Brain Research 1051 (2005) 72 – 80 www.elsevier.com/locate/brainres
Research Report
Diazoxide preconditioning attenuates global cerebral ischemia-induced blood–brain barrier permeability Ga´bor Lenzse´r a,b,*, Be´la Kis a, Ferenc Bari c, David W. Busija a a Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston-Salem, NC 27157-1010, USA Institute of Clinical Experimental Research and Human Physiology, Semmelweis University, Budapest, U¨llo˝i u´t 78/a, H-1083, Hungary c Department of Physiology, Faculty of Medicine, University of Szeged, Szeged, Do´m te´r 10, H-6720, Hungary
b
Accepted 23 May 2005 Available online 6 July 2005
Abstract Brain edema formation due to blood – brain barrier (BBB) disruption is a major consequence of cerebral ischemia. Previously, we demonstrated that targeting mitochondrial ATP-sensitive potassium channels (mitoKATP) protects neuronal tissues in vivo and in vitro, however, the effects of mitoKATP openers on cerebral endothelial cells and on BBB functions have never been examined. We investigated the effects of mitoKATP channel opener diazoxide on BBB functions during ischemia/reperfusion injury (I/R). Rats were treated with 6, 20 or 40 mg/kg diazoxide ip for 3 days then exposed to global cerebral ischemia for 30 min. BBB permeability was assessed by administering Evan’sblue (EB) and Na-fluorescein (NaF) at the beginning of the 30 min reperfusion. I/R increased BBB permeability for the large molecular weight EB (ng/mg) in the cortex (control: 146 T 12, n = 7; I/R: 1049 T 152, n = 11) which was significantly attenuated in diazoxide-treated rats (575 T 99, n = 9; 582 T 104, n = 8; 20 and 40 mg/kg doses). Diazoxide pretreatment also significantly inhibited the extravasation of the low molecular weight NaF. Edema formation in the cortex was also decreased after diazoxide pretreatment. In cultured cerebral endothelial cells, diazoxide depolarized the mitochondrial membrane, suggesting a direct diazoxide effect on the endothelial mitochondria. Our results demonstrate that preconditioning of cerebral endothelium with diazoxide protects the BBB against ischemic stress. D 2005 Elsevier B.V. All rights reserved. Theme: Disorder of the nervous system Topic: Ischemia Keywords: Global cerebral ischemia; Blood – brain barrier; Preconditioning; Diazoxide
1. Introduction The blood – brain barrier (BBB) is responsible for the limited and regulated movement of plasma constituents into the brain parenchyma. During pathophysiological conditions, damage to endothelial cells and alterations in BBB function can have adverse effects on the brain. BBB disruption was found to precede and may be the initiating event of focal ischemic lesions in hypertensive encephal-
* Corresponding author. Medical Center Boulevard, Winston-Salem, NC 27157, USA. Fax: +1 336 7160237. E-mail address:
[email protected] (G. Lenzse´r). 0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2005.05.064
opathy [47]. BBB injury can worsen the outcome after cerebral ischemia as well [26]. Therefore, it is important to develop approaches to limit BBB dysfunction after cerebral ischemia. One effective approach that has been first shown to protect neurons is ischemic preconditioning (IPC). Thus, subjecting the brain to a short duration of an ischemic event can induce tolerance to a longer ischemic insult applied later [7,21]. Ischemic preconditioning also can protect aortic and coronary endothelial cells from ischemia/reperfusion (I/R)induced injury, showing that the vasculature is not an exception of this type of protection [3,48]. In the cerebrovascular compartment, IPC was shown to attenuate brain edema formation and BBB disruption after permanent middle cerebral artery occlusion (MCAO) [31] and to
G. Lenzse´r et al. / Brain Research 1051 (2005) 72 – 80
reduce injury in primary cerebral endothelial cells during and following oxygen-glucose deprivation (OGD) [2]. The exact underlying mechanisms of IPC are unknown; however, IPC can be triggered by a variety of other stimuli, including heat shock, exercise, opioids and cytokines. In addition, activation of mitochondrial ATP-sensitive potassium (mitoKATP) channels has been proposed to play a pivotal role in preconditioning [37]. Pharmacological agents that open mitoKATP channels reproduce preconditioning without any other intervention [15,46]. Moreover, physiological or chemical preconditioning is prevented by blockers of the mitoKATP channels [46]. The beneficial effects of the prototype mitoKATP channel opener diazoxide have been well demonstrated in the heart and other organs and have been shown in experimental neurological preparations by our laboratory [10,40,44] and by others [6,28,35,43]. The purpose of our study was to determine whether diazoxide pretreatment affects BBB function following ischemic stress in rats. We tested the hypothesis that diazoxide would reduce BBB permeability and decrease brain water content following global cerebral I/R and that mitoKATP channel opening would be involved in its action. Furthermore, we examined whether diazoxide depolarizes cultured endothelial cells.
2. Methods 2.1. Surgery and experimental groups Experiments were carried out on male Wistar rats, weighing 270– 320 g. Prior to the surgical procedures, food was deprived for 2 h. Anesthesia was induced with 3.5% halothane in a mixture of 72% nitrous oxide and 28% oxygen and maintained on 1.2% halothane in the spontaneously breathing animals. The femoral artery and vein were cannulated for physiological monitoring, infusion of drugs and blood withdrawal. Via a midline cervical incision, both common carotid arteries were isolated from adhering tissue and nerves, and silk threads (3-0) were placed loosely around them. Arterial blood pressure was monitored continuously, and blood gases and pH were measured serially. The animals were exposed to global
73
cerebral ischemia using combined bilateral common carotid artery occlusion and arterial hypotension. Following administration of heparin (500 IU/kg iv.), mean arterial blood pressure was lowered to 35 – 40 mm Hg by blood withdrawal (1.7 – 2.4 ml/100 g body weight) through the femoral vein, and both common carotid arteries were occluded by placing microaneurysm clips around the vessels for 30 min. Reperfusion was achieved by removing the clips and giving back the shed blood with a speed of 1.2 ml/min. The body temperature was measured and maintained at 37.5 T 0.2 -C. Animals were divided into seven experimental groups. Animals in Group 1 (n = 7) served as sham controls. Six other groups were exposed to I/R. Group 2 (n = 11) received vehicle (n = 11), and Groups 3 (n = 5), 4 (n = 9) and 5 (n = 8) received diazoxide administered ip at doses 6, 20 and 40 mg/kg on three consecutive days, respectively. The final diazoxide or vehicle injection was given 24 h before the global cerebral ischemia. Groups 6 (n = 4) and 7 (n = 4) were treated by the diazoxide antagonist 5-hydroxydecanoic acid (5-HD) at the dose of 40 or 100 mg/kg (ip) 20 min before the DIAZ (40 mg/kg) injection. 2.2. Quantitation of BBB opening Blood –brain barrier permeability was assessed by measurement of EB and NaF content in brain [1,4]. EB dye or NaF (2% solution for each, given at 4 ml/kg) were administered with the shed blood at the start of reperfusion and allowed to circulate for 30 min. Then, the animals were transcardially perfused with 1000 ml/kg heparinized PBS at 100 mm Hg pressure at room temperature. The brains were removed and rinsed with PBS, and two 4 mm wide coronal slices were made, starting at bregma levels +1.80 mm and 2.20 mm. After removing the meninges, the cerebral cortex above the rhinal fissure from the first slice and both hippocampi from the second slice were dissected as shown (Fig. 1). Each brain piece was weighed. Then, the samples were homogenized in 10 times volume of 50% trichloroacetic acid to precipitate protein and centrifuged for 10 min at 13,600 g. For the EB measurement, the supernatant was diluted with ethanol (1:3), and its fluorescence was determined (excitation at 620 nm and emission at 680 nm) [4]. For
Fig. 1. TTC stained coronal brain slices after 30 min global cerebral ischemia followed by 24 h reperfusion. Left: bregma 0.2 mm; right: bregma 4.2 mm; central section of the investigated 4-mm-wide coronal slices. The marked areas depict the sampled cortical and hippocampal regions. Decreased staining indicates ischemia/reperfusion damage.
74
G. Lenzse´r et al. / Brain Research 1051 (2005) 72 – 80
the NaF measurement, the supernatant was diluted with 5 M NaOH (1:0.8) to increase the pH, and its fluorescence was determined (excitation 440 nm and emission at 525 nm). A microplate spectrophotometer (BMG FLUOstar Optima) was used for the fluorescence measurements. Calculations were based on external standards in the same solvent (10 – 200 ng/ ml). The tissue contents of the dyes were quantified from linear standard curves derived for each of the dyes and were expressed per gram of tissue. 2.3. Determination of brain water content Cortical water content was determined by the dry –wet weight method on sham (n = 3), vehicle (n = 3) and 20 mg/ kg diazoxide (n = 3)-treated rats [13]. Following 30 min of ischemia, reperfusion was allowed for 4 h. After decapitation, cortical samples were obtained, as described and weighed immediately to obtain the wet weight. The tissue was then dried in an oven at 100 -C for 24 h and reweighed to obtain the dry weight. Brain water content was calculated as (wet weight dry weight) / wet weight 100%.
2.6. Cerebral blood flow (CBF) measurement Application of diazoxide lowers the blood pressure, and this could affect the cerebral perfusion consequently, we tested the effects of the highest dose we used on these parameters. In an additional group (Group 8) of animals (n = 7), we measured cerebral blood flow and arterial blood pressure changes just after ip administration of 40 mg/kg diazoxide. The head of the animals was secured in a stereotactic frame. The cerebral blood flow was measured continuously before and after diazoxide injection with a laser-Doppler probe (Perimed; Probe 403; Sweden) with a diameter of 0.25 mm. The probes were placed over the thinned skull 2 mm caudal to bregma, 4 mm lateral to midline. The baseline value of 100% CBF was defined as the average laser-Doppler flux of a 5 min period, thereafter 1 min periods were averaged in every 3 min before and after ip diazoxide application. The parameters were checked for 30 min following the injection. Blood flow values were expressed as percentage of the baseline. 2.7. Drugs
2.4. Rat cerebral endothelial cell culture Primary rat cerebral endothelial cells (CECs) were isolated as previously described [22] and were seeded onto collagen type IV and fibronectin-coated glass cover slips. The endothelial culture medium consisted of Dulbecco’s Modified Eagles medium (DMEM) supplemented with 20% fetal bovine plasma derived serum (Animal Technologies Inc., Tyler, TX), 2 mM glutamine, 1 ng/ml basic fibroblast growth factor, 100 Ag/ml heparin, 5 Ag/ml vitamin C and antibiotics. Confluent cultures (4 –5th day in vitro) consisted of more than 95% of RCECs verified by positive immunohistochemistry for von Willebrand factor and negative immunochemistry for glial fibrillary acidic protein (GFAP) and a-smooth muscle actin.
All chemicals used were obtained from Sigma if not mentioned otherwise. Diazoxide was diluted in 0.2 M NaOH solution prior to further dilution in physiologic saline for intraperitoneal injection. 2.7.1. Data analysis Data are presented as means T SEM. Group differences were determined by a one way ANOVA followed by pairwise comparisons using the Student –Newman –Keuls test. In the case of cerebral blood flow and blood pressure measurement, values of different time points were compared by one way repeated measures ANOVA. Differences between values were considered significant when P < 0.05.
2.5. Analysis of mitochondrial membrane potential (DWm )
3. Results
DCm was monitored using the DCm-sensitive dye, tetramethylrhodamine ethyl ester (TMRE, Molecular Probes, Eugene, OR, USA), as described previously [23]. CEC cultures were loaded in the dark at 37 -C in a 5% CO2 incubator with 1.75 AM TMRE in DMEM for 1 h. After loading, the cells were washed three times with PBS. Experiments were carried out at 22 -C in PBS. Confocal images of cellular TMRE fluorescence were acquired on a Zeiss LSM 510 laser scanning microscope using a 63 water immersion objective (Zeiss, Jena, Germany). Fields of cells were randomly selected. The cells were treated with diazoxide, and fluorescent images were recorded every 20 s for 5 min after treatment (k ex = 543, k em > 560 nm). The average pixel intensity in individual cell bodies was determined using the software supplied by the manufacturer (Zeiss).
3.1. Physiological variables Rectal temperatures, blood gases and pH and plasma glucose levels showed no significant differences among groups (Table 1). Mean arterial blood pressure was significantly elevated after 40 mg/kg diazoxide pretreatment compared to vehicle, but there was no difference among the sham, vehicle, 6 mg/kg diazoxide or 20 mg/kg diazoxide groups (Table 1). In the 100 mg/kg 5-HD group, the standard base excess and the hematocrit were slightly elevated compared to the diazoxide group. 3.2. Blood – brain barrier permeability In the cortical tissue, relatively little EB was present in the sham rats (Fig. 2A). However, the amount of EB
G. Lenzse´r et al. / Brain Research 1051 (2005) 72 – 80
75
Table 1 Physiological parameters for rats before ischemia or sham operation after vehicle or diazoxide pretreatment Control
6 mg/kg
20 mg/kg
40 mg/kg
n 18 (7+11) 5 9 8 Blood glucose 95.3 T 15.7 98.5 T 4.4 89.3 T 8.4 90.4 T 18.8 MBP 90.9 T 5.1 88.4 T 8.1 92.6 T 3.8 101.7 T 5.6* pH 7.39 T 0.03 7.40 T 0.02 7.40 T 0.02 7.40 T 0.03 PCO2 46.1 T 3.5 45.6 T 3.5 45.0 T 1.6 46.4 T 5.2 PO2 132.9 T 7.6 129.0 T 3.7 131.7 T 7.9 126.8 T 6.5 SBE 2.9 T 1.4 3.1 T 2.3 2.3 T 1.2 3.0 T 1.4 Hct 40.2 T 2.4 42 T 2.3 41 T 1.1 41.8 T 2.9 Data given as mean T SEM; *P < 0.001 different from control (control includes the combined data of the sham and vehicle-treated ischemic groups). Blood glucose in mg/dl. Mean blood pressure (MBP), pCO2 and pO2 in mmHg. Standard base excess (SBE) in mmol/l. Hematocrit (Hct) in percent.
increased dramatically following ischemia in the vehicletreated animals. Diazoxide administration reduced the amount of EB in a dose-dependent fashion. Qualitatively similar results were obtained for NaF, but the magnitude
Fig. 2. (A,B) Evan’s blue (A) and sodium fluorescein (B) concentrations in the cerebral cortical tissue after sham operation and following ischemia/ reperfusion in vehicle or diazoxide-pretreated animals. Data are shown as mean T SEM, *P < 0.05 compared with vehicle treatment (A,B); ## < 0.001, # < 0.05 compared with sham (A). All groups are significantly different from the sham group on panel (B), P < 0.001.
Fig. 3. Evan’s blue and sodium fluorescein concentrations in the cerebral cortex after ischemia/reperfusion. Pretreatment with diazoxide (40 mg/kg) or diazoxide (40 mg/kg) following 5-hydroxydecanoic acid (5-HD) (40 mg/ kg or 100 mg/kg). Data are shown as mean T SEM. All groups are significantly different from the sham group, P 0.001.
of the effects of diazoxide was not as great as with the EB, in spite of the statistically significant difference, the changes were not remarkable (Fig. 2B). However, the two higher doses of diazoxide reduced the EB content of the cortex. Similar to the cortex, EB and NaF contents increased in hippocampus following ischemia from levels observed in sham animals (438.9 T 105.3 vs. 107.6 T 14.4 ng/mg for EB and 850.3 T 60.5 vs. 447.6 T 31.2 ng/mg for NaF). However, diazoxide, even at the highest concentration, did not significantly reduce levels of either dye (419.9 T 154.9, 321.7 T 153.9 and 379.4 T 92.8 ng/mg for EB at 6, 20 and 40 mg/kg diazoxide doses and 867.7 T 103.4, 840.7 T 87.1 and 804.7 T 48.1 ng/mg for NaF respectively). 5-HD, a mitoKATP channel blocker, failed to increase EB and NaF permeability in the cortex given before diazoxide treatment (Fig. 3).
Fig. 4. Water content of cortical brain tissue in the sham, vehicle or 20 mg/ kg diazoxide-pretreated ischemic groups. Water content given as percentage of the wet weight. Data are shown as mean T SEM, *P < 0.05 compared with vehicle-treated ischemic group. Both ischemic groups are significantly different from the sham one, P < 0.001.
76
G. Lenzse´r et al. / Brain Research 1051 (2005) 72 – 80
3.3. Water content After ischemia/reperfusion, the water content of the cortical brain tissue elevated both in the vehicle and diazoxide-treated animals. However, diazoxide pretreatment significantly decreased the cortical edema (Fig. 4). 3.4. Mitochondrial membrane potential in cultured endothelial cells The intensity of TMRE fluorescence is proportional to the mitochondrial membrane potential. Fluorescent intensity at the starting point was regarded as 100%. 250 AM diazoxide caused the loss of mitochondrial membrane potential in cultured primary rat endothelial cells (Fig. 5A,B).
3.5. Acute effects of diazoxide on blood pressure and cerebral blood flow Diazoxide reduced mean arterial pressure and cerebral blood flow in the investigated 30 min period. The maximum decrease was 13.2% in laser-Doppler flux (86.8% from 100%) and was 18.5% in mean blood pressure (70.4 mm Hg from 86.4 mm Hg.) compared to baseline. Both showed a slight increase towards control baseline values at the end of the 30 min (Fig. 6).
4. Discussion The main findings of our experiments are that: (1) diazoxide preconditioning (3 consecutive days before
Fig. 5. (A,B) Effect of diazoxide on mitochondrial membrane potential in cultured rat cerebral endothelial cells. Representative confocal images showing tetramethylrhodamine ethyl ester (TMRE) fluorescence in cultured rat cerebral endothelial cells treated with vehicle or diazoxide (250 AM). Numbers on the images indicate the elapsed time after diazoxide application (A). TMRE fluorescence in cultured rat cerebral endothelial cells, measured by a confocal microscope after exposure to vehicle or 250 AM diazoxide (B). Data are shown as mean T SEM, the corresponding values of each time point were significantly different from each other from 20 s until the end of the experiment (5 min).
G. Lenzse´r et al. / Brain Research 1051 (2005) 72 – 80
Fig. 6. Changes in the cerebrocortical blood flow and the mean arterial blood pressure after intraperitoneal injection of 40 mg/kg diazoxide with time. Data points are averages of 1 min periods. Cerebral blood flow is expressed as percentage of the measured laser-Doppler flux compared to baseline. Data are shown as mean T SEM, post-application values are significantly different from pre-application values from the marked time point until the end of the experiment, P < 0.05 (one way repeated measures ANOVA; n = 7).
ischemia) gave partial protection against BBB opening and brain edema in a severe global ischemic model in rat. (2) Diazoxide depolarizes the mitochondria of brain endothelial cells, raising the possibility of a direct endothelium-derived effect in the preconditioning. The BBB opens in a biphasic fashion in reperfusion, the first opening just after the restoration of blood flow [32,45]. We focused on the early period of BBB disruption because it is easier to predict when it will occur, and the mechanisms are more fully understood. Disruption of the BBB causes extravasation of albumin and other high molecular weight compounds [14,24,30], and passage of low molecular weight substances is increased as well [34]. Extravasation of proteins into extracellular spaces is correlated with the development of vasogenic edema [25], which causes increased intracranial pressure, one of the main life-threatening complications of ischemia/reperfusion injury. In addition, Dietrich et al. found a spatial correlation between increased vascular permeability and neuronal death showing the close interrelationship between neuronal injury and microvascular defects [9]. Since increased BBB permeability could worsen the outcome after ischemia/reperfusion in several ways, searching for protective treatments is important. Ischemic preconditioning, a brief exposure to ischemia before a long period of injurious ischemia, is a powerful method to provide robust protection for cells. After preconditioning, tolerance can be observed within 2 –3 h (acute) or between 24 – 72 h (delayed) after the initial conditioning injury [5]. From a clinical point of view, delayed preconditioning may have additional therapeutic benefits because of its greater window of opportunity. Although the concept of preconditioning has been widely studied, little attention has focused on the effects of preconditioning on BBB disruption. Masada reported BBB protection and decreased brain edema formation in perma-
77
nent MCAO in rats after preconditioning with a brief period of ischemia 3 days before the injurious ischemia [31], and Ikeda showed decreased IgG extravasation after hypoxia– ischemia in newborn rats following heat preconditioning [19]. Diazoxide, an effective mitoKATP channel opener, was extensively used to exert acute tolerance against ischemia in the heart and in the brain [10,15,40,44,46]. Moreover, it has been reported to exert delayed protective effects against I/R in heart [36,49] as well as in neurons and astrocytes [23,33,41]. In the present study, we investigated whether pretreatment with diazoxide could elicit delayed protection against I/R-induced BBB disruption. In our global ischemia reperfusion model, we found opening of the barrier for a low (NaF) as well as for a high (EB-albumin) molecular weight marker in early reperfusion as it was found in similar studies. Diazoxide pretreatment at the two highest doses decreased EB leakage, showing that the BBB was less permeable to albumin after preconditioning. The decrease in NaF permeability was more subtle, perhaps reflecting the known differences in movement of the two dyes through the BBB and different degrees of protection. Small molecules such as NaF may pass through the entire interendothelial cleft into the extracellular space, while leakage of the macromolecules may happen through pinocytotic vesicles, which can be seen in increased number in brain endothelial cells following ischemia [9,34]. A major concern about diazoxide is that, in addition to its potential effects on the mitoKATP channel, diazoxide also inhibits succinate dehydrogenase. Thus, it has been difficult to precisely define the role of mitoKATP channel activation and to separate these effects from events subsequent to succinate dehydrogenase inhibition. In previous studies, we have found that 3-nitropropionic acid (3-NPA), a specific inhibitor of SDH, also protected neurons both in vivo [18] and in vitro [23]. It is not known whether effects of SDH inhibition are additive or counter productive to mitoKATP channel related cellular protection, but neuroprotective effects of 3-NPA are substantially less than what we see with either diazoxide or BMS-191095, a selective mitoKATP channel opener which is free from the known side effects of the prototype mitoKATP channel opener diazoxide. To clarify this issue, we used the putative mitoKATP channel blocker, 5-HD [15], however, it did not inhibit the protective effect of diazoxide in our experiments. In contrast to previous implications that 5-HD is a selective mitoKATP channel blocker, recent studies suggest that 5-HD should no longer be considered as a suitable pharmacological tool to selectively inhibit the mitoKATP channels [8,12,16,17,27]. 5HD is metabolized and may be completely oxidized in mitochondria [12,16,17,27]. In addition, 5-HD was shown very recently to have neuroprotective effect against ischemic and apoptotic cell death [29]. In this context, it is not surprising that 5-HD did not inhibit the BBB protective effect of diazoxide in our experiments.
78
G. Lenzse´r et al. / Brain Research 1051 (2005) 72 – 80
Although the mechanism of protection remains to be elucidated, according to our inquiry, this would be the first evidence of BBB tolerance against I/R induced by pharmacological preconditioning. In addition, our treatment could attenuate the edema formation in the cortex as well, which may be partly due to the decreased extravasation of proteins, but because in ischemic brain edema both cytotoxic and vasogenic edema are involved, effects on cellular functions cannot be excluded. In in vitro, we investigated whether the mitochondria of primary cerebral endothelial cells can be depolarized by diazoxide. Endothelial effects of diazoxide on the cerebrovasculature have been shown recently. Application of diazoxide before ischemia preserved endothelium-dependent (hypercapnia) but not independent (iloprost) cerebrovascular dilator responses following I/R in newborn pigs [11]. The cerebral endothelial cells have a very important role in the establishment and maintenance of the BBB as well. In cooperation with other cell types, principally with the astrocytes, they are responsible for the distinct barrier properties of the cerebrovascular bed [42]. The astrocytes are proved to be capable of showing ischemic tolerance upon diazoxide preconditioning [41]. Our finding that diazoxide affects mitochondrial membrane potential in cerebral endothelial cells raises the possibility of direct endothelial protection in the cerebrovascular bed by diazoxide preconditioning. This is in good agreement with recent findings that primary brain endothelial cells can be preconditioned by short duration oxygen-glucose deprivation [2]. The direct endothelial protection could be especially important in regard to the BBB in the light of new findings that intact cerebral endothelial cells can establish a functioning barrier even after selective astrocytic damage in vivo [50]. Cerebral endothelial protection may affect ischemic brain damage in several ways, in addition to reduced BBB damage and edema formation. For example, protection may be due to inhibition of hemorrhagic transformation, reduced postischemic inflammation and/or alterations in the release of vasoactive substances [2]. Since diazoxide is a well known antihypertensive agent and this may affect cerebral blood flow, we checked the effects of acute application of the drug on these parameters. Both blood pressure and cerebral blood flow decreased after application of the highest dose of diazoxide we used. Will this hypotension and hypoperfusion induced by diazoxide precondition the rat against the ischemia? To address this issue, Kapinya et al. have injected intraperitoneally dihydralazine to decrease the rat arterial blood pressure to
