Regulation of extracellular signal-regulated kinase 1/2 influences hippocampal neuronal survival in a rat model of diabetic cerebral ischemia.

Activation of extracellular signal-regulated kinase 1/2 has been demonstrated in acute brain ischemia. We hypothesized that activated extracellular signal-regulated kinase 1/2 can protect hippocampal neurons from injury in a diabetic model after cerebral ischemia/reperfusion. In this study, transient whole-brain ischemia was induced by four-vessel occlusion in normal and diabetic rats, and extracellular signal-regulated kinase 1/2 inhibitor (U0126) was administered into diabetic rats 30 minutes before ischemia as a pretreatment. Results showed that the number of surviving neurons in the hippocampal CA1 region was reduced, extracellular signal-regulated kinase 1/2 phosphorylation and Ku70 activity were decreased, and pro-apoptotic Bax expression was upregulated after intervention using U0126. These findings demonstrate that inhibition of extracellular signal-regulated kinase 1/2 activity aggravated neuronal loss in the hippocampus in a diabetic rat after cerebral ischemia/reperfusion, further decreased DNA repairing ability and accelerated apoptosis in hippocampal neurons. Extracellular signal-regulated kinase 1/2 activation plays a neuroprotective role in hippocampal neurons in a diabetic rat after cerebral ischemia/reperfusion.

Introduction

Epidemiological data show that diabetes is closely related to propensity to stroke. These studies have not only elicited the statistical association between diabetes and stroke, but also explored possible reasons behind this association (Gerstein et al., 1997; Bell, 1999; Stegmayr et al., 2002). Previous studies using monkey models have found that, compared with the non-hyperglycemic group, the high glucose group showed a large area of brain injury and extensive necrosis (including the cerebral cortex, basal ganglia, brainstem and cerebellum) after cerebral ischemia (Lin et al., 1998). In fact, the cellular decision to undergo apoptosis is determined by the integration of multiple survival and death signals. Extracellular signal-regulated kinase 1/2 (ERK1/2) signaling pathway is a major cell survival pathway that has been extensively studied recently (Irving et al., 2000; Chu et al., 2004). We would like to determine if this nerve cell loss seen when hyperglycemia aggravates brain ischemia is associated with the ERK1/2 signaling pathway, and elucidate the exact roles and underlying mechanisms of this process.

ERK1/2 is a key member of the mitogen activated protein kinase family, and plays a pivotal role in multiple factors involved in cerebral ischemic injury. In mammalian cells, ERK1/2 can be activated by a variety of stimuli, and in turn, the activated ERK1/2 can typically phosphorylate a number of downstream substrates, such as c-Raf-1, MEK, Elk-1, and c-Fos (Satoh et al., 2000). These substrates regulate a number of genes and proteins involved in neuronal apoptosis, such as Bcl-2 and caspase-3, which eventually determine neuronal survival or death (Irving et al., 2000; Satoh et al., 2000; Chu et al., 2004; Jin et al., 2011). Previous studies compared the changes of activated ERK1/2 between normal rats and hyperglycemic rats under ischemic conditions and mapped the mechanisms of the ERK1/2 pathway under diabetic and ischemic conditions (Zhang et al., 2006). However, this approach created some ambiguous conclusions as the activation of ERK1/2 creates a double-edged sword during the cerebral ischemic pathological process. If there is no corresponding intervention to change or mediate the degree of ERK1/2 activation, it is difficult to delineate the specific roles and mechanisms of ERK1/2 activation. Here, we take the first step in that direction by preconditioning hyperglycemic rats with the ERK1/2 inhibitor (U0126), in order to elicit the specific role of ERK1/2 activation in diabetes mellitus combined with cerebral ischemia.

Accumulating evidence suggests that increased cytotoxic substances, such as oxygen free radicals, can cause DNA damage. Impaired DNA in turn activates DNA repairing enzymes to repair and improve the tolerance of neurons to ischemia after insult (Mimeault et al., 2010; Hori et al., 2012). However, if the DNA repair function is decreased, the impaired DNA can produce mutations and neuronal death. Does ERK1/2 activation associate with DNA repair function during the pathological process of nerve cell loss induced by hyperglycemia aggravating brain ischemia? Ku70, a regulatory subunit of DNA dependent protein kinase, is mainly involved in the repair of DNA double faults, and increases the efficiency and accuracy of DNA repair (Rajakumar Mandrau et al., 2011). Previous studies showed that DNA damage was associated with mitogen activated protein kinase activation in vitro (Yan et al., 2008; Marampon et al., 2011). However, there is a lack of data surrounding whether ERK1/2 activation influences DNA repair (Ku70 activity) under cerebral ischemia combined with diabetic conditions. We hypothesized that ERK1/2 activation plays a neuro-protective role in nerve cell loss induced by diabetic global cerebral ischemia/reperfusion, and that inhibition of ERK1/2 activation can inhibit Ku70 activity and aggravate the injury of nerve cells.

To verify this hypothesis, the present study established a cerebral ischemic model in normal rats and hyperglycemic rats using a typical four-vessel occlusion method. Hyperglycemic rats were also preconditioned with the ERK1/2 inhibitor (U0126). The goal of this study was to observe dynamic changes in neuronal loss, ERK1/2 and Ku70 activity, and pro-apoptotic gene Bax expression in the hippocampus. Results may provide new data on the possible mechanisms of ERK1/2 activation in DNA repair following cerebral ischemia in a diabetic state.

Materials and Methods

Animals

A total of 160 specific pathogen-free male Sprague-Dawley rats, aged 2.5 months, weighing 250 ± 20 g, were purchased from Beijing Experimental Animal Center, Chinese Academy of Science, China (License No. SCXK (Jing) 2011-2012). All rats were housed in a laboratory room at controlled temperature (22–26°C) and relative humidity (40–70%), and were allowed adequate light and diet. Experimental procedures were approved by the Animal Use and Care Advisory Committee of Health Science Center of Hebei United University, China.

Grouping and ischemia induction

One hundred and sixty rats were randomly divided into four groups: sham group, normoglycemia global cerebral ischemia/reperfusion (NI/R) group, diabetic global cerebral ischemia/reperfusion (DCI) group, and DCI + U0126 group. There were 40 rats in each group and 10 in each time point subgroup (1, 6, 24, 48 hours after reperfusion). U0126 was purchased from Santa Cruz Biotechnology, Santa Cruz, CA, USA.

Transient brain ischemia (30 minutes) was induced using the four-vessel occlusion method, as previously described (Kim et al., 2011). In brief, under anesthesia with chloral hydrate (300–350 mg/kg, i.p), vertebral arteries were electrocauterized and common carotid arteries were exposed. Rats were allowed to recover for 24 hours and fasted overnight. Ischemia was then induced by occluding the common carotid arteries using aneurysm clips. Animals were selected for the present experiment during ischemia, if they had: (1) a completely flat bitemporal electroencephalogram; (2) maintenance of dilated pupils; and (3) absence of corneal reflex. Carotid artery blood flow was restored by releasing the clips (Li et al., 2005). Rectal temperature was maintained at 37°C during ischemia and 2 hours after ischemia. Twelve hours after fasting, animals in the DCI group received 55 mg/kg intraperitoneal streptozocin (Xin Ran Biological Technology Co., Ltd., Shanghai, China). If 72 hours after fasting, animals presented with polyuria, decreased body mass and blood glucose more than 16.7 mmol/L, they were defined as diabetic animals before ischemia. The diabetic animals in the DCI + U0126 group received 0.01 mg/kg U0126 (dissolved in 1% dimethyl sulfoxide) intravenously 30 minutes before ischemia via the tail vein. The same surgical procedures were carried out in the DCI and DCI + U0126 groups. The sham group also underwent the above surgery, but occlusion of the carotid artery was omitted. At the indicated time points (1, 6, 24 and 48 hours after 30 minutes of ischemia), animals were decapitated, and brain tissue was processed for further analysis.

Tissue preparation

Five rats at 1, 6, 24, and 48 hours after brain injury were decapitated under anesthesia, and brain tissue was separated and fixed with 4% paraformaldehyde solution and embedded in paraffin in preparation for histology and immunohistochemistry.

Five rats at 1, 6, 24, and 48 hours after brain injury were decapitated under anesthesia, the hippocampal region was dissected away and rapidly frozen in liquid nitrogen. The frozen hippocampal tissue samples were homogenized in 1:10 (w/v) ice-cold homogenization buffer A containing 10 mmol/L HEPES (pH 7.9), 0.5 mmol/L MgCl2, 10 mmol/L KCl, 0.1 mmol/L EDTA, 0.1 mmol/L EGTA, 50 mmol/L NaF, 5 mmol/L dithiothreitol, 10 mmol/L β-glycerophosphate, 1 mmol/L sodium orthovanadate, 1% NP-40, enzyme inhibitors (1 mmol/L benzamidine, p-nitrodomains phenyl phosphate, phenylmethylsulfonyl fluoride), and 5 µg/mL each of aprotinin, leupeptin, and pepstatin A. Homogenate was then centrifuged at 1,000 × g for 15 minutes at 4°C. The supernatant containing cytosolic material was collected for the determination of protein concentration. Reagents were purchased from Xin Ran Biological Technology Co., Ltd., Shanghai, China.

Neurological deficit scores

At 24 and 48 hours after reperfusion, five rats underwent neurological evaluation (before decapitaition), and were scored on an 18-point scale (Rancan et al., 2003). The neurobehavioral scale consisted of the following six tests: (1) spontaneous activity (0 to 3 points); (2) symmetry in the movement of four limbs (0 to 3 points); (3) forepaw outstretching (0 to 3 points); (4) climbing (1 to 3 points); (5) body proprioception (1 to 3 points); and (6) response to vibrissae touch (1 to 3 points). The score given to each rat at the completion of the evaluation is the summation of all six individual test scores. Three is the minimum neurological score and eighteen is the score exhibited by normal animals.

Hematoxylin-eosin staining

Brains were quickly removed and further fixed with the same fixation solution at 4°C overnight. Post-fixed brains were embedded in paraffin, and cut into 5 µm coronal sections using a microtome. Paraffin-embedded brain sections were deparaffinized with xylene and rehydrated using an ethanol gradient (100% to 70% v/v), followed by washing with water. The sections were stained with 0.1% (w/v) hematoxylin and eosin, and were examined with light microscopy (Olympus, Tokyo, Japan). The number of surviving hippocampal CA1 pyramidal cells was counted as the neuronal density. Specifically, the hippocampal CA1 region was divided into three equal parts and a place was chosen on the section to count the surviving cells using Motic-6.0 Image Acquisition and Image Analysis System (200 × magnification; LabWorks Software, UVP Inc., Upland, CA, USA).

Immunohistochemistry and western blot analysis

After being incubated in 5% normal goat serum, coronal sections were incubated with polyclonal rat-anti-mouse phosphorylated ERK1/2 (1:2,000; Cell Signaling Biotechnology, Danvers, MA, USA) or polyclonal rat-anti-mouse Ku70 (1:250; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or polyclonal rat-anti-mouse Bax (1:200; Santa Cruz Biotechnology) antibodies overnight at 4°C. An equivalent dilution of rat IgG was used as primary antibody for negative control. Subsequently, the sections were incubated in biotinylated rabbit anti-rat secondary antibody (1:500; Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China) for 90 minutes at room temperature followed by incubation in Avidin-Biotin complex for 90 minutes. Finally, the sections were developed with stable 3,3′-diaminobenzidine and nuclei were counterstained with hematoxylin. Brown-stained positive cells were observed microscopically (Laishi Electronic Technology, Shanghai, China). Samples were separated by 10% or 7.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis and electrotransferred onto nitrocellulose membranes. After blocking with 3% bovine serum albumin for 3 hours, the membrane was probed with primary antibodies at 4°C overnight. The primary antibodies used were mouse monoclonal anti-phosphorylated ERK1/2 (1:1,000; Cell Signaling Biotechnology), rabbit anti-Bax polyclonal antibody (1:1,000; Sigma, St Louis, MO, USA), rabbit anti-Ku70 polyclonal antibody (1:1,000; Sigma), and rabbit anti β-Actin polyclonal antibody (1:1,000; Cell Signaling Biotechnology) antibodies. Bound antibodies were detected using alkaline phosphatase-conjugated goat anti-mouse IgG (1:10,000; Sigma) or goat anti-rabbit IgG (1:10,000; Sigma) for 2 hours at room temperature. Immunoreactivity was detected by a NBT/BCIP assay kit (Kexing Biological Technology Co., Ltd., Shanghai, China) according to the manufacturer's instructions. The bands on the membranes were scanned and analyzed with an image analyzer (Lab Works Software; UVP Inc, Upland, CA, USA). The ratio of absorbance of the sample target protein bands and reference gene band β-Actin after calibration (the relative expression ratio) reflect protein level.

Statistical analysis

Values from five independent rats were expressed as mean ± SD. Statistical analysis was performed using one-way analysis of variance, followed by Duncan's new multiple range method (SPSS, Chicago, IL, USA). P values of less than 0.05 were considered statistically significant.

Results

ERK1/2 pathway inhibition led to a decrease in the number of surviving neurons in the hippocampal CA1 region following cerebral ischemia in diabetic rats

Hematoxylin-eosin staining demonstrated a large number of dead cells in the hippocampal CA1 region from cerebral ischemia/reperfusion groups (DCI group, NI/R group, DCI + U0126 group). These cells had an irregular morphology coupled with shrinking nuclei. Conversely, only few dead cells were encountered in the sham group. Statistical analysis revealed that diabetes aggravated cell death, and neuron survival rate was lower in the DCI group than in the NI/R group (P < 0.05). Pretreatment with U0126 before ischemia significantly increased neuronal degeneration in diabetic rats (P < 0.05; Figure 1).

Figure 1

Effects of extracellular signal-regulated kinase 1/2 pathway inhibition on the number of surviving neurons in the hippocampal CA1 region following cerebral ischemia in diabetic rats.

(A–D) Hematoxylin-eosin stained sections of nerve cell morphological changes in hippocampal CA1 regions 48 hours after injury in the sham (NSO) group (A), normoglycemia global cerebral ischemia/reperfusion (NI/R) group (B), diabetic global cerebral ischemia/reperfusion (DCI) group (C) and DCI + U0126 group (D); arrows indicate dead cells (× 400). (E) The number of surviving neurons in the hippocampal CA1 region from each group. Data are expressed as mean ± SD from five rats at each time point, n = 40. Differences between the groups were compared using one-way analysis of variance and the Student-Newman-Keuls test. *P < 0.05, vs. NSO group; #P < 0.05, vs. NI/R group; †P < 0.05, vs. DCI group.

ERK1/2 pathway inhibition led to a decrease in neurological deficit scores following cerebral ischemia in diabetic rats

Neurological deficit is an important outcome after ischemia/reperfusion. At 1 and 6 hours after injury, animals were in a very poor state, and failed to complete neurological deficit evaluation, so we chose to conduct the evaluation at 24 and 48 hours after injury. As shown in Figure 2, the neurological deficit scores at 24 and 48 hours in the DCI group were significantly lower than in the NI/R group (P < 0.05). Moreover, pretreatment with U0126 before ischemia led to a significant neurological deficit and the neurological deficit scores in the DCI + U0126 group were significantly lower than that in the DCI group (P < 0.05). The results of neuronal loss and neurological deficit scores showed that U0126 aggravated the pathological injury induced by DCI, suggesting that ERK1/2 activation has potentially protective effects under diabetic and ischemic conditions.

Figure 2

Effects of extracellular signal-regulated kinase 1/2 pathway inhibition on the neurological function of diabetic rats following cerebral ischemia.

A lower score shows more serious functional damage. Data are expressed as mean ± SD from five rats at each time point. There were forty rats in each group. Differences between the groups were compared with one-way analysis of variance and the Student-Newman-Keuls test. *P < 0.05, vs. sham (NSO) group; #P < 0.05, vs. normoglycemia global cerebral ischemia-reperfusion (NI/R) group; †P < 0.05, vs. diabetic global cerebral ischemia-reperfusion (DCI) group.

Phosphorylated ERK1/2 expression decreased following cerebral ischemia in diabetic rats

Immunohistochemical staining results demonstrate that a large number of phosphorylated ERK1/2 positive cells were found in the hippocampal CA1 region from cerebral ischemia/reperfusion groups (DCI group, NI/R group, DCI + U0126 group) 1, 6, 24 and 48 hours after ischemia. These cells stained yellow and most positive products were located in the nucleus. Cells retained regular morphology. Conversely, only a few ERK1/2 phosphorylated cells were encountered in the control group (Figure 3A–D). The number of phosphorylated ERK1/2 positive cells in the U0126 group was lower than that in the DCI group.

Figure 3

Expression of phosphorylated extracellular signal-regulated kinase 1/2 (ERK1/2) in diabetic rats following cerebral ischemia (absorbance ratio of phosphorylated ERK1/2/β-actin).

(A–D) Immunohistochemistry of phosphorylated ERK1/2 positive cells in CA1 region 24 hours after injury in sham (NSO), normoglycemia global cerebral ischemia/reperfusion (NI/R), diabetic global cerebral ischemia/reperfusion (DCI) and DCI + U0126 groups, respectively (arrows refer to positive cells, × 400). (E) Western blot photomicrographs of phosphorylated ERK1/2 protein expression in hippocampal CA1 region in the NSO, NI/R, DCI and DCI + U0126 groups, respectively. (F) ERK1/2 phosphorylation from each group was detected using western blot analysis. Data are expressed as mean ± SD from five rats at each time point, 40 rats per group. Differences between the groups were compared using one-way analysis of variance and the Student-Newman-Keuls test. *P < 0.05, vs. NSO group; #P < 0.05, vs. NI/R group; †P < 0.05, vs. DCI group. h: Hours.

Western blot analysis suggested that ERK1/2 phosphorylation in the NI/R group was significantly higher than the sham group (P < 0.05), reached a peak at 24 hours after injury and decreased obviously at 48 hours after injury. ERK1/2 phosphorylation in the DCI group was significantly lower than that in the NI/R group (P < 0.05). Moreover, pretreatment with U0126 before ischemia led to a significant ERK1/2 phosphorylation decrease in diabetic rats, and ERK1/2 phosphorylation in the DCI + U0126 group was significantly lower than that in DCI group (P < 0.05; Figure 3E, F).

ERK1/2 pathway inhibition led to a decrease of Ku70 activity and an increase in Bax expression following cerebral ischemia in diabetic rats

Immunohistochemistry demonstrated that some Ku70 positive cells and a large number of Bax positive cells were found in the hippocampal CA1 region from cerebral ischemia/reperfusion groups (NI/R, DCI, DCI + U0126 groups) at 1, 6, 24 and 48 hours after ischemia. Ku70 positive cells were stained yellow and most positive products were located in the nucleus. Cells retained regular morphology (Figure 4A–D). Bax positive cells were stained yellow and most positive products were located in the cytoplasm. These cells displayed an irregular morphology (Figure 5A–D). Generally, the number of Ku70 positive cells in the U0126 group was lower than in the DCI group, while the number of Bax positive cells in the U0126 group was higher than in the DCI group.

Figure 4

Effect of extracellular signal-regulated kinase 1/2 pathway inhibition on Ku70 activity in the hippocampal CA1 region following cerebral ischemia in diabetic rats (absorbance ratio of Ku70/β-actin).

(A–D) Immunohistochemistry of Ku70 positive cells in hippocampal CA1 region 24 hours after injury in the sham (NSO), normoglycemia global cerebral ischemia/reperfusion (NI/R), diabetic global cerebral ischemia/reperfusion (DCI), and DCI + U0126 groups, respectively (arrows refer to positive cells, × 400). (E) Western blot photomicrographs of Ku70 protein expression in hippocampal CA1 regions in the NSO group, NI/R group, DCI group and DCI + U0126 group. (F) Ku70 expressions from each group were detected by western blot analysis. Data are expressed as mean ± SD from five rats at each time point, 40 rats per group. Differences between the groups were compared using one-way analysis of variance and the Student-Newman-Keuls test. *P < 0.05, vs. NSO group; #P < 0.05, vs. NI/R group; †P < 0.05, vs. DCI group. h: Hours.

Figure 5

Effect of extracellular signal-regulated kinase 1/2 pathway inhibition on Bax expression in the hippocampal CA1 region following cerebral ischemia in diabetic rats (absorbance ratio of Bax/β-actin).

(A–D) Immunohistochemistry of Bax positive cells in hippocampal CA1 region 24 hours after injury in sham (NSO), normoglycemia global ce-rebral ischemia/reperfusion (NI/R), diabetic global cerebral ischemia/reperfusion (DCI), and DCI + U0126 groups, respectively (arrows refer to positive cells, × 400). (E) Western blot photomicrographs of Bax protein expression in hippocampal CA1 region in the NSO group, NI/R group, DCI group and DCI + U0126 group. (F) Bax expression from each group was detected by western blot analysis. Data are expressed as mean ± SD from five rats at each time point, 40 rats per group. Differences between the groups were compared using one-way analysis of variance and the Stu-dent-Newman-Keuls test. *P < 0.05, vs. NSO group; #P < 0.05, vs. NI/R group; †P < 0.05, vs. DCI group. h: Hour.

Western blot analysis suggested that Ku70 activity was increased at 1 and 6 hours, but decreased at 24 and 48 hours after injury in the NI/R group. Ku70 protein expression in the DCI group was lower than that in NI/R group (P < 0.05). Moreover, pretreatment with U0126 before ischemia led to a significant decrease in Ku70 protein expression, with Ku70 protein expression in the DCI + U0126 group significantly lower than the the DCI group (P < 0.05; Figure 4E, F). Bax protein expression in the NI/R group was significantly higher than the sham group (P < 0.05). Bax protein expression in the DCI group was also significantly higher than the NI/R group (P < 0.05). Moreover, pretreatment with U0126 before ischemia led to a significant increase in Bax protein expression. Bax protein expression in the DCI + U0126 group was significantly higher than the DCI group (P < 0.05; Figure 5E, F).

Discussion

In the present study, hematoxylin-eosin staining results showed that the number of surviving neurons and neurological deficit scores in the DCI + U0126 group were significantly lower than those in the DCI group. This evidence suggested that ERK1/2 activation is potentially protective under diabetic, ischemic conditions. Furthermore, a decrease in ERK1/2 activity is involved in the pathogenesis of diabetes aggravating nerve cell injury following ischemia. Our findings are not consistent with research performed by Ma et al. (2009).

We demonstrate that the phosphorylation of ERK1/2 was significantly higher at 1 and 3 hours after ischemia when compared with the NI/R group, suggesting that increasing ERK1/2 activity could enhance neuronal death induced by DCI. Previous studies revealed that ERK1/2 activation induced by cerebral ischemia plays either a protective or a destructive role in ischemia (Mori et al., 2002; Abe et al., 2003; Clausen et al., 2004; Wang et al., 2004). The reason behind these “double effects” in ERK1/2 activation is due to differences in a stressed environment. In the current study, a global cerebral ischemia model was established in diabetic rats using the modified four-vessels occlusion method. This technique is a well-developed method requiring the occlusion of the artery with a noninvasive arterial clip, and also places fewer stressful stimuli on brain tissue (Kim et al., 2011). In Ma's study (2009), the diabetes combined with cerebral ischemia model was established using a double-vessel occlusion and femoral artery bloodletting method, which can lead to hypotension and low brain perfusion in addition to ischemic stress induced by vessel occlusion. These differences may account for the disparity between results. The Ras/ERK1/2 cascade can be activated by growth factors, such as epidermal growth factor and vascular endothelial growth factor. Studies showed that these endogenous protective factors obviously decreased under diabetic/ischemic conditions (Uchino et al., 1997; Fan et al., 2005), which may be a reason for weakened ERK1/2 activity.

Ku70, a regulatory subunit of DNA dependent protein kinase, is thought to play a key role in DNA repair during the process of nerve cell loss in cerebral ischemic injury (Schackerford et al., 1999). To verify whether ERK1/2 activation associates with DNA repair, we observed Ku70 activity. Results showed that Ku70 expression was decreased at each time point during DCI, and was not increased at 1 and 6 hours, which occurred in the NI/R group. This evidence suggested that decreased Ku70 activity is the main factor in nerve cell loss after cerebral ischemia was aggravated by diabetes. We also found that Ku70 expression in the DCI + U0126 group was significantly lower than that in the DCI group, suggesting that ERK1/2 activation could regulate Ku70 expression during DCI.

Recently, various mechanisms have been proposed to explain the close relationship between DNA-repair and the activation of the mitogen activated protein kinase signal pathway (Li et al., 2012). Some studies suggested that ischemia-induced mitogen activated protein kinase signal pathway activation could reduce gene mutation and maintain the stability of the genome in vivo (Hayakawa et al., 2009). Other studies demonstrated that radiation-induced DNA and DNA-dependent protein kinase damage could change the activation of mitogen activated protein kinases, including ERK1/2 (Ying et al., 2008; Marampon et al., 2011). Bax is a proapoptotic molecule from the Bcl-2 family and promotes ischemic cell death via releasing apoptotic factors, such as cytochrome C.

In the current study, Bax protein expression was significantly increased in the DCI + U0126 group compared with the DCI group. The decreased activity of ERK1/2 can increase the expression of Bax, in turn, leading to more loss of nerve cells. Interestingly, we also found that there was a relationship between Ku70 and Bax, with Ku70 activity decreasing while Bax level increased. Under the diabetic cerebral ischemia conditions, Ku70 activity was strongly inhibited, in turn the abnormal expression of Bax was increased. Recently, studies showed that inhibition of Ku70 activity could enhance the expression of p53 or phosphorylation of p53, and in turn, Bax increased (Avigailb et al., 2008; Gregory et al., 2010; Chitra Subramanian et al., 2011). It was also reported that Ku70 itself could combine with Bax and form a Ku70-Bax complex (Avigail et al., 2008; Ioannis et al., 2009).

In summary, we speculate that in the pathological processes behind diabetes mellitus aggravated ischemia/reperfusion injury, a decrease in ERK1/2 activity resulted in a decrease in Ku70 activity and weakened DNA repairing ability, thus Bax expression was increased, which led to the loss of nerve cells. The limitations of this research lie in using streptozotocin injection to establish diabetic models. The high blood glucose response in this model is similar with type I diabetes. However, most patients with type II diabetes are clinically prone to cerebral infarction. So some deviations from our conclusions may exist. However, in the current study, we simulated cerebral ischemia/reperfusion injury under high glucose conditions and we confirmed that decreasing ERK1/2 activation plays a protective role in the pathological process of diabetes aggravating cerebral ischemia/reperfusion injury, decreasing Ku70 activity and increasing Bax expression. The present study provides useful information for understanding the pathogenesis and a promising approach to clinical diabetes combined with stroke. In the further, some works should be done to observe the ERK1/2 activation trends in other brain regions such as in cortex, and whether having difference in region and time, so as to better understand the underlying effects of ERK1/2 activation and identify new potential therapeutic targets for cerebral ischemia.