Free radicals have been shown to play an important role in ischemia-reperfusion injury in several organ systems; however, the role of free radicals in central nervous system ischemia has been less well studied. Many potential free radical-generating systems exist. The primary products of these reactions, superoxide and hydrogen peroxide, may combine to produce hydroxyl radicals. Of the many potential sources of free radical generation, the enzyme xanthine oxidase has been shown to be important in ischemia in noncerebral tissue. We investigated the effect of the hydroxyl radical scavenger dimethylthiourea and the xanthine oxidase inhibitor allopurinol on infarct volume in a model of continuous partial ischemia. Male Sprague-Dawley rats were treated with dimethylthiourea or allopurinol before middle cerebral artery occlusion. Infarct volume was measured by triphenyltetrazolium chloride staining of brains removed 3 or 24 hours after occlusion. Stroke volume was reduced by 30% after dimethylthiourea treatment and by 32-35% after allopurinol treatment. At 24 hours after stroke, cortical tissue was more effectively protected than caudate tissue with both agents. Pretreatment with dimethylthiourea and allopurinol also significantly reduced cerebral edema formation and improved blood-brain barrier function as measured by fluorescein uptake. Our results imply that hydroxyl radicals are important in tissue injury secondary to partial cerebral ischemia and that xanthine oxidase may be the primary source of these radicals. (Stroke 1989;20:488-494)

Summary: Oxygen free radicals have been implicated as mediators of tissue damage in ischemic brain . We previ ously demonstrated that the hydroxyl radical scavenger 1,3-dimethyl-2-thiourea (DMTU) reduces infarct size af ter middle cerebral artery occlusion (MCAO) in rats. The present study was undertaken to determine whether this protection results from a preservation of the CBP . Adult male Sprague-Dawley rats were treated with DMTU (750 mg/kg i. There is increasing evidence that oxygen free rad icals are important mediators of brain injury during cerebral ischemia (Yamamoto et aI., 1983; Asano et aI., 1984; Watson et aI., 1984; Abe et aI., 1988; Hall et aI., 1988; Patt et aI., 1988; Young et aI., 1988; Liu et aI., 1989; Martz et aI., 1989); however, the mech anisms by which these toxic compounds damage brain tissue are unknown. Some studies suggest that the microvasculature may be a major site of injury. Free radicals clearly cause increased vascu lar permeability in intestine (Parks and Granger, 1983), skeletal muscle (Korthuis et aI., 1985), and hamster cheek pouch (Del Maestro et aI., 1981). In brain, administration of exogenous free radicals produces increased cerebrovascular permeability (Chan et aI., 1984;Olesen, 1987) and brain edemaReceived May 29, 1989; revised October 16, 1989; accepted October 18, 1989. Address correspondence and reprint requests to Dr. A. L. Betz at D3227 Medical Professional Building, University of Michigan, Ann Arbor, MI 48109-0718, U. S. A.Abbreviations used: AIB, a-aminoiso-butyric acid; BBB, blood-brain barrier; DMTU, 1,3-dimethyl-2-thiourea; MeAO, middle cerebral artery occlusion. 352([3H]a-aminoisobutyric acid) . CBF was reduced in a graded fashion in the ischemic cortex: 0.169 ± 0.020, 0.261 ± 0.017, and 0.435 ± 0.023 mUg/min (mean ± SEM, n = 8) after 4 h in the central, intermediate, and outer zones, respectively . Brain edema was present in a similar pattern, while blood-brain barrier permeability remained normal . Treatment with DMTU significantly reduced brain edema in the central and intermediate zones at both 4 and 24 h. However, CBP in the DMTU-treated animals was identical to that of the vehicle-treated animals . These results suggest that hydroxyl radicals play a role in the development of ischemic brain edema, but the mecha nism does not appear to involve a direct effect on CBP .

Xanthine oxidase (XO) has been proposed as an important source of free radicals during ischemia. This enzyme normally exists as a dehydrogenase (XD), but it is converted to XO in some ischemic tissues. Recently, treatment of animals with the XD and XO inhibitor allopurinol or with free radical scavengers before cerebral ischemia has been shown to reduce brain injury. Therefore, we studied conversion of XD to XO in three ischemic and nonischemic brain regions during focal cerebral ischemia resulting from permanent occlusion of the middle cerebral artery (MCAO) in anesthetized rats. In nonischemic brain, 16-22% of the enzyme was in the XO form. After 24 h of ischemia this value was not significantly different (10-15%). Neither the total activity of XO nor that of XD changed, indicating that there was no irreversible conversion of XD to XO. To further explore the possible role of XO, we examined the effect of various doses of allopurinol (5, 20, or 100 mg/kg given 1 h before MCAO or 100 mg/kg given 48, 24, and 1 h before MCAO) on uric acid accumulation, brain edema formation, and cerebral blood flow (CBF) 24 h after MCAO. All but the lowest dose of allopurinol greatly reduced the appearance of uric acid in the ischemic brain; however, only the highest dose of allopurinol had any beneficial effect on brain edema. This reduction in brain edema occurred without a significant improvement in CBF. Thus XO is probably not an important source of free radicals in this model of focal cerebral ischemia.