Changes of cortisol content in plasma and hippocampus after focal cerebral ischemia-reperfusion in rats

AIM: To investigate the relationship between glucocorticoid (Gc) and injury of hippocampus neurons and the effect of Gc on dementia episode after cerebral ischemia-reperfusion. METHODS: The rat model of middle cerebral artery occlusion (MACO) was established. Cortisol contents in hippocampus and plasma of the model rats were examined by means of the radioimmunoassay at 2 h, 6 h, 12 h, 24 h after reperfusion. RESULTS: The levels of cortisol content in model group were significantly higher than those in sham group and normal group both in hippocampus and plasma. The highest cortisol content was observed at 6 hours after reperfusion. HE staining showed that the impairment of hippocampus neurons was aggravated progressively with reperfusion interval elongating. CONCLUSION: The increased cortisol in hippocampus and plasma, after 2 h cerebral ischemia and 24 h reperfusion, could aggravate the injury of hippocampus neurons and lead to dementia post stroke.     

Changes of Ski expression levels in rat activated astrocytes

AIM: To explore the time-dependent change of Ski protein expression in normal and activated astrocytes in rats.METHODS: The astrocytes were obtained from rat cerebral cortex and cultured in vitro. The astrocytes were treated with LPS and scratch injury for activation. Western blot analysis was used to determine glial fibrillary acidic protein (GFAP) and Ski protein levels in activated astrocytes at a series of time points. The indirect immunofluorescence staining method was performed to detect the location of Ski protein in the astrocytes.RESULTS: The protein of GFAP was naturally expressed in the astrocytes, beginning to increase after treated with LPS and scratch injury. Little protein expression of Ski in the normal astrocytes was observed. The Ski protein expression began to increase after treated with 1 mg/L LPS, peaked at 4 d (P<0.05) and then deceased, but was stills higher than that in the normal cells. The protein expression level of Ski after scratch injury was highly consistent with above mentioned. Ski was mainly observed in the nucleus of the normal cells and the cells treated with LPS for 6 d, while it was observed in the cytoplasm 2 and 4 d after treated with LPS.CONCLUSION: The protein of Ski is expressed in the astrocytes, and the expression level is increased in activated astrocytes, mainly located in the nucelus. Ski may plays an essential roles in the processes of activation and proliferation of astrocytes.     

Effect of dexmedetomidine on astrocytes in rats with focal cerebral ischemia-reperfusion

AIM: To investigate the effects of dexmedetomidine on astrocytes in rats with focal cerebral ischemia-reperfusion. METHODS: Sixty female SD rats, weighing 230~250 g, were randomly divided into sham operation group, ischemia-reperfusion group, dexmedetomidine preconditioning group 1 and dexmedetomidine preconditioning group 2. The model of middle cerebral artery occlusion (MCAO) was established by thread embolism of middle cerebral artery. In sham operation group, the carotid arteries were exposed without performing MCAO. In ischemia-reperfusion group, NS was injected intraperitoneally 30 min before focal cerebral ischemia-reperfusion. The rats in dexmedetomidine preconditioning group 1 and dexmedetomidine preconditioning group 2 received intraperitoneal injection of dexmedetomidine at doses of 20 μg/kg and 40 μg/kg, respectively. The neurological scores were studied, and the pathological changes were observed under microscope with HE staining. The expression of glial fibrillary acidic protein (GFAP) and tumor necrosis factor α (TNF-α) in astrocytes was detected by the methods of immunohistochemistry and immunoblotting 24 h after cerebral ischemia-reperfusion. RESULTS: No neurological change was observed in sham operation group. The neurological deficiency scores in ischemia-reperfusion group were markedly higher than those in dexmedetomidine preconditioning group 1 and group 2 (P<0.05). Compared with sham operation group, the expression of GFAP and TNF-α in astrocytes and the level of GFAP increased significantly 24 h after focal cerebral ischemia-reperfusion. Pretreatment with dexmedetomidine significantly attenuated the expression of GFAP and reduced the infarct size and inflammation. CONCLUSION: Dexmedetomidine has a neuroprotective effect on focal cerebral ischemia-reperfusion injury by inhibiting the astrocytes.     

Expression of glial fibrillary acidic protein in developing rat brain after intrauterine infection

AIM: To investigate the alteration of glial fibrillary acidic protein (GFAP) immunoreactivity in developing rat brain after intrauterine infection. METHODS: Escherichia coli (E.coli) was inoculated into uterine horn of pregnant rats when gestation was 15 days and the control group was inoculated with normal saline. Immunohistochemistry was used for evaluation of GFAP expression in pup brains at postnatal day 1 (P1), P3, P7, P14, P21. RESULTS: GFAP-immunopositive cells was scarce in the periventricular white matter at P1 and P3 in two groups (P>0.05), but not in other brain regions. The number of GFAP-immunopositive cells of the E.coli-treated pups was markedly increased in periventricular white matter and hippocampus at P7 compared with the control group (P<0.05). The E.coli-treated pups at P14 showed a marked increase in GFAP expression in periventricular white matter, corpus callosum and cortex (P<0.01). However, no significant different levels of GFAP expression in any brain regions were found at P21 between two groups (P>0.05). CONCLUSION: Intrauterine infection induces an increased expression of GFAP in the neonatal brain.     

Effects of baicalin on glial fibrillary acidic protein and nuclear factor-κB expression in juvenile rat hippocampus after status convulsion

AIM: To study the effects of baicalin (BC) on glial fibrillary acidic protein (GFAP) and nuclear factor-κB (NF-κB) expression and neuronal apoptosis in juvenile rat hippocampus after status convulsion (SC). METHODS: One hundred and ninety five juvenile male Sprague-Dawley rats were randomly divided into 3 groups: normal saline pretreatment group (NS group), SC group and SC with BC pretreatment group (BC group). Each of these 3 groups would be subdivided into 5 subgroups sacrificed at 4 h, 12 h, 24 h, 48 h and 72 h after SC. The rat SC model was prepared by lithium-pilocarpine chemical method. The protein expression of GFAP and NF-κB was detected by the method of immunohistochemistry. The mRNA expression of GFAP was detected by RT-PCR. The neuronal apoptosis was observed by TdT-mediated dUTP nick end labeling (TUNEL). RESULTS: Compared with NS group, the GFAP positive cells was increased in SC group (P<0.05). Compared with SC group, the expression of GFAP was significantly reduced in BC group (P<0.05). Compared with NS group, the NF-κB positive cells was increased in SC group (P<0.05). Compared with SC group, the expression of NF-κB was significantly reduced in BC group. RT-PCR showed that the expression trend of GFAP mRNA was similar to that of the protein. Compared with NS group, the TUNEL positive cells in the hippocampus CA1 area in SC group increased significantly 12 h after SC (P<0.01), and reached a peak at 48 h. After the intervention with BC, the TUNEL positive cells decreased significantly between 12~48 h after SC (P<0.05 or P<0.01), but the number of TUNEL positive cells remained significantly greater than that in NS group (P<0.05). CONCLUSION: The expression of GFAP and NF-κB in the hippocampus increased after SC in rats. Baicalin decreases the expression of GFAP and NF-κB in hippocampus of rats with pilocarpine-induced seizures, and reduces the number of neuronal apoptosis, suggesting that baicalin may protect against the brain damage caused by status convulsion.     

Effects of adrenal gland on the expression of bax and bcl-2 in hippocampus after cerebral ischemia

AIM: To observe the effects of adrenal gland on the hippocampus responses to cerebral ischemia. METHODS: 36 Wistar rats were randomly divided into three groups: sham-operated control group (sham), unilateral adrenalectomy were performed in ADX and GC group, and GC group were injected with 5 mg/per rat of dexamethasone before cerebral ischemia. Fourteen days after the first operation, all animals were performed occlusion of bilateral carotid artery for 15 min, and then reperfusion. 3 rats of each group were sacrificed at 1 h, 4 h, 8 h and 24 h after reperfusion and hippocampus were dissected. The total RNA was rapidly extracted from hippocampus tissue. The expressions of c-fos, bcl-2 and bax gene were quantified with the method of semiquantitive RT-PCR. RESULTS: The expressions of c-fos and bax in three groups showed no statistical differences (P>0.05). The expression of bcl-2 in sham group was significantly higher than that in GC and ADX groups (P<0.05). However, no differences of bcl-2 expression between GC and ADX group (P>0.05) was observed. The ratio of bax to bcl-2 in sham group was significantly lower than that in GC and ADX groups (P<0.05), no significant differences of the ratio displayed between ADX and GC group. CONCLUSION: The expression of c-fos and bax in hippocampus after cerebral ischemia is not affected by adrenal gland. The excision of unilateral adrenal gland downregulates bcl-2 expression and raises the ratio of bax to bcl-2 in rat hippocampus after cerebral ischemia. Dexamethasone treatment does not alter the expression of bcl-2 in ADX and GC groups. The results indicate that the adrenal gland can counteract cell apoptosis in hippocampus tissue induced by cerebral ischemia. Adrenal steroids are not sufficient to enable the compensatory increase in bcl-2 expression in steroid-deficient animal, some other mechanism may exist.     

Therapeutic action of transplantation of marrow mesenchymal stem cells into infarction area after cerebral middle artery occlusion in rats

AIM: To explore the differentiation and the functional behavior of marrow mesenchymal stem cells (MSC) transplanted into the cerebral infarction area after cerebral middle artery ischemia in rats. METHODS: MSC were isolated from human rib marrow and cultured in L DMEM medium in vitro. The model of rat cerebral infarction by cerebra middle artery occlusion was established, and the identified MSC were transplanted intracerebrally 10 days later. Immunohistochemistry technique was used to identify the cell survivor and its differentiation to the neurogenesis in the transplantation site, and at 2 weeks and 6 weeks after transplantation, the functional tests were comparatively studied. RESULTS: The results showed that the survivor of transplanted MSC was differentiated to neural phenotype cells, and the functional behavior of the injury rats was recovered significantly after MSC transplantation (P<0.05). CONCLUSION: Our data suggest that transplantation of MSC may be a powerful autoplastic therapy for the stroke.     

Losartan inhibits LPS-induced GFAP expression via AMPK activation in mouse hippocampus

AIM: To investigate the effects of losartan on lipopolysaccharide (LPS)-induced glial fibrillary acidic protein (GFAP) expression, and to determine whether adenosine 5'-monophosphate (AMP)-activated protein kinase (AMPK) activation is involved in the mechanism.METHODS: Adult male KM mice were divided into control group, LPS model group, losartan treatment group, and losartan and Compound C co-treatment group. To establish a model of central nervous system inflammation, the mice received daily intracerebroventricular injection of LPS (24 μg/d) for 2 d. Daily losartan administration (0.5, 1 or 5 mg·kg^-1·d^-1, ip) initiated at 14 d prior to LPS injection. Compound C (10 mg/kg, ip), a selective AMPK inhibitor, started to be injected daily at 2 d prior to LPS injection. The hippocampal tissues in each group were isolated at 3 d after the last LPS injection, and then the protein levels of GFAP, AMPK, p-AMPK, mammalian target of rapamycin (mTOR) and p-mTOR were determined by Western blot.RESULTS: Twice LPS injections significantly increased the expression of GFAP in the hippocampus (P<0.01). Losartan inhibited LPS-induced GFAP expression in a concentration-dependent way, and losartan at 5 mg·kg^-1·d^-1 significantly inhibited GFAP expression and AMPK activation (P<0.05), but it had no obvious effect on mTOR activation. Furthermore, Compound C significantly reversed the effect of losartan treatment on LPS-induced GFAP expression and AMPK phosphorylation (P<0.05).CONCLUSION: Losartan inhibits LPS-induced GFAP expression in the mouse hippocampus, and AMPK activation but not mTOR, is involved in the mechanism.     

Selenium-enriched Spirulina platensis promotes proliferation of hepatocytes in rat partial hepatectomy

AIM: To explore the effects of selenium-enriched Spirulina platensis (Se-SP) on proliferation of hepatocytes in rat hepatectomy. METHODS: Rat hepaectomy model was conducted using male Wistar rats. The rats were randomized into five groups: operation groups with 150 (H), 50 (M) and 15 (L) mg·kg^-1·d^-1 of Se-SP, placebo-control (P) and sham operation group (F). Activities of glutathione peroxidase (GPx) and thioredoxin reductases (TR) in hepatocytes were determined by chemical colorimetry. The expression index of proliferating cell nuclear antigen (PCNA) in hepatocytes was detected by immunohistochemistry, and the level of ［³H］-TDR incorporation in regenerative hepatocytes was analyzed by radio-immunity. RESULTS: Activity of GPx and TR, PCNA expression index as well as ［3H］-TDR insertion in hepatocytes (in vitro) were obviously higher (P<0.05) in L groups than those in P and F groups. All parameters were significant changed (P<0.05) after operation in H, M, L and P groups whereas some slightly change in F group. Furthermore, correlation analysis showed that the levels of GPx and TR in hepatocytes all showed positive correlation with PCNA expression (r²=0.77 and 0.87, respectively) and with ［³H］-TDR incorporation level (r² = 0.73 and 0.84, respectively) in hepatocytes. CONCLUSION: Se-SP enhances hepatocyte proliferation in rat hepatectomy, up-regulation of selenoenzymes might be responsible for this effect.     

Relation between astroglial activation state and ischemic tolerance in the gerbil hippocampus

AIM:To observe the relation between astroglial activation state and ischemic tolerance in the gerbil hippocampus. METHOD: Bilateral occlusion of common carotid arteries and immunofluorescent methods in the gerbil hippocampal tissue. Slices were used. The morphological changes of the neurons were observed by light microscopy.RESULTS:Pretreatment with 2 minute bilateral carotid artery occlusion produced protective effects of CA₁ neurons. Glial fibrillary acidic protein (GFAP) staining showed weak positive cells in control group. Most of GFAP positive cells showed weak and middle positive cells after recirculation 1d and 2d following ischemic 3.5 min and preconditioning of the brain with sublethal ischemia respectively.CONCLUSION:Astroglial played an important role in cerebral ischemia. It is possible that state of astroglial activation related to neuronal survival in ischemic tolerance.     

Effects of β-ME and RA on expression of GFAP in mesenchymal cells derived from mouse fetal liver

AIM: To investigate the effects of β-mercaptoethanol (β-ME) and all-trans rentinal acid (RA) on glial fibrillary acidic protein (GFAP) expression in mesenchymal cells derived from mouse fetal liver in vitro. METHODS: Cells suspension from 14.5-days-old mouse fetal liver were cultured in DMEM/HEPES/F12 supplemented with 20％ FCS and mesenchymal cells were acquired after discarding nonadherent cells. The 5th passage cells were induced by β-ME and RA. The characteristics of treated cells were assayed by immunocytochemistry staining at 5 hours and 5 days after induction. β-actin as an internal control, GFAP gene expression of mesenchyal cells was detected with semi-quantitative RT-PCR. RESULTS: After being inducted by β-ME and RA, 80％ approximately of the cells exhibited typical neural morphology and about 85％ expressed GFAP phenotype. Semi-quantitative RT-PCR showed that mRNA expression of GFAP increased in treated cells versus untreated cells (P<0.01). CONCLUSION: GFAP expression in mesenchymal cells derived from mouse fetal liver in vitro increases after being treated with β-ME and RA.     

Effects of ischemic postconditioning on hippocampus CA1 area rCBF and astrocyte activation after cerebral ischemia in tree shrews

AIM: To study the effects of ischemic postconditioning (PC) on regional cerebral blood flow (rCBF) and astrocyte (AS) activation in hippocampus CA1 area and to explore the possible mechanism of ischemic PC affecting glial fibrillary acidic protein (GFAP) expression during focal cerebral thrombosis. METHODS: The thrombotic focal cerebral ischemia was induced by photochemical reaction in tree shrews, and ischemic postconditioning was established by cliped ipsilateral carotid of the animal at 4 h after cerebral ischemia. The rCBF and GFAP expressions in hippocampus CA1 area were detected, respectively, by laser-Doppler (LD) fowmeter and immunohistochemistry. RESULTS: The numbers of GFAP positive cells were increased markedly and GFAP expression enhanced (P<0.01). AS oncosis was apparent 24 h after cerebral ischemia. Postconditioning increased hippocampus rCBF from (2.55±0.28) PU to (10.42±3.75) PU (P<0.05) at 24 h and from (9.84±1.22) PU to (18.74±1.60) PU (P<0.05) at 72 h after the cerebral ischemia, and AS oncosis was inhibited markedly. CONCLUSION: Multiple, short, regional carotid occlusions may prolong “time window” of therapeutic cerebral ischemia. The protection mechanism of the ischemic postconditioning may be associated with the increase in rCBF and improvement of hippocampus microenvironment by regulating AS activation.     

Effects of betaine on expression of GFAP,glycine and glycine receptor in hippocampus of pentylenetetrazole-induced epileptic rats

AIM：To study the effects of betaine on glial fibrillary acidic protein (GFAP), glycine (Gly) and glycine receptor (GlyR) expression in the hippocampus of rats with epilepsy induced by pentylenetetrazole (PTZ). METHODS：Forty-eight healthy male Wistar rats were randomly divided into control group, PTZ (35 mg·kg-1·d-1, intraperitoneal injection) group, PTZ+betaine (450 mg·kg-1·d-1, intragastric administration) group, PTZ+betaine (225 mg·kg-1·d-1, intragastric administration) group, PTZ+betaine (112.5 mg·kg-1·d-1, intragastric administration) group and PTZ+sodium valproate (200 mg·kg-1·d-1, intragastric administration) group. The rats in control group were intraperitoneally injected with saline at the same volume as PTZ injection, and those in control group and PTZ group received intragastric administration of saline at 1.0 mL·d-1. Rat behavior was recorded. Serum homocysteine (Hcy) level was measured. The expression of GFAP in the hippocampus was measured by immunofluorescence. Hippocampal Gly content was measured by an amino acid analysis system. The expression of GlyR was detected by immunofluorescence and Western blotting. RESULTS：There was no difference in the latency of grand mal seizures among groups (P>0.05). However, betaine treatment significantly decreased the duration of the first grand mal seizure compared with PTZ group (P<0.01). Serum Hcy level in PTZ group was significantly lowered compared with control group (P<0.01), and further decreased after betaine treatment (P<0.05). GFAP in PTZ group was significantly higher than that in control group (P<0.01), and decreased after betaine treatment (P<0.05). Gly in PTZ group was significantly lowered compared with control group (P<0.01), and increased after betaine treatment (P<0.05). The content of GlyR among groups showed the same trend as Gly. CONCLUSION：Betaine treatment shows antiepileptic effect, which may be related to its effects on the metabolites of Hcy and Gly.     

Astragalus injection inhibits expression of Apaf-1 in rat hippocampus after cerebral ischemia and reperfusion

AIM: To investigate the effect of Astragalus injection on the expression of apoptotic protease-activating factor 1 (Apaf-1) in the hippocampus of global cerebral ische-mia-reperfusion rats. METHODS: Male SD rats were randomly divided into 4 groups with 30 each: sham operation group, cerebral ischemia-reperfusion group, cerebral ischemia-reperfusion+Astragalus injection group, and cerebral ischemia-reperfusion+vehicle group. The global cerebral ischemia-reperfusion model of the rats was established by 4-vessel occlusion. The rats in cerebral ischemia-reperfusion group, cerebral ischemia-reperfusion+Astragalus injection group and cerebral ischemia-reperfusion+vehicle group were further divided into 7 subsets, according to the reperfusion time of 0 h, 0.5 h, 2 h, 6 h, 24 h, 72 h and 120 h. After reperfusion, the brains were removed at the corresponding time points. The protein expression of Apaf-1 in hippocampal neurons was detected by immunohistochemistry and Western blotting. The mRNA expression of Apaf-1 was observed by RT-PCR. RESULTS: Compared with sham operation group, the expression of Apaf-1 at mRNA and protein levels at all time points except 0 h and 120 h increased obviously in cerebral ischemia-reperfusion group (P<0.05). Compared with cerebral ischemia-reperfusion group, the expression of Apaf-1 at mRNA and protein levels at all time points except 0 h and 120 h decreased obviously in cerebral ischemia-reperfusion+Astragalus injection group (P<0.05). However, those in cerebral ischemia-reperfusion+vehicle group had no obvious change (P>0.05). CONCLUSION: Astragalus injection inhibits the expression of Apaf-1 at mRNA and protein levels in hippocampus of global cerebral ischemia-reperfusion rats, thus inhibiting the apoptosis of hippocampal neurons.     

Role of ERK on proliferation of airway smooth muscle cells in chronic asthmatic rats

AIM: To investigate the roles of extracellular signal-regulated kinase(ERK) signaling pathway on regulating proliferation of airway smooth muscle by observing the expression of ERK in airway smooth muscle(ASM) in chronic asthmatic rats.METHODS: Airway remodeling was detected in chronic asthmatic rats by using image analysis system. The expressions of ERK and proliferating cell nuclear antigen(PCNA) in lung tissue from chronic asthmatic rats were observed by immuocytochemistry staining. The expressions of ERK1/2, p ERK1/2 and PCNA were detected in airway smooth muscle (ASM) by immunofluorescence double staining with confocal microscopy, and the expressions of protein or mRNA of ERK and PCNA in ASM were also detected by immunoblotting and hybridization in situ,respectively.RESULTS: The thickening of smooth muscle and structural remodeling in airway were observed in chronic asthmatic rats by image analysis. The enhanced expressions of ERK and PCNA appeared obviously increased in same lung tissue and the expressions of protein or mRNA of ERK and PCNA were significantly increased in ASM.CONCLUSION: ERK signal pathway might be an important pathway on regulating cell proliferation of ASM resulting in asthmatic airway remodeling.     

Influence of hypoxia-inducible factor-1 alpha on lung cancer cell A549 growth in vitro and in vivo

AIM: To study the role of hypoxia-inducible factor-1alpha(HIF-1α) on lung cancer cells A549 growth in vitro and in vivo. METHODS: To observe the growth rate of A549 cells after HIF-1α transfected, A549 cells (1×10⁶/mouse) were inoculated subcutaneously into 20 nude mice, which were randomly divided into two groups: the control group (group A, n=10), the HIF-1α transfected group(group B, n=10). The weights of subcutaneous tumor were detected. The resected specimens were made into paraffin-embedded sections. The proliferating cell nuclear antigen (PCNA) was identified by immunohistochemistry(ISH). The expressions of HIF-1α、 apoptosis-related protein survivin and bcl-2 were analyzed by Western blot. RESULTS: The growth rates of the HIF-1α transfected lung cancer cells A549 were significantly increased, and more importantly, the HIF-1α transfected lung cancer cells A549 was able to enhance lung cancer growth in nude mice(P＜0.05). The PCNA were increased significantly in group B, compared with group A. The expressions of HIF-1α, survivin and bcl-2 in group B were increased significantly than that of group A. CONCLUSIONS: HIF-1α increases lung cancer cells A549 growth in vitro and in vivo and its mechanism may be due to promotion of proliferation and inhibition of apoptosis.     

Postconditioning modulates TLR4 expression in hippocampus of tree shrews with thrombotic cerebral ischemia

AIM: To observe the effects of thrombotic cerebral ischemia and postconditioning on the expression of toll-like receptor 4 (TLR4) in hippocampus of tree shrews.METHODS: The model of thrombotic focal cerebral ischemia was established by photochemical reaction.Four hours after the onset of photochemical reaction, ischemic postconditioning was induced by 3 repeated cycles of carotid artery occlusion for 5 min and reperfusion for 5 min. The histological changes of hippocampus (by HE staining), TLR4 protein level (by Western blotting) and TLR4 mRNA expression (by semiquantitative RT-PCR) were observed.RESULTS: The extensive neuronal degeneration in hippocampus was observed from 4 h to 72 h and peaked at 24 h after cerebral ischemia, but was significantly attenuated after postconditioning. Cerebral ischemia caused a progressive increase in the expression of TLR4 protein at 4 h and 24 h (P<0.05), and decreased at 72 h (P<0.05). In contrast to ischemia groups, postconditioning decreased the expression of TLR4 protein at 4 h and 24 h (P<0.05), but an increase in the expression of TLR4 at 72 h (P<0.05) was observed. Simultaneously, the level of TLR4 mRNA in hippocampus showed the tendency of approximate variation in accordance with the protein expression.CONCLUSION: The expression of TLR4 increases by cerebral ischemia. The protection mechanisms of postconditioning may be associated with modulating TLR4 expression.     

Effects of Astragalus injection on neuronal apoptosis and expression of JNK3 in rat hippocampus after cerebral ischemia reperfusion

AIM:To investigate the effects of Astragalus injection on neuronal apoptosis and expression of c-Jun N-terminal kinase 3(JNK3) in the rat hippocampus after cerebral ischemia reperfusion. METHODS:The rat model of cerebral ischemia reperfusion was set up by a four-vessel occlusion method. The SD rats were randomly divided into 4 groups:sham operation group, cerebral ischemia reperfusion group(model group), cerebral ischemia reperfusion+Astragalus injection group(Astragalus injection group) and cerebral ischemia reperfusion+vehicle group(vehicle group). The rats in model group, Astragalus injection group and vehicle group after transient global cerebral ischemia(30 min) were then divided into 7 subgroups according to the reperfusion time of 0 h, 0.5 h, 2 h, 6 h, 24 h, 72 h and 120 h. The apoptosis of the neuron in the hippocampus was measured by the method of TUNEL staining. The expression of JNK3 at mRNA and protein levels was determined by real-time PCR and Western blotting,respectively. RESULTS:Compared with sham operation group, the number of apoptotic neurons increased in model group(P<0.05). Compared with model group, the number of apoptotic neurons decreased obviously in Astragalus injection group(P<0.05). Compared with sham operation group, the expression of JNK3 at mRNA and protein levels in the hippocampus increased obviously in model group at all time points except 120 h(P<0.05). Compared with model group, the expression of JNK3 at mRNA and protein levels in the hippocampus decreased obviously in Astragalus injection group at all time points except 120 h(P<0.05). CONCLUSION:Astragalus injection decreases neuronal apoptosis in rat hippocampus after cerebral ischemia reperfusion by inhibiting the expression of JNK3 at mRNA and protein levels.