INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

WHILE THE MAIN FUNCTION of mitochondria is ATP synthesis, these intracellular organelles also play a major role in mediating cellular responses to ischemic stress. For example, calcium influx into mitochondria and the opening of mitochondrial permeability transition pores appear to be early steps leading to necrotic and apoptotic cell death in various tissues and organs including the brain (19, 29). Additionally, numerous studies have shown that the pharmacological targeting of mitochondrial ATP-sensitive K⁺ (mitoK_ATP) channels, with acute administration of drugs such as diazoxide, protects ischemic myocardium against injury (17, 30). Diazoxide is a selective opener of mitoK_ATP channels and has minimal effects on sarcolemmal K_ATP channels (10, 20, 31). Furthermore, transient activation of mitoK_ATP channels with sublethal ischemia appears to be a crucial step in the development of ischemic preconditioning, which protects myocytes from subsequent prolonged ischemia (15, 17). Protective effects of acute administration of diazoxide or ischemic preconditioning can be prevented by 5-hydroxydecanoate (5-HD), a selective antagonist of mitoK_ATP channels (15, 20, 31), whereas a selective inhibitor of sarcolemmal K_ATP channels has no protective effect (15). While the mechanism of acute cellular protection by diazoxide is not completely understood, one possible explanation is that the opening of the mitoK_ATP channel minimizes calcium influx, mitochondrial swelling, and subsequent injury by depolarizing the membranes of these organelles (15, 21, 35).

In contrast to the heart, little is known about possible cellular protective effects of mitoK_ATP channel activation in the brain. For example, only two studies have specifically targeted mitoK_ATP channels in an attempt to protect the brain against ischemic or anoxic stresses. Thus Domoki et al. (13) have shown that diazoxide administration preserves neuronal function after 10 min of ischemic stress in the piglet cerebral cortex. This protective effect was reversed with coapplication of 5-HD. In the other study, diazoxide prevented cell injury in brain slices during 30 min of anoxia (4). We are unaware of any previous reports except for our own preliminary findings (7, 38) concerning the effects of mitoK_ATP channel opening on infarct volume after ischemia-reperfusion injury in the brain.

In the present study, we explored the potential for the mitoK_ATP channel to protect the brain against ischemic insult. First, we used immunoblotting techniques to establish the presence of essential subunits (inward rectifying K⁺ channel and sulfonylurea receptor) of the K_ATP channel on brain mitochondria. We chose to examine the Kir6.1 and sulfonylurea receptor (SUR)2 subunits based on previous studies (5, 47) and pilot experiments. Second, we evaluated the direct effect of diazoxide on membrane potential (_m) of rat brain isolated mitochondria using fluorescent approaches. Third, we investigated whether pharmacological opening of the mitoK_ATP channel would limit ischemic damage after transient middle cerebral artery (MCA) occlusion (MCAO) in the rat. Finally, we examined the effects of diazoxide on cerebral blood flow (CBF) and arterial blood pressure during control, ischemic, and postischemic conditions.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

All experiments were performed on male Wistar rats weighing 190-210 g. The procedures were approved by the Institutional Animal Care and Use Committee.

Mitochondria preparation. Pure, isolated mitochondria preparations from the rat brain were collected using a discontinuous percoll gradient method (27, 37, 41). This protocol offers sufficient yield with very low contamination from other organelles (1-4%), as demonstrated by electron microscopy and by Western blotting for proteins associated with contaminating structures (27, 34). Additionally, we have shown that mitochondria isolated by this method maintain a _m, have normal respiratory values, and continue to import proteins via an energy-dependent mechanism (34). Adult male Wistar rats were anesthetized with thiopental sodium (250 mg/kg ip) and decapitated. The brain was removed and homogenized by eight strokes of a Teflon-glass motorized tissue grinder (Wheaton; Milville, NJ) in isolation buffer containing 12% percoll (Amersham Pharmacia; Uppsala, Sweden). Mitochondria were isolated in sucrose buffer containing 0.32 M sucrose, 1 mM EDTA, and 10 mM Tris · HCl; pH 7.4. The resulting suspension was centrifuged for 3 min at 3,000 rpm, and the middle layer was saved and layered on top of a discontinuous percoll gradient (24/40%). The gradient was centrifuged for 5 min at 19,000 rpm, and the third layer containing the purified mitochondria was collected. The preparation was washed in isolation buffer and centrifuged for 10 min at 14,000 rpm, and the pellet was used for further investigations. All procedures were performed on ice.

Western blotting. Protein was extracted from freshly isolated mitochondria by the addition of boiling lysis buffer [containing 1% (vol/vol) of 1 mol/l Tris and 1% (wt/vol) sodium dodecyl sulphate]. The samples were sonicated, heated at 95°C for 5 min, and centrifuged for 20 min at 12,000 rpm at 4°C. The supernatant was used for immunoblotting. Protein concentration was measured by a Bio-Rad DC protein assay kit (Bio-Rad Laboratories; Hercules, CA). An equal volume of sample buffer [45% (vol/vol) of 0.5 mol/l Tris (pH 6.8), 45% (vol/vol) glycerol, 10% (vol/vol) H₂O, 0.75 mg/ml bromophenol blue, and 90 mg/ml sodium dodecyl sulphate] was added to each sample. Equal amounts of protein were separated on a 4-20% gradient mini gel (Bio-Rad) and transferred to a polyvinylidene difluoride membrane. After the membrane was blocked with 3% bovine serum albumin, primary antibodies against the K_ATP receptor subunits Kir6.1 or SUR2 (Santa Cruz Biotechnology; Santa Cruz, CA) were applied followed by horseradish peroxidase-conjugated secondary antibody. Chemiluminescence was used to visualize the bands. Molecular mass markers (Bio-Rad) were included on each blot. Specificity of the method was tested by omitting the primary antibody from the procedure or preabsorption with the respective blocking peptides, which resulted in the disappearance of the specific bands (data not shown).

Fluorescent microscopy. Freshly isolated mitochondria were dispersed in buffer containing 125 mM KCl, 2 mM K₂HPO₄, 5 mM MgCl₂, 10 mM HEPES, 10 µM EGTA, 5 mM Na-malate, and 5 mM Na-glutamate at pH 7.0. The mitochondria were visualized using a Zeiss scanning confocal microscope with a fluorescein filter set. Mitochondrial _m was monitored by the addition of 2 µg/ml rhodamine-1,2,3 (36). Three subsequent images were recorded from each preparation: 1) baseline, 2) with the addition of diazoxide (0.1 mM), and 3) with the addition of nigericin (0.01 mM). The ionophore nigericin served as a positive control, because it fully decreases the transmembrane proton gradient (pH) and converts it to _m. Mitochondrial autofluorescence was routinely recorded in each preparation before adding rhodamine-1,2,3. The intensity of autofluorescence of the mitochondria was negligible compared with the intensity after addition of rhodamine-1,2,3.

MCAO experiments. Animals were fasted overnight with access to water before surgery. After anesthesia was induced with 5% halothane in 95% oxygen, endotracheal intubation was performed, and all the animals were mechanically ventilated with 1.0-1.5% halothane in a 70:30 gas mixture of N₂O and O₂ (SAR-830 Ventilator, CWE). The tail artery was cannulated for continuous monitoring of mean arterial blood pressure (MABP; model 300, Blood Pressure Analyzer, Digi-Med) and intermittent sampling of arterial blood for determination of gases and pH. The head was then stabilized in a stereotactic frame, and a hole was made in the skull. A total amount of 30 µl diazoxide (0.4 mM, n = 6, or 2 mM, n = 13) was infused into the right lateral ventricle (AP, 0.8 mm, lateral 1.5 mm, and dorsoventral 3.5 mm), according to the rat brain atlas (32) 15 min before the induction of ischemia (9). Diazoxide was initially dissolved in a small amount of dimethyl sulfoxide (DMSO) and further diluted in buffered saline. Additionally, we studied a sham-treated group (n = 12) in which one-half of the rats were treated with 30 µl buffered saline alone and the other half were treated with an appropriate amount of DMSO in 30 µl buffered saline. This amount of DMSO corresponded to that administered with the higher dose of diazoxide. Because there were no substantial differences in total infarct volume between these two vehicle groups (41.9 ± 6.7% saline alone vs. 45.3 ± 7.1% DMSO in saline), the data were combined into one sham group. Finally, we examined the specificity of the effect of diazoxide using the mitochondrial K_ATP channel inhibitor 5-HD. Ten microliters of 100 mM 5-HD was infused into the right lateral ventricle over 20 s, 5 min before the treatment with the high dose of diazoxide (n = 6).

CBF assessment during MCAO. CBF was measured on the right front parietal cortex with a laser Doppler flowmeter (Multichannel Laser Doppler System, PERIMED) during resting conditions, during ischemia, and during reperfusion in two groups of rats. MABP was also measured during the same conditions. We have used these procedures previously (37). One group of rats received 2 mM in 30 µl saline infused into the right lateral ventricle before MCAO (n = 6). Another six sham animals received either saline alone (n = 2) or DMSO in saline (n = 4) infused into the right lateral ventricle before MCAO.

Statistical analysis. Data are expressed as means ± SE. The physiological parameters among different treatment groups were compared by one-way ANOVA. Neurological score was compared among all treatment groups 24 h after ischemia by one-way ANOVA followed by Fisher least-significant-difference post hoc analysis. The percent infarction was compared among different treatments by one-way ANOVA followed by Fisher least-significant-difference post hoc analysis or t-test, whichever was appropriate. A P value <0.05 was regarded as statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Immunoblotting. Both necessary subunits, which are required for a functional K_ATP channel, were found in isolated brain mitochondria. The specific antibody for the inwardly rectifying K⁺ channel (Kir6.1) recognized a single band of ~50 kDa (Fig. 1B). This band is enriched in mitochondria, showing that the primary localization of the Kir6.1 subunit in the brain is in this organelle. Similarly, the anti-SUR2 antibody recognized a specific band of ~130 kDa, which was also enriched in the mitochondria compared with the full brain tissue level (Fig. 1A). We also compared the amounts of intact Kir6.1- and SUR2-immunoreactive bands in the territory affected by MCAO with the contralateral, nonischemic side. However, we were unable to detect any substantial differences in the intensity of the bands from either side for either subunit (data not presented).

View larger version (85K): [in this window] [in a new window]   Fig. 1.   Western blots using antibodies against sulfonylurea receptor (SUR)2 (A) or Kir6.1 (B) for the whole brain and isolated brain mitochondria (mito). Immunopositive bands were present for each protein in both brain and brain mitochondria, and, on a per weight basis, both bands showed enrichment of respective proteins in mitochondria compared with the whole brain.

Fluorescent study. Autofluorescence was negligible before application of rhodamine-1,2,3. During baseline conditions, the mitochondria showed a weak fluorescent signal with rhodamine-1,2,3 (Fig. 2A). The addition of diazoxide increased the intensity of fluorescence, indicating a decrease in pH (Fig. 2B). As a positive control, we added the ionophore nigericin. Nigericin fully converts pH to _m and maximizes the fluorescent signal (data not shown). Thus the addition of diazoxide markedly alters mitochondrial function through the depolarization of the inner membrane.

View larger version (121K): [in this window] [in a new window]   Fig. 2.   Confocal fluorescent images of isolated rat brain mitochondria using the fluorophore rhodamine-1,2,3. Left, mitochondria in transmitted light with DIC optics; right, fluorescence of the same field. A: baseline condition; B: in the presence of diazoxide (0.1 mM).

MCAO. In all animals studied, the systemic physiological parameters measured were kept within the normal range throughout the manipulations. The treatment groups did not differ with respect to rectal temperature, blood gases, glucose, or blood pressure (data not shown).

View larger version (9K): [in this window] [in a new window]   Fig. 3.   Neurological scores 24 h after transient middle cerebral artery occlusion (MCAO). Values are means ± SE. Treatment with 2 mM diazoxide significantly improved the neurological scores 24 h after transient MCAO compared with sham treatment. Although not significant, treatment with 0.4 mM diazoxide tended to increase the neurological score. Administration of 5-hydroxydecanoate (2 mM diazoxide + 5-HD) completely abolished the protective effect of 2 mM diazoxide. *P < 0.01 vs. sham and P < 0.05 vs. 2 mM diazoxide + 5-HD.

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Representative coronal sections of 2,3,5-triphenyltetrazolium chloride-stained brains 24 h after MCAO. Left, slice from the brain of a sham-treated animal; right, slice from the brain of an animal receiving 2 mM diazoxide. Infarction size was less in the diazoxide-treated animal.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Infarct volume 24 h after transient MCAO. Values are means ± SE. For the cortex and basal ganglia, the treatment with 2 mM diazoxide reduced the infarction size compared with the saline treatment. Coapplication of 5-HD with 2 mM diazoxide (2 mM diazoxide + 5-HD) prevented this reduction. Treatment with 0.4 mM diazoxide did not affect infarct volume. *P < 0.05 vs. sham and P < 0.01 vs. 2 mM diazoxide + 5-HD; **P < 0.05 vs. sham and P < 0.01 vs. 2 mM diazoxide + 5-HD.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Cerebral blood flow (CBF) and mean arterial blood pressure (MABP) in animals treated with diazoxide (2 mM in 30 µl saline, n = 6) or in sham animals (n = 6) during the ischemia and reperfusion periods. Values are means ± SE. The left y-axis indicates CBF expressed as a percentage of the control value. The right y-axis shows MABP expressed as a percentage of the control value. There were no significant differences in CBF and MABP between the two groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The major new finding from the present experiments is that the pretreatment with the mitoK_ATP channel opener diazoxide had a neuroprotective effect against transient focal cerebral ischemia in the rat. While decreases in infarct volume were particularly prominent in the cerebral cortex, significant neuroprotection also occurred in subcortical areas. The neuroprotective effects of diazoxide were reversed by coadministration of 5-HD. Additionally, data from isolated mitochondria provide evidence that the obligatory subunits of K_ATP channels are present in these organelles and that diazoxide-induced dissipation of _m may be involved in this protective effect. Finally, CBF responses to ischemia and reperfusion were unaffected by diazoxide treatment. Thus the selective targeting of the mitoK_ATP channel appears to lessen the impact of ischemic insult to the brain.

The current study supports and extends earlier results from studies in our laboratory and from another laboratory. In the first study, we found that diazoxide treatment given before 10 min of global ischemia preserved normal dilator responses of cortical arterioles to topical application of N-methyl-D-asparate (NMDA) (13). Application of NMDA induces dilation via production of nitric oxide (NO) by cortical neurons and subsequent diffusion of NO to surface arterioles, where NO relaxes vascular smooth muscle cells (13). Typically, ischemia-reperfusion dramatically decreases arterial dilation to NMDA via a mechanism involving altered function of NMDA receptors located on neurons. In the second study, diazoxide treatment protected cortical cells in brain slices during 30 min of anoxia (4). Limitations of these studies are that only immediate effects (30-60 min) and not long-term effects of anoxia were examined and the responses of noncortical cells were not investigated. Furthermore, in the latter study, it is unclear whether diazoxide treatment prevented or simply prolonged survival of neocortical cells, because under anoxic conditions without reoxygenation it is expected that all neurons would eventually die. These initial studies provide circumstantial evidence for the participation of mitoK_ATP channels in neuroprotection.

The combination of several different approaches in the present study provides the first direct evidence for an important role of the mitoK_ATP channel in limiting the neurological injury due to MCAO. Administration of diazoxide, a specific activator of mitoK_ATP channels, limited the extent of death of both cortical and subcortical neurons against 90 min of ischemia induced by transient MCAO (10, 20, 31). Thus, 24 h after reperfusion, dramatic reductions in infarct volume were seen in diazoxide-treated animals compared with sham animals. Furthermore, coadministration of 5-HD, a highly selective antagonist of mitoK_ATP channels, completely reversed neuroprotection (15, 20, 31). Brain infarct volume was similar in animals receiving vehicle or 5-HD alone before ischemia (unpublished observations). In addition, we found that both obligatory subunits required for a functional K_ATP channel were found in isolated brain mitochondria. The specific antibody for the inwardly rectifying K⁺ channel (Kir6.1) recognized a single band of ~50 kDa. Similarly, the specific antibody against SUR2 recognized a specific band of ~130 kDa. These bands are enriched in mitochondria compared with brain tissue, indicating the importance of these K⁺ channels to function of these organelles. The molecular masses for these subunits are comparable with values reported by other investigators (1, 4, 5, 47). Additionally, fluorescent approaches established the ability of diazoxide to depolarize mitochondria.

Although we (45) and other investigators (4, 18, 43) have shown that nonspecific K⁺ channel agonists have similar protective effects, we believe that this is the first report of the administration of a selective mitoK_ATP channel opener increasing neuronal survival after transient brain ischemia (43). However, precise mechanisms by which mitoK_ATP channel activation protects the brain are unclear. We believe that indirect protective effects via increases in CBF are unlikely, because our previous studies indicate that diazoxide does not affect tone of cerebral arteries (13), and the present results indicate that CBF responses during and after ischemia are similar in diazoxide-treated and sham animals. One possible mechanism is that acute activation of mitoK_ATP channel results in K⁺ influx, organelle depolarization, and expansion of mitochondrial matrix volume. The results of our fluorescent experiments provide direct evidence that diazoxide can decrease _m in brain mitochondria. Regulation of matrix volume is an essential element in the regulation of mitochondrial energy production. For example, matrix expansion has been postulated to activate electron transport and stimulate mitochondrial metabolism, which may "cushion" cells against transient ischemic events (15, 22). Additionally, mitochondrial membrane depolarization induced by K⁺ influx through mitoK_ATP channels is expected to dissipate _m and thereby decrease Ca²⁺ influx during ischemia. Thus the driving force for Ca²⁺ influx through the Ca²⁺ uniport is reduced, consequently attenuating mitochondrial Ca²⁺ overload (12, 14, 23). Mitochondrial Ca²⁺ overload has been closely correlated with mitochondrial damage, which results in both necrotic and apoptotic forms of cell death (2, 8, 23, 40, 44). For example, the accumulation of Ca²⁺ in mitochrondria damages complex I of the electron transport chain (22), increases production of reactive oxygen species, and causes formation of mitochondrial permeability transition pores, all of which can lead to cell death (6, 11, 25, 33). Thus, in addition to increasing energy production before ischemia, diazoxide treatment may also protect neurons by attenuating or delaying mitochondrial Ca²⁺ overload.

In conclusion, the present study has provided the first direct evidence that selective opening of the mitoK_ATP channel provides neuroprotection against ischemia-reperfusion in the rat brain. Thus, despite profound differences in cellular physiology and sensitivity to anoxic injury between myocardial cells and neurons, the same pharmacological approach has been shown to protect the heart and brain, respectively, against ischemia-reperfusion injury.