Temporal characteristics of the protective effect of aminoguanidine on cerebral ischemic damage

Brain Research 802 Ž1998. 104–110

Research report

Temporal characteristics of the protective effect of aminoguanidine on cerebral ischemic damage Fangyi Zhang, Costantino Iadecola

)

Laboratory of CerebroÕascular Biology and Stroke, Department of Neurology, UniÕersity of Minnesota, Box 295 UMHC, 420 Delaware Street S.E., Minneapolis, MN 55455, USA Accepted 19 May 1998

Abstract We investigated the temporal profile of the reduction in focal cerebral ischemic damage exerted by aminoguanidine ŽAG., an inhibitor of inducible nitric oxide synthase ŽiNOS.. In anesthetized spontaneously hypertensive rats, the middle cerebral artery ŽMCA. was occluded distal to the origin of the lenticulostriate arteries. Rats were treated with vehicle Žsaline. or AG Ž100 mg kgy1, i.p.. immediately after MCA occlusion and, thereafter, two times per day. Rats were sacrificed 1 Ž n s 7., 2 Ž n s 8., 3 Ž n s 6. or 4 days Ž n s 5. after MCA occlusion. Injury volume Žmm3 . was determined in thionin-stained sections using an image analyzer. Volumes were corrected for ischemic swelling. Administration of AG up to 2 days after MCA occlusion did not reduce cerebral ischemic damage Ž p ) 0.05 from vehicle; t-test.. Treatment for a longer period decreased injury volume, the reduction averaging 21 " 5% at 3 days Ž p - 0.05. and 30 " 9% at 4 days Ž p - 0.05.. Aminoguanidine did not affect ischemic brain swelling Ž p ) 0.05.. Administration of AG did not substantially modify arterial pressure, arterial blood gases, pH, hematocrit, plasma glucose and rectal temperature. We conclude that the protective effect of AG is time-dependent and occurs only when the drug is administered for longer than 2 days, starting after induction of ischemia. Because iNOS enzymatic activity develops more than 24 h after MCA occlusion wC. Iadecola, X. Xu, F. Zhang, E.E. El-Fakahany, M.E. Ross, Marked induction of calcium-independent nitric oxide synthase activity after focal cerebral ischemia, J. Cereb. Blood Flow. Metab. 14 Ž1995. 52–59; C. Iadecola, F. Zhang, X. Xu, R. Casey, M.E. Ross, Inducible nitric oxide synthase gene expression in brain following cerebral ischemia, J. Cereb. Blood Flow Metab. 15 Ž1995. 378–384.x, the data support the hypothesis that the protective effect of AG is mediated by inhibition of iNOS in the post-ischemic brain. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Nitric oxide; Inducible nitric oxide synthase; Rat; Stroke; Middle cerebral artery

1. Introduction We have previously demonstrated that cerebral ischemia is followed by development of inducible nitric oxide synthase ŽiNOS. activity in the ischemic brain between 24 and 96 h after permanent occlusion of the rat middle cerebral artery ŽMCA. w14,17x. Treatment with the relatively selective iNOS inhibitor, aminoguanidine ŽAG., for 3 days, starting 24 after MCA occlusion, attenuates iNOS activity, reduces infarct size and improves neurological recovery w16,20x. Because iNOS produces NO in large, potentially toxic, amounts w21x, the findings raise the possibility that NO produced by iNOS contributes to the development of cerebral ischemic damage. The recent finding

)

Corresponding author. Fax: q1-612-625-7950

0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 5 5 7 - 5

that null mice lacking the iNOS gene are relatively protected from cerebral ischemic damage adds further support to the hypothesis that iNOS expression is deleterious to the ischemic brain w15x. AG has other pharmacological actions that could potentially contribute to its protective effect. These include inhibition of diamine oxidase w5x, an enzyme that participates in histamine and polyamine catabolism w23x and inhibition of formation of advanced glycation endproducts ŽAGEs. w6x. The latter effect is particularly important because it has been recently demonstrated that exogenous AGEs enhance the ischemic damage produced by focal cerebral ischemia in the rat, an effect that is prevented by AG w27x. In light of the potential therapeutic value of pharmacological iNOS inhibition in stroke, it would be important to obtain additional information on the mechanisms of the protective effect of AG.

F. Zhang, C. Iadecolar Brain Research 802 (1998) 104–110

In this study we, therefore, sought to provide additional insights into the role of iNOS in the protective effect of AG. We reasoned that, if the action of AG on cerebral ischemic damage is related to iNOS inhibition, then AG should be effective only when administered in the period during which there is iNOS expression in the post-ischemic brain. Accordingly, since the expression of iNOS enzymatic activity occurs between 24 and 96 h after permanent MCA occlusion w14,17x, administration of AG during the first day after ischemia should not be effective, whereas treatments extending beyond 1 day after ischemia should reduce ischemic damage.

105

100% oxygen., the left femoral artery was cannulated and rats were placed on a stereotaxic frame ŽD. Kopf Instruments, model 1404, Tujunga, CA.. Body temperature was maintained at 37 " 0.58C by a thermostatically controlled infrared lamp ŽYSI, model 73A-TA, Yellow Springs, OH.. The arterial catheter was connected to a pressure transducer for recording of mean arterial pressure. Plasma glucose was measured by a glucose analyzer ŽBeckman. and arterial blood gases were monitored by a blood gas analyzer Žmodel 178, Ciba-Corning, Medfield, MA.. After completion of the surgical procedures, the arterial catheter was tunnelled under the skin and exteriorized at the level of the tail. The catheter was used for recording of arterial pressure and for determination of plasma glucose, hematocrit and arterial blood gases at different times after MCA occlusion.

2. Materials and methods Methods for MCA occlusion with monitoring of physiological parameters and for determination of injury volume have been described in detail in previous publications w16,17,24x and will only be summarized.

2.2. MCA occlusion and measurement of ischemic injury Õolume Procedures for MCA occlusion and for determination of ischemic injury volume are identical to those published previously w24x. Briefly, a 3–4-mm hole was drilled at a site superior and lateral to the left foramen ovale to expose the left MCA. The MCA was elevated and cauterized distal to the origin of the lenticulostriate arteries and

2.1. General surgical procedures Studies were conducted on 47 male spontaneously hypertensive rats ŽHarland. weighing 300–400 g. Under halothane anesthesia Žinduction: 5%; maintenance: 1%, in

Table 1 The pO 2 , pCO 2 ŽmmHg. and pH in the groups of rats studied Hours after MCAO 0

4

8

12

24

48

72

96

24 h

pH pCO 2 pO 2 pH pCO 2 pO 2 pH pCO 2 pO 2 pH pCO 2 pO 2 pH pCO 2 pO 2 pH pCO 2 pO 2 pH pCO 2 pO 2 pH pCO 2 pO 2 n

48 h

72 h

96 h

Vehicle

AG

Vehicle

AG

Vehicle

AG

Vehicle

AG

7.40 " 0.02 46 " 1 76 " 6 7.43 " 0.01 42 " 1 85 " 3 7.36 " 0.01 45 " 1 81 " 4 7.48 " 0.02 43 " 1 85 " 5 7.39 " 0.01 42 " 2 94 " 3 – – – – – – – – – 6

7.40 " 0.02 47 " 1 75 " 5 7.49 " 0.02 40 " 1 92 " 4 7.42 " 0.02 44 " 2 82 " 8 7.48 " 0.02 43 " 2 87 " 6 7.39 " 0.01 44 " 1 81 " 5 – – – – – – – – – 7

7.40 " 0.01 45 " 1 75 " 4 7.44 " 0.02 43 " 2 82 " 7 7.40 " 0.02 44 " 2 77 " 7 7.41 " 0.01 45 " 1 80 " 6 7.40 " 0.01 46 " 1 69 " 5) 7.40 " 0.03 39 " 2 84 " 7 – – – – – – 6

7.41 " 0.01 44 " 1 83 " 4 7.43 " 0.02 43 " 2 87 " 4 7.42 " 0.01 44 " 1 80 " 4 7.40 " 0.01 45 " 1 77 " 3 7.39 " 0.01 44 " 1 83 " 2 7.41 " 0.01 43 " 2 89 " 4 – – – – – – 8

7.39 " 0.01 47 " 1 82 " 13 7.39 " 0.01 45 " 2 82 " 5 7.37 " 0.02 46 " 2 80 " 11 7.45 " 0.02 42 " 1 94 " 4 7.44 " 0.02 43 " 3 80 " 5 7.44 " 0.02 43 " 3 77 " 6 7.47 " 0.04 44 " 3 76 " 10 – – – 5

7.44 " 0.04 43 " 2 84 " 7 7.44 " 0.02 42 " 1 84 " 9 7.39 " 0.03 43 " 3 92 " 3 7.41 " 0.01 43 " 2 89 " 4 7.46 " 0.03 40 " 3 86 " 6 7.45 " 0.02 42 " 3 86 " 8 7.48 " 0.01 36 " 1 96 " 3 – – – 6

7.38 " 0.02 45 " 1 89 " 1 7.39 " 0.01 46 " 3 95 " 3 7.41 " 0.03 46 " 1 87 " 7 7.43 " 0.03 40 " 1 91 " 11 7.44 " 0.04 43 " 4 90 " 2 7.47 " 0.07 46 " 1 91 " 1 7.47 " 0.02 44 " 2 89 " 1 7.54 " 0.02 45 " 1 81 " 2 4

7.44 " 0.01 42 " 2 86 " 4 7.45 " 0.02 44 " 1 89 " 4 7.43 " 0.01 41 " 2 87 " 5 7.42 " 0.01 43 " 2 83 " 5 7.42 " 0.02 44 " 2 85 " 6 7.48 " 0.02 40 " 3 83 " 5 7.53 " 0.05 42 " 3 85 " 7 7.50 " 0.05 41 " 2 82 " 3 5

) p - 0.05 from vehicle, t-test. AG: aminoguanidine; MCAO: middle cerebral artery occlusion.

F. Zhang, C. Iadecolar Brain Research 802 (1998) 104–110

106

medial to the inferior cerebral vein w24x. Animals were then returned to their cages and closely monitored until they recovered from anesthesia completely. Rats were sacrificed at different time points after induction of ischemia and their brains were removed and processed for determination of ischemic injury volume. The forebrain was frozen in cooled isopentane Žy308C.. Coronal forebrain sections Žthickness, 30 mm. were cut serially in a cryostat, collected at 300-mm intervals and stained with thionin. As described in detail elsewhere w24x, injury volume, represented by the area of pallor, was determined using an image analyzer ŽImaging Research, MCID, St. Catharines, Ontario, Canada.. Injury volume in cerebral cortex was corrected for swelling according to the method of Lin et al. w18x, as previously described w16,25x. 2.3. Experimental protocol After insertion of the femoral arterial catheter, rats were returned to their cages. When rats were fully recovered from the anesthesia, baseline arterial pressure, blood gases, plasma glucose, rectal temperature and hematocrit were measured. Rats were then re-anesthetized for MCA occlusion. One set of rats received i.p. injections of AG

hemisulphate ŽSigma; 100 mg kgy1 in 1 ml of saline. Ž n s 7. or saline Žvehicle. Ž n s 6. immediately after MCA occlusion and 8 h later. The dose of AG used Ž100 mg kgy1 . was found in previous studies to selectively inhibit post-ischemic iNOS activity without affecting brain calcium-dependent NOS activity, cerebral blood flow or cerebrovascular reactivity to hypercapnia w16x. The pH of the solutions injected was adjusted to 7.0. These rats were sacrificed 24 h after MCA occlusion for determination of infarct size. A second set of rats received AG Ž n s 8. or saline Ž n s 6. immediately after MCA occlusion and 8, 24, and 32 h later. These animals were sacrificed 48 h after MCA occlusion. A third set of rats received AG Ž n s 6. or saline Ž n s 5. immediately after MCA occlusion and 8, 24, 32, 48 and 56 h later. These animals were sacrificed 72 h after MCA occlusion. A fourth set of rats received AG Ž n s 5. or saline Ž n s 4. immediately after MCA occlusion and 8, 24, 32, 48, 56, 72, and 80 h later. These animals were sacrificed 96 h after MCA occlusion. Arterial pressure, rectal temperature and plasma glucose were measured before each AG administration and 1 h later. Arterial blood gases were measured before and after MCA occlusion and before and after the first AG administration. Thereafter, blood gases were measured once per day. Arterial hematocrit was also measured once daily.

Table 2 Plasma glucose Žmg dly1 . and rectal temperature ŽC8. in the groups of rats studied Hours after MCAO 0 4 8 12 24 36 48 60 72 80 84 96

24 h

Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature Glucose Temperature n

48 h

72 h

96 h

Vehicle

AG

Vehicle

AG

Vehicle

AG

Vehicle

AG

146 " 8 37.5 " 0.2 138 " 2 38.2 " 0.1 148 " 7 38.4 " 0.1 158 " 10 38.3 " 0.1 153 " 9 38.3 " 0.1 – – –

138 " 6 37.5 " 0.2 132 " 3 38.3 " 0.1 138 " 6 38.3 " 0.1 126 " 8 38.4 " 0.1 141 " 11 38.3 " 0.1 – – –

– – – – – – – – – – 6

– – – – – – – – – – 7

147 " 12 37.8 " 0.1 141 " 6 38.2 " 0.2 136 " 7 38.1 " 0.3 136 " 8 38.1 " 0.1 135 " 10 38.1 " 0.1 124 " 7 38.3 " 0.1 128 " 2 38.0 " 0.1 – – – – – – – – – – 6

159 " 12 37.7 " 0.1 148 " 14 37.9 " 0.1 143 " 11 38.2 " 0.1 129 " 8 38.2 " 0.2 141 " 7 38.3 " 0.1 124 " 5 38.2 " 0.1 123 " 6 38.1 " 0.1 – – – – – – – – – – 8

150 " 7 37.1 " 0.2 132 " 6 37.8 " 0.2 142 " 7 37.9 " 0.1 141 " 6) 38.1 " 0.1 137 " 3 38.1 " 0.1 122 " 2 38.1 " 0.1 142 " 13 37.9 " 0.1 133 " 7 37.7 " 0.1 127 " 5 37.7 " 0.2 – – – – – – 5

142 " 10 37.3 " 0.1 131 " 12 37.9 " 0.1 127 " 6 38.3 " 0.1 122 " 2 38.1 " 0.1 132 " 4 38.2 " 0.1 118 " 7 38.2 " 0.1 148 " 14 38.0 " 0.2 148 " 13 37.8 " 0.1 134 " 6 37.7 " 0.1 – – – – – – 6

145 " 5 37.2 " 0.1 126 " 6 37.8 " 0.1 130 " 3 37.8 " 0.1 138 " 8 37.9 " 0.1 145 " 3 38.0 " 0.1 126 " 4 38.1 " 0.1 130 " 1 38.1 " 0.1 139 " 22 37.9 " 0.1 110 " 1 37.9 " 0.1 140 " 1) 37.9 " 0.1 124 " 16 37.8 " 0.1 120 " 1.2 37.4 " 0.1 4

161 " 15 37.1 " 0.1 127 " 5 37.9 " 0.1 127 " 7 37.9 " 0.2 122 " 4 38.0 " 0.1 130 " 4 38.2 " 0.1 130 " 5 38.2 " 0.1 135 " 7 38.2 " 0.2 123 " 9 38.1 " 0.1 124 " 7 37.8 " 0.1 127 " 3 37.7 " 0.1 115 " 6 37.7 " 0.1 118 " 6 37.5 " 0.3 5

) p - 0.05 from aminoguanidine group Ž t-test.. Values at 28, 32, 52, 56, 76 h were measured but are not reported in this table Ž p ) 0.05 between groups.. AG: aminoguanidine; MCAO: middle cerebral artery occlusion.

F. Zhang, C. Iadecolar Brain Research 802 (1998) 104–110

107

Fig. 1. Mean arterial pressure in vehicle-treated rats and in rats receiving aminoguanidine Žsee Section 2 for administration schedule.. The middle cerebral artery was occluded at time 0. No differences in arterial pressure were observed between treated and untreated rats Ž p ) 0.05, t-test.. MCAO: middle cerebral artery occlusion.

Table 3 Effect of aminoguanidine on injury volume in spontaneously hypertensive rats with middle cerebral artery occlusion Volume Žmm3 . 24 h

Total injury Neocortical injury Neocortical injury Žcorrected for swelling. Swelling Striatum n Mean " S.E. ) p - 0.05 from vehicle Ž t-test.. AG: aminoguanidine.

48 h

72 h

96 h

Vehicle

AG

Vehicle

AG

Vehicle

AG

Vehicle

AG

209 " 12 188 " 12 125 " 6 63 " 7 10 " 2 6

204 " 11 185 " 9 123 " 5 62 " 7 8 "2 7

219 " 7 202 " 5 110 " 3 92 " 5 6 "1 6

212 " 11 190 " 10 96 " 5) 94 " 8 6 "1 8

216 " 15 189 " 10 113 " 4 76 " 6 9 "3 5

190 " 12 153 " 8) 89 " 2) 65 " 8 18 " 4 6

190 " 11 165 " 8 121 " 6 44 " 2 11 " 3 4

152 " 10) 135 " 8) 85 " 5) 50 " 4 7 "2 5

108

F. Zhang, C. Iadecolar Brain Research 802 (1998) 104–110

2.4. Data analysis Data presented in the text, tables and figures are expressed as means " S.E. Comparisons among multiple groups were statistically evaluated by the analysis of variance and the Tukey–Kramer modification of Tukey’s test ŽSystat, Evanston, IL.. Comparisons between two groups were evaluated by the Student’s t-test. Differences were considered significant for p - 0.05.

In rats treated for 48 h, injury volume was slightly reduced Žy10 " 15%., but the reduction did not reach statistical significance ŽTable 3; Fig. 2.. A significant reduction in injury volume was also observed in rats treated with AG for 72 and 96 h Ž p - 0.05; Table 3; Fig. 2.. The MCA occlusion resulted in swelling of the ischemic hemisphere, secondary to edema formation. The volume of swelling was maximal 48 h after MCA occlusion ŽTable 3.. Treatment with AG did not affect the magnitude of the tissue swelling Ž p ) 0.05..

3. Results 3.1. Effect of AG on arterial blood pressure, blood gases, plasma glucose, hematocrit and rectal temperature The arterial blood gases of the rats studied are presented in Table 1. With the exception of the pO 2 at one time point in the group of rat sacrificed at 48 h, differences in pH, pCO 2 and pO 2 were not statistically significant Ž p ) 0.05 from corresponding vehicle; t-test.. Plasma glucose and rectal temperature are presented in Table 2. Except for two time points ŽTable 2., plasma glucose did not differ between treated and untreated rats Ž p ) 0.05.. No differences in rectal temperature or arterial pressure ŽFig. 1. were observed between treated and untreated rats Ž p ) 0.05.. 3.2. Effect of AG on infarct Õolume In vehicle-treated rats, MCA occlusion produced a lesion involving predominantly the cerebral cortex ŽTable 3.. Volume and regional distribution of the area of injury were comparable to those previously reported in this model Že.g., Ref. w24x.. In rats treated with AG for 24 h, injury volume was not different from controls Ž p ) 0.05, t-test..

Fig. 2. Time-dependence of the effect of aminoguanidine on the injury produced by occlusion of the rat middle cerebral artery ŽMCA. in cerebral cortex. Data were corrected for swelling as described in Section 2. A significant reduction in injury volume is observed only in rats treated for 72 or 96 h after MCA occlusion Ž p- 0.05, analysis of variance and Tukey’s test..

4. Discussion Focal cerebral ischemia is associated with delayed expression of iNOS in the affected brain. The iNOS mRNA and iNOS enzymatic activity reach a maximum at 48 h after ischemia and return to baseline by 7 days w17x. The iNOS inhibitor, AG, when administered between 24 and 96 h after ischemia, attenuates post-ischemic iNOS activity and reduces the size of the infarct, suggesting that NO produced by iNOS contributes to the development of tissue damage w16x. In the present study, we sought to investigate further the effect of AG on cerebral ischemic damage. If the protective effect depends on iNOS inhibition, treatment with AG in the early stages of cerebral ischemia, when iNOS expression is nil or minimal, would not be expected to reduce ischemic damage, whereas administration at a time when there is iNOS expression should reduce tissue damage. We found that AG does not reduce injury volume when it is administered up to 48 h after MCA occlusion. However, if the administration of AG was prolonged beyond 48 h, a reduction in injury volume was observed. The effect of AG on cerebral ischemic damage cannot be due to actions of the drug on arterial pressure, blood gases, plasma glucose or rectal temperature because these parameters were monitored and did not differ substantially between groups. It is also unlikely that the effect of AG is related to improvement of cerebral blood flow because AG administration does not affect cerebral blood flow w16x. Similarly, the effect of AG cannot be attributed to inhibition of NOS isoforms other than iNOS. First, we have previously demonstrated that AG, at the dose used in the present study, does not inhibit brain calcium-dependent NOS activity, which reflects predominantly neuronal NOS w16x. Second, effects of AG on endothelial NOS are unlikely in our preparation because NO produced by endothelial NOS is thought to be protective in focal cerebral ischemia w13x. Accordingly, iNOS inhibition would be expected to enlarge injury volume. Finally, the reduction in injury volume is unlikely to be a consequence of a reduction in ischemic swelling rather than tissue damage. We have demonstrated here that treatment with AG does not affect the magnitude of ischemia-induced brain swelling.

F. Zhang, C. Iadecolar Brain Research 802 (1998) 104–110

AG is endowed with pharmacological actions distinct from iNOS inhibition that could potentially contribute to its protective effect in ischemia. For example, AG is a potent inhibitor of diamine oxidase w5x, an enzyme that participates in the degradation of histamine and polyamines w23x. The terminal oxidative deamination of polyamines by diamine oxidase leads to the production of toxic aldehydes w23x, compounds that could potentially contribute to cerebral ischemic damage. Because increased polyamine synthesis and degradation begins early after an ischemic insult w22x, it does not seem likely that, under the experimental conditions of the present study, inhibition of diamine oxidase contributes to the protective effect of AG. Further studies are needed to exclude this possibility more firmly. Another potential mechanism by which AG could reduce tissue damage is by inhibition of AGEs production or cross-linking w6x. Systemic administration of AGEs prior, but not after ischemia, exacerbates cerebral ischemic damage, an effect that is blocked by AG w27x. However, it has not been demonstrated whether ischemia leads to AGEs formation. The observation that the deleterious effects of AGEs occur only when AGEs are administered prior to induction of ischemia w27x suggests that formation of these products after ischemia may not participate in the tissue damage. However, little is known about the formation of AGEs in cerebral ischemia and the matter requires further investigation. At variance with the present study, pre-treatment or early post-treatment with AG Ž160 mg kgy1 . has been reported to reduce infarct size in a model in which the right common carotid artery ŽCCA. is ligated, the right MCA severed and then the left CCA transiently occluded in Lewis rats w7,26x. Although the reasons for this discrepancy are unclear at the present time, differences in the ischemic models and in the dose of AG are likely to play a role. One possibility is that in the MCA–CCA ‘tandem’ occlusion model, iNOS is expressed earlier than in our model and that the protective effect of AG is also mediated via iNOS inhibition. Alternatively, other pharmacological actions of AG could be involved. AG does not cross well the blood-brain barrier and does not enter the normal brain w4,5x. However, because MCA occlusion is associated with an increase in blood-brain barrier permeability to small hydrophilic molecules Že.g., Ref. w1x., it is likely that AG gains access to the ischemic brain, probably via collateral perfusion. One potential problem with our experimental design is that rats sacrificed 24–48 h after ischemia received smaller cumulative amounts of AG than rats sacrificed at later time points. Although after a single i.v. administration Ž74 mg kgy1 . AG is cleared from plasma with a half-life of 90 min w5x, AG is cleared more slowly from tissues, particularly salivary glands and kidneys w5x. However, after multiple administrations Ž50 mg kgy1 dayy1 for 7 days., the levels of AG in these organs were lower than those observed 2 h

109

after a single injection Ž74 mg kgy1 . w4,5x. Therefore, it is unlikely that 4 days of treatment will result in brain concentrations greater than those produced by a single injection. It is however unknown whether at the higher doses used in the present study Ž100 mg kg twicey1 dayy1 ., the pharmacokinetic characteristics of AG are comparable to those of the lower doses used in these previous studies. Furthermore, the rate of AG uptake and wash-out from the ischemic brain has not been established. Therefore, additional studies will be required to completely rule out the possibility that accumulation of the drug in the ischemic brain contributes to the delayed nature of the protective effect of AG. In addition, future studies will have to address the possibility that early sacrifice in the groups treated for 24 or 48 h prevented the ‘maturation’ of the protective effect of AG. The observation that AG reduces injury volume 96 but not 24 h after ischemia is in agreement with the finding that in iNOS, null mice injury volume is reduced at 96 h following MCA occlusion and not earlier w15x. These observations indicate that NO produced by iNOS is involved in the mechanisms of the delayed evolution of the damage that occurs in the post-ischemic period. Recent evidence suggests that ischemic brain injury progresses over several days after induction of ischemia w2,3,8,9,11,19x. In apparent conflict with this hypothesis is the observation that, in our experiments, injury volume in vehicle-treated rats did not increase over time. However, it must be kept in mind that the area of pallor in histological stains does not necessarily reflect a completed infarct, i.e., pan-necrosis w8x. Twenty-four hours after ischemia, the tissue at the periphery of the area of pallor is still potentially viable w8x. However, 96 h after ischemia, the area of pallor overlaps with the area of pan-necrosis w8,9x. Therefore, the area of pallor at 24 h includes both retrievable and irretrievable brain tissue, whereas at 96 h, the are of pallor reflects only irretrievable tissue damage. The observation that iNOS inhibition or gene deletion reduces injury volume at 96 but not 24 h after induction of ischemia is consistent with the hypothesis that iNOS-derived NO contributes to the progression of ischemic injury during the post-ischemic period. The delayed protective effect conferred by AG Žf 30%. is smaller than that afforded by treatment modalities directed at the early stages of cerebral ischemic injury, e.g., glutamate receptor antagonists Žf 50%. Žsee Ref. w12x for a review.. However, the therapeutic window of glutamate receptor antagonists Žf 1 h; w12x. is smaller than that offered by AG Ž12–24 h; w16,20x.. Treatment based on iNOS inhibition could be combined with reperfusion therapy, e.g., thrombolysis, or with glutamate receptor antagonists w10x. Such multimodal therapeutic approach is likely to be most effective because it will address distinct pathogenic events which play important roles in ischemic brain injury.

110

F. Zhang, C. Iadecolar Brain Research 802 (1998) 104–110

5. Conclusion We have demonstrated that AG reduces infarct size only when treatments are administered for longer than 48 h, starting after induction of ischemia. The time period during which AG administration is protective corresponds well to the period of time when iNOS is expressed in the post-ischemic brain. While the findings provide additional support to the hypothesis that NO production by iNOS contributes to cerebral ischemic damage, they also indicate that such damage continues to evolve well beyond the confines of the initial ischemic insult. Assuming that the results of the present study are applicable to human stroke, the data raise the possibility that rational therapeutic interventions could be instituted even 1 day after the onset of cerebral ischemia. Acknowledgements Supported by NIH grant NS34179. C.I. is an established investigator of the American Heart Association. References w1x M. Anwar, O. Costa, A.K. Shina, H.R. Weiss, Middle cerebral artery occlusion increases cerebral capillary permeability, Neurol. Res. 15 Ž1993. 232–236. w2x A.E. Baird, A. Benfield, G. Schlaug, B. Siewert, K.-O. Lovblad, ¨ R.R. Edelman, S. Warach, Enlargement of human cerebral ischemic lesion volumes measured by diffusion-weighted magnetic resonance imaging, Ann. Neurol. 41 Ž1997. 581–589. w3x J.-C. Baron, R. von Kummer, G.J. del Zoppo, Treatment of acute ischemic stroke. Challenging the concept of a rigid and universal time window, Stroke 26 Ž1995. 2219–2221. w4x S. Baylin, Z. Horakova, M.A. Beaven, Increase in food consumption and growth after treatment with aminoguanidine, Experientia 31 Ž1975. 562–564. w5x M.A. Beaven, J.W. Gordon, S. Jacobsen, W.B. Severs, A specific and sensitive assay for aminoguanidine: its application to a study of the disposition of aminoguanidine in animal tissues, J. Pharmacol. Exp. Ther. 165 Ž1969. 14–22. w6x M. Brownlee, H. Vlassara, A. Kooney, P. Ulrich, A. Cerami, Aminoguanidine prevents diabetes-induced arterial wall protein cross-linking, Science 232 Ž1986. 1629–1632. w7x K.M. Cockroft, M.I. Meistrell, G.A. Zimmerman, D. Risucci, O. Bloom, A. Cerami, K.J. Tracey, Cerebroprotective effects of aminoguanidine in a rodent model of stroke, Stroke 27 Ž1996. 1393–1398. w8x M.O. Dereski, M. Chopp, R.A. Knight, L.C. Rodolosi, J.H. Garcia, The heterogeneous temporal evolution of focal ischemic neuronal damage in the rat, Acta Neuropathol. 85 Ž1993. 327–333. w9x J.H. Garcia, Y. Yoshida, H. Chen, Y. Li, Z.G. Zhang, J. Lian, S. Chen, M. Chopp, Progression from ischemic injury to infarct following middle cerebral artery occlusion in the rat, Am. J. Pathol. 142 Ž1993. 623–635.

w10x M.D. Ginsberg, M. Globus, R. Busto, W.D. Dietrich, The potential of combination pharmacotherapy in cerebral ischemia, in: J. Krieglstein, H. Oberpichler ŽEd.., Pharmacology of Cerebral Ischemia, Wissenschaftliche Verlagsgesellschaft, Stuttgard, 1990, pp. 499–510. w11x W.-D. Heiss, M. Huber, G.R. Fink, K. Herloz, U. Pietrzyk, R. Wagner, K. Weinhard, Progressive derangement of peri-infarct viable tissue in ischemic stroke, J. Cereb. Blood Flow Metab. 12 Ž1992. 193–203. w12x K.-A. Hossmann, Glutamate-mediated injury in focal cerebral ischemia: the excitotoxin hypothesis revised, Brain Pathology 4 Ž1994. 23–36. w13x Z. Huang, P.L. Huang, J. Ma, W. Meng, C. Ayata, M.C. Fishman, M.A. Moskowitz, Enlarged infarcts in endothelial nitric oxide synthase knockout mice are attenuated by nitro-L-arginine, J. Cereb. Blood Flow Metab. 16 Ž1996. 981–987. w14x C. Iadecola, X. Xu, F. Zhang, E.E. El-Fakahany, M.E. Ross, Marked induction of calcium-independent nitric oxide synthase activity after focal cerebral ischemia, J. Cereb. Blood Flow Metab. 14 Ž1995. 52–59. w15x C. Iadecola, F. Zhang, R. Casey, M. Nagayama, M.E. Ross, Delayed reduction in ischemic brain injury and neurological deficits in mice lacking the inducible nitric oxide synthase gene, J. Neurosci. 17 Ž1997. 9157–9164. w16x C. Iadecola, F. Zhang, X. Xu, Inhibition of inducible nitric oxide synthase ameliorates cerebral ischemic damage, Am. J. Physiol. 268 Ž1995. R286–R292. w17x C. Iadecola, F. Zhang, X. Xu, R. Casey, M.E. Ross, Inducible nitric oxide synthase gene expression in brain following cerebral ischemia, J. Cereb. Blood Flow Metab. 15 Ž1995. 378–384. w18x T.-N. Lin, Y.Y. He, G. Wu, M. Khan, C.Y. Hsu, Effect of brain edema on infarct volume in a focal cerebral ischemia model in rats, Stroke 24 Ž1993. 117–121. w19x G. Marchal, V. Beaudouin, P. Rioux, V. de la Sayette, F. Le Doze, F. Viader, J.-M. Derlon, J.-C. Baron, Prolonged persistence of substantial volumes of potentially viable brain tissue after stroke, Stroke 27 Ž1996. 599–606. w20x M. Nagayama, F. Zhang, C. Iadecola, Delayed treatment with aminoguanidine reduces focal cerebral ischemic damage and enhances neurological recovery in rats, J. Cereb. Blood Flow Metab. Ž1998. in press. w21x C. Nathan, Inducible nitric oxide synthase: what difference does it make?, J. Clin. Invest. 100 Ž1997. 2417–2423. w22x W. Paschen, M. Cleef, G. Rohn, M. Muller, A.E.I. Pajunen, Is¨ ¨ chemia-induced disturbances of polyamine synthesis, Progress in Brain Research 96 Ž1993. 147–160. w23x A. Sessa, A. Perin, Diamine oxidase in relation to diamine and polyamine metabolism, Agents Actions 43 Ž1994. 69–77. w24x F. Zhang, C. Iadecola, Stimulation of the fastigial nucleus enhances EEG recovery and reduces tissue damage after focal cerebral ischemia, J. Cereb. Blood Flow Metab. 12 Ž1992. 962–970. w25x F. Zhang, C. Iadecola, Infarct measurement methodology, J. Cereb. Blood Flow Metab. 14 Ž1994. 697–698. w26x G.A. Zimmerman, O. Bloom, M.I. Meistrell, D. Ford, M. Bianchi, K.J. Tracey, Aminoguanidine attenuates focal cerebral ischemia, Surg. Forum 45 Ž1994. 600–603. w27x G.A. Zimmerman, M.I. Meistrell, O. Bloom, K.M. Cockroft, M. Bianchi, D. Risucci, J. Broome, P. Farmer, A. Cerami, H. Vlassara, K.J. Tracey, Neurotoxicity of advanced glycation endproducts during focal stroke and neuroprotective effects of aminoguanidine, Proc. Natl. Acad. Sci. USA 92 Ž1995. 3744–3748.