All studies were carried out in a 9-week-old male Sprague-Dawley rats (n=16, 250~280 g) that had free access to drinking water and standard rodent food pellets. The experimental procedures were reviewed and approved by the Animal Care and Use Committee of Dongguk University (IRB: 09-45). Further, animal care and use were in accordance with the guidelines of the National Institutes of Health (Bethesda, MD). Focal cerebral ischemia was induced by occlusion of the left middle cerebral artery as described previously (Hasegawa et al., 1994).

Anesthesia was induced with 3% isoflurane in a mixture of oxygen/nitrous oxide (30 : 70) and rats were maintained with 1% isoflurane in the oxygen/nitrous oxide gas mixture. A catheter was inserted and positioned in the femoral artery and arterial blood pressure was measured and recorded continuously throughout the procedures. Body temperature was monitored continuously during all procedures using a rectal thermometer probe. Temperature control was accomplished with the aid of a heating pad which was kept at 37℃.

Under the dissecting microscope, left middle cerebral artery was occluded for 3 h, 6 h, and 24 h using a 4-0 mono filament (3 cm in length) coated with a mixture of silicone resin. Sham-operated controls rats were subjected to middle cerebral artery surgery without occlusion. After 3 h, 6 h, and 24 h of occlusion of the middle cerebral artery, rats were anesthetized with isoflurane again and the brain tissues were removed for 2% 2,3,5-triphenyltetrazolium chloride (TTC; Sigma Aldrich Corp., St Louis, MO) staining.

Rats were sacrificed and their brains were quickly removed and sectioned into 2-mm-thick slices starting from the frontal pole using a Brain Matrix Slicer (Vibratome Co.) (n=4). Slices were then immersed in TTC in a Petri dish and incubated at 37℃ for 20 minutes. Slices were flipped at the 10-minute mark to ensure staining of anterior and posterior faces.

At the scheduled time, sham-operated (n=3) and ischemic rats (n=3) were reanesthetized and their brain were fixed with a transcardiac infusion of 4% paraformaldehyde following perfusion with isotonic saline to remove blood from the cerebral vasculature. The brain was removed and post-fixed in the same fixative for 12 hours. Perfused brains were then paraffin-embedded and serial coronal sections 5 µm thick were obtained at the level of dorsal third ventricle (bregma-4.16 mm). Paraffin wax was removed in xylene over night at room temperature (RT) and the sections were rehydrated with ethanol (99%, 96%, 70%). After washing in distilled water, the sections were then stained with cresyl violet for 30 minutes at RT. The sections were then treated successively with ethanol (50%, 70%, 95%, 100%) and a differentiator (glacial acetic acid and 95% ethanol).

Penumbral or control tissues were removed from the TTC-stained brains of ischemic (n=4) and sham-operated rats (n=3), respectively for immunoblotting analysis. For protein extraction, the tissue was homogenized in homogenizing buffer (0.32 mM sucrose, 25 mM imidazole, 1 mM ethylenediaminetetraacetate (EDTA), pH 7.2 containing 8.5 mM leupeptin, 1 mM phenylmethylsulfonyl fluoride). Samples of homogenates were run on 9~15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Bio-Rad Mini Protean II) in duplicates in which one gel was run in parallel and subjected to Coomassie blue (Coomassie brilliant blue 0.3 g, 2-propanol 200 ml, acetic acid) staining to assure identical loading. The other gel was subjected to immunoblotting.

After electrophoresis, the separated protein was transferred to nitrocellulose membrane in a buffer solution containing 50 mM Tris-base, 380 mM glycine, and 20% methanol. Membranes were blocked with 5% milk in phosphate buffer solution-T (80 mM Na₂HPO₄, 20 mM NaH₂PO₄, 100 mM NaCl, 0.1% Tween 20, pH 7.5) for 1 h and incubated overnight at 4℃ with rabbit anti-Na⁺/K⁺-ATPase affinity purified polyclonal antibody (Chemicon, Temecula CA, USA, 1: 1,000), rabbit anti-NCX1 affinity purified polyclonal antibody (Chemicon, Temecula CA, USA, 1 : 2,000), rabbit anti-NMDA receptor subunits (NR2A, NR2B) polyclonal antibody (gifted from Moon, 1 : 5,000), mouse anti-neuronal nuclei (NeuN) monoclonal antibody (Chemicon, Temecula CA, USA, 1 : 1,000), mouse anti-glial fibrillary acidic protein (GFAP) monoclonal antibody (Boehringer Mannheim Biochemica, Philadelphia, USA, 1 : 2,000), and mouse anti-2',3'-cyclic nucleotide 3'-phosphodiesterase (CNPase) monoclonal antibody (Chemicon, Temecula CA, USA, 1: 1,000). The sites of antibody-antigen reaction were visualized with horseradish peroxidase (HRP)-conjugated secondary antibodies (P447 or P448, diluted 1 : 3.000; DAKO, Glostrup, Denmark), an enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech, Little Chalfont, UK) system and exposure to photographic film (Hyperfilm ECL, RPN3103K, Amersham Pharmacia Biotech, Little Chalfont, UK). The immunoblot signal developed by ECL system was quantified using Scion Image software (version 1.59).

Quantitative data are presented as mean±standard error of the mean (SEM). Comparisons between groups were made by unpaired student t-test. P values<0.05 were considered significant.

In the current study, experimental focal cerebral ischemia was induced in rats by the permanent middle cerebral artery occlusion for 3, 6, and 24 hours. Slices were divided into two zones, i.e., infarction zone (marked in black arrow) and penumbral zone (marked in white arrows) in the ipsilateral hemisphere according to the TTC staining pattern (Fig. 1A, pMCAO-3h). The penumbra was defined in static terms as the cellular interface between the infracted core cells that were committed to die and unaffected area of normal blood flow. The - numbers represent distances from the bregma. Cresyl violet staining was undertaken to examine whether the severity of neuronal injury in the penumbra was associated with duration of focal cerebral ischemia. Viable cells (arrows) were significantly decreased from 3 to 6 h after focal cerebral ischemia (Fig. 1B, C). Moreover, viable cells were not detected 24 h after focal cerebral ischemia which was very similar to that of the ischemic core (Fig. 1D). This finding indicated that the extent of neuronal damage was associated with the duration of focal cerebral ischemia.

To evaluate the effect of focal cerebral ischemia in the penumbra, immunoblotting analyses of Na⁺/K⁺-ATPase were performed (Fig. 2A). The expression of Na⁺/K⁺-ATPase was not significantly altered as compared with that of the sham-operated controls at 3 or 6 h following focal cerebral ischemia. However, the expression of Na⁺/K⁺-ATPase was significantly decreased at 24 h of focal cerebral ischemia (78±6% of sham-operated controls, n=4, ^*P<0.05) (Fig. 2B). Furthermore, the expression of NCX1 was significantly increased 24 h after focal cerebral ischemia as compared with sham-operated controls (144±2% of sham-operated controls, n=4, ^*P<0.05) (Fig. 3A, B).

Three hours after focal cerebral ischemia, the expression of NR2A-containing NMDA receptor in the penumbra was significantly increased while the expression of NR2B-containing NMDA receptor was significantly decreased (NR2A: 153±9%; NR2B: 37±2%, n=4, respectively, ^*P<0.05) as compared with those of sham-operated controls. However, the expression of NR2A and NR2B in the penumbra was reversed 24 h after focal cerebral ischemia (NR2A: 40±7% of sham-operated controls; NR2B: 120±16% of sham-operated controls, respectively, n=4, ^*P<0.05) (Fig. 4A, B).

Immunoblotting analyses demonstrated that expression of neural (NeuN) and astroglial (GFAP) cell proteins was affected in the penumbra three to 24 h after focal cerebral ischemia (Fig. 5A). The expression of NeuN was significantly decreased and their expression was closely associated with the duration of ischemia (3 h: 52±6%; 6 h: 38±12%; 24 h: 29±8% of sham-operated controls, respectively, n=4, ^*P<0.05). The expression of GFAP was also directly affected by the duration of ischemia (3 h: 69±5%; 6 h: 56±12%; 24 h: 53±14% of sham-operated controls, respectively, n=4, ^*P<0.05) (Fig. 5B). In contrast, the expression level of CNPase for oligodendrocytes in the penumbra was unchanged following focal cerebral ischemia (Fig. 5A, B).

Middle cerebral artery occlusion produces regions of brain with near complete and incomplete ischemia (reduced blood flow). In general, areas of mild ischemic injury occur where Na⁺/K⁺ ATPase is preserved, while in areas of more severe ischemia, ATP levels are low and Na⁺/K⁺-ATPase activity is reduced (D'Ambrosio et al., 2002). Therefore, significant reduction of Na⁺/K⁺-ATPase in the penumbra at 24 h as compared with those of 3 and 6 h indicates a deleterious effect of 24 h of ischemia. Furthermore, the current study verified the histological findings demonstrating that the extent of neuronal damage depends on the duration of ischemia.

Several studies have examined the role of Na⁺/K⁺ ATPase ion channels in hypoxic-ischemic neuronal damage and have concluded that Na⁺ influx is an important initiating event leading to anoxic damage (Stys et al., 1992; Tasker et al., 1992). The decreased Na⁺/K⁺ ATPase expression in the present study postulates the disturbance of NCX1 and/or NMDA-induced Ca²⁺ influx by enhanced cellular K⁺ efflux and Na⁺ influx which result in ischemic depolarization in neurons and astrocytes. In general, Ca²⁺ influx or extrusion of NCX1 Depends on the ability of Na⁺/K⁺ ATPase to pump K⁺ and Depolarized the plasma membrane (Silver et al., 1997). The reduction of Na⁺/K⁺ ATPase on NMDA-induced Ca²⁺ influx also might be related to an enhancement of Ca²⁺ permeation of NMDA channels (Czyz et al., 2002). Furthermore, intracellular Na⁺ accumulation in astrocytes can contribute to glutamate release, which occurs by reversal of the Na⁺/glutamate cotransporter (Anderson & Swanson, 2000). This cotransporter normally mediates the entry of two Na⁺ ions along with one molecule of glutamate and represents an important mechanism of glutamate uptake by astrocytes, which ensures neuronal survival (Storck et al., 1992). If it is reversed due to excessive intracellular Na⁺, astrocytes begin to promote glutamate release and might contribute to neuronal damage during ischemia. Therefore, intracellular Na⁺ accumulation plays a critical role in NCX1 and/or NMDA-induced neuronal cell death by participating in mechanisms that brings about Ca²⁺ overload and accelerates glutamate release from the astrocytes.

In a normal brain, NCX1 is thought to be important in buffering neuronal intracellular Ca²⁺ by transporting one Ca²⁺ out of cells and three Na⁺ into the cells (Blaustein & Lederer, 1999). However, based on vitro and in vivo studies, it appears that during and following cerebral ischemia, it is likely that under depolarizing conditions, NCX1 can contribute to Ca²⁺ influx and neuronal injury. After NCX1 reverses, any further depolarization of the plasma membrane increases the electrochemical driving force of Ca²⁺ influx via this pathway (Hansen & Zeuthen, 1981; Benveniste et al., 1984). The role of reverse NCX1 in mediating toxic Ca²⁺ influx is supported by neuroprotective effects of NCX1 inhibitors (Schröder et al., 1999; Matsuda et al., 2001). The neurotoxic mechanism that leads to penumbra cell death in response to elevated intracellular Na⁺ in the current study represent the reverse operation of the plasma membrane NCX1, engaged by plasma membrane depolarization and intracellular Na⁺ overload through a decreased Na⁺/K⁺-ATPase. However, reverse NCX1 does not significantly contribute to Ca²⁺ influx after inactivation of NMDA receptors (Kiedrowski 2001). The mechanism of Na⁺-dependent Ca²⁺ influx requires open NMDA channels, because occlusion of the channels with MK-801 almost completely inhibited Ca²⁺ accumulation. Our results suggest that increased reverse NCX1 in the penumbra participates in Na⁺/K⁺-ATPase-dependent amplification of NMDA-induced Ca²⁺ influx in ischemic depolarized neuronal cells.

Our results indicate that increased NR2B in the penumbra after 24 h of focal cerebral ischemia may be comprised of combinations of NR2B subunit, along with the NR1 subunit (Monyer et al., 1992). Pharmacologic studies show that NR2B-containing NMDA receptor channels, expressed in Xenopus oocytes, exhibits a higher affinity for L-glutamate and considerably longer offset decay time courses following brief application of L-glutamate than the NR1-NR2A channel (Meguro et al., 1992). In particular, the offset decay time course is thought to be crucial for the determination of intracellular Ca²⁺ concentration (Perkel et al., 1993). These reports suggest that NR2B-containing NMDA receptors are more efficient than receptors containing NR2A in the process of Ca²⁺ influx. Chronic incubation with GA resulted in a down-regulation particularly of the NR2B subunit and reduced the NMDA receptor-mediated increase in intracellular Ca²⁺ (Kölker et al., 2002). In the present study, we demonstrated a reduction of Na⁺/K⁺ ATPase and increased expression of NCX1 and NR2B in the penumbra by focal cerebral ischemia. Because the Na⁺/K⁺ ATPase is particularly important to evoke the ischemic depolarization, decrease in Na⁺/K⁺ ATPase would result in a relief of the voltage-dependent Mg²⁺ block of NMDA receptors (Gegelashvili & Schousboe, 1997) and further increase of Na⁺-dependent Ca²⁺ influx mediated by NR2B-containing NMDA receptors and NCX1. In addition, extrasynaptic NR2B-containing NMDA receptors antagonize nuclear signaling to cAMP response element binding protein (CREB), block induction of brain derived neurotrophic factor (BDNF) expression, and are involved in mitochondrial dysfunction and cell death (Hardinghan et al., 2002).

Significant increase of NR2A 3 h after focal cerebral ischemia might couple with the compensatory response to diminish the Ca²⁺ which is associated with lower affinity for glutamate and considerably shorter offset decay time for Ca2+ compared with those of NR2B-containing NMDA receptors. This result suggests the possibility that the other Ca²⁺ ion channels beyond the NMDA receptors play a major role in the neuronal injury 3 h following focal cerebral ischemia. Moreover, depending on the type of NMDA receptor, Ca²⁺ entry can determine the biological outcome of Ca²⁺ signaling, as shown by site-specific differences in the regulation of CREB-mediated transcription (Hardinghan et al., 2002). Increases in the synaptic NMDA receptor subunit, NR2A, 3 h after focal cerebral ischemia may be associated with neuronal survival in the penumbra given that synaptic NMDA receptors promote nuclear signaling to CREB, induce BDNF gene expression, and activate an anti-apoptotic pathway (Rumbaugh & Vicini, 1999).

There are many instances in which focal brain lesions also seem to have an impact on the function of surrounding or remote brain areas due to the fact that the brain can be considered as a network with multiple and intricate connections (Beck et al., 1996). Therefore, it is very important to analyze in detail which prognosis of focal cerebral ischemia is a direct consequence of the lesion, the perilesional area, or a reaction of the surrounding brain to the lesion. The most remarkable decrease of neural (NeuN) and astroglial protein (GFAP) 24 h after focal cerebral ischemia supports evidence demonstrating that increased NCX1 and NR2B-containing NMDA receptor is closely associated with Ca²⁺-dependent neuronal injury in the penumbra. Moreover, further declines of NeuN than that of GFAP in the penumbra may suggest that neurons in the penumbra are likely to be more susceptible to ischemic injury than astroglia. The underlying mechanisms for the susceptibility of neurons to ischemic insults may be explained in terms of glucose metabolism. Recent studies using multiphoton microscopy demonstrated that neurons use primarily oxidative metabolism, whereas astrocytes are glycolytic (Kasischke et al., 2004). In addition, glucose uptake in primary cultured astroglias was increased in response to the elevated extracellular K⁺ than that of neurons (Yu et al., 1989).

In conclusion, the current study suggest the new intracellular Ca²⁺ overloading mechanisms after focal cerebral ischemia in which Ca²⁺ influx through the reverse mode of NCX1, Intracelluler Ca²⁺ overhead is positively reinforced by glutamate-dependent NR2B-containing NMDA receptors which is triggered by deranged transport of Na⁺ through decreased Na⁺/K⁺-ATPase.