GLIAL REACTION IN THE HIPPOCAMPUS AFTER GLOBAL CARDIOGENIC ISCHEMIA

ABSTRACT: Many experimental surgerical procedures have been perfomed in the analyse of the phenomenon of brain trophism and plasticity, however undesirable intercorrence can occour leading to specific changes in the results that should be taken into attention. To study this issue we have promoted a transient cardiogenic interruption of the blood flow together with a transient occlusion of the bilateral common carotid arteries (2VO) in rats and analysed the state of activation of astrocyte and microglia by means of the glial fibrillary acidic protein (GFAP) and OX42 immunohistochemistry, respectively. Rats were submitted to incomplete global cerebral ischemia (IGCI) by occlusion of the bilateral carotid arteries for 30 minutes. During the IGCI surgical, some rats received a higher dose of the chloral hydrate anaesthesia which promoted a cardiogenic interruption of the blood flow (CIBF) for a period of 10 minutes followed by and prompt reperfusion. During that period, animals were submited to a cardiac ventilated. Sham operation were made in control animals. Rats were killed and their brains processed 14 days after the surgery. The animals that have received a IGCI showed a slight astroglial and microglial reaction in all the hippocampal formation, however the animal submitted to CIBF showed a massive infiltration of the reactive astrocyte and microglia in CA1 subfield. This results demonstrated that a transient occlusion of the bilateral carotid arteries leads to activation of glial cells in the hippocampus, however this response can be remarkable changed in animal developing a transient systemic hypoperfusion during surgery. Thus, an accurated monitoration of the hemodinamic condition of the animal has to be done in experimental models of brain ischemia and the results have to be analysed in view of this aspect. SUBJECT HEADINGS: Astrocyte. Microglia. Cerebral ischemia. Imunohistochemistry. Image analysis. Página 1 de 15 Acta Cirurgica Brasileira Glial reaction in the hippocampus after global cardiogeni...


INTRODUCTION
In the last 10 years the neuroscience field of research has been improved substantially with the development of t new techniques, which have been employed in many experiments involving the study of neuronal trophism and plasticity.Many neurological sequeles after experimental models of neurotrauma and ischemia have been be improved by neuronal trophic and plastic responses 1,2,3,4,5 .
It has been showing that the paracrine trophic responses promoted by activated glial cells such as reactive astroc and microglia are substantially important in the mechanisms leading neurons to support injury as well as to deve subsequently neuronal plasticity 6,7 .It was shown by SANTIAGO RAMON & CAJAL, DEL-RIO HORTEGA a many other recent investigators that the modulation in the function of reactive glial cells trigger an ideal environ for a proper neuronal function 8 .Astrocyte and microglia readly react to neuronal lesion as well as to the breakdo the neuronal homestasis 9,10,11,12 .Long lasting glial reaction has also been described depending on the type and magnitude of the lesion 13,14 .Actived glial cells in an injured brain region can tigger secretion of many molecule which regulate the local inflamation.The later trophic and plastic neuronal events are dependent on the magnitud the fashion of earlier glial reaction 6,15,16,10,17,18 .Specific subfields of the hippocampal formation such as CA1 region have been showen to be particularly vunera ischemic damage 19 .Four vessel occlusion (4VO) model followed by reperfusion leads to a specific degeneration CA1 subfield of the hippocampal formation three days after the insult 20 .Furtheremore, the transient bilateral occlusion of common carotid arteries (2VO) model of brain ischemia in addition with systemic hypotension also to injury in specific brain region 21 .It has been described that glial reaction following ischemic damage also inter with the neuronal maintenance or degeneration 22,23 .
Many experimental manipulations like microneurosurgeries, stereotaxical injection, microdialisis have been perf in rodents as well as primates in order to analyse the phenomenum of brain trophism and plasticity 24,25 becomes necessary to demonstrate how undesirable events, i. e. systemic hypotension during neurosurgerical experimental procedures can promote specific lesions in the brain.
To study this issue we have promoted a 2VO of transient ischemia with or without cardiogenic interruption of flow in rats and analysed the activation of astrocyte and microglia by means of well defined markers for these gl cells such as the immunohistochemistry of the glial fibrillary acidic protein (GFAP) and OX42, respectively.Th degree of the changes was quantified by means of microdensitometric image analysis.

Incomplete global cerebral ischemia (IGCI)
Adult male Wistar rats [body weight (b.w.) 240-280 g] from the Institute of Biomedical Science (São Paulo, Bra were used in the present study.The rats were kept under controlled temperature and humidity conditions with a standardized light and dark cycle (lights on at 0700 h and off at 1900 h) and with free access to food pellets and water.Under chloral hydrate anesthesia (Merck, Germany, 0.42 mg/g, b.w.), animals were placed in a apparatus and by means of a neck midline incision, the two common carotid arteries were exposed and wound w threads without damaging the vessels and the vagus nerves.Bilateral incomplete cerebral ischemia (ICI) was promoted by looping the threads wound around the common carotid arteries for 30 minutes 21,26 .After that a com reperfusion was promptly promoted.

Cardiogenic interruption of the blood flow (CIBF)
Twenty minutes after IGCI some rats received a higher dose chloral hydrate anesthesia (Merck, Germany, 0.65 m b.w.) which promoted a cardiogenic interruption of the blood flow for a period of 10 minutes 5 .During that perio Página 2 de 15 Acta Cirurgica Brasileira -Glial reaction in the hippocampus after global cardiogeni... 13/09/2012 http://www.scielo.br/scielo.php?pid=S0102-86502001000100003&script=sci_arttext animals were submited to a cardiac massage and ventilation.Reperfusion occurs immediately after the reanimati Thus, the total period of ICI in this groups was also 30 minutes.

Immunohistochemical procedures
Fourteen days after the global ischemia, the animals were deeply anesthetized and sacrificed by a transcardiac perfusion with 70 ml isotonic saline at room temperature followed by 350 ml of fixation fluid (4€C) over a period minutes.The fixative consisted of paraformaldehyde in 0.1 M phosphate buffer, pH 6.9.The brains were remove kept in the fixative solution at 4€C for 90 minutes, rinsed in 20% sucrose (Synth, S•o Paulo, Brazil) dissolved in phosphate buffered saline (PBS), pH 7.4, for 48 h, frozen in dry ice-cooled (-40€C) isopentane (Sigma) and store 70€C until use.
Adjacent serial 60 mm thick coronal brain sections were obtained with a cryostat (Leica, CM 3000, Germany) at rostrocaudal levels -2.30 mm to -5.80 mm according to the atlas of PAXINOS AND WATSON 27 .The sections sampled systematically during sectioning.Ten series in a rostrocaudal order including every ten section were use immunohistochemistry.
Immunoreactivity was detected by the avidin-biotin peroxidase technique 28 .Floating sections were washed 2 x 1 minutes in 0.1M PBS, pH 7.4.Sections from series one and two were used to label astrocyte and microglia, respectively.The series of sections were incubated for 48 h at 4€C under shaking with a rabbit polyclonal antiser against glial fibrillary acidic protein (GFAP, Dakoparts, Denmark) diluted 1:1200 or with a mouse monoclonal antiserum against OX 42 (Harlan, USA) (Chadi et al., 1993; Cerutti and Chadi, 2000).The antibodies were dilut PBS containing 0.5% Triton X-100 (Sigma) and 1% BSA (Sigma).After that, the series of the sections were was again in PBS (2 x 10 minutes) and incubated with biotinylated either goat anti-rabbit or horse anti-mouse immunoglobulins, both diluted 1:250 (Vector, USA) for 2 hours.The sections were washed again in PBS incubated with an avidin-biotin peroxidase complex (both diluted 1:125, Vectastain, Vector, for 90 minutes).Immunoreactivity was visualized using 3-3'-diaminobenzidine tetrahydrocloride (DAB, Sigma) as a chromogen H 2 O 2 (0.05%, v/v, Sigma) for 8 minutes.The GFAP and OX 42 immunostained sections were counterstained wi cresyl violet to allow interalia the visualization of the glial cell nuclei and the neuronal cell bodies.

Semiquantitative microdensitometric analysis of the GFAP and OX42 imunoreactivities
The microdensitometric analysis was made in 3 sections on both right and left sides of the hippocampus by mean an IBAS image analyser (Zeiss-Kontron).The subfields CA1, CA2, CA3 of the piramidal cell layers and gyrus (DG) were specifically analysed.The image analysis procedures have been described previously the image was acquired by a television camera from the microscope (x 63.5 objective).After shading discrimination procedure was performed according to the following: the mean grey value (MGV) and s.e.m. of g matter in the area of the hippocampus devoided of specific labeling (background, bg) was measured.Grey darker than MGV -3 s.e.m. were considered as belonging to specific labeling and thus discriminated.The speci (sp) MGV was then defined as the difference between the bg MGV and the MGV of the discriminated profiles.I present analysis, this parameter reflects the amount per cell of GFAP or OX42 immunoreactivities plus the cresy violet staning.It must be remembered that in the absence of a standart curve, MGV only gives semiquantitative evaluations of the intensity of the immunoreactivity.

Analysis of the GFAP immunoreactivity in the sham operated rat
We found GFAP immunoreactive astroglial profiles homogeneously distributed throughout the cerebral cortical of the neocortex of sham operated rats.In the hippocampal formation, astroglial profiles were also found in all la of the CA1, CA2 and CA3 subfields as well as in the DG (Fig. 1A).The GFAP immunoreactive profiles showed moderate amount of GFAP immunoreactvity in the cytoplasm (Fig. 1B).It was possible to see thin GFAP immunoreactive processes projecting from the citoplasm of the labeled astrocytes (Fig. 1B).

Analysis of the GFAP immunoreactivity in the IGCI and CIBF rats
The GFAP immunoreactivity was not changed in the cerebral cortical layers of the neocortex of IGCI rats.Homogeneous distribution was also observed in all subfields (CA1, CA2, CA3 and DG) of the hippocampal form of the ICGI rats, however a slight astroglial reaction characterized by a small increase in the number and in the s this glial cell was found in these regions of ICGI rats (Fig. 1C).
The GFAP immunoreactive profiles was also homogeneously distributed throughout the cerebral cortical layers neocortex of the CIBF rats and they were very similar those found in that region of sham operated rats.However all subregions of the hippocampal formation of the CIBF rats the GFAP immunoreactive profiles had a cytoplasm which accumulated large amount of GFAP immunoreactive (Fig. 2B).Massive increases in the number of the GF immunoreactive astroglial profiles with enlarged cytoplasmic processes were found in the hippocampal formatio the CIBF rats (Fig. 2B).Those findings were more prominent in the CA1 subfield of the hippocampal formation CIBF rats where a massive infliltration of reactive GFAP immunoreactive astrocytes was seen (Fig. The microdensitometric analysis of the GFAP immunoreactivity demonstrated that the cerebral cardiogenic isch increases the spMGV of the GFAP immunoreactive astroglial profiles by 12.52% in the region CA2, 15.53% in CA3 and 7.43% in the DG 14 days after the ischemic insult compared to the corespondent regions of operated (Fig. 3).However, the major increase of 32% was observed in CA1 subfield of the hippocampal formati the CIBF rats (Fig. 3).

Analysis of the OX42 immunoreactivity in the sham operated rats
We found the presence of the OX42 immunoreactive microglial profiles homogeneously distributed throughout cerebral cortical layers of the neocortex and all subfields of the hippocampal formation of the sham operated rats OX42 immunoreactive profiles showed small cytoplasm which accumulated low amount of OX42 immunoreact (Fig. 4B).It was observed several delicate OX42 immunoreactive processes projecting from the cytoplasm (

Analysis of the OX42 immunoreactivity in the IGCI and CIBF rats
OX42 immunoreactive microglial profiles was seen homogeneously distributed throughout the cerebral cortical in the neocortex and throughout the hippocampal formation of the IGCI rats, however a slight microglial reactio characterized by an increased number of profiles could to be observed in those regions (Fig. 4C).
The OX42 immunoreactivity in the cerebral cortical layer of the neocortex of the CIBF rats was similar to that fo in the IGCI.However, an increased number of OX42 immunoreactive profiles showing enlarged cytoplasm and higher amount of OX42 immunoreactivity was found in all subregions of the hippocampal formation of the CIBF (Fig. 5A).Many OX 42 immunoreactive profiles had round shape and short process in the haippocampal formati the CIBF (Fig. 5B).A massive OX42 immunoreactivity was observed in the CA1 subfield of the hyppocampus o CIBF rat (Fig. 5A).The analysis of the cresyl violet stained neuronal profiles in the subregions of the hippocamp formation showed no changes in the pyramidal cell layer of the CIBF rats 14 days after surgery, however a mass disappearance of the pyramidal neurons of the CA1 region was found after CIBF (Fig. 2A and 5A).The microdensitometric analysis of the OX 42 immunoreactivity demonstrated that the cerebral cardiogenic isch increases the spMGV of the OX 42 immunoreactive microglial profiles by 9.38% in the CA2, 9.09% in the CA3 2.59% in the DG 14 days after the ischemic insult compared to the corespondent regions of the sham (Fig. 6).However, the major increase of 22,21% was observed in CA1 subfield of the hippocampal formation of CIBF rats (Fig. 6).

DISCUSSION
In this study an incomplete global cerebral ischemia was performed by means of a transient occlusion of the bila commum carotid arteries in a well characterized experimental model of brain hypoperfusion called 2 effects of the transient 2-VO performed here on the forebrain astroglial and microglial activation as well as the disappearance of pyramidal neurons of CA1 region of the hippocampal formation were remarkably potentiated b temporary cardiogenic interruption of the systemic blood flow.Using the advantage of immunohistochemistry to specifically label glial cells combined with quantitative microdensitometric image analysis we have described degree of glial activation in the most vulnerable brain regions to ischemia i.e. subfields of the hyppocampal form The disappearance of neurons stained by cresyl violet was also analysed following the transient global ischemia.
Following experimental transient brain ischemia with reperfusion (IGCI procedure), morphological and neurochemical changes take place in degenerative and survival neurons as well as in the close by glial 35, 36, 37, 38, 39 .The degree of the changes can varie depending on the resistance of a particular neuronal cell population 40 as well as on the locally inflammatory-mediated responses 41 .
In an experimental point of view, it has also to be emphasized that regarding the effects of a transient global isch the hemodinamic conditions of the laboratory animals during surgery may interfere substantially with the analys the results.
The 2-VO model of brain schemia in rats employed in this study has been extensively used in order to analyse th mechanisms triggering neuronal death or maintenance 21 .
Another model of transient global brain ischemia called 4-VO has been also performed in rats when higher degre lesion is desier 42 .In this model, a permanent occlusion of the vertebral arteries is followed by a transient occlusi the common carotid arteries, bilaterally.It has also to be mentioned that the high level of mortality (80%) accompanied 4-VO procedure may sometimes make it difficult to elaborate more complex biochemical and beha experiments.
It has to be considered that anesthetic agents favorably effect outcome from brain ischemia 43 even though this m be the case of chloral hydrate employed in the present work which could not prevent further damage in the CA1 neurons of the hyppocampal formation after cardiogenic ischemia.
In the case of more severe ischemia followed by reperfusion, it is well known that neurons of neocortical and 6, small to medium striatal neurons and hippocampal pyramidal neurons of the CA1 and CA4 regions are m susceptible to schemic damaged 20,26 .
Ischemia produced by bilateral carotid artery occlusion as performed in the present analysis is able to increase th concentration of the extracellular amino acids glutamate, aspartate, GABA and taurune which in turn is modulat the stimulation of adenosine A1 receptors 44 .Furthermore, a permanent occlusion of both common carotid arterie reduces the muscanirric acetylcoholine receptor binding in the frontal cortex and hyppocampus 12 weeks after su with learning impairment showed by the hypoperfusioned rats 45 .The heat shock protein 70 that is associated several celular processes, including DNA replication and transport of proteins across membrans, is expressed in CA1, CA3 and CA4 pyramidal neurons of the hippocampus following a transient forebrain ischemia promoted b VO model 46 .
It has been described that prior the death of CA1 neurons i.e. 24 hours post ischemia (four vessel occlusion), sub of the hippocampus show calpain mediated and spectrum breakdown products, an increased silver staining and a decreased neurophysiological response to afferent stimulation 47 .
Lipid peroxidation takes place in brain regions where iron is deposited late after transient forebrain ischemia Because an accumulation of calcium is implicated in excitotoxic cell death, many studies have attempted to corr the vulnerability of neurons with the presence or absence of the calcium binding proteins parvalbumin and calbin because of their calcium-buffering abilities 40 .
Other fact to be considered is that the different model/intensity of brain schemia regimes may lead to a differenti pore-like opening of the blood-brain barrier 48 which in turn may also be correlated with the selective neuronal de by changing the clearance and/or diffusion of neurotrophic and neurotoxic substances at the ischemic A massive diminution of the pyramidal neurons stained by cresyl violet of the CA1 region of the hyppocampus together with a remarkable astroglial and microglial activation in this region of the CIBF rats observed in study demonstrated that the intensity of the effects promoted by the 2-VO model of ischemia may be potentiated systemic hypoperfusion.The reduction of the mean arterial blood pressure to 40 mmHg by hypovolenic has been associated with a 15 minutes occlusion of both common carotid arteries to perform a experimental bilat incomplete cerebral ischemia 21 .Página 10 de 15 Acta Cirurgica Brasileira -Glial reaction in the hippocampus after global cardiog... 13/09/2012 http://www.scielo.br/scielo.php?pid=S0102-86502001000100003&script=sci_arttext The reaction of glial cells, i.e. the astrocytic response, has commonly been described following an injury of the c nervous system 14 .Animals submitted to cerebral ischemia models have showed astroglial and microglial activati selective vulnerable brain regions 51,49,52 .Following a global cerebral ischemia, the insult of CA1 subfield of the hippocampal formation leads to a local infiltration of microglia and astrocyte 51, 49 .
In the present analysis a higher degree of astroglial and microglial reaction was found in the CA1 subfield of the ischemic rat submitted an additional cardiogenic hypoperfusion of the blood flow, which can be correlated with degree of CA1 lesion, since a major disappearance of CA1 neurons was in the CIBF rats.
The upregulation of the synthesis of basic fibroblast growth factor (bFGF) by reactive astrocytes, a neurotrophic factor with actions on hippocampal neurons 53 was described in the ischemic hippocampus following ischemia the other hand, reactive astrocytes can synthesize increased amount of endotelin (ET) 1 and 3 in the CA1 damag region after ischemia as an increased binding of ET is seen in activated microglial aggregation on damaged pyra cell layer of this region 52 .These observation may help to explain the massive glial activation in the CA1 subfield transient global ischemia potentiated by cardiogenic hypoperfusion.

FIGURE 1 Figure 1 -
FIGURE 1Figure 1 -Digital images showing the glial fibrillary acid protein (GFAP) immunoreactivity counterstained with cresyl violet (A-C) in the right hippocampal formation of a sham operated rat (A,B) and of an animal submitted to a transient occlusion of the bilateral common carotid arteries for 30 minutes followed by reperfusion (IGCI group, C).The animal was sacrificed 14 days after the surgery.Quiescent (B) and reactive (C) astrocytes are pointed.Barrs: 200 mm (A) and 10 mm (B,C).

FIGURE 2 Figure 2 -
FIGURE 2 Figure 2 -Digital images showing the glial fibrillary acid protein (GFAP) immunoreactivity counterstained with cresyl violet (A-B) in the right hippocampal formantion of an animal submitted to a 20 minutes occlusion of the bilateral common carotid arteries plus 10 minutes of cardiogenic interruption of the systemic blood flow (CIBF).The animal was sacrificed 14 days after the surgery.A strong GFAP immunoreactivity is observed in the CA1 region of the hippocampus arrows (A).Reactived astrocytes are showed in the CA1 subfield of the hippocampus in B (arrowsheads).Bars: 200 mm (A) and 10 mm (B).

FIGURE 3 Figure 3 -
FIGURE 3 Figure 3 -Figure shows the increase of the mean grey value (MGV) of the astroglial profiles in the CA1, CA2, CA3 subregions and the denteate gyrus (DG) of the hippocampal formation of the rat submitted to a 20 minutes occlusion of the bilateral common carotid plus a 10 minutes of cardiogenic interruption of the systemic blood flow (CIBF) compared to sham operated.Astrocytes were labeled with glial fibrillary acid protein (GFAP) immunohistochemistry. Microdensitometric analysis was performed in an image analyser (for details see text).

Figure 4 -
Figure 4 -Digital images showing the OX 42 immunoreactivity counterstained with cresyl violet (A-C) in the right hippocampal formation of a sham operated rat (A,B) and of an animal submitted to a transient occlusion of the bilateral common carotid arteries for 30 minutes followed by reperfusion (IGCI group, C).The animal was sacrificed 14 days after the surgery.Quiescent (B) and reactive (C) microglias are pointed.Barrs: 200 mm (A) and 10 mm (B,C).

FIGURE 5 Figure 5 -
FIGURE 5Figure 5 -Digital images showing the OX 42 immunoreactivity counterstained with cresyl violet (A-B) in the right hippocampal formantion of an animal submitted to a 20 minutes occlusion of the bilateral common carotid arteries plus 10 minutes of cardiogenic interruption of the systemic blood flow (CIBF).The animal was sacrificed 14 days after the surgery.A strong OX 42 immunoreactivity is observed in the CA1 region of the hippocampus arrows (A).Reactived microglial profiles are showed in the CA1 subfield of the hippocampus in B (arrowsheads).Barrs: 200 mm (A) and 10 mm (B).

FIGURE 6 Figure 6 -
FIGURE 6 Figure 6 -Figure shows the increase of the mean grey value (MGV) of the microglial profiles in the CA1, CA2, CA3 subregions and the denteate gyrus (DG) of the hippocampal formation of the rat submitted to a 20 minutes occlusion of the bilateral common carotid plus a 10 minutes of cardiogenic interruption of the systemic blood flow CIBF compared to sham operated.Microglias were labeled with OX 42 immunohistochemistry.Microdensitometric analysis was performed in an image analyser (for details see text).