N.R.S., 5 rue Blaise Pascal, 67084 13trasbourg, France.

503

REACTIVE GLIOSIS AFTER bFGF INJECTION

ferase (ChAT) activity in these neurons (Unsicker et al., 1987). In vivo studies performed more recently have emphasized the fact that bFGF promotes survival of brain neurons subjected to axonal transection (Anderson et al., 1988; Sievers et al., 1987) and also attenuates the decrease of hippocampal ChAT activity induced by partial fimbria transection (Barotte et al., 1989). The present study was designed t o elucidate whether a lesion performed in the neonatal rat brain would result in a reactive gliosis. The increase in the total number of reactive astrocytes, by using an antiserum to GFAP, was analyzed after stab wound or electrolytic lesion administered to various structures such as cerebral cortex, hippocampus, striatum, or corpus callosum. We wanted also to know whether bFGF injection after an early electrolytic lesion would result in an increase of the number of reactive astrocytes. The delays in survival of the rats after brain injury were 3,7, or 20 days. MATERIALS AND METHODS Seventy Wistar rat pups born in our laboratory were used in this study. On their sixth postnatal day, they were anesthetized by an i.p. injection of Nembutal(22 mgkg) and loosely fixed in a David-Kopf stereotaxic apparatus. The rats were divided in groups of three, which received the same cerebral intervention (stab, lesion, injection, or injection after lesion) performed in various structures: cortex, striatum, hippocampus, or corpus callosum). The various lesions were made by lowering an epoxylite-coated stainless steel electrode through holes drilled in the cartilaginous skull. For the electrolytic lesions, a 2 mA DC current was passed through the electrode for 10 s. Two microliters of bFGF (about 2 ng) or buffer alone (20 mM Tris, 2 M NaC1, pH 7.2) were injected in some of the structures after the lesion at controlled low speed. Basic FGF was prepared from bovine brain by using heparin-Sepharose affinity chromatography, as previously described (Pettmann et al., 1985). After completion of the surgery, the pups were returned to the dam and allowed to remain with it until they were sacrificed, either 3, 7, or 20 days later. The animals were then anesthetized with a lethal dose of Nembutal and perfused intracardially with a saline solution, followed by pure methanol at 4°C. The brains were removed, stored for 1 day in methanol a t 4"C, embedded in paraffin, and sectioned at 10 pm. After removal of the paraffin, coronal sections were incubated with rabbit anti-GFAP antiserum (Dakopatts, Denmark, diluted 1:400) for 2 h at room temperature. After washing with phosphate-buffered saline (PBS), they were incubated with 1%sheep anti-rabbit a-globulin conjugated with peroxidase (Institut Pasteur Production, France) for 1 h. After three washes with PBS, a mixture containing 0.018% (v/v) 4-chloro-l-naphthol and 0.002% (v/v)hydrogen peroxide in PBS was applied

for 20 min. The sections were washed with PBS and mounted with Kaiser's glycerol-gelatin (Merck). Controls were performed by using serum from a non-immunized rabbit at a 1:400 dilution. Sections were observed under light microscopy with a 1 0 objective, ~ coupled with a video-camera and a computer. Reactive cells were counted by means of an image analysis system (TAS, Leitz) in several fields of 25 pm2 surface. Results are given in total number (n) of cells in all measured fields that covered a surface of 0.4 mm2, including the strongest reaction site. RESULTS In the normal intact brain of a young rat, at 9 days of age, there were only a few GFAP-positive astrocytes around the ventricules, in the white matter such as the corpus callosum (Fig. 11,and also in the external layer of the cerebral cortex. In the other layers of the cortex as well as in the other cerebral structures no GFAP-positive astrocytes were observable. At the age of 13 days, the number of fibrous GFAP-positive astrocytes as well as their distribution in the brain was about the same as at 9 days. At the age of 26 days, the GFAP-positive astrocytes were localized in the same areas as those described at the previous ages but were more numerous, and also appeared in the hippocampus. They presented the same stellate morphology as well as the same size compared to the younger ages. When the 6-day-old rats received a stab in their brain, GFAP-positive astrocytes were already present 3 days later, in the region of the stab wound (striatum, hippocampus, or cortex) (Fig. 2). The numbers of astrocytes counted in these two latter areas were 153.3 ? 8.7 and 185.6 ? 13.2 per field, respectively, and were not significantly different (P > 0.05 by t test). The surface covered by this gliosis corresponded to about 0.4 mm2. An electrolytic lesion made in these same cerebral areas resulted in the appearance of reactive GFAPpositive astrocytes 3 days later. The cytoplasm of the cell bodies and of their thick processes was intensely stained. ANOVA analysis of results obtained for the cortex revealed a lesion effect (F1,36 = 4,950.4, P < 0.001) as well as a delay effect (F2,36 = 76.8, P < 0.001). The astrocytes were dispersed throughout the lesion area, which generally covered 1mm2 (Fig. 3a, in the cortex and Fig. 3b, in the striatum). Numbers in the cortex are reported in Table 1. The comparison of results revealed that for the other cerebral areas changes obtained for the number of astrocytes following an electrolytic lesion were about the same as for the cortex. When the rats were killed 7 days after stabbing or after having received an electrolytic lesion, there was no increase in the number of reactive astrocytes as revealed by the quantitative evaluations reported in Table 1(P > 0.05between the number of astrocytes measured after a delay of 3 and 7 days, respectively, following an

504

ECLANCHER ET AL.

Fig. 1. Coronal section of a brain from a 9-day-old control rat. GFAP immunostaining showing the normal occurrence of fibrous astrocytes in the corpus callosum. Bar = 100 pm. x 100.

Fig. 2. Fibrous astrocytes revealed by GFAP immunostaining 3 days after a stab made in the cortex of a 6-day-old rat. Bar = 100 pm. x 100.

electrolytic cortical lesion). After 20 days of delay, the astrocytes were still dispersed throughout the wound area and their number remained unchanged (Table 1, P > 0.05) When bFGF was injected into the cortex, it induced the appearance 3 days later of a higher number (n = 328) of reactive astrocytes localized in the injection site in comparison to the number of astrocytes obtained

either after a single stab wound (n = 185.6) (P < 0.01) or after an injection of Tris in the cortex (n = 144.7) (P < 0.01) (Table 1).The area cowered by this gliosis was identical (0.4mm2)to that measured after a stab wound. The injection of bFGF in the other cerebral structures such as corpus callosum (Fig. 4a) resulted after the same delay of 3 days in the same reactive gliosis in terms of extension and morphology of the GFAP-posi-

505

REACTIVE GLIOSIS AFTER bFGF INJECTION

Fig. 3. Many fibrous astrocytes revealed by GFAP immunostaining, 3 days after an electrolytic lesion made in the cortex (a) or in the striaturn (b) of a 6-day-old rat. Bar = 100 pm. x 100.

TABLE 1. Number of fibrous astrocytes in the rat cerebral cortex (layers II-VI) counted in a surface of 800 X 500 pm, which included the strongest reaction site (mean i SD of three rats)

astrocytes was still increased although the gliosis area was not enhanced and was limited to the injection site (Fig. 4b, in the corpus callosum); the number of reactive astrocytes was also significantly higher than the number of astrocytes counted after a delay of 3 days, in the cortex (P < 0.05) (Table 1).This increased density of the astrocytes was also observed in comparison with values obtained after the same delay following a Tris injection (P < 0.01) (Table 1). After a delay of 20 days, the number of fibrous astrocytes following the bFGF injection in the cortex was significantly reduced in comparison to that obtained after 7 days ( P < 0.05). The extension of the gliosis reaction was reduced to 700 x 300 pm and the size of the astrocytes was also reduced by half (diameter of the cell body was about 5 pm), as shown in Fig. 4c. The microinjection of bFGF just after an electrolytic lesion performed in the cortex resulted in the appearance of many reactive astrocytes with thick processes localized in the lesion site 3 days later (Fig. 5a). The number was significantly higher than the value obtained after electrolytic lesion alone (P < 0.01) (Table 1).The number of reactive astrocytes was still increased (significantly,P < 0.05) (Table 1)after a delay of 7 days (Fig. 5b). This intensely reactive gliosis was, however, restricted to the electrolytic lesion site in which the bFGF was injected and covered the same surface (about 1mm2)as in the case of the electrolytic lesion alone. It never invaded the neighboring structures nor the contralateral hemisphere and the density of the astrocytes was rather homogenousin the gliotic area. After 20 days of survival, the number of reactive astrocytes was not significantly different from that counted after a delay of 7 days (P < 0.05) (Table 1).The reactive astrocytes were no longer dispersed throughout the electrolytic lesion area, but were confined to three or four layers on both sides of the lesion. Their processes were thinner than those of astrocytes visible after a delay of 7 days and the star shape of these astrocytes was not as obvious, as shown in the striatum (Fig. 512). DISCUSSION

In the brain of the young rat (9 days old) we have been able to visualize astrocytes by immunocytochemical Rats staining, using an antiserum against GFAP that is Control 0.0 f 0.0 0.0 + 0.0 15.7 i 4.0 considered to be exclusively localized in astrocytes (BigCortical lesion 756.7 + 19.4 760.0 * 10.0 766.3 f 20.3 nami and Dahl, 1974; Schiffer et al., 1986). The GFAPTris injection 144.7 + 7.6 185.0 i 5.0 158.3 f 12.6 bFGF injection 328.0 i 19.3 551.0 i 11.5 273.0 + 8.2 positive astrocytes displayed the typical stellate apCortical 870.0 * 18.0 880.0 + 10.0 863.3 + 32.1 pearance of fibrous astroglia, as also mentioned by lesion + Tris Cortical 1,078.0 f 128.2 1,961.3 + 58.8 1,915.7 f 133.2 McDermott and Lantos (1989). Essentially localized in lesion + bFGF the white matter, and around the ventricules, they were also found at the cortical surface. This continuous external glial membrane (“theglia limitans”)was positive for tive astrocytes as that obtained in the cortex. Indeed, GFAP and was also reported by Bignami and Dahl the gliosis area was about the same (0.4 mm2) and the (1974).With increasing ages (from 9 to 26 days), GFAPastrocytes had the same fibrous aspect; their cell body positive cells became more numerous in the corpus diameter was about 10 pm. After a delay of 7 days callosum and at 26 days also appeared in the hippocamfollowing the bFGF injection, the number of reactive pus. Vaughn and Peters (1967) have shown that by day PostoDerative delavs before sacrifice (davs) 3 7 20

506

ECLANCHER ET AL.

Fig. 4. Intense GFAP immunoreactivity following bFGF injection in the corpus callosum of a 6-day-old rat after 3 (a) and 7 days (b). Reduction of the gliosis reaction in the cortex after 20 days of survival (c). Bar = 100 wm. ~ 1 0 0 .

REACTIVE GLIOSIS AFTER bFGF INJECTION

Fig. 5. Reactive astrocytes revealed by GFAP immunostaining, following bFGF injection just after an electrolytic lesion performed in the cortex of a 6-day-old rat, after a delay o f 3 (a) and 7 (b) days and in the striatum after 20 days (c). x 100. Note the progressive increase in the number of GFAP-positive astrocytes from 3 to 7 following bFGF addition in the lesion site and the enlarged positively stained astrocytic processes. Bar = 100 km.

507

20 in the rat, the astrocytes appear to be mature and contain many glial intermediate filaments but few microtubules. Bjorklund et al. (1985) have suggested that in developing animals gliofilaments are growing within an essentially already existing cytoplasm and thus gradually larger areas of the cells become visible using GFAP immunohistochemistry. The astrogliosis we obtained as early as 3 days after stabbing or after an electrolytic lesion made in the 6-day-old rat brain was about the same after postoperative delays of 7 or 20 days. Moore et al. (1987)have also observed an astrocytic response in the fetal as well as neonatal rat brain after cold lesions. The amount of GFAP expressed varied according to the gestational age at which the damage was inflicted and the postnatal maturation of the brain, Other authors have described a moderate gliosis at the lesion site of neonatal rats by the first week that intensified somewhat after 12 (Moore et al., 1987),20(Politis and Houle, 1985),30 (Bignami and Dahl, 1976),and 60 days followingthe lesion (Barrett et al., 1984). For Bignami and Dahl (1976) a delayed astroglial response to injury is a common feature of the developing brain. On the other hand, Berry et al. (1983) reported that the small number of scattered astrocytes at the lesion edges, observed 4 days after lesioning rats at day 5, disappeared with time. It was only when the lesion was made later than day 8 after birth that a typical scar could be obtained. The astrogliosis we observed in the brain of the 9-day-old rat (which had been injured 3 days earlier) is in contradiction to other works reporting the absence of reactive astrogliosis in injured neonatal rats (Berry and Henry, 1975; Mathewson and Berry, 1985; Osterberg and Wattenberg, 1963; Sumi and Hager, 1968). These authors did not find the typical hypertrophy and stellate appearance of astrocytes after lesions produced in the immature rat brain prior to 8,10, or 12 days of age. It is quite likely that these negative results could be ascribed to the tissue fixation method. Indeed, in previous experiments in which we used paraformaldehyde instead of methanol as the fixative, we could not obtain evidence of GFAP-positive cells in brains of neonatal rats, either before or after injury. We found that the extent of the astroglial reaction and the number of reactive astrocytes were the same no matter which structures were damaged (white matter or gray matter). For the striatum, we noticed that the gliosis area that appeared after the lesion contained reactive astrocytes with rather long GFAP-positiveprocesses, which were sometimes intermingled. The fact that the astrogliosis remained localized in the vicinity of the lesion site and never invaded the hemisphere contralateral to the lesion is in agreement with the reports of Mathewson and Berry (1985) and Barrett et al. (1984). In contrast, Bignami and Dahl (1976) reported that the astrocytic reaction following newborn rat brain stabbing became remarkably extensiveafter 1 month, involving the non-stabbed hemisphere as well. In regard to the intensity of the astrocytic response that occurred after damage performed in white o r gray mat-

508

ECLANCHER ET AL.

ter, Mathewson and Berry (1985) and Miyake et al. (1988) reported a dependence of this response on location of damage (cerebral cortex, corpus callosum, or deep structures of the cerebral hemisphere). Posttraumatic astrocyte hyperplasia was thought to be brought about by amitotic division (Lapham, 1962). Later it was generally agreed that the increase in the number of reactive astrocytes that appeared after a stab or an electrolytic lesion performed in the brain is due to proliferation of astrocytes as detected by radioactive labeling after tritiated thymidine incorporation in the adult injured rat (Adrian et al., 1978; Azmitia and Whitaker, 1983; Cavanagh, 1970; Fujita et al., 1975; Latov et al., 1979; Ludwin, 1985; Murray and Walker, 1973; Persson, 1976; Privat, 1982; Skoff, 1975) as well as in the rat injured at birth (Janeczko, 1988). Recently, Miyake et al. (1988) and Takamiya et al. (1988) noticed that in the 2nd to 6th layers of the cortex, proliferation of reactive astrocytes was not the major reason for the marked increase in their number. After injecting [3H]thymidine successively for 6 days after stabbing, only 17% of GFAP-positive astrocytes of the cortical layers were labeled (Miyake et al., 1988).More recently, Miyake et al. (1989)combined i3H1thymidineincorporation with immunohistochemistry for 5-100 protein and concluded that reactive proliferation of astrocytes is a minor phenomenon in gliosis of injured cerebral cortex, in contrast with their remarkable reactive hypertrophy. Another recent study (Kreutzberg et al., 1989) gave evidence that, following axotomy of facial motoneurons in the rat, GFAP-immunoreactive astrocytes appeared that very likely correspond to a non-mitotic transformation rather than a proliferation of astrocyte precursors or of GFAP-positive fibrous astrocytes. The mitotic stimulus elaborated in response to axotomy appears to be selective in affecting microglia but not astrocytes, as also reported by Kitamura (1980). From our observations, since we have not performed radioactive labeling experiments, we cannot affirm whether the gliosis observed within the region of injury resulted from the proliferation of the astroglial cell population, from the transformation of macroglial cells into reactive astroglia, or from both cell proliferation and transformation. The astrogliosis we observed as early as 3 days after a bFGF microinjection made in the 6-day-old rat brain was enhanced after a post-operative delay of 7 days. After 20 days of survival, the reactive astrocytes were dispersed throughout the injection site. When bFGF was injected just after the electrolytic lesion in the same site, it resulted in a strong astrogliosis after 3 days, which increased after 7 days. After 20 days of survival, however, the fibrous astrocytes were no longer dispersed but were packed along the borders of the lesion, thus reducing its extension. At this time there was no apparent strong difference in the astrogliosis observed among the various lesioned structures after bFGF microinjection. As a general rule, the extent of the gliosis area was dependent on the size of the area destroyed by the electrolytic lesion. Astrocytic density was rather homogenous in the whole injury site.

The formation of a scar by ristroglial reaction following injury has generally been regarded as a major obstacle to neurite regeneration. More recently, however, the idea of the inhibitory influence of gliosis on central nervous system regeneration has been questioned. The astrogliosis that occurred after an early brain lesion was enhanced when bFGF was injected after this lesion, as observed in the present study. The consequence was the formation of a scar that could probably be beneficial for wound repair and neural sprouting in the damaged brain. The injection of exogenous bFGF in the electrolytic lesion site may activate the processes in the glial cicatrix involved with mesodermal elements such as hematogenous cells and fibroblasts producing collagen (Davidson et al., 1985). Schwartz et al. (1987) have reported that after injury both the neurons and their surrounding non-neuronal cells are activated. The lack of regeneration in mammals may stem from the inability of the glial cells to achieve a reactive state and thereby t o acquire the appropriate properties at the right time. Thus, the activation of astrocytes by bFGF that we observed may contribute toward improving the survival of neurons reported by Anderson et al. (1988) and by our group (Barotte et al., 1989) after damage of the septohippocampal system followed by bFGF injections in the adult animal. In the present study we found that the microinjection of bFGF in the lesion site induced (sometimes after 7 days) the appearance of entangled long processes that filled the wound. This result could be correlated with those of Kreutzberg's group. Indeed, Graeber and Kreutzberg (1988) observed that the astrocytic processes take over the perineurond positions of the microglia at later stages. About 3 weeks after facial nerve axotomy, the reactive astrocytes begin to form thin, lamellar processes covering all neuronal surfaces (Kreutzberg et al., 1989). Finally, bFGF plays a major role in the enhanced expression of GFAP in astrocytes known to be involved in transport of water and many electrolytes and metabolites within the brain (De Robertis and Gerschenfeld, 1961) and thus contributes toward improving the formation of a growth-supportive environment for neurons.

ACKNOWLEDGMENTS The present study was supported by a grant from the Fondation pour la Recherche Mddicale Frangaise. The skillful technical assistance of :Mrs. M.J. Angst was much appreciated.

REFERENCES Adrian, E.K. Jr., Williams, M.G., and George, F.C. (1978) Fine structure of reactive cells in injured nervous tissue labeled with 3HThymidine injected before injury. J. Conip. Neurol., 180:815-840. Anderson, K.J., Dam, D., Lee, S., and Cotman, C.W. (1988) Basic

REACTIVE GLIOSIS AFTER bFGF INJECTION fibroblast growth factor prevents death of lesioned cholinergic neurons in vivo. Nature, 332:360-361. Azmitia, E.C. and Whitaker, P.M. (1983) Formation of a glia scar following microinjection of fetal neurons into the hip ocampus or midbrain of adult rat: An immunocytochemical stufy. Neurosci. Lett., 38~145-150. Barotte, C., Eclancher, F., Ebel, A,, Labourdette, G., Sensenbrenner, M., and Will, B. (1989) Effects of basic fibroblast growth factor (bFGF) on choline acetyltransferase activity and astroglial reaction in adult rats after partial fimbria transection. Neurosci. Lett., 101:197-202. Barrett, C.P., Donati, E.J., and Guth, L. (1984) Differences between adult and neonatal rats in their astroglial response to spinal injury. Exp. Neurol., 84:374-385. Berry, M. and Henry, J . (1975) Response of the neonatal central nervous system to injury. Neuropathol. Appl. Neurobiol., 2:166. Berry, M., Maxwell, W.L., Logan, A., Mathewson, A,, McConnell, P., Ashhurst, D.E., and Thomas, G.H. (1983)Deposition of scar tissue in the central nervous system. Acta Neurochir. [Suppl.] (Wien), 32:31-53. Bignami, A. and Dahl, D. (1974)Astrocyte-specificprotein and neuroglial differentiation. An immunofluorescence study with antibodies to the glial fibrillary acidic protein. J . Comp. Neurol., 153:27-38. Bignami, A. and Dahl, D. (1976) The astroglial response to stabbing. Jmmunofluorescence studies with antibodies to astrocyte-specific protein (GFA)in mammalian and submammalian vertebrates. Neuropathol. Appl. Neurobiol., 2:99-110. Bjorklund, H., Eriksdotter-Nilsson, M., Dahl, D., Rose, G., Hoffer, B., and Olson, L. (1985)Image analysis of GFA-positive astrocytes from adolescence to senescence.Exp. Brain Res., 58:163-170. Cavanagh, J.B. (1970)The proliferation of astrocytes around a needle wound in the rat brain. J.Anat., 106:471-487. Cotman, C.W. and Nieto-Sampedro, M. (1985) Progress in facilitating the recovery of function after central nervous system trauma. Ann. N.Y. Acad. Sci., 457233-104. Davidson, J.M., Klagsbrun, M., Hill, K.E., Buckley, A,, Sullivan, R., Brewer, P.S., and Woodward, S.C. (1985) Accelerated wound repair cell proliferation, and collagen accumulation are produced by a cartilage-derived growth factor. J. Cell. Biol., 100:1219-1227. De Robertis, E. and Gerschenfeld, H.M. (1961) Submicroscopic morphology and function of glial cells. Int. Rev. Biol., 3:1-65. Finklestein, S.P., Apostolides, P.J., Caday, C.G., Prosser, J., Philips, M.F., and Klagsbrun, M. (1988) Increased basic fibroblast growth factor (bFGF) immunoreactivity at the site of focal brain wounds. Brain Res., 460:253-259. Fujita, S., Hattori, H., Hattori, T., and Kitamura, T. (1975) Studies on origin and morphologyof the so-calledmicroglia. IX.Nature and fate of reactive neuroglia in the damaged brain. Prog. Neur. Sci., 19:102-107. Gensburger, C., Labourdette, G., and Sensenbrenner, M. (1987)Brain basic fibroblast growth factor stimulates the proliferation of rat neuronal precursor cells in vitro. FEBS Lett., 287:l-5. Gospodarowicz, D., Cheng, J.,Liu, G., Baird, A,, and Bohlent, P. (1984) Isolation of brain fibroblast growth factor by heparin-Sepharose affinity chromatography: Identity with pituitary fibroblast growth factor. Proc. Natl. Acad. Sci. USA, 81:6963-6967. Graeber, M.B. and Kreutzberg, G.W. (1988)Delayed astrocyte reaction following facial nerve axotomy. J . Neurocytol., 17:209-220. Janeczko, K. (1988)The proliferation response of astrocytes to injury in neonatal rat brain. A combined immunocytochemicaland autoradiographic study. Brain Res., 456:280-285. Kitamura, T. (1980) Dynamic aspects of glial reactions in altered brains. Pathol. Res. Pract., 168:301-343. Kreutzberg, G.W., Graeber, M.B., and Streit, W.J. (1989)Neuron-glial relationship during regeneration of motorneurons. Metab. Brain Dis.,4231-85. Lapham, L.W. (1962)Cytologic and cytochemical studies of neuroglia. I. A study of the problem of amitosis in reactive protoplasmic astrocytes. A m . J . Pathol., 41:l-21. Latov, N., Nilaver, G., Zimmerman, E.A., Johnson, W.G., Silverman, A.J.,Defendini, R., and Cote, L. (1979)Fibrillary astrocytes proliferate in response t o brain injury. Deu. Biol., 72:381-384. Ludwin, S.K. (1985) Reaction of oligodendrocytes and astrocytes to trauma and implantation. A combined autoradiographic and immunohistochemical study. Lab. Invest., 52:20-30. Mathewson, A.J. and Berry, M. (1985) Observations on the astrocyte response to a cerebral stab wound in adult rats. Brain Res., 327:61-69.

509

McDermott, K.W.G., and Lantos, P.L. (1989) The distribution of glial fibrillary acidic protein and vimentin in postnatal marmoset (Callithrixjacchus) brain. Deu. Brain Res., 45~169-177. Miyake, T., Hattori, T., Fukuda,M., Kitamura, T., and Fujita, S. (1988) Quantitative studies on proliferative changes of reactive astrocytes in mouse cerebral cortex. Brain Res., 451:133-138. Miyake,T., Hattori, T., Fukuda, M., and Kitamura, T. (1989)Reactions of S-100-positiveglia after injury of mouse cerebral cortex. Brain Res., 489:3140. Moore, I.E., Bountempo,J.M., and Weller, R.O. (1987)Response offetal and neonatal rat brain to injury. Neuropathol. Appl. Neurobiol., 13:219-228. Morrison, R.S., Sharma, A,, de Vellis, J., and Bradshaw, R.A. (1986) Basic fibroblast growth factor supports the survival of cerebral cortical neurons in primary culture. Proc. Natl. Acad. Sci. USA, 83:7537-7541. Murray, H.M. and Walker, B.E. (1973)Com arative study of astrocytes and mononuclear leukocytes reacting to grain trauma in mice. Exp. Neurol., 41:290-302. Nathaniel, E.J.H. and Nathaniel, D.R. (1981) The reactive astrocyte. Adu. Cell. Neurobiol., 2:249-301. Osterberg, K.A. and Wattenberg, L.W. (1963)The age ofdependency of enzymes in reactive glia. Proc. Sac. Exp. Biol. Med., 113~145-147. Persson, L. (1976)Cellular reaction to small cerebral stab wound in rat frontal lobe. An ultrastructural study. VirchowsArch. (Cell Pathol.), 2212137. Pettmann, B., Weibel, M., Sensenbrenner, M., and Labourdette, G. (1985) Purification of two astroglial growth factors from bovine brain. FEBS Lett., 189:102-108. Pettmann, B., Labourdette, G., Weibel, M., and Sensenbrenner, M. (1986) The brain fibroblast growth factor FGF is localized in neurons. Neurosci. Lett., 68:175-180. Politis, M.J. and Houle, J.D. (1985)Effect of cytosine arabino-furanoside (AraC)on reactive gliosis in vivo. An immunohistochemicaland morphometric study. Brain Res., 328:291-300. Privat, A. (1982)Gliogenesis in the central nervous system of rodents. IIIrd ISDN Meeting, Patras, Greece. Schiffer, D., Giordana, M.T., Migheli, A., Giaccone, G., Pezzota, S.,and Mauro, A. (1986)Glial fibrillary acidic protein and vimentin in the experimental glial reaction ofthe rat brain. Brain Res., 374:llO-118. Schwartz, M., Harel, A,, Solomon, A,, Lavie, V., Savion, N., Stein-Izsak, C., Baronik, Y., Zak, N., Vogel, Z., Cohen, A., and Belkin, M. (1987) Molecular and cellular aspects of axon-glia interaction in CNS regeneration. J . Physiol. (Paris), 82:314-321. Sensenbrenner, M., Pettmann, B., Labourdette, G., and Weibel, M. (1985)Properties of a brain growth factor promoting proliferation and maturation of rat astroglial cells in culture. In: Hormones and Cell Regulation. J.E. Dumont, B. Hamprecht, and J. Nunez, eds. Inserm European Symposium,Vol. 9, pp. 345-360. Sievers, J., Hausmann, B., Unsicker, K., and Berry, M. (1987) Fibroblast growth factors promote the survival of adult rat retinal ganglion cells after transection of the optic nerve. Neurosci. Lett., 761157-162. Skoff, R.P. (1975) The fine structure of pulse labelled (3H-thymidine) cells in degenerating rat optic nerve. J . Comp. Neurol., 161595-612. Sumi, S.M. and Hager, H. (1968) Electron microscopic study of the reaction of the newborn rat brain to injury. Acta Neuropathol. (Ber1.1, 10:324-335. Takamiya,Y., Kohsaka, S.,Toya, S.,Otani, M., andTsukada, Y. (1988) Immunohistochemical studies on the proliferation of reactive astrocytes and the expression of cytoskeletal proteins following brain injury in rats. Deu. Brain Res., 38:201-210. Unsicker, K., Reichert-Preibsch, H., Schmidt, R., Pettmann, B., Labourdette, G., and Sensenbrenner, M. (1987) Astroglial and fibroblast growth factors have neurotrophic functions for cultured peripheral and central nervous system neurons. Proc. Natl. Acad. Sci. USA, 84:5459-5463. Vaughn, J.E. and Peters, A. (1967) Electron microsco y of the early postnatal development of fibrous astrocytes. A)m. J . Anat., 121~131-152. Walicke, P.A., Cowan, W.M., Ueno, N., Baird, A,, and Guillemin, R. (1986) Fibroblast growth factor promotes survival of dissociated hippocampal neurons and enhances neurite extension. Proc. Natl. Acad. Sci. USA, 83:3012-3016. Weibel, M., Pettmann, B., Labourdette, G., Miehe, M., Bock, E., and Sensenbrenner, M. (1985) Morphological and biochemical maturation of rat astroglial cells grown in a chemically defined medium: Influence of an astroglial growth factor. Deu. Neurosci., 3:617-630.