1 Introduction Cerebral ischemia can lead to the decline of learning and memory function,but the brain areas associated with learning and memory and their roles in these functions have not yet been clarified. Results of some biochemical studies have indicated that acetylcholine (AchE),γ-amino butyric acid (GABA),superoxide dismutase (SOD),and malondialdehyde (MDA) in the frontal lobe and hippocampus as well as the changing levels of certain amino acids are involved in the mechanism underlying the decline of learning and memory function during cerebral ischemia^[1]. Protein synthesis decreased after cerebral ischemia,and the expression of some genes and gene products such as glucose-binding proteins,calcium-binding protein D28 as well as immediate early genes such as c-fos,c-jun, jun B,and jun D significantly increased. At the early stage of impermanency cerebral ischemia,the expression of the protein Bcl-2 was increased in the hippocampus^{[2, 3]}; however,there are also some contradictory views ^[4]. The aim of this study was to gain a better understanding of the effect of reduced regional cerebral blood flow (rCBF) on learning and memory function and its mechanism.

2 Materials and methods 2.1 Animals Sixty-four healthy male Sprague Dawley (SD) rats that were active,sensitive to electrical re-activity,with swift escape responses during the Y type electric labyrinth test,and an average weight of 308 ± 54 g were selected. These rats were randomly divided into 2 groups,an operation group (model group) and a false operation group (control group); each group consisted of 32 rats. The operation group and the false operation group were randomly divided into A,B,C, D and A₀,B₀,C₀,D₀,respectively,with each subgroup consisting of 8 rats.

2.2 Methods 2.2.1 Establishing of animal models All selected animals were fasted for 12 h and deprived of water for 4 h before the experiment,and all were given ventral anesthesia using 1% of pentobarbital (40 mg/kg). Then,the rats were placed on an operation table on their backs,and were dissected from the middle of their necks. The rats in the operation group had their common carotid arteries on both sides exposed and carefully ligated with a silk thread of size “0”,which were then imbedded subcutaneously, and the incisions were sutured. However,the rats in the false operation group were only dissected to expose their common carotid artery,and then,their incisions were sutured.

2.2.2 Measurement of learning and memory function The learning and memory function was measured as previously reported. The following indexes were recorded: training times,error number (EN),time to reach the target,total reaction time (TRT),and initiative evasion rate(IER).

2.2.3 Measurement of rCBF After the evaluation of the learning and memory function in each group,fiber probes of type Periflux PF3 laser Doppler perfusion (LDP) were inserted into the corresponding cerebral regions; the orientation of the cerebral regions was according to the cerebral orientation atlas for rat. A continuous measurement of rCBF was recorded for 10 min in the frontal lobe and hippocampus of each rat and the output signals of LDP were noted,and the average 10 min rCBF for each rat was calculated using the PERISOFT procedure.

2.2.4 Detection of c-jun,c-fos,and Bcl-2 (1) Preparation of tissue slices: After anesthesia with pentobarbital (1%,40 mg/kg),the rats were perfused transcardially. After perfusion,the right hippocampus and frontal lobe were carefully excised and fixed in the polyformaldehyde (4%) at 4 ℃ overnight. Then, the brain tissues were cryoprotected with 20% and 30% sugar in 0.1 mol/L phosphate buffer saline (PBS) at 4 ℃,respectively. The 40 μm thick cryosections were prepared for the analysis of c-jun,c-fos,Bcl-2,and Bax expression levels.

(2) Immunohistochemistry for c-fos and c-jun: The experimental steps were as follows: (a) the dewaxed slices were treated with 3% hydrogen peroxide (H₂O₂) for 10 min to destroy the endogenous peroxidase activity; (b) antigen retrieval was performed by boiling the slices for 10 min in EDTA in a microwave,and then,naturally cooling it to room temperature; (c) the slices were placed in natural sheep blood serum (10%) at room temperature for 30 min; (d) the slices were then incubated in rabbit anti-c-jun (1:200) and anti-c-fos (1:50) antibodies for 24 h at 4 ℃; (e) incubation with the secondary antibody consisting of 1:100 biotin according to the direction of SP reagent box was performed for 120 min at room temperature with shaking,followed by washing with PBS for 10 min each,3 times; (f) incubation with 1:100 streptavidinperoxidase was for 60 min at room temperature with shaking,followed by washing with PBS 4 times for 10 min each; (g) the slices were counterstained with hematoxylin for 1 min; (h) the slices were mounted on slides,dehydrated,and coverslipped,followed by analysis under a microscope with computer image processing system. PBS (0.01 mol/L) was applied as the negative contrast.

(3) Quantitative measurement of c-fos and c-jun positive cells: c-fos and c-jun positive cells were randomly measured in 20 units of areas on each slice under 400× magnification (units of areas are the window of areas provide by the computer system),and the average numbers of positive cells in the 20 units for c-fos and c-jun were calculated per slice. Semi-quantitative analysis of the randomly selected 20 slices from each group was done by testing the pigmentation average absorbency with the MPIAS-500 colorized multimedia image processing system.

(4) The detection of Bcl-2 and Bax: The experimental steps were as follows: The dewaxed coronary slices were treated with 3% hydrogen peroxide (H2O2) for 15 min to destroy the endogenous peroxidase activity, then washed with PBS,incubated in rabbit anti-rat Bcl-2 or Bax antibodies for 2 h at 37 ℃,and then washed with PBS. In order to eliminate nonspecific pigmentation,some of the slices were replaced with PBS and incubated in a similar manner. This was followed by incubation with instant type Ⅱ antibody for 30 min at 37 ℃,washing with PBS,staining with 3,3’ diamino benzidine (DAB),dehydration,and coverslipping. Randomly,10 slices from each rat were used to count the number of Bcl-2 or Bax positive cells in 100 cells in each hippocampus CA1 region of the same visual field under the microscope.

The positive cells were identified by the brown yellow granules in the cytoplasm or on the karyotheca under the optical microscope.

2.3 Statistical analysis The results are shown as mean ± standard deviation. The students t-test was used to compare the differences between two groups,and F-analysis was used to compare the mean between multiple groups. Analysis was performed with SPSS 12.0 software.

3 Results 3.1 The effect of the decreased rCBF in the right frontal lobe and hippocampus on learning and memory function Learning and memory performance of rats in the A,B, C,and D groups was lower than that of rats in the A0, B₀,C₀ and D₀ groups at 4 h,8 h,24 h,and 3 d after operation. The indexes of training times,EN,time to reach the mark,and TRT also increased gradually in the A,B,C,and D groups,and the initiative evasion rate (IER) decreased gradually (trend test P < 0.05). However,there were no significant differences in learning and memory indexes among the A₀,B₀,C₀, and D₀ groups (Table 1).

3.2 Comparison of the rCBF in the right frontal lobe and hippocampus between the two groups The rCBF in the right frontal lobe and hippocampus of rats in the operation group was significantly lower than that of rats in the false operation group at 4 h, 8 h,24 h,and 3 d after operation; the difference between each group was significant (P < 0.05). The rCBF of the operation group had obvious difference at different times (P < 0.05),and the rCBF decreased gradually with time after the operation (trend test P < 0.05). However,the distinction of the rCBF at different times after the operation in the false operation group was not significant (P > 0.05). The details are shown in Table 2.

3.3 The effect of the decreased rCBF on the expression of c-fos and c-jun in the right hippocampus and frontal lobe The pigmentation positive rate and average absorbency of c-fos and c-jun in the right hippocampus and frontal lobe in the operation group were higher than those in the false operation group at different times after the operation (P > 0.05). In the operation group, the difference between the pigmentation positive rate and average absorbency of c-fos and c-jun at different times after the operation was significant,and increasing with time after the operation (trend test P < 0.05). In the false operation group,the distinctions between the pigmentation positive rate and average absorbency of c-fos and c-jun at different times after the operation were not significant (P > 0.05). The details are shown in Tables 3 and 4 and Figures 1 and 2 Although the expression of c-fos and c-jun positive cells in the operation group was more obvious,there were only a few positive cells in the false operation group.

3.4 The effect of the decreased rCBF on the expression of Bcl-2 and Bax in the right hippocampus and frontal lobe Bcl-2- and Bax-positive cells were found in the operation group,and the expression levels of Bcl-2 and Bax were different at different times after the operation, that is,the expression of Bcl-2 and Bax increased with the extending of time (trend test,P < 0.05). The Bax/Bcl-2 protein expression ratio also increased with the extending of time after the operation,but the difference in the frontal lobe was not significant. In the false operation group,no expression of Bcl-2- and Bax-positive cells could be found at different times. The detailed results are displayed in Tables 5 and 6, and Figures 3 and 4.

4 Discussion In the central nervous system,neurons in the hippocampus and frontal lobe are the most sensitive to ischemia and anoxia. In this study,our data showed that the rCBF in the right frontal lobe and hippocampus of rats in the operation group was significantly lower than that of rats in the false operation group. Furthermore,the rCBF gradually decreased with time after the operation. The learning capability of rats in the operation group was also lower than that of rats in the false operation group,and similarly decreased with time.

There are usually two types of ischemic brain tissue injuries: necrosis and apoptosis. There is significant apoptosis-related protein expression during cerebral ischemia,which supports the occurrence of apoptosis in ischemic trauma. However,the mechanism of neuronal apoptosis after cerebral ischemia remains unknown. It has been demonstrated that the active expression of apoptosis-related genes plays a critical role in the mechanism of neuron apoptosis. There are usually two types of apoptosis related genes: the apoptosis promoting genes like c-fos,c-jun,c-myc, p53,bax,and fas; the anti-apoptosis genes,such as Bcl-2 and Bcl-x1^[5]. c-fos and c-jun can be expressed transiently when the cell is stimulated by a set of physical,chemical,and biological factors,to produce c-fos and c-Jun protein^{[6, 7, 8]}; therefore,the levels of c-fos and c-jun expression are considered as a good index of activation of the brain. It has been well established that c-fos and c-jun are activated during stress in broad areas of the brain. This study shows that the pigmentation positive rate and average absorbency of c-fos and c-jun in the right hippocampus and frontal lobe in the operation group are higher than those of the false operation group at different times after the operation (P > 0.05). In the operation group,the distinctions between the pigmentation positive rate and average absorbency of c-fos and c-jun at different times after the operation were significant,and increased with time after the operation (trend test P < 0.05). In the false operation group,the distinctions between the pigmentation positive rate and average absorbency of c-fos and c-jun at different times after the operation were insignificant (P > 0.05). This suggests that the expression of c-fos and c-jun has a negative correlation with the rCBF in the memory region.

The upregulation of c-jun and c-fos expression by cerebral ischemia may be due to the following factors: c-fos codes for a certain phosphoryl nucleoprotein, which acts as a transcription factor to activate the expression of target genes by binding to specific DNA sites; when new or intense stimulation is lacking,the transcription level of c-fos is low,but its expression is swiftly upregulated when the neuron is excited or its state is changed. Additionally,because the expression subsides within 5-6 h,c-fos immunohistochemistry can mark an excited neuron in multisynaptic neural pathways^[9]. Thus,high level of c-fos detected by immunohistochemical methods can be regarded as the sign of changing neuronal activity,and it is widely applied in studies of functional nerve anatomy and neural pathways. Therefore,the decline of learning and memory function caused by cerebral ischemia may have a relationship with the enhanced expression of c-fos and c-jun in the hippocampus and frontal lobe caused by ischemia.

However,the mechanism of c-fos and c-jun promoting apoptosis has not been clarified yet,but presumably it is related with the effect of AP-1^[10]. AP-1,which is activated by the expression of c-fos and c-jun (nucleus DNA-binding protein),is a heterodimer joined by a leucine zipper,and forms the structural base for transferring of apoptosis signals^[11]. TPA response element (TRE) is present in its DNA binding region,and binds to DNA by each of its regions rich in alkalescent amino acid,regulating the replication and transcription of DNA. At the same time, it induces the expression of genes containing TRE, and acts as a gene regulating protein,regulating the transcriptions of target genes or the late onset genes, thereby,inducing apoptosis or promoting cell cycle. Alternatively,c-fos protein is regarded as a sort of membrane protein,which cross-links with antibodies to induce apoptosis. Therefore,the enhanced expression of c-fos and c-jun in the neurons of the regions involved in learning and memory can lead to the decline of learning and memory function.

Study reports show that Bcl-2 and Bcl-xl proteins are anti-apoptotic,and Bax and Bcl-xs proteins promote apoptosis^[12]. The excessive expression of Bcl-2 together with Bax forms isodimers,which inhibit apoptosis^[13]. However,Bax-Bax isodimers can promote apoptosis, since Bax is an apoptosis promoting gene. It causes the releasing of some molecules,which leads to the activation of cysteine proteinase,and simultaneously, hastens apoptosis by resisting the protecting effect of Bcl-2 or induces apoptosis by conversion of mitochondrial transparency^[14]. In this study,Bcl-2 and Bax positive cells are expressed in the operation group in the right hippocampus and frontal lobe at different time after the operation. The expression of Bcl-2,Bax, and the Bax/Bcl-2 protein expression ratio increased gradually (trend test P＜0.05) with time after the operation. However,there was no expression of Bcl-2 and Bax positive cell in the false operation group at different time after the operation. This suggests that when the rCBF of the hippocampus and frontal lobe decreases,cerebral cell ischemia can lead to the increased expression of Bcl-2 and Bax positive cells. The enhanced Bax/Bcl-2 protein expression ratio promotes the cerebral cell apoptosis,which causes the decline of learning and memory function. Considering that the changes in the Bax/Bcl-2 ratio are linked to the survival of the cell and the increased Bax/Bcl-2 protein expression ratio can promote neuronal apoptosis,it is clear that the reduction of the rCBF can cause the elevation of the expression of Bcl-2,and Bax,and Bax/Bcl-2 ratio,which leads to decreased learning and memory function.

In conclusion,the mechanism underlying the learning and memory function decline caused by decreased rCBF is very complex. It not only involves the relationship of the above genes with apoptosis, but may also involve other causations,which are yet to be investigated. The next step should be studying that relationship between blood flow perfueion levels and other relevent gene of learning and memory function in strict control of various confounding factors,so improves the learning and memory function of patients with cerebral blood perfusion deficiency by controlling gene expression.

Acknowledgment We are grateful to M Yang,M Huang and M Wei for their technical assistance.

Conflict of interests The authors have no financial interest to disclose regarding the article.