Microglial and astroglial reactions to anterograde axonal …histoserver.uab.cat/PDFs/Publicaciones/1994_Microglial... · 2007. 5. 29. · Exp Brain Res (1994) 98:245-260 9 Springer-Verlag

Exp Brain Res (1994) 98:245-260 �9 Springer-Verlag 1994

M. B. Jensen �9 B. Gonz/tlez B. Casteltano �9 J. Zimmer

Microgliai and astroglial reactions to anterograde axonal degeneration: a histochemical and immunocytochemical study of the adult rat fascia dentata after entorhinal perforant path lesions

Received: 5 July 1993 / Accepted: 1 October 1993

Abstract The reaction of microglial and a stroglial cells to anterograde axonal degeneration was studied in the fascia dentata of adult rats at various timepoints after removal of the entorhinaI perforant path projection. Microglial cells were identified by histochemical staining for nucleoside diphosphatase (NDPase) at light and electron microscopical levels. Astroglial cells were stained immunocytochemically for glial fibrillary acidic protein (GFAP). Activated astroglial cells and some mi- crogliaI cells also stained immunocytochemically for the intermediate filament protein vimentin. Phagocytotic activity was detected by histochemical staining for acid phosphatase. The postlesional connective reorganization of the cholinergic septohippocampal projection was monitored by histochemical staining for acetylcholinesterase. Twenty-four hours after entorhinal cortex ablation, microglial cells in the perforant path zones of the fascia dentata and the adjacent neuropil reacted by shortening and coarsening of processes and an increase in NDPase reactivity. These changes occurred prior to a noticeable increase in GFAP immunoreactivity and hypertrophy of astroglial cells (first evident on postlesional day 2) or sprouting of cholinergic septohippocampal fibres (first evident on day 3). There was evidence of an early, local proliferation of microglial cells in the denervated perforant path zones and migration into these zones of microglial cells from adjacent intact areas. The specific accumulation of strongly stained microglial cells within the denervated parts of

M. B. Jensen PharmaBiotec, Institute of Neurobiology, University of Aarhus, DK-8000 Aarhus

B. Gonzalez - B. Castellano Department of Cell Biology and Physiology, Autonomous University of Barcelona, E-08193 Bellaterra (Barcelona), Spain

M. B. Jensen (l~) �9 J. Zimmer PharmaBiotec, Department of Anatomy, University of Odense, DK-5000 Odense, Denmark

the dentate molecular layer persisted for at least 4 weeks, while the astroglial reaction subsided at 3 weeks. The results demonstrate an early activation of microglial cells by axonal degeneration, and indicate that these cells may play a pivotal, inductive role in the subsequent glial and neural events.

Key words Hippocampus �9 Synaptic plasticity Glial cells �9 Rat

Introduction

Brain injury elicits degenerative as well as reparative reactions, involving both neuronal and glial elements. Astroglial and microglial cells react both at the site of lesion and more distantly as response to lesion-induced anterograde axonal degeneration or retrograde neuronal changes (Kerns and Hinesman 1973ab; Lynch et al. 1975; Kreutzberg et al. 1989; Finsen et at. 1991; Poirer et al. 1991; Jorgensen et al. 1993). Besides the well-known retrograde neuronal reactions to axotomy (Duchen 1992), adult neurons are also capable of axonal regrowth, as for instance collateral axonal sprouting in- to adjacent denervated areas (Raisman t969; Zimmer 1973, 1974; Lynch and Cotman 1975) and regenerative axonal growth into grafts of peripheral nerves or fetal brain tissue (Tonder etal. 1989; Wictorin and Bj6rklund 1990; Zimmer et al. 1992). Regarding the role of glial cells in these reactive processes, several in vitro and in vivo experiments have shown that both microglial cells and astrocytes produce substances affect- ing the other cell type (Frei et al. 1986; Giulian et at. 1986, 1988ab, 1991; Lindsay 1986), just as there are ex- amples of structural interactions between microglial cells and neurons (Murabe and Sano 1982; Thanos 1991) and production of neurotrophic factors by astrocytes (Lindsay 1979; Lindsay etal. 1982; McCaffery et al. 1984; Rudge et al. 1985, 1992; Manthorpe et al. 1986; Kreutzberg et at. 1989; Fagan and Gage 1990; Spranger et al. 1990; Yoshida and Gage 1991).

246

In the present study we used a well-known experimental paradigm and lesion model to demonstrate se- quential, and hence possibly causally related, reactions of microglia and astroglial cells to anterograde axonal degeneration. By ablating the entorhinal cortex in adult rats we induced a distinct, dense, but distantly located anterograde degeneration of perforant path (PP) fibres in the molecular layer of the fascia dentata. At different postlesional survival times thereafter we monitored the reactions of microglial and astroglial cells and the lesion-induced collateral sprouting of the cholinergic septohippocampal projection by histochemical and immunocytochemical methods.

The observed temporal sequence of cellular reactions, with an early microglial activation preceding other observable changes, suggests a pivotal role of these cells.

Materials and methods

Adult Wistar rats (180-240 g) of both sexes were used. Aspiration lesions of the dorsal one half of the right entorhinal

cortex are known to induce anterograde terminal degeneration of the PP projection to the outer parts of the molecular layer of the fascia dentata at mid-posterior to septal levels. For lesioning, the rats (n=40) were anaesthetised with pentobarbital and a 1.5- to 2-mm-wide burr hole made in the skull in front of and lateral to lambda. After opening of the dura, the dorsal occipital cortex and the underlying dorso-posterior parts of the entorhinal cortex were removed by aspiration. Spongostan was placed in the lesion cavity before closing and suturing of the skin. At postlesional survival times of 12 h, 24 h and 2, 3, 7, 14, 21 and 28 days, the rats were again anaesthetised, perfused and their brains removed and processed histologically as described below. At least four rats were available for each survival time. For control, adult rats (n = 8) without entorhinal lesions were perfused and processed similarly. Structures on the unoperated side of the experimental animals not receiving projections from the lesioned entorhinal cortex also served as a control.

For histological processing and staining, the rats were killed with an overdose of pentobarbital and perfused transcardially for 7 rain with 4% paraformaldehyde and 0.5% glutaraldehyde in a 0.1 M cacodylate buffer, pH 7.4, after which the brains were removed in toto and placed in the fixative for additional 2 h. After a rinse in 0.1 M cacodylate buffer, pH 7.4, the brains were cut in 30-gm-thick sections in the frontal or the horizontal plane on an Oxford Vibratome. The sections were collected as three parallel series in 0.1 M cacodylate buffer, pH 7.4, for immediate histochemical and immunohistochemical processing.

For the demonstration of microglial cells we used a histochemical reaction for nucleoside diphosphatase (NDPase), while astroglial cells were visualised by immunocytochemical staining for glial fibrillary acidic protein (GFAP). For activated astroglial cells, and possibly also activated microglial cells (Schnitzer et al. 1981, Schiffer et al. 1986; Graeber et al. 1988a), we used an immunocytochemical staining for the intermediate filament protein vimentin.

For histochemical demonstration of nucleoside diphosphatase (NDPase), Vibratome sections were incubated in the Novikoff and Goldfisher medium (Novikoff and Goldfisher 1961) at 37 ~ C for 1 h using as substrate either thiamine pyrophosphate (TPP, Sig- ma) or inosine 5'diphosphate. For control, sections were incubated in the same medium without substrate. After incubation the sections were placed in 2% ammonium sulphide for 2 min to visualise and stabilise the reaction product, and then mounted on slides in Dammar resin.

For electron microscopy of NDPase activity, stained Vi- bratome sections were postfixed in 1% OsO4 in cacodylate buffer, pH 7.4, for 1 h at room temperature, dehydrated in graded con- centrations of ethanol and embedded in Epon. Ultrathin sections were stained with uranyl citrate and lead nitrate before electron microscopic analysis.

For glial fibrillary acidic protein (GFAP) immunostaining, Vi- bratome sections were reacted according to the unlabelled antibody peroxidase-antiperoxidase (PAP) method of Sternberger (1986), as previously employed (Zimmer and Sunde 1984). After a wash in Tris buffer 0.05 M, pH 7,4, with 1% Triton X-100 the sections were incubated in normal swine serum for 30 min, and incubated with the primary antibody overnight at 4 ~ C. The GFAP antibody was obtained from DAKO (Copenhagen, Den- mark, Z 334) and used in a 1:2400 dilution. After wash, the sections were incubated in swine anti-rabbit immunoglobulin (DAKO, Z 196, dilution 1:30) for 30 rain at room temperature, washed again, and incubated with a PAP complex (DAKO, Z 113, dilution 1:75) for 30 min. For visualisation of the peroxidase we used diaminobenzidine (DAB) (Bie and Berntsen, D 5637) as the chromogen (50 mg in 100 ml, 0.05 M Tris buffer, pH 7.4) with 0.033 ml H202 added immediately before use. The immunocytochemically stained sections were finally dehydrated in ethanol, mounted and coverslipped in Dammar resin.

For immunocytochemical demonstration of vimentin, Vi- bratome slices were washed in Tris buffer 0.05 M, pH 7.4, with 1% Triton X-100, and then for 30 rain in Tris buffer 0.05 M, pH 7.4, with 10% fetal calf serum. The sections were incubated with the primary antibody directed against vimentin (DAKO, Copen- hagen, M 725, dilution 1 : 1000). After rinsing in buffer, the sections were incubated in biotinylated anti-mouse immunoglobulins (Amersham, UK, RPN 1001, dilution 1:200) for 60 min, washed twice and incubated with Avidin-peroxidase complex (Sigma, USA, A 3151, diluted 1:70) for 60 min. From then the immunocytochemical staining was performed as described above for GFAP.

For demonstration of the distribution and reorganization of the Acetylcholinesterase (AChE)-positive, cholinergic septohippocampal projection at the different postlesional survival times some Vibratome sections were collected from each rat for AChE staining. After mounting on gelatine-coated glass slides, the sections were stained according to the thiocholine method used by Geneser-Jensen and Blackstad (1971).

For histochemical demonstration of acid phosphatate (Ac- Pase), the method of Soufleris et al. (1983) was adapted. The incubation medium was prepared by dissolving 10 mg of naphtol AS- BI phosphate in 1 ml dimethylformamide, adding 16 ml of 0.1 M Trizma buffer, pH 5.0, and 1.6 ml of hexasotized pararosaniline, after which the pH of the medium was adjusted to 5.0 with 1.0 N NaOH. The hexasotised pararosaniline was prepared just before use by mixing equal volumes of a freshly prepared 4% solution of sodium nitrate in distilled water with a 4% pararosaniline acid solution. The pararosaniline acid solution was prepared by dissolving 1 g of pararosaniline in 20 ml of distilled water, adding 5 ml of 10 N hydrochloride and then filtering the solution. Vi- bratome sections were incubated for 15 min at 37 ~ C and then rinsed in 0.1 M cacodylate buffer, pH 7.4.

For histochemical demonstration of 5'-nucleotidase, the incubation medium was the same as for the demonstration of NDPase, but without MnCI2, and with 5'-adenosine-monophosphate as substrate instead of 5'-inosine-diphosphate. The sections were incubated for 30 min. at 37 ~ C before treatment with ammonium sulphide and mounting.

Results

Normal hippocampus and fascia dentata

Before dealing with the lesion-induced changes in the fascia dentata, a brief description of the appearance and

247

Fig. 1 NDPase staining of microglial cells in FD, CA3 and CA1 of the normal hippocampus (A, B) and 24 h (C, D) and 3 days (E, F) after perforant path (PP) lesion. NDPase reactive microglial cells are almost evenly distributed within the normal hippocampus (A) with additional staining of endothelial cells (B). Twenty- four hours after the lesion, the PP zones in the molecular layers of

FD, CA3 and CA1 stand out (C, D). The staining density of the processes from the individual cells increases, and each process appears coarser than normal (D). After 3 days these changes are more pronounced (E, F), FD Fascia dentata, g granule cell layer, lrn stratum lacunosum moleculare of CA1, m dentate molecular layer. Bars: A, C, E 750 gm; B, D, F 100 gm

248

distribution of the NDPase-positive microglial cells in the hippocampus and fascia dentata of the normal adult rat is given. This is followed by a short description of the normal astroglial staining for GFAP and the normal appearance of the staining for vimentin.

NDPase staining of microglial cells

Vibratome sections showed a dense TPPase and IDPase staining of cells with all the morphological characteris- tics of the classical del Rio Hortega microglial cell. In addition, there was a vascular staining of endothelial cells (Figs. lAB, 4A). There were no significant differences in the histochemical staining patterns between sections stained with the TPPase and the IDPase method. With some exceptions, NDPase positive microglial cells were almost evenly distributed within the hippocampus and fascia dentata without obvious differences along the septotemporal axis. At a given level, the density of microglial cells within the hippocampal pyra- midal and dentate granule cell layers was lower than in the adjacent neuropil (Fig. 1A). In the fascia dentata there was an accumulation of stained cells along the deep, hilar border of the granule cell layer (Fig. 1A), and along vessels in the obliterated part of the hippocampal fissure (Fig. 1B). The stained microglial cells were nowhere specifically related to blood vessels. The individual cells had small polymorphic cell bodies, which in sections counterstained with toluidine blue contained a darkly stained, elongated nucleus. The enzymatic reaction product of TPPase or IDPase could be located, even at the light-microscopic level, to the outer cell membrane. This was particularly evident in the larger proximal processes (arrow, Fig. 4A). Each cell usually had from three to five such processes, which had smaller processes branching off again in all directions. Each cell with all recognisable processes covered a sphere of ap- proximately 70 m in diameter.

Electron microscopy of TPPase or IDPase stained sections confirmed the presence of the enzyme reaction product in cells with small, slightly elongated dark nuclei with a dense chromatin pattern. The reaction product was localised to the outer cell membrane of both the cell body and the larger and smaller processes that arose from the cell bodies or from other processes (Fig. 4B). The staining of endothelial cells and what we, according to the terminology of Graeber and Streit (1990), believe to be perivascular cells (see Discussion) was also confirmed at the ultrastructural level (Fig. 4C).

With the present staining method employed on Vi- bratome sections, and care being taken that the sections did not dry at any time, we usually did not observe staining of the Golgi apparatus of neurons or glial cells (Castellano et al. 1990).

GFAP staining of astroglial cells

The distribution of GFAP-stained astroglia in the normal hippocampus and fascia dentata has been thor-

Fig. 2 NDPase staining of microglial cells 7 (A) and 14 (B) days after a PP lesion. The specific accumulation of strongly stained microglial cells in the zones with axonal degeneration persists for more than 3 weeks. Bar 750 gm

oughly described (Zimmer and Sunde 1984; Schmidt- Kastner and Szymas 1990). The present observations in normal adult rats agreed with these descriptions. At this point we therefore only wish to point out the regular, polarised distribution of the astroglial processes in the normal fascia dentata (Fig. 6A).

Vimentin-immunoreactiv cells

The vimentin reactivity in normal adults rats consisted of: (1) a slight staining of endothelial cells, (2) a slight staining in most neuropil areas of process-bearing cells with an appearance and distribution like astroglial cells, and (3) a strong staining of process-bearing fibrous cells located in the hippocampal white matter (alveus and angular bundle) and along the deep, hilar border of the dentate granule cell layer. The latter cells were, as also observed by Schmidt-Kastner and Szymas (1990), polarised with stained processes directed mainly into the granule cell layer (Fig. 7A). Single vimentin-reactive cells were occasionally found in the dentate molecular layer or the hippocampal neuropil layers.

249

Fig. 3 Nissl (A) and NDPase (B, C) stains 3 days after a PP lesion, showing intimately related cell pairs, indicative for cell division. A Many cell pairs in FD (arrows). B Intimately related cell bodies in inner zone of the dentate molecular layer. C Electron micrograph of two closely related NDPase-positive cells with a mitotic chromatin pattern, g Granule cell layer. Bars: A 100 gm, B 25 gin, C 5 gm

Lesion-induced changes

Based on the distribution of the glial reactions in the perforant path zones of the fascia dentata, it was found that most entorhinal lesions induced almost complete denervation of the septal and dorsal levels of the fascia dentata. More posteriorly towards temporal levels, the denervation usually became partial, that is only involving the lateral or the medial perforant path, before finally to leave the posterior to temporal levels of the dentate unaffected.

The effects of PP lesions are described separately for each cellular marker. These include TPPase, IDPase and AcPase for microglial cells, GFAP for astroglial cells, vimentin for reactive astroglial cells and possibly reactive microglial cells, and AChE for cholinergic septohippocampal nerve fibres. Changes will be described as they occurred during the postlesional period, ranging from 12 h to 28 days. First there will be an initial brief presentation of the main changes over time, followed by a more detailed account for each postlesional period.

TPPase and IDPase staining of microgtiat cells

The lesion-induced changes in TPPase and IDPase staining consisted of increased staining and accumulation of microglial cells in the denervated PP zones of the ipsilateral dentate molecular layer and CA3 molecular layer, and bilaterally in the CA 1 molecular layer, corresponding to the termination of the lesioned direct and crossed temporo-ammonic tracts. The changes were evident already 24 h after the lesion, at which time there also was a depletion of the adjacent, non-denervated, commissural-associational zone in the fascia dentata for NDPase-reactive cells. From then on, and for all survival times used, the NDPase reactive cells in the denervated neuropil remained more numerous and more densely stained than the cells in the intact neuropil layers.

Postlesional day 1. Already 24 h after the entorhinal lesion, but not at 12 h, the PP zones in the molecular layers of fascia dentata and CA3 stood out against the surrounding layers with an increased TPPase and ID- Pase staining (Figs. 1C, D). In the dentate molecular layer there was both an increase in staining within the PP zones relative to unoperated controls and the non-lesioned contralateral side and relative to the subjacent inner zone of the molecular layer next to the granule cells. This part of the molecular layer and the granule cell layer contained fewer stained cells than normal, while the number of cells had increased in the denervated PP zones. Here the individual microglial cells stood

251

Fig. 5 AcPase staining pattern in the normal hippocampus (A) and 3 days after a PP lesion (B). Nem-ons, in particular large dentate hilar neurons, together with perivascular cells display a positive AcPase reaction located to the cytoplasm (A, B). Three days after the entorhinal lesion (B) AcPase-positive microglial cells occur in the denervated PP zones in the dentate molecular layer, g Granule cell layer. Bar 90 gm

out more compact than normally, as the staining density of the processes had increased, and each process appeared coarser and darker than normal. In the inner zone of the dentate molecular layer, with intact commissural-associational hilo-dentate projections, some of the cells still present were polarised with processes extending superficially towards and into the PP zone. At the same time the main processes had fewer small branches.

Microglial reactions similar to those in the denervated PP zones in fascia dentata were observed bilaterally in the molecular layer of CA1 in response to the degeneration of the temporo-ammonic tract projections from the lesioned entorhinal area.

Postlesional days 2 and 3. The increases in TPPase and IDPase staining and the accumulation of cells in the denervated PP zones were even more evident at day 2 and day 3 after the lesion (Figs. 1E, F). Below the dener-

Fig. 4 NDPase staining of microglial cells (A, B) and a perivascu- Iar cell (C) in the normal hippocampus, and a perivascular microglial cell (D) 3 days after a PP lesion. Notice the location of NDPase in the outer microglial cell membrane (arrow, A, B). Three days after the lesion we found no signs of phagocytotic activity at the ultrastructuraI level (D). e Endothelial cell, p perivascular cell, pm perivascular microglial cell. Bars: A 25 gin, B--D 5 gm

vated PP zones, in the intact commissural-associational zone where the number of NDPase-reactive cells was subnormal, pairs of stained microgliaI cells with intimately related celt bodies were observed (Fig. 3B). Tolu- idine blue staining revealed that many such "cell pairs" were also present in the denervated PP zones (Fig. 3A), where the dense NDPase staining otherwise disclosed their presence. Also, electron microscopy confirmed the presence of several intimately related, NDPase-reactive cells here (Fig. 3C). Both in counterstained Vibratome sections and in semithin and ultrathin plastic sections the small nuclei of the cells had a coarser, more patchy chromatin pattern than in the normal fascia dentata. At the ultrastructural level, the nuclei of many of the closely related cell bodies were highly irregular and indented, and a few cells displayed a clearly mitotic chromatin pattern (Fig. 3C).

NDPase-reactive cells with a close structural relation to vessels were observed at both the light and the electron microscopical level, but this feature was clearly atypical for most of NDPase-reactive microglial cells. At the ultrastructural level, NDPase-reactive cells enclosed within the parenchymal basal lamina of blood vessels had the appearance of perivascular cells as desig- nated by Graeber and Streit (1990) (Fig. 4C). Other ND- Pase-positive cells found next to vessels were located outside the basal lamina (Fig. 4D), and in agreement with Graeber and Streit we consider these cells to be "perivascular microglial cells". None of the NDPase-reactive cells or processes near vessels or in the neuropil displayed signs of active phagocytotic activity at the ultrastructural level.

PostIesional days 7, 14, 21 and 28. The specific accumulation of strongly stained microglial cells within the den-

252

Fig. 6 Astroglial staining pattern (GFAP) in the normal hippocampus (A) and 3 (B) and 14 (C) days after a PP lesion. A Normal fascia dentata with regular, polarised distribution of the astroglial processes. B Three days after the PP lesion, well-stained, hypertrophic, GFAP-reactive astrocytes are prominent in the PP zones. Hypertrophic, superficially directed processes mainly arise from astroglial cell bodies located within the PP zones or at the transition towards the commisural-associational zone. C Four- teen days after the lesion, the GFAP reaction in the PP zones has descreased. Due to shrinkage of the denervated neuropil and/or vascular reordering, the astroglial processes have become irregular and distorted, g Granule cell layer, im inner molecular layer of the fascia dentata, om outer molecular layer of the fascia dentata. Bar 50 gm

ervated parts of the dentate molecular layer persisted for 1, 2, 3 and 4 weeks after the lesion (Fig. 2A, B). Dur- ing the same time, the denervated zones of the molecular layer were shrinking. Together with a widening of the inner zone, with microglial cells showing normal TP- Pase and IDPase staining, this caused the band of neuropil with the densely stained microglial cells to narrow. One week or more after the lesion, the previously observed pairs of TPPase/IDPase positive microglial cells were no longer observed.

The widening of the inner zone coincided with expan- sion of the zone with normal low AChE activity (see below). Simultaneously, the number of NDPase-stained microglial cells in this zone increased to normal levels.

AcPase staining

In control rats, AcPase reactivity in the hippocampal region was observed in some perivascular cells and in neurons, as described by Vijayan and Cotman (1983)

(Fig. 5A). Within their cytoplasm, perivascular cells - as defined by Graeber and Streit (1990) - displayed a strong reaction, localised as clusters of cytoplasmatic red granules. Also neurons, and most clearly large dentate hilar neurons, displayed a faint, but positive AcPase reaction located to the cytoplasm. During the first 2 days after the entorhinal lesion, the AcPase staining did not change. It was only present in some perivascular cells and neurons, with no discernable increase in the denervated areas. At 3 days after the entorhinM lesion there was, however, an increased number of AcPase- positive cells in the areas with axonal degeneration and an increased, heavy staining of perivascular cells (Fig. 5B). The stained cells contained a few strongly positive granules in their cytoplasm. Without employing double stains at this survival time, it was difficult to identify the AcPase-positive cells in the neuropil as either astrocytes or microglial cells. At 7 days, the areas with degeneration still contained AcPase-positive glial cells, but the staining had weakened. Double labelling for NDPase and AcPase allowed us to identify the Ac- Pase-positive cells as being NDPase positive and hence microglial cells. At 7 days the number of AcPase-posi- tire perivascular cells had increased slightly in the denervated areas. No AcPase staining was performed at longer survival times,

AstrogIial staining for GFAP

The reaction of the astrocytes to anterograde axonal degeneration followed the same pattern as that of the microglial cells, except that the increase in GFAP immunoreactivity was slightly delayed relative to the increase in NDPase staining of the microglial cells. The

Fig. 7 Astroglial staining for vimentin (VIM) in the normal hippocampus (A) and 3 days after a PP lesion (B). A Note the strong staining of process- bearing fibrous cells along the deep, hilar border of the dentate granule cell layer (g) in the normal hippocampus and the upwards direction of their processes. B At 3 days there was specific occurrence of vimentin-positive cells in the denervated zones, g Granule cell layer. Bar 100 gm

253

astroglial reaction thus became evident between 1 and 2 days after the entorhinal lesion, peaked at about 7 days postlesionally and then declined during the com- ing weeks.

Postlesional day 1. Both 12 and 24 h after the entorhinal lesion the GFAP staining for astroglial cells appeared normal in the dentate PP zones. At most there was a very slight increase in reactivity of the radiating astroglial processes in some rats with a survival time of 24 h.

Postlesional days 2 and 3. Well-stained, hypertrophic, GFAP-reactive astrocytes were now prominent in the denervated PP zones in fascia dentata and CA3 as well as in the temporo-ammonic tract zones in CA1. In the fascia dentata, hypertrophic, superficially directed processes mainly arose from astroglial cell bodies located within the PP zones or along the deep border towards the commissural-associational zone (Fig. 6B). Some hypertrophic, densely stained processes did, however, also arise from astroglial cell bodies located within the inner zone of the molecular layer. In such cases particularly the superficially directed processes were hypertrophic and more densely stained than normal. The remaining astroglial cells in the inner molecular layer, the granule cell layer and the immediate subgranular zone of the granule cell layer displayed some increase in staining density, but appeared normal with regard to distribution and polarity of the processes.

At posterior to temporal levels of the fascia dentata, where only the medial or the lateral PP was lesioned, the astroglial hypertrophy and increase in GFAP staining

became confined largely to the respective zones of denervation in the middle or outer parts of the dentate molecular layer.

Postlesional days 7, 14 and 21. One week after the lesion the astrocytes displayed maximum GFAP immunoreactivity, with a distribution and morphology similar to that observed after 3 days. Due to the subsequent shrinkage of the denervated PP zones, the astroglial processes did, however, become increasingly irregular and distorted, as seen most markedly at 14 days after the denervation (Fig. 6C). As it is well known that astrocytes are intimately related to the vascular system, the distortion of the astroglial processes could also be asso- ciated with changes in blood capillary orientation. It has been shown that a reorganisation of the vascular system takes place in the dentate gyrus already 2 days after entorhinal lesions (Scheff et al. 1978). At this time the density of the GFAP reaction was still pronounced in the denervated areas, although weaker than at 7 days. At 21 days the GFAP immunoreactivity had attained almost normal levels.

Immunocytochemical staining for vimentin

Besides the normal staining of some hippocampal endothelial cells, white matter fibrous astrocytes and radi- al cells in the subgranular zone of the fascia dentata, no new vimentin-immunoreactivity was detected at 24 h after the lesion. At 3 and 7 days there was, however, specific occurrence of vimentin-positive cells in the denervated zones on the lesioned side (Fig. 7B). Some of

254

Fig. 8 Consecutive sections stained in parallel for AChE (A), NDPase (B) and GFAP (C) 7 days after an ipsilateral lesion of the medial perforant path, causing terminal and axonal degeneration in the middle one third of the molecular layer. By comparison, it is seen that the AChE-rich band (arrows, A) indicative of sprouting of cholinergic septo-hippocampal afferents, corresponds to the denervated PP zones with increased microglial (B) and astroglial (C) activity. The section stained for AChE (A) displayed more shrinkage during processing than the other two. g Granule cell layer. Bar 100 gm

topic we shall only confirm that our present results cor- responded to those obtained in other studies. Beginning at 3 days after the entorhinal lesion, the AchE staining in the outer parts of the dentate molecular layer increased dramatically. From comparisons between parallel series of sections, we found that induced AChE staining was localised in the zones with increased microglial NDPase and astroglial GFAP activity (Fig. 8A, B, C).

these cells had a typical stellate astroglial morphology; others resembled microglial cells.

With regard to location and extent, the changes in vimentin-immunoreactivity displayed some variability between cases with the same postlesional survival time. In some rats the changes were restricted to the denervated PP zones. In others, larger parts of the hippocampus on the lesioned side as well as overlying neocortical areas on the lesioned side showed in.creased levels of immunoreactivity. Other differences between rats included the number of vimentin-reactive cells along the deep, hitar border of the dentate granule cell layer.

AChE staining

Removal of the PP projection is known to induce increased AChE staining along the apical parts of the dentate granule cell dendrites, corresponding to sprouting of cholinergic septohippocampal afferents (Lynch et al. 1972, 1976; Zimmer 1973; Cotman et al. 1977; Scheff et al. 1980). In view of the extensive literature on this

5 '-Nucleotidase staining

Only 7 and 14 days postlesion did the 5'-nucleotidase staining and distribution differ from that observed in control rats described by Kreutzberg et al. (1978a). Dur- ing the 2nd week after the lesion, a diffuse but specific staining of the neuropil did, however, develop in the denervated PP zones. Some staining was located in a few very lightly stained cells, which could not, however, be clearly identified as microglial cells.

Discussion

NDPase staining

Under normal conditions, the ecto-enzyme NDPase is located in the plasmalemma of microgtial cells (Castel- lano et al. 1984, 1989a; Castellano 1987). This suggests that the enzyme might be involved in the regulation of nucleoside phosphate levels in the extracellular space. Increased NDPase activity of activated microglia might

therefore reflect a change in extracellular levels or turnover of nucleoside phosphates. Although substances liberated from degenerating axons and terminals have not been identified so far, they might well include thiamine compounds or nucleoside phosphates. Thiamine and thiamine phosphates have been reported to be released by axons during hyper-excitation and after the administration of different types of excitatory drugs (Cooper et al. 1963; Itokawa and Cooper 1970). Studies in progress have shown moreover that local in- fusion of thiamine pyrophosphate induces a rapid increase in NDPase activity and migration of microglial cells (Castellano et al. 1989b). Other findings suggest that thiamine or thiamine phosphoesters have non- coenzymatic roles in the central nervous system (Berman and Fishman 1975; Cooper and Pincus 1979; Matsuda et al. 1989), and that the intracerebral levels of thiamine are regulated within very narrow limits (Rindi et al. 1984).

Regarding functions of the nucleosides and nu- cleosidephosphates, several studies have demonstrated a neuromodulatory effect (Burnstock 1977; Phillips and Wu 1983) and induction of cell proliferation and differ- entiation (Fox and Kelley 1978). The proliferation of microglial cells in the areas with axonal and terminal degeneration might accordingly be induced by the release of these substances, possibly in conjunction with plasma constitutes able to leak through the blood-brain barrier under these conditions (Jensen et al., in prepara- tion).

The first microglial reactions occurred within the neuropil layers with axonal degeneration. The initial polarisation of processes from microglial cells located in adjacent intact parts of the dentate molecular layer was also directed towards the degenerating PP zones. When passing into the denervated PP zones, these processes became coarser with increased levels of NDPase activity. Focusing on the initial period after entorhinal lesioning, it accordingly appeared that not only cells located within the area of degeneration, but also parts of cells extending into these zones displayed enhanced NDPase activity. Later, microglial cells from adjacent areas appeared to migrate into the denervated areas.

Both the localised increase in NDPase activity and the directed migration of cells, strongly suggest that the lesioned axons and terminals triggered the microglial response by a local release (or the arrest of normal release) of some, so far unknown substances. The attrac- tion of the microglial processes towards the areas with axonal and terminal degeneration moreover indicates that membrane-bound receptors are involved in this "guided" extension. At present the substances and receptors involved are not known, but microglial cells do express several surface receptors, both normally and after various stimulations, for example Fc (IgG 1/2b) and complement receptor type 3b(C3b), and Mac-l , when activated (Frei et al. 1987; Finsen et al. 1990).

Microglial reactions

255

Microglial cells are some of the cells most responsive to injury within the CNS in terms of migration and proliferation. They react to mechanical trauma (Boya et al. 1986; Streit et al. 1988), vascular insults (Jorgensen et al. 1993), infections (Giulian 1987) and immune responses (Lassman et al. 1986; Matsumoto and Fujiwara 1987), as well as exhibiting neuronal responses to distant axotomy (Sj6strand 1965; Kreutzberg 1966; Kerns and Hi- nesman 1973ab; Graeber et al. 1988b). Migratory and proliferative properties of microglial cells have been reported previously in the dentate gyrus of the adult rat hippocampus in relation to anterograde axonal degeneration after entorhinal lesions (Lynch et al. 1975; Gall etal. 1979; Gehrmann etal. 1991). Using a lesion paradigm similar to the present one, the first dividing microglia-like cells were observed 20 h after the entorhi- hal lesion, while the exact onset of the migratory activity was uncertain.

In the present study, the first evidence of altered microglial activity was in the zones with terminal degeneration 24 h after the entorhinal lesion. In other animals, sacrificed after only 12 h, no reactions were evident. At 24 h, the microglial cells displayed both increased ND- Pase activity and a change in morphology, with coarser and more densely stained processes than normal. In addition there were strong indications of local proliferation of microglial cells, as well as migration of microglial cells from adjacent intact neuropil areas into the dener- rated PP zones.

Regarding migration, there was an evident loss of microglial cells in the intact inner zone of the dentate molecular layer concomitant with an increase of ND- Pase-positive cells in the denervated PP zones. In direct relation to this, Gall et al. (1979) showed by timed triti- ated thymidine labelling and subsequent autoradiography that migration of small microglia-like cells does contribute to the accumulation of cells in the denervated parts of the dentate molecular layer. In relation to the migratory activity of microglial cells, Graeber et al. (1988a) showed that vimentin was expressed in activated, but not in resting rat microglial cells after axotomy of the facial nerve. These observations correspond well with our findings of vimentin-positive cells with microglial morphology in the denervated PP zones 3 days after the lesion. Since the cytoskeletal protein vimentin appears to be part of an "intermediate filament lattice anchored at nuclear membrane sites" (Kreutzberg et al. 1989), the appearance of vimentin might well be related to migratory activity and changes in the microglial morphology.

Regarding proliferation of microglial cells, the present observations of NDPase-reactive microglial cells with intimately related cell bodies at, in particular, 2 days postlesional, are indicative of cell division. This is supported by the finding that the cells in the cell pairs displayed a patchy chromatin pattern, verified at the ultrastructural level and not found in normal microglial cells.

256

AcPase

The general occurrence of AcPase in hippocampal neurons will not be discussed here, as the presence of this hydrolytic enzyme in the Golgi apparatus and lyso- somes is well known. Moreover, we did not observe any changes in the neuronal AcPase staining in the hippocampus of the lesioned animals compared to normal rats. The presence of AcPase in perivascular cells and reactive glial cells is discussed below.

The appearance of AcPase-reactive microglial cells in the areas of axonal degeneration, demonstrated by double staining for AcPase and NDPase 7 days after the lesion, suggested an active phagocytic role of these cells or an increased intracellular degradation of structural or functional proteins. AcPase and other hydrolytic enzymes as ATPase, 5'-nucleotidase, aryl sulphatase and non-specific stearase have been demonstrated in amoeboid microglial cells in the perinatal rat brain (Ling 1977; Ling et al. 1982; Kaur et al. 1984; Boya et al. 1986). In such developing brains, the amoeboid cells have been demonstrated primarily in regions with degeneration of neurons and rearrangement of fibres as part of the normal development (Ling 1977; Valentino and Jones 1981). When amoeboid cells transform into resting microglial cells, they become ramified, loose their hydrolytic enzymatic activity and in consequence of this their phagocytic role (Boya et al. 1979; Perry and Gordon 1988). Although microglial cells may act as ac- tively phagocytising cells in the denervated PP zones (Gehrmann et al. 1991), it is astroglial cells which have been repeatedly shown to have a predilection for phago- cytosis of degenerating axon terminals (Hoff et al. 1982; Phelps et al. 1991), often engulfed together with the still attached postsynaptic spines (Matthews et al. 1976).

In astroglial and neuronal primary cell cultures and organotypic slice cultures, where the presence of microglial cells have been shown by NDPase staining (Castellano et al. 1990, 1991), amoeboid-like microglial cells showed a strong reaction for AcPase. In contrast to the in vivo situation, ramified-like microglial cells also expressed a faint AcPase staining in the cultures, sug- gesting persistence of some degree of immaturity or phagocytic activity.

The AcPase-positive, perivascular cells deserve some comment as they were strongly positive for AcPase with their cytoplasm loaded with AcPase positive granules. The nature of the~e cells was thoroughly discussed by Ibrahim and coworkers (Ibrahim 1974; Ibrahim et al. 1974) who suggested that the AcPase-positive cells ("granular pericytes" or '~ cells") were mast cells, of a type peculiar to the CNS. There is some evidence that mast cells can facilitate the entry of cells from the blood stream not only into the skin, but also into the brain (Griffin and Mendoza 1986). Under certain conditions and mediated by T lymphocytes, mast cells may thus release vasoactive amines which can open endothelial tight junctions and facilitate the entry of inflammatory cells (Askenase et al. 1980). Also, Graeber and Stre-

it (1990) recently distinguished between two types of perivascularly located cells, namely "pericytes" and "perivascular cells". They defined perivascular cells as fitting the morphological definition of pericytes as being enclosed within the basal lamina, but yet belonging to a distinct population of resident CNS macrophages due to their immunophenotype (ED-2 positive, Ia positive). We found that the AcPase-positive, perivascularly located cells also were NDPase positive, i.e. expressing a typical marker of macrophages/microglial cells. In accor- dance with this, we would classify the AcPase-positive cells as perivascular cells with macrophage properties.

The slight increase in number of perivascular Ac- Pase-positive cells in the denervated PP zones 7 days after the entorhinal lesion, leads us to speculate whether at least some blood-borne cells invade the denervated fascia dentata during the axonal degeneration.

5'-Nucleotidase staining

In an ultrastructural study, Kreutzberg and Barron (1978b) reported high activity of 5'-nucleotidase in the plasmalemma of microglial cells in the facial nucleus 2-5 days after peripheral transection of the facial nerve. They suggested that the increased enzymatic activity was related to increased production of adenosine, an important intercellular messenger, which might be of significance for the regenerating motor neuron. For the first days after the entorhinal lesion, we did not observe any light microscopical changes in 5'-nucleotidase activity in the neuropil layers with axonal degeneration or the adjacent intact neuropil and cell layers. After 1 and 2 weeks a diffuse staining was confined to the denervated PP zones, with very light staining of a few microglia- like cells. Both the time course and the pattern of 5'-nucleotidase expression after anterograde axonal degeneration are accordingly different from those observed in relation to retrograde, axotomy-induced, neuronal changes.

Astroglial reactions

The interplay between neurons, astroglial and microglial cells in the developing and the mature CNS is highly complicated both in terms of timing and action of intercellular messengers and receptors. Several potential regulatory molecules have, however, evolved. Astroglial cells are thus capable of producing and releasing substances with effects on neuronal survival and neurite extension (Lindsay 1986; Wujek and Akeson 1987; Ga- dient et al. 1990; Yoshida and Gage 1991), as well as regenerative processes in the injured rat brain (Nieto- Sampedro et al. 1982; Crutcher and Collins 1986; Needels et al. 1986; Barde et al. 1987). With direct rele- vance to this study, Heacock et al. (1984) demonstrated after entorhinal lesions an increase in the trophic activity in the hippocampus which correlated to the degree of

astrogliosis. Giulian and collaborators have shown in several studies that interleukin-1 (IL-1), produced and released by macrophages and microglial cells, can stimulate the growth of astrocytes in vitro, and induce astrogliosis and neovascularisation after intracerebrat injections in vivo (Giulian et al. 1986, 1988b). Based on these and other studies (Nieto-Sampedro and Berman 1987), IL-1 is now considered to play an important role in the the CNS reaction to injury. Along this line, in vitro experiments have shown that CNS-derived (microglia-derived) IL-1, as well as recombinant IL-1, in addition to stimulating astrocyte proliferation also increases astroglial glutamine synthetase activity and the amount of GFAP in reactive astrocytes. Besides its mi- togenic activities, IL-1 may accordingly act as a glial cell activator and modulator. Regarding a possible feed- back from astrocytes to microgliat cells, Giulian et al. (1991) recently described the existence of two microglial mitogens (MMs) which were produced by astrocytes in the traumatised rat brain and shown to influence the growth of microglial cells in vitro and in vivo.

The observations that IL-1 is released in vitro by amoeboid microglia and acts on astrocytes suggest that similar mechanisms are operative in vivo. The present ob- servation of a slight delay in the astroglial response relative to the microglial reaction is compatible with this, although it may be argued that different cellular mark- ers were used to detect the microglial and the astroglial response. So far no one has reported an astroglial response in the denervated fascia dentata less than 24 h after the lesion, but at lesion sites in the rat cerebral cortex Condorelli et al. (1990) did find a strong increase of mRNA encoding for GFAP as early as 6 h after the lesion. An increase in GFAP mRNA has also been reported in the fascia dentata after entorhinat lesions, but the time of onset of this increase was not determined (Poirier et al. 1990).

The present study provides no evidence of astroglial celt proliferation in the denervated PP zones. We are therefore left to conclude that the PP denervation elicit- ed hypertrophy, but not hyperplasia of the dentate astroglial cells. A similar pattern was observed in brain stem motor nuclei in response to peripheral axotomy (Graeber et al. 1988b) and in the hippocampus after intraventricular injections of kainic acid (Jorgensen et al. 1993).

Concluding remarks

The glial response to perforant part denervation consisted of a primary microglial activation with an increase in number and NDPase reactivity, parallelled by a change in morphology of the microgliat cells. The changes must be assumed to be triggered by the release of substances from the degenerating axons and terminals, and we hypothesise that nucleosides or nucleoside phosphates might be among these substances. The as-

257

troglial reactions are suggested to be stimulated by the initial activation of microglial cells, possibly through IL-1 and other factors. Furthermore, we suggest that the reactive astrocytes take part in reactive synaptogenesis and that they, by means of MMs, take part in regu- lating the ongoing microglial response. At day 3 we observed the induction of an AChE-rich band in the zones with increased microglial and astroglial activity. We also observed increased AcPase reactivity in microglial and perivascular cells in the denervated areas, but no significant increase in 5'-nucleotidase activity.

As the microglial change s occurred slightly ahead of noticeable reactive changes in astroglial cells, we point to the microglial cells as having a pivotal role in the subsequent glial and neuron-glia interactions (Finsen et al. 1993; Jorgensen et al. 1993). Exactly which substances are the mediators of the intercellular signalling among glial cells are at present unknown, but cytokines like I L l are obvious candidates. Correlative studies, making use of the well-defined hippocampal structure and connective organisation and including the degenerative and reactive stages of the neural elements (axons, target cells), will be most helpful in identifying and char- acterising the exact function of these signal molecules.

Acknowledgements Special thanks are due to Dr. Bente R. Finsen for informative discussions and helpful comments. The technical assistance provided by Dorete Jensen, Anette Bottrup and Thork- ild Nielsen is gratefully acknowledged, as is the photographic help from Albert Meier and the financial support from the Lundbeck Foundation, the Danish MRC and the Danish State Biotechnolo- gy Programme.

References

Askenase PW, Bursztajn P, Gershon MD, Gershon RK (1980) T-cell-dependent mast cell degranulation and release of sero- tonin in murine delayed-type hypersensitivity. J Exp Med 152:1358-1374

Barde YA, Davies AM, Johnson JE, Lindsay RM, Thoenen H (1987) Brain-derived neurotrophic factor. Prog Brain Res 71:185-189

Berman K, Fishman RA (1975) Thiamine phosphate metabolism and possible co enzyme-independent functions of thiamine in brain. J Neurochem 24:457M65

Boya J, Calvo J, Prado A (1979) The origin of microglial cells. J Anat 129(1):177-186

Boya J, Calvo J, Garcia-Maurifio E (1986) Nature of macrophages in rat brain. A histochemical study. Acta Anat 127:142-145

Burnstock G (1977) Purine nucleotides and nucleosides as neuro- transmitters or neuromodulators in the central nervous system. In: Usdin E et al. (ed) Neuroregulators and psychiatric disorders. Oxford University Press, New York, pp 470-477

Casteltano B (1987) Estudio histoquimico de la actividad TPPasa/ NDPasa en neuronas y c61ulas gliales del sistema nervioso central. Bellaterra (Barcelona): Universitat Aut6noma de Bar- celona

Castellano B, Palacios G, Gonzfilez B (1984) A comparative study on NDPase and TPPase activities of glial cells in vertebrates and invertebrates. Trab Inst Cajal Invest Biol 75:81

Castellano B, Gonzglez B, Palacios G (1989a) Cytochemical demonstration of TPPase in myelinated fibers in the central and perypheral nervous system of the rat. Brain Res 492:203-210

258

Castellano B, Gonz/tlez B, Vela JM (1989b) Glial response to intracerebral injection of thiamine pyrophosphate. A prelimina- ry study. Eur J Neurosci [Suppl J 2]: 249

Castellano B, Gonz~tlez B, Pedersen EB, Finsen BR, Zimmer J (1990) Amoeboid and ramified microglial cells are present in primary astroglial and neuronal cell cultures. Eur J Neurosci [Suppl J 3]: $3259

Castellano B, Gonz/dez B, Jensen MB, Pedersen EB, Finsen BR, Zimmer J (1991) A double staining technique for simultaneous demonstration of astrocytes and microglia in brain sections and astroglial cell cultures. J Histochem Cytochem 39(5):561- 8

Condorelli DF, Dell'Albani P, Kaczmarek L, Messina L, Spampi- nato G, Avola R, Messina A, Giuffrida-Stella AM (1990) Glial fibrillary acidic protein messenger RNA and gtutamin synthetase activity after nervous system injury. J Neurosci Res 26: 251-257

Cooper JR, Pincus JH (1979) The role of thiamine in nervous tissue. Neurochem Res 4(2):223-39

Cooper JR, Roth RH, Kini MM (1963) Biochemical and physio- logical function of thiamine in nervous tissue. Nature 199:609-610

Cotman C, Gentry C, Steward O (t977) Synaptic replacement in the dentate gyms after unilateral entorhinal lesion: electron microscopic analysis of the extent of replacement of synapses by the remaining entorhinal cortex. J Neurocytol 6:455-464

Crutcher KA, Collins F (1986) Entorhinal lesions result in increased nerve growth factor-like activity in medium conditioned by hippocampal slices. Brain Res 399:383-389

Duchen LW (1992) General pathology of neurons and neuroglia. In: Hume Adams J, Duchen LW (eds) Greenfild's neuropatho- logy, 5th edn. Arnold, London, pp 1-68

Fagan AM, Gage FH (1990) Cholinergic sprouting in the hippocampus: a proposed role for IL-1. Exp Neurol 110:105-120

Finsen BR, Pedersen EB, Sorensen T, Hokland M, Zimmer J (1990) Immune reactions against intracerebral murine xeno- grafts of fetal hippocm~pal tissue and cultured cortical astrocytes in the adult rat. Prog Brain Res 82:111-128

Finsen BR, Jorgensen MB, Castellano B, Jensen MB, Diemer NII, Zimmer J (1991) Glial reactions in the rat hippocampus following intraventricular kainic acid injection. Eur J Neurosci [Suppl J 4]: 98

Finsen BR, Jorgensen MB, Diemer NH, Zimmer J (1993) Micro- glial MHC antigen expression after ischemic and kainic acid lesions of the adult rat hippocampus. Glia 7:41--49

Rox IH, Kelley WN (1978) The role of adenosine and 2'-deoxya- denosine in mammalian cells. Ann Rev Biochem 47:655-686

Frei K, Bodmer S, Schwerdel C, Fontana A (1986) Astrocyte-deri- red interleukin 3 as a growth factor for microglial cells and peritoneal macrophages. J Immunol 137:3521-3527

Frei K, Siepl C, Groscurth P, Bodmer S, Schwerdel C, Fontana A (1987) Antigen presentation and tumor cytotoxicity by interfe- ron-treated microglial cell. Eur J Immunol 17:1271-1278

Gadient RA, Cron KC, Otten U (1990) Interleukin-1 and tumor necrosis factor- synergistically stimulate nerve growth factor (NGF) release from cultured rat astrocytes. Neurosci Lett 117: 335-340

Gall C, Rose G, Lynch G (1979) Proliferative and migratory activity of glial cells in the partially deafferented hippocampus. J Comp Neurol 183:539-550

Gehrmann J, Schoen SW, Kreutzberg GW (t99t) Lesion of the rat entorhinal cortex leads to a rapid microglial reaction in the dentate gyrus. A light and electron microscopical study. Acta Neuropathol 82:442-455

Geneser-Jensen FA, Blackstad TW (1971) Distribution of acetyl cholinesterase in the hippocampal region of the guinea pig. I. Entorhinal area, parasubiculum, and presubiculum. Z Zelt- forsch 114:460-481

Giulian D (1987) Amoeboid microglia as effectors of inflamma- tion in the central nervous system. J Neurosci Res 18:185-171

Giulian D, Baker TJ, Shih L-CN, Lachman LB (1986) Interleukin 1 of the central nervous system is produced by ameboid microglia. J Exp IVied 164:594-604

Giulian D, Vaca K, Johnson B (1988a) Secreted peptides as regu- lators of neuron-glia and glia-glia interactions in the developing nervous system. J Neurosci Res 21:487-500

Giulian D, Young DG, Woodward J, Brown DC, Lachman LB (1988b) Interleukin-1 is an astroglial growth factor in the developing brain. J Neurosci 8(2):709-714

Giulian D, Johnson B, Krebs JF, George JK, Tapscott M (1991) Microglial mitogens are produced in the developing and injured mammalian brain. J Cell Biol 112:323-333

Graeber MB, Streit WJ (1990) Perivascular microglia defined (Let- ters to the editor). Trends Neurosci 13(9): 366

Graeber MB, Streit WJ, Kreutzberg GW (1988a) The microglial cytoskeleton: vimentin is localized within activated cells in situ. J Neurocytol 17(4):573-580

Graeber MB, Tetzlaff W, Streit WJ, Kreutzberg GW (1988b) Mi- croglial cells but not astrocytes undergo mitosis following rat facial nerve axotomy. Neurosci Lett 85:317-321

Griffin DE, Mendoza QP (1986) Identification of the inflammatory cells present in the central nervous system of normal and mast ceU-deficient mice during sindibis virus encephalitis. Cell Immunol 97:454-459

Heacock AM, Schonfeld AP, Katzman R (1984) Relation of hippocampal trophic activity to cholinergic nerve sprouting. Soc Neurosci Abstr 10:1052

Hoff SF, Scheff S\V, Benardo LS, Cotman CW (1982) Lesion-induced synaptogenesis in the dentate gyms of aged rats: I. Loss and reacquisition of normal synaptic density. J Comp Neurol 205:246-252

Ibrahim MZM (1974) The mast cells of the mammalian CNS. Part 1 Morphology, distribution and histochemistry. J Neurol Sci 21:431-478

Ibrahim MZM, Khreis Y, Koshayan DS (1974) The histochemical identification of microglia. J Neurol Sci 22:2 t 1-233

Itokawa Y, Cooper JR (1970) Ion movements and thiamine in nervous tissue. I. Intact nerve preparations. Biochem Pharma- col t9:98%992

Jorgensen MB, Finsen BR, Jensen MB, Castellano B, Diemer NH, Zimmer J (1993) Microglial and astroglial reactions to ischemic and kainic-acid induced lesions of the adult rat hippocampus. Exp Neurol 120:70-88

Kaur C, Ling EA, Wong WC (1984) Cytochemical localization of 5'-nucleotidase in amoeboid microglial cells in postnatal rats. J Anat 139:1-7

Kerns JM, Hinesman EJ (t973a) Neuroglial response to sciatic neurectomy: I. Light microscopy and autoradiography. J Comp Neurol 151:237-254

Kerns JM, Hinesman EJ (1973b) Neuroglial response to sciatic neuroctemy. II. Etectron microscopy. J Comp Neurol 151:255-280

Kreutzberg GW (1966) Autoradiographische Untersuchung tiber die Beteiligung yon Gliazellen an der axonalen Reaktion im Facialiskern der Ratte. Acta Neuropathol (Berl) 7:149-161

Kreutzberg GW, Barron KD, Schubert P (1978a) Cytochemical localization of 5'-nucleotidase in glial plasma membranes. Brain Res 158(2):247-257

Kreutzberg GW, Barron KD (1978b) 5'-Nucleotidase of microglial cells in the facial nucleus during axonat reaction. J Neurocytol 7(5):601-610

Kreutzberg GW, Graeber MB, Streit WJ, Tetzlaff W (1989) Glial reactions accompanying the regeneration process in motoneu- rons. In: Scarpini E, Fiori MG, Pleasure D, Scarlato G (eds) Peripheral nerve development and regeneration: recent advan- ces and clinical applications, voI 19. Fidia Research Series, Liviana Press, Padova, pp 199-206

Lassman H, Vass K, Brunner Ch, Seitelberger F (1986) Characte- rization of inflammatory infiltrates in experimental allergic en- cephalomyelitis. Prog Neuropathot 6: 33-62

Lindsay RM (1979) Adult rat brain astrocytes support survival of both NGF-dependent and NGF-insensitive neurones. Nature 282:80-82

259

Lindsay RM (1986) Reactive gliosis. In: Fedoroff S, Vernadakis A (eds) Astrocytes. Cell biology" and pathology of astrocytes, vol 3. Academic Press, New York, pp 231-262

Lindsay RM, Barber PC, Sherwood MR, Zimmer J, Raisman G (1982) Astrocyte cultm-es from adult rat brain. Derivation, characterization and neurotrophic properties of pure astroglial cells from corpus callosum. Brain Res 243:329-343

Ling EA (1977) Light and electron microscopic demonstration of some lysosomal enzymes in the amoeboid microglia in neonatal rat brain. J Anat 123:637-648

Ling EA, Kaur C, Wong WC (1982) Light and electron microscopic demonstration of non-specific esterase in amoeboid microglial cells in the corpus callosum in postnatal rats: a cytochemical link to monocytes. J Anat 135:385 394

Lynch G, Cotman CW (1975) The Hippocampus as a model for studying anatomical plasticity in the adult brain. In: Isaacson RL, Pribram KH (eds) The hippocampus, vol I. Plenum Press, New York, pp 123-155

Lynch G, Matthews DA, Mosko S, Parks T, Cotman C (1972) Induced acetylcholinesterase-rich layer in rat dentate gyrus following entorhinal lesions. Brain Res 42: 311-318

Lynch G, Rose G, Gall C, Cotman CW (1975) The response of the dentate gyrus to partial deafferentation. In: Santini M (ed) Golgi centennial s~nposium. Proceedings. Raven Press, New York, pp 305-317

Lynch G, Gall C, Rose G, Cotman CW (1976) Changes in the distribution of the dentate gyms associational system following unilateral or bilateral entorhinal lesions in the adult rat. Brain Res 110:57-71

Manthorpe M, Rudge JS, Varon S (1986) Astroglial cell contribu- tions to neuronal survival and neuritic growth. In: Fedoroff S, Vernadakis A (eds) Astrocytes, vol I. Academic Press, New York, pp 315-376

Matthews DA, Cotman C, Lynch G (1976) An electron microscopic study of lesion-induced synaptogenesis in the dentate gyrus of the adult rat. I. Magnitude and time course of degeneration. Brain Res 115:1-21

Matsuda T, Doi T, Tonomura H, Baba A, Iwata H (1989) Postna- tal development of thiamine metabolism in rat brain. J Neuro- chem 52:842--846

Matsumoto Y, Fujiwara M (1987) In situ detection of class I and II major histocompatibility complex antigens in the rat central nervous system during experimental allergic encephalomyeli- tis. An immunohistochemical study. J Neuroimmunol 12:265- 277

McCaffery CA, Raju TR, Bennett MR (1984) Effects of cultured astroglia on the survival of neonatal rat retinal ganglion cells in vitro. Dev Biol 104:441-448

Murabe Y, Sano Y (1982) Morphological studies on neuroglia. V. Microglial cells in the cerebral cortex of the rat, with special reference to their possible involvement in synaptic function. Cell Tissue Res 223(3):493-506

Needels DL, Nieto-Sampedre M, Cotman CW (1986) Induction of a neurite-promoting factor in rat brain following injury or deafferentation. Neuroscience 18:517-526

Nieto-Sampedro M, Berman MA (1987) Interleukin-l-like activity in rat brain: sources, targets, and effect of injury. J Neurosci Res 17:214-219

Nieto-Sampedro M, Lewis ER, Cotman CW, Manthorpe M, Ska- per SD, Barbin G, Longo FM, Varon S (1982) Brain injury causes a time-dependent increase in neuronotrophic activity at the lesion site. Science 217:86(~861

Nieto-Sampedro M, Whittemore SR, Needels DL, Larson J, Cot- man CW (1984) The survival of brain transplants is enhanced by extracts from injured brain. Proc Natl Acad Sci USA 81: 6250--6254

Novikoff AB, Goldfischer S (I 961) Nucleoside diphosphatase activity in the Golgi apparatus and its usefulness for cytological studies. Proc Natl Acad Sci USA 47:802

Perry VH, Gordon S (1988) Macrophages and microglia in the nervous system. Trends Neurosci 11(6):273-277

Phelps S, Mitchell J, Wheal HV (1991) Changes to synaptic ultra- structure in field CAt of the rat hippocampus following intra- cerebroventricular injections of kainic acid. Neuroscience 40:687-699

Philips JW, Wu PH (1983) Roles of adenosine and adenine nucleotides in the central nervous system. In: Daly JW; Phillis JW, Kuroda Y, Shimizu H, Vi M (eds) Physiology and pharmacolo- gy of adenosine derivatives. Raven Press, New York, pp 219- 236

Poirier J, May PC, Osterburg HH, Geddes J Cotman C, Finch CE (1990) Selective alterations of RNA in rat hippocampus after entorhinal cortex lesioning. Proc Natl Acad Sci USA 87(1): 303-307

Poirier J, Hess M, May PC, Finch CE (1991) Cloning of hippocampal poly(A) RNA sequences that increase after entorhinal cortex lesion in adult rat. Mol Brain Res 9(3):191-195

Raisman G (1969) Neuronal plasticity in the septal nuclei of the adult rat. Brain Res 14:25-48

Rindi G, Cominciolo V, Regiani C, Patrini C (1984) Nervous tissue thiamine metabolism in vivo. II. Thiamine and its phosphoesters: dynamics in different brain regions and sciatic nerve of the rat. Brain Res 293:329-342

Rudge JS, Manthorpe M, Varon S (1985) The output of neuronotrophic and neurite-promoting agents from rat brain astroglial cells: a microculture method for screening potential regulatory molecules. Dev Brain Res 19:161-172

Rudge JS, Alderson RF, Pasnikowski E, McClain J, Ip NY, Lind- say RM (1992) Expression of ciliary neurotrophic factor and the neurotrophins-nerve growth factor, brain-derived neurotrophic factor and neurotrophin 3-in cultured rat hippocampal astrocytes. Eur J Neurosci 4:459-471

Scheff SW, Benardo LS, Cotman CW (1978) Lesion-induced vascular changes in the dentate gyms following removal of the entorhinal cortex in adult rats. Exp Neurol 62:815-820

Scheff SW, Benardo LS, Cotman CW (1980) Decline in reactive fiber growth in the dentate gyrus of aged rats compared to young adult rats following entorhinal cortex removal. Brain Res 199:21-38

Schiffer D, Giordana MT, Migheli A, Giaccone G, Pezzotta S, Mauro A (1986) Glial fibrillary acidic protein and vimentin in the experimental glial reaction of the rat brain. Brain Res 374(1):110-t18

Schmidt-Kastner R, Szymas J (1990) Immunohistochemistry of glial fibrillary acidic protein, vimentin and S--100 protein for study of astrocytes in hippocampus of rat. J Chem Neuroanat 3:179-192

Schnitzer J, Franke WW, Schachner M (1981) Immunocytochemi- cal demonstration of vimentin in astrocytes and ependymal cells of developing and adult mouse nervous system. J Cell Biol 90(2): 435-47

Sj6strand J (1965) Proliferative changes in glial cells during nerve regeneration�9 Z Zellforsch 68:481-493

Soufleris AJ, Pretlow TP, Bartolucci AA, Pitts AM, MacFadyen AJ, Boohaker EA, Pretlow TG (1983) Cytologic characterization of pulmonary alveolar macrophages by enzyme histochemistry in plastic. J Histochem Cytochem 31(12): 1412-1418

Spranger M, Lindholm D, Bandtlow C, Heumann R, Gnahn H, N/iher-No6 M, Thoenen H (1990) Regulation of nerve growth factor (NGF) synthesis in the rat central nervous system: comparison between the effects of interleukin-1 and various growth factors in astrocyte cultures and in vivo. Eur J Neuros- ci 2: 6%76

Sternberger LA (1986) Immunocytochemistry, 3rd edn. J Wiley, New York, pp 90-209

Streit WJ, Graeber MB, Kreutzberg GW (t 988) Functional plasticity of microglia: a review. Glia 1:301-307

Thanos S (1991) Specific transcellular carbocyanine-labelling of rat retinal microglia during iniury-induced neuronal degeneration. Neurosci Lett 127(1):108-112

Tonder N, Sorensen T, Zimmer J (1989) Enhanced host perforant path innervation of neonatal dentate tissue after grafting to axon sparing, ibotenic acid lesions in adult rats. Exp Brain Rcs 75:483-496

260

Valentino KL, Jones EG (1981) Morphological and immunocytochemical identification of macrophages in the developing corpus callosum. Anat Embryol 163:157-172

Vijayan VK, Cotman CW (1983) Lysosomal enzyme changes in young and aged control and entorhinal-lesioned rats. Neuro- biol Aging 4(1):13-23

Wictorin K, Bj6rklund A (1990) Fetal striatal transplants in the ibotenate lesioned striatum of adult rats: specific anatomical integration between graft and host. In: Bj6rklund A, Aguayo AJ, Ottoson D (eds) Brain repair. (Wenner-Gren International Symposium Series, vol 56) Stockton Press, New York, pp 341- 353

Wujek JR, Akeson RA (1987) Extracellular matrix derived from astrocytes stimulates neuritic outgrowth from PC12 cells in vitro. Brain Res 431:87-97

Yoshida K, Gage FH (1991) Fibroblast growth factor stimulate nerve growth factor synthesis and secretion by astrocytes. Brain Res 538:118-126

Zimmer J (1973) Extended commissural and ipsilateral projections in postnatally deentorhinated hippocampus and fascia dentata demonstrated in rats by silver impregnation. Brain Res 64:293-311

Zimmer J (1974) Long term synaptic reorganization in rat fascia dentata deafferented adolescent and adult stages: observations with the Timm method. Brain Res 76:336-342

Zimmer J, Sunde N (1984) Neuropeptides and astroglia in intracerebral hippocampal transplants: an immunohistochemical study in the rat. J Comp Neurol 227:331-347

Zimmer J, Tonder N, Sorensen T (1992) Regenerative axonal growth studied by neural transplantation. In: Schousboe A, Diemer NH, Kofod H (eds) Drug research related to neuroac- tive amino acids. (Alfred Benzon Symposium 32) Munksgaard, Copenhagen, pp 431~444

Documents

Microglial and astroglial reactions to anterograde axonal …histoserver.uab.cat/PDFs/Publicaciones/1994_Microglial... · 2007. 5. 29. · Exp Brain Res (1994) 98:245-260 9 Springer-Verlag