- Email: [email protected]
The effects of post-ischemic hypothermia on the neuronal injury and brain metabolism after forebrain ischemia in the rat
191
Journal of the Neurological Sciences, 107 (1992) 191-198 © 1992 Elsevier Science Publishers B.V. All rights reserved 0022-510X/92/$05.00
JNS 03690
The effects of post-ischemic hypothermia on the neuronal injury and brain metabolism after forebrain ischemia in the rat Hua Chen
1,4
Michael Chopp
1,4
Ana M.Q. Vande Linde 1, Mary O. Dereski and K.M.A. Welch 1,4
2,
Julio H. Garcia 3
Departments of I Neurology, 2 Radiation Oncology and 3 Pathology, Henry Ford Hospital, Detroit, M1 48202 (U.S.A.) and 4 Department of Physics, Oakland University, Rochester, MI 48309 (U.S.A.)
(Received 21 May, 1991) (Revised, received 3 September, 1991) (Accepted 6 September, 1991)
Key words: Rat forebrain ischemia; Post-ischemia hypothermia; Neuronal damage; In vivo Sip NMR spectroscopy; Intracellular pH; Intracellular [Mg 2+] Summary We investigated the effect of moderate post-ischemic hypothermia on neuropathological outcome and cerebral high energy phosphate metabolism, intracellular pH and Mg z÷ concentration in the rat. Three groups of animals were investigated: (1) Wistar rats subjected to 12 min of forebrain ischemia under normothermic conditions (n = 17), (2) rats subjected to the identical procedure of ischemia, except that 30°C hypothermia was induced post-ischemia and maintained for 2 h of reperfusion (n = 6), and (3) control hypothermic rats not subjected to ischemia (n = 4). In vivo 31p NMR spectroscopy was performed prior to ischemia, and at intervals up to 168 h after ischemia. Histological analysis of brain tissues was performed 7 days after ischemia. No significant differences in cortical and hippocampal neuronal damage was detected between the two experimental groups. Significantly lower pH values were detected in the hypothermic ischemic animals at 24 h (P---0.0001) and 48 h (P = 0.018) post-ischemia compared to the normothermic ischemic animals. Normothermic ischemic animals exhibited significantly lower [Mg 2÷] at 72 h (P < 0.006) compared to the pre-ischemia level. Our data indicate that post-ischemic hypothermia modifies the profiles of post-ischemic brain tissue pH and Mg 2÷ concentration, and this modification is not associated with histopathological outcome 7 days after ischemia.
Introduction H y p o t h e r m i a has been studied as a protective treatment against experimental ischemic brain damage, when induced either during or after ischemia (Berntman et al. 1981; Busto et al. 1987; Minamisawa et al. 1990a, b; Welsh et al. 1990). Mild 34°C post-ischemic hypothermia lessens neuronal damage after 8 min of forebrain ischemia but not after 12 min of ischemia (Chopp et al. 1991), while induction of 30°C whole-body hypothermia after a 10-min episode of forebrain ischemia protects neurons from ischemic cell damage (Busto et al. 1989a). The metabolic response of brain to cerebral ischemia is also affected by hypothermic intervention. In vivo 31p nuclear magnetic resonance ( N M R ) spec-
Correspondence to: H. Chen, Henry Ford Hospital, 2799 W. Grand Blvd., Detroit, MI 48202, U.S.A. Tel.: (313) 876-3936; Fax: (313) 8761318.
troscopy m e a s u r e m e n t s of cerebral metabolism during hypothermic transient global cerebral ischemia and reperfusion in the cat revealed a reduction of the rate of loss of high energy phosphate metabolites and intracellular acidosis during ischemia, and a more rapid rate of return of metabolites and p H upon reperfusion compared to the normothermic global ischemia (Chopp et al. 1989). Distinct post-ischemic profiles of cerebral p H and free Mg 2÷ concentration were exhibited after normothermic forebrain ischemia (Chopp et al. 1990a; Vande Linde et al. 1990). To our knowledge, however, there have been no reports on the effect of postischemic hypothermia on the time course of cerebral energy metabolism, p H and Mg 2+ concentration after ischemia. The objective of the present study was to determine whether 30°C hypothermia affects neuronal damage and metabolic response, when induced immediately after 12 rain of forebrain ischemia in the rat. We tested the hypothesis that p H and Mg 2+ concentration profiles measured by in vivo 31p N M R spectroscopy reflect
192 the neuropathological outcome after normothermic and hypothermic ischemia.
Materials and methods
Twenty-seven male Wistar rats, weighing 200-300 g, were used in the experiment. Twelve min forebrain ischemia was induced using bilateral carotid artery occlusion in conjunction with systemic hypotension. This ischemia model has been described by Smith et al. (Smith et al. 1984a,b) and has been employed in our laboratory (Chopp et al. 1990a,b, 1991). Animals were divided into 3 groups: (1) normothermic ischemic animals (n = 17), in which rectal temperature was controlled and maintained at 37°C throughout the experimental procedure. Selected in vivo 31p NMR data from 12 of these animals have been previously reported (Chopp et al. 1990a), (2) hypothermic ischemic animals (n = 6), in which rectal temperature was maintained at 37°C before and during ischemia, and 30°C whole-body hypothermia was induced immediately after ischemia and maintained for 2 h of reperfusion, (3) hypothermia without ischemia (n = 4), in which rectal temperature was reduced to 30°C for 2 h without ischemia. Animals were fasted overnight before surgery, but allowed free access to water. Rectal temperature was controlled in all the experiments using a feed-back heating pad. Hypothermia was instituted 5 rain after ischemia, or after induction of anesthesia in the control hypothermic animals, by spraying alcohol onto the skin and fanning the animal with room air (22°C). 30°C hypothermia was achieved within 15 min. The heating pad was set to prevent the animal's temperature from falling below 30°C. After 2 h of reperfusion, the animal was rewarmed to 37°C over a 30-rain period using a heating pad. In this ischemia model, brain temperature is closely coupled ( +__l°C) to rectal temperature during control and reperfusion conditions (Chopp et al. 1991). Each animal was anesthetized 24-48 h prior to ischemia with 0.5% halothane in 70% N20 and 30% 0 2 using a face mask, and placed in the magnet to obtain a baseline pre-ischemia 31p NMR spectrum. Baseline and post-ischemia (24, 48, 72, 96 and 168 h) NMR measurements were performed using the method previously reported (Chopp et al. 1990a,b). In vivo 3~p NMR spectroscopy was performed at 161.9 MHz using an 89-mm diameter wide bore, vertical magnet and Varian VXR4000 spectrometer. A two-turn surface coil (6 mm o.d.) was centered over the rat brain. The NMR sensitive volume approximates a hemisphere with a 3-mm radius, encompassing cortex and extending into hippocampus. Each spectrum was acquired with a spectral width of 40 tzs at a repetition rate of 0.6 s, and was the sum of 1000 transients. NMR1 software (New Method Research, Syracuse, NY, U.S.A.) was used to
process spectra to obtain area ratios of phosphocreatine (PCr) to inorganic phosphate (Pi), /3-adenosine triphosphate (/3-ATP) to total phosphate, and PCr to /3-ATP, as metabolic energy parameters. Chemical shifts for Pi, PCr, 7-ATP, a-ATP, and /3-ATP were recorded for subsequent pH and [Mg 2~] analysis, lntracellular pH was calculated prior to the NMR1 analysis from the chemical shift between phosphocreatinc and inorganic phosphate. Free magnesium ion concentrations, [Mg2*], were assessed using the chemical shift of Pi, PCr, 3,-ATP, a-ATP, and /3-ATP in combination with an in vitro calibration scheme (Halvorson et al. in press). Seven days after the ischemic insult, the animals were anesthetized i.m. with ketamine (44 mg/kg) and xylazine (13 mg/kg). Brain perfusion was performed transcardially with saline and 10% buffered formalin. Each brain was carefully removed 1 h after perfusion, and immersed in formalin for one week before sectioning. Brains were cut coronally into 3-mm thick slabs with a rodent brain matrix. The tissues were processed, embedded in paraffin, and 6-1zm thick slices were stained with hematoxylin and eosin. Brain tissue damage was evaluated using light microscopy from regions of tissue from which NMR data are obtained. A standard section, identified by the hippocampal shape and size, containing subiculum, hippocampus CA 1/2, CA3, was selected from each rat. Fig. 1 illustrates the demarcation of hippocampal regions in which cellular damage was evaluated using the GFAP immunohistochemistry staining method. Neurocortical damage was evaluated in a 3 x 2 mm 2 area in each hemisphere, and the hippocampus CA 1/ 2 was further divided into medial, central, and lateral subregions. Semiquantification of damaged neurons in the cortex was performed on the following scale: 0 = no ischemic neurons, I - - f e w ischemic neurons, 2 = many ischemic neurons, and 3-most neurons are ischemic. Quantitative calculation of neuronal damage in hippocampus was evaluated by direct counting of the neurons in the CA1/2 sections of hippocampus, and the data expressed as the total numbers of the intact neurons and the percentage of necrotic neurons to total neurons. The examiners were blinded to the animal groups.
Data analysis Brain tissue pH, [Mg 2+] and high energy phosphate values were analyzed with a repeated measures analysis of variance. If a significant interaction was detected between groups and time, then two sample t-tests were performed at each time point to compare the group differences. Paired t-tests were performed within each group to compare each time point to the pre-ischemic values. All analyses were corrected for multiple comparisons using Bonferroni correction. Data from the control hypothermic group were analyzed quantita-
193 tively. A Wilcoxon r a n k sum test was used to c o m p a r e the h y p o t h e r m i c a n d n o r m o t h e r m i c ischemic groups o n cortical d a m a g e a n d two sample t-tests were c o n d u c t e d to c o m p a r e intact n e u r o n s b e t w e e n the two groups. A n arc-sine t r a n s f o r m a t i o n was p e r f o r m e d to n o r m a l i z e h i p p o c a m p a l n e u r o n a l c o u n t data, t h e n S t u d e n t ' s ttests were p e r f o r m e d on the p e r c e n t a g e of necrotic n e u r o n s verses total n e u r o n s b e t w e e n n o r m o t h e r m i c a n d h y p o t h e r m i c groups. All data are p r e s e n t e d as m e a n _+ SD.
TABLE 1 MEAN-+SD OF BLOOD GAS AND SERUM GLUCOSE VALUES FOR NORMOTHERMIC ANIMALS AND HYPOTHERMIC ANIMALS pH (ram Hg)
pCO 2 (mm Hg)
pO 2 (mm Hg)
Glucose (mg/dl)
Normothermia (n = 17) Before ischemia 7.36+_0.03 35+5 After ischemia 7.34_+0.02 31 +5
136+24 142_+25 124-+29 -
Hypothermia (n = 6) Before ischemia 7.35-+0.04 39+-7 After ischemia 7.30_+0.01 4 0 + _ 8
139-+26 129_+18 133+_36 -
Results
Physiological p a r a m e t e r s are given in T a b l e 1. A r t e rial blood gases a n d s e r u m glucose values prior to ischemia a n d 30 m i n after ischemia were within n o r m a l physiological ranges for the two groups of animals. No arterial b l o o d gases were t a k e n d u r i n g N M R acquisition of spectra, b e c a u s e o u r previous study had shown that blood gas values were within n o r m a l physiologic r a n g e while a n i m a l s were placed vertically in the N M R p r o b e ( C h o p p et al. 1990a).
Fig. 2 shows 31p N M R spectra from a r e p r e s e n t a t i v e a n i m a l o b t a i n e d before, 24 h, a n d 168 h after 12 m i n f o r e b r a i n ischemia. Relative intensities of spectral peaks were similar at all time points. No i n t e r a c t i o n s for m e t a b o l i c p a r a m e t e r s ( / 3 - A T P / t o t a l p h o s p h a t e , P C r / / 3 - A T P , P C r / P i ) were d e t e c t e d b e t w e e n groups a n d time ( P > 0.4, data not shown), i n d i c a t i n g b r a i n
~ii~i~i~ii:(/
Fig. 1. Glial fibrillary acidic protein (GFAP) immunohistochemical staining of a 6-/zm coronal section from a rat subjected to 12 min forebrain ischemia. Astrocytic reaction with GFAP indicates a sharp demarcation between CA1/2 and CA3 regions of hippocampus (arrow). A 3 x 2 mm2 box indicates the area in one hemisphere in which cortical damage is evaluated. Bars divide CA1/2 into medial, central, and lateral regions for determination of neuronal counts.
194 TABLE 2 MEAN_+SD O F [Mg z+] F O R N O R M O T H E R M I ( " POTHERMIC ISCHEMIC ANIMALS
Baseline 24 h 48 h 72 h 96 h 168 h
A N D HY-
Normothermia (n = 17)
Hypothermia (n = 6)
0.27 + 0.08 0.25 +0.05 0.27_+0.10 (I.22 +_0.04 0.21 ± 0.05 0.25 _+0.07
0.23 ± 0.03 0.39±0.22 0.42±0.18 0.25 ± 0.07 0.21 ± 0,11 0.24 + 0.18
mM mM mM * mM ~ mM mM
mM mM mM mM mM mM
* P = 0.02 compare to hypothermia; + P = 0.006 compare to baseline.
3
12 14 ,5
3O
0
-]0
PPM Fig. 2. 3~p N M R spectra from a representative rat; pre-ischemia (bottom), 24 h after 12 min forebrain ischemia (middle) and 168 h after ischemia (top). Spectral peaks: 1. primarily phosphomonoesters, 2. inorganic phosphate, 3. phosphocreatine, 4. primarily ~,-ATP, 5. primarily a - A T P , 6. ~-ATP.
metabolism has returned to the pre-ischemia value in both normothermic and hypothermic ischemic animals. Fig. 3 shows the time course of N M R pH values for the hypothermic and normothermic ischemic groups. Control hypothermic animals exhibited constant pH values over time (data not shown). No difference was 7.50
7.36
7.22 Q, 7.08
6.94
6.80
J
I
I
J
I
0
24
48
72
96
I
J
I
120
144
168
TIME (I.IOIJRS)
Fig. 3. Mean brain tissue pH + SD m e a s u r e d prior to ischemia (time 0) and at 24, 48, 72, 96 and 168 h after 12 min forebrain ischemia, zx, normothermic animals (n = 17), o, hypothermic animals (n =6).
detected ( P > 0.1) between pre-ischemic pH values in normothermic and hypothermic ischemic animals. An interaction between the normothermic and hypothermic ischemic groups and time was detected ( P = 0.004), indicating that the two groups behave differently over time. Significantly lower pH was detected in the hypothermic ischemic group, at 24 h ( P = 0.0001) and 48 h ( P = 0.018) post-ischemia, compared to the normothermic ischemic group. In the normothermic ischemic animals, alkalosis was exhibited both at 24 h (pH = 7.24 + 0.14, P = 0.0002) and 48 h (pH = 7.20 + 0.14, P = 0.002) post-ischemia, compared to the pre-ischemic values (pH = 7.06 + 0.05). Conversely, hypothermic ischemic animals revealed acidosis (pH = 6.99 __+0.05, P = 0.0002) at 24 h compared to its pre-ischemic values (pH = 7.09 _+ 0.05). Although statistically significant mean alkalosis was not detected in the hypothermic ischemic rats, 5 out of 6 animals exhibited atkalosis either at 72, 96, or 168 h after the ischemic insult (pH from 7.17 to 7.36). Table 2 shows the time course of free [Mg 2+] for the normothermic and hypothermic ischemic animals. [Mg 2+] were constant in the control hypothermic animals (data not shown). No difference was detected ( P > 0.3) between pre-ischemic [Mg 2+] in normothermic and hypothermic ischemic animals. An interaction between groups and time was detected ( P < 0.01). At 48 h post-ischemia, the hypothermic [Mg ÷2 ] was significantly elevated compared to the normothermie [Mg 2+ ]. In the normothermic ischemic animals, [Mg 2+] decreased at 72 h ([Mg z÷] = 0.22 + 0.04 mM, P = 0.006) compared to the pre-ischemic [Mg z+ ] value ([Mg 2+ ] = 0.27 _ 0.08 mM). No neuronal damage was detected in the control hypothermic group. Cerebral cortex revealed few scattered dying neurons in both normothermic and hypotherrnic ischemic animals. No difference in neuronal damage of cerebral cortex was detected between the normothermic and hypothermic ischemic groups (scores =0.81 +__0.66 in normothermia verses 0.67 + 0.52 in hypothermia). Fig. 4 shows the sections of
195 hippocampus for the control hypothermic animals and the normothermic and hypothermic ischemic animals. Control hypothermic animals demonstrated no neuronal damage, while both ischemic groups exhibited severe neuronal injury in the hippocampal region. The numbers of intact neurons and the percentage of necrotic neurons in the medial, central, lateral and overall hippocampus CA1/2 are shown in Table 3. A marginal difference was detected between the two groups only in the medial region (P = 0.056), showing that hypothermic ischemic animals had fewer necrotic neurons (85.8 + 9.7%) compared to the normothermic ischemic animals (93.2 + 5.8%).
Discussion
Our data indicate that lowering the body temperature to 30°C post-ischemia does not significantly protect against or alter the pattern of neuronal injury in hippocampus and cortex after 12 min of forebrain ischemia. However, post-ischemic hypothermia significantly modifies dynamic profiles of change in the corresponding brain tissue pH and [Mg2+]. The data thus suggest that these profiles have little or no predictive value for neuronal damage in this model of forebrain ischemia. Post-ischemic intracellular cerebral alkalosis has been reported both in stroke patients (Syrota et al. 1985; Welch et al. 1989) and in experimental animals (Mabe et al. 1983; Chopp et ah 1990a, b). The profile of cerebral alkalosis has been associated with the duration of ischemia, and alkalosis per se has been thought to mediate ischemic cell damage (Chopp et al. 1990a). Among stroke patients, cerebral alkalosis has been associated with ischemic irreversibility brain infarct (Welch et al. 1989). Our data do not contradict clinical data indicating that cerebral infarction is associated with intracellular alkalosis. Our data, however, suggest that the presence (normothermia) or absence (hypothermia) of post-ischemic intracellular alkalosis is not an indication of mild ischemic cell damage in transient forebrain ischemia. NMR pH measurement is an average of pH values from all cells within the detected volume of the surface coil. In brain tissue, the cells include neurons, glia, endothelial cells and inflammatory cells. The presence of inflammatory cells or reactive glia may contribute to the brain tissue alkalosis. Proliferating cells are activated in an alkalotic environment, and themselves exhibit intracellular alkalosis (McCord and Roy 1982; Schuldiner and Rozengurt 1982; Chesler and Kraig 1989). Possibly, hypothermia alters the time course of reactive response to tissue injury or delays the neuronal injury process. Analysis of the neuronal, glial and
inflammatory cell activity for the normothermic and hypothermic animals at the 24-h or 48-h time points are necessary to resolve the possible role of glia and inflammatory cells in the altering acid/base balance of ischemic brain tissue and delayed neuronal injury by hypothermia. The reason for brain tissue acidosis 24 h after post-ischemia hypothermia is unknown. The effect of intracellular pH on free intracellular [Mg 2+] may play a role in promoting ischemic cell damage. Concomitant reduction of [Mg 2+] and tissue alkalosis has been reported in brain, muscle, and heart (Adam et ah 1989). Mg 2+ may stabilize the plasma membrane, compete with Ca 2÷ to reduce Ca 2+ entry into the cell, and inhibit release of excitatory neurotransmitters (Rothman 1983; Mayer et ah 1984; Kass et al. 1988). Thus, magnesium depletion, paralleling alkalosis, may evoke tissue damage in response to an ischemic insult. In the present study, the level of decrease in [Mg 2÷] observed in normothermic ischemia animals does not evoke severe tissue damage. Likewise maintenance of [Mg 2+] at control levels by hypothermia fails to improve histological outcome. Thus the modifications of [Mg 2+] as measured in the present experiment does not significantly alter neuronal damage. Hypothermic protection in cerebral ischemia has been attributed to a combination of reduction of excitatory neurotransmitter release (Busto et al. 1989b), mitigation of abnormal ion fluxes (Astrup et al. 1981), reduction of edema (Dempsey et al. 1987) and lactate (Schuldiner and Rozengurt 1982), and reduction of leukotrienes (Dempsey et ah 1987). Our previous findings indicated that 34°C post-ischemia hypothermia only protected neuronal damage after 8 min ischemia, but not after 12 min ischemia (Chopp et al. 1991). Other investigators have demonstrated that a lower degree of hypothermia (27°C) during ischemia completely protects against neuronal damage after 20 min of forebrain ischemia, while animals subjected to postischemic hypothermia exhibited variable neuronal damage (Boris-Moiler et al. 1989). Busto et al. found neuronal protection by 30°C post-ischemic hypothermia in animals subjected to 10 min of forebrain ischemia (Busto et al. 1989a). Thus, the neuronal protection by post-ischemic hypothermia depends on both duration of ischemia and depth of hypothermia. The presence of only selective marginal protection by postischemic hypothermia in the present model may be attributed to the complex interplay of a "long" 12-min duration of ischemia coupled with a moderate postischemia 30°C hypothermia. 30°C hypothermia is apparently not low enough to exert strong neuronal protection against 12 min of forebrain ischemia. In summary, our data suggest that, although postischemic hypothermia significantly modifies the profiles of intracellular pH and [Mg 2÷] after a transient
196
Fig. 4. H / E staining of 6 - # m thick paraffin sections of hippocampal C A 1 / 2 region illustrating neuronal damage of (a) control hypothermic animal, (b) normothermic 12-min ischemic animal, and (c) hypothermie 12-rain ischemic animal. Neurons are intact in the control hypothermic animals. Necrotic neurons were exhibited in both ischemic groups of animals (noted by arrows) ~ × 300).
197
Fig. 4. (continued). TABLE 3 MEAN_+ SD OF THE NUMBERS OF INTACT NEURONS AND THE PERCENTAGE OF NECROTIC NEURONS TO TOTAL NEURONS OF HIPPOCAMPUS CA1/2 REGION
Medial Central Lateral Overall
Normothermia (n = 16)
Hypothermia (n = 6)
Numbers
%
Numbers
%
11.7+ 8.1 11.2_+ 8.3 17.3_+ 9.7 40.1 _+24.8
93.2_+5.8 89.6_+5.4 86.8_+8.5 90.0_+6.5
19.6_+14.5 16.9_+12.2 23.9_+14.9 60.4+40.3
85.8_+ 9.7 * 85.8_+10.7 83.3_+11.3 84.2_+11.1
* P = 0.056.
forebrain ischemia, no significant effect of hypothermia on histopathological outcome is detected. Acknowledgements The authors wish to thank Dr. Quan Jiang and Ms. Sue Ann Lee for technical assistance and Ms. Mary L. Rexroad for manuscript preparation. This work was supported by NINDS Grants NS23393 and NS29463.
References Adam, W.R., Craik, D.J., Kneen, M. and Wellard, R.M. (1989) Effect of magnesium depletion and potassium depletion and chlorothiazide on intracellular pH in the rat, studied by 3lp NMR. Clin. Exp. Pharm. Physiol., 16: 33-40.
Astrup, J., Moller-Sorenson, P. and Rahbeck-Sorenson, H. (1981) Inhibition of cerebral oxygen and glucose consumption in the dog by hypothermia, pentobarbital and lidocaine. Anesthesiology, 55: 263-268. Berntman, L., Welsh, F.A. and Harp, J.R. (1981) Cerebral protective effect of low-grade hypothermia. Anesthesiology, 55: 495-498. Boris-Moiler, F., Smith, M.L. and Siesj6, B.K. (1989) Effects of hypothermia on ischemic brain damage: a comparison between preischemic and post-ischemic cooling. Neurosci. Res. Commun., 5: 87-94. Busto, R., Dietrich, W.D., Globus, M.Y.T, Valdes, I., Scheinberg, P. and Ginsberg, M.D. (1987) Small differences in intra-ischemic brain temperature critically determine the extent of ischemic neuronal injury. J. Cereb. Blood Flow Metab., 7: 729-738. Busto, R., Dietrich, W.D., Globus, M.Y. and Ginsberg, M.D. (1989a) Post-ischemic moderate hypothermia inhibits, CA1 hippocampal ischemic neuronal injury. Neurosci. Lett., 101: 299-304. Busto, R., Globus, M.Y.T, Dietrich, D., Martinez, E., Valdes, I. and Ginsberg, M.D. (1989b) Ischemia-induced release of neurontransmitters and free fatty acids in the rat brain: effects of mild intra-ischemic hypothermia. Stroke, 20: 904-910. Chesler, M. and Kraig, R.P. (1989) Intracellular pH transients of mammalian astrocytes. J. Neurosci., 9: 2011-2019. Chopp, M., Knight, R., Tidwell, C.D., Helpern, J.A., Brown, E. and Welch, K.M.A. (1989) The metabolic effects of mild hypothermia on global cerebral ischemia and recirculation in the cat: comparison to normothermia and hyperthermia. J. Cereb. Blood Flow Metab., 9: 141-148. Chopp, M., Chen, H., Vande Linde. A.M.Q, Brown, E. and Welch, K.M.A. (1990a) The time course of post-ischemic intracellular alkalosis reflects the duration of ischemia. J. Cereb. Blood Flow Metab., 10: 860-865.
198 Chopp, M., Vande Linde, A.M.Q, Chen, H., Knight, R., Helpern, J.A. and Welch, K.M.A. (1990b) Chronic intracellular cerebral alkalosis after an 8-minute forebrain ischemic insult in the rat. Stroke, 21(3): 463-466. Chopp, M., Chen, H., Dereski, M.O. and Garcia, J.H. (1991) Mild hypothermie intervention after graded ischemic stress in rats. Stroke, 22: 37-43. Dempsey, A.L., Combs, D.J., Edwards Maley, M., Cowen, D.E., Roy, M.W. and Conaldson, D.L. (1987) Moderate hypothermia reduces post-ischemic edema development and leukotrine production. Surgery, 21: 177-181. Halvorson, H., Vande Linde, A.M.Q, Helpern, J.A. and Welch, K.M.A. (in press) Assessment of magnesium concentrations by P-31 NMR in vivo. NMR in Biomedicine. Kass, I.S., Cottrell, J.E. and Chambers, G. (1988) Magnesium and cobalt, not nimodipine, protect neurons against anoxic damage in the rat hippocampal slice. Anesthsiology, 69: 710-715. Mabe, H., Blomqvist, P. and Siesj6, B.K. (1983) Intracellular pH in the brain following transient ischemia. J, Cereb. Blood Flow Metab., 3: 109-114. Mayer, M.L., Westbrook, G.L. and Cuthrie, P.B. (1984) Voltage-dependent block by magnesium ion of NMDA receptors in spinal cord neurones. Nature, 309: 261-263. McCord, J.M. and Roy, R.S. (1982) The pathophysiology of superoxides: roles in inflammation and ischemia. Can. J. Physiol. Pharmacol., 60: 1346-1352. Minamisawa, H., Nordstrom, C.-H., Smith, M.-L. and Siesj6, B.K. (1990a) The influence of mild body and brain hypothermia on ischemic brain damage. J. Cereb. Blood Flow Metab., 10: 365-374. Minamisawa, H., Smit M.-L. and Siesj6, B.K. (1990b) The effect of mild hyperthermia and hypothermia on brain damage following 5, 10 and 15 minutes of forebrain ischemia. Ann. Neurol., 28: 26-33.
Rothman, S.R. (1983) Synaptic activity mediates death ol hypoxic neurons. Science, 536-537. Schuldiner, S. and Rozengurt, E. (1982) N a / H antiport and swiss 3T3 cells. Mitogenic stimulation leads to cytoplasmic alkatinization. Proc. Natl. Acad. Sci. U.S.A., 79:7778 7782. Smith, M.L., Auer, R.N. and Siesj6, BK. (1984a) The density of distribution of ischemic brain injury in the rat following 2-10 minutes of forebrain ischemia. Acta Neuropathol. (Berl.), 64: 319-332. Smith, M.L., Bendek, G., Dahlgren, N., Rosen, I., Wiek)ch, T. and Siesj6, B.K. (1984b) Models for studying long-term recovery following forebrain ischemia in the rat. 2. A 2-vessel occlusion model. Acta Neurol. Scand., 69: 385-401. Syrota, A., Samson, Y., Boullais, C., Wajnberg, P., Loc'h, C., Crouzwl, C., Maziere, B., Soussaline, F. and Baron, J.C. (1985) Tomographic mapping of brain intracellular pH and extracellular water space on stroke patients. J. Cereb. Blood Flow Metab., 5: 358-368. Vande Linde, A.M.Q., Chopp, M., Chen, H.. Halvorson, tt.R., Helpern, J.A., Brown, E. and Welch, K.M.A. (1990) Decline of cerebral free Mg 2 concentration after forebrain ischemia. Society of Magnetic Resonance in Medicine, Ninth Annual Scientific Meeting and Exhibition, 2: 980. Welsh, F.A., Sims, R.E. and Harris, V.A. (199tl) Mild hypothermia prevents ischemic injury in gerbil hippocampus: J. Cereb. Blood Flow Metab., 10: 557-563. Welch, KM.A., Levine, S.R. and Helpern, J.A. (1989) Pathophysio[ogicla correlates of cerebral ischemia: the significance of cellular acid base shifts. In: Richichi, I. (Ed.), Proceedings of the International Conference of Cardiology and Neurology. Palmi, Italy, Calabro, pp. 205-217.
Journal of the Neurological Sciences, 107 (1992) 191-198 © 1992 Elsevier Science Publishers B.V. All rights reserved 0022-510X/92/$05.00
JNS 03690
The effects of post-ischemic hypothermia on the neuronal injury and brain metabolism after forebrain ischemia in the rat Hua Chen
1,4
Michael Chopp
1,4
Ana M.Q. Vande Linde 1, Mary O. Dereski and K.M.A. Welch 1,4
2,
Julio H. Garcia 3
Departments of I Neurology, 2 Radiation Oncology and 3 Pathology, Henry Ford Hospital, Detroit, M1 48202 (U.S.A.) and 4 Department of Physics, Oakland University, Rochester, MI 48309 (U.S.A.)
(Received 21 May, 1991) (Revised, received 3 September, 1991) (Accepted 6 September, 1991)
Key words: Rat forebrain ischemia; Post-ischemia hypothermia; Neuronal damage; In vivo Sip NMR spectroscopy; Intracellular pH; Intracellular [Mg 2+] Summary We investigated the effect of moderate post-ischemic hypothermia on neuropathological outcome and cerebral high energy phosphate metabolism, intracellular pH and Mg z÷ concentration in the rat. Three groups of animals were investigated: (1) Wistar rats subjected to 12 min of forebrain ischemia under normothermic conditions (n = 17), (2) rats subjected to the identical procedure of ischemia, except that 30°C hypothermia was induced post-ischemia and maintained for 2 h of reperfusion (n = 6), and (3) control hypothermic rats not subjected to ischemia (n = 4). In vivo 31p NMR spectroscopy was performed prior to ischemia, and at intervals up to 168 h after ischemia. Histological analysis of brain tissues was performed 7 days after ischemia. No significant differences in cortical and hippocampal neuronal damage was detected between the two experimental groups. Significantly lower pH values were detected in the hypothermic ischemic animals at 24 h (P---0.0001) and 48 h (P = 0.018) post-ischemia compared to the normothermic ischemic animals. Normothermic ischemic animals exhibited significantly lower [Mg 2÷] at 72 h (P < 0.006) compared to the pre-ischemia level. Our data indicate that post-ischemic hypothermia modifies the profiles of post-ischemic brain tissue pH and Mg 2÷ concentration, and this modification is not associated with histopathological outcome 7 days after ischemia.
Introduction H y p o t h e r m i a has been studied as a protective treatment against experimental ischemic brain damage, when induced either during or after ischemia (Berntman et al. 1981; Busto et al. 1987; Minamisawa et al. 1990a, b; Welsh et al. 1990). Mild 34°C post-ischemic hypothermia lessens neuronal damage after 8 min of forebrain ischemia but not after 12 min of ischemia (Chopp et al. 1991), while induction of 30°C whole-body hypothermia after a 10-min episode of forebrain ischemia protects neurons from ischemic cell damage (Busto et al. 1989a). The metabolic response of brain to cerebral ischemia is also affected by hypothermic intervention. In vivo 31p nuclear magnetic resonance ( N M R ) spec-
Correspondence to: H. Chen, Henry Ford Hospital, 2799 W. Grand Blvd., Detroit, MI 48202, U.S.A. Tel.: (313) 876-3936; Fax: (313) 8761318.
troscopy m e a s u r e m e n t s of cerebral metabolism during hypothermic transient global cerebral ischemia and reperfusion in the cat revealed a reduction of the rate of loss of high energy phosphate metabolites and intracellular acidosis during ischemia, and a more rapid rate of return of metabolites and p H upon reperfusion compared to the normothermic global ischemia (Chopp et al. 1989). Distinct post-ischemic profiles of cerebral p H and free Mg 2÷ concentration were exhibited after normothermic forebrain ischemia (Chopp et al. 1990a; Vande Linde et al. 1990). To our knowledge, however, there have been no reports on the effect of postischemic hypothermia on the time course of cerebral energy metabolism, p H and Mg 2+ concentration after ischemia. The objective of the present study was to determine whether 30°C hypothermia affects neuronal damage and metabolic response, when induced immediately after 12 rain of forebrain ischemia in the rat. We tested the hypothesis that p H and Mg 2+ concentration profiles measured by in vivo 31p N M R spectroscopy reflect
192 the neuropathological outcome after normothermic and hypothermic ischemia.
Materials and methods
Twenty-seven male Wistar rats, weighing 200-300 g, were used in the experiment. Twelve min forebrain ischemia was induced using bilateral carotid artery occlusion in conjunction with systemic hypotension. This ischemia model has been described by Smith et al. (Smith et al. 1984a,b) and has been employed in our laboratory (Chopp et al. 1990a,b, 1991). Animals were divided into 3 groups: (1) normothermic ischemic animals (n = 17), in which rectal temperature was controlled and maintained at 37°C throughout the experimental procedure. Selected in vivo 31p NMR data from 12 of these animals have been previously reported (Chopp et al. 1990a), (2) hypothermic ischemic animals (n = 6), in which rectal temperature was maintained at 37°C before and during ischemia, and 30°C whole-body hypothermia was induced immediately after ischemia and maintained for 2 h of reperfusion, (3) hypothermia without ischemia (n = 4), in which rectal temperature was reduced to 30°C for 2 h without ischemia. Animals were fasted overnight before surgery, but allowed free access to water. Rectal temperature was controlled in all the experiments using a feed-back heating pad. Hypothermia was instituted 5 rain after ischemia, or after induction of anesthesia in the control hypothermic animals, by spraying alcohol onto the skin and fanning the animal with room air (22°C). 30°C hypothermia was achieved within 15 min. The heating pad was set to prevent the animal's temperature from falling below 30°C. After 2 h of reperfusion, the animal was rewarmed to 37°C over a 30-rain period using a heating pad. In this ischemia model, brain temperature is closely coupled ( +__l°C) to rectal temperature during control and reperfusion conditions (Chopp et al. 1991). Each animal was anesthetized 24-48 h prior to ischemia with 0.5% halothane in 70% N20 and 30% 0 2 using a face mask, and placed in the magnet to obtain a baseline pre-ischemia 31p NMR spectrum. Baseline and post-ischemia (24, 48, 72, 96 and 168 h) NMR measurements were performed using the method previously reported (Chopp et al. 1990a,b). In vivo 3~p NMR spectroscopy was performed at 161.9 MHz using an 89-mm diameter wide bore, vertical magnet and Varian VXR4000 spectrometer. A two-turn surface coil (6 mm o.d.) was centered over the rat brain. The NMR sensitive volume approximates a hemisphere with a 3-mm radius, encompassing cortex and extending into hippocampus. Each spectrum was acquired with a spectral width of 40 tzs at a repetition rate of 0.6 s, and was the sum of 1000 transients. NMR1 software (New Method Research, Syracuse, NY, U.S.A.) was used to
process spectra to obtain area ratios of phosphocreatine (PCr) to inorganic phosphate (Pi), /3-adenosine triphosphate (/3-ATP) to total phosphate, and PCr to /3-ATP, as metabolic energy parameters. Chemical shifts for Pi, PCr, 7-ATP, a-ATP, and /3-ATP were recorded for subsequent pH and [Mg 2~] analysis, lntracellular pH was calculated prior to the NMR1 analysis from the chemical shift between phosphocreatinc and inorganic phosphate. Free magnesium ion concentrations, [Mg2*], were assessed using the chemical shift of Pi, PCr, 3,-ATP, a-ATP, and /3-ATP in combination with an in vitro calibration scheme (Halvorson et al. in press). Seven days after the ischemic insult, the animals were anesthetized i.m. with ketamine (44 mg/kg) and xylazine (13 mg/kg). Brain perfusion was performed transcardially with saline and 10% buffered formalin. Each brain was carefully removed 1 h after perfusion, and immersed in formalin for one week before sectioning. Brains were cut coronally into 3-mm thick slabs with a rodent brain matrix. The tissues were processed, embedded in paraffin, and 6-1zm thick slices were stained with hematoxylin and eosin. Brain tissue damage was evaluated using light microscopy from regions of tissue from which NMR data are obtained. A standard section, identified by the hippocampal shape and size, containing subiculum, hippocampus CA 1/2, CA3, was selected from each rat. Fig. 1 illustrates the demarcation of hippocampal regions in which cellular damage was evaluated using the GFAP immunohistochemistry staining method. Neurocortical damage was evaluated in a 3 x 2 mm 2 area in each hemisphere, and the hippocampus CA 1/ 2 was further divided into medial, central, and lateral subregions. Semiquantification of damaged neurons in the cortex was performed on the following scale: 0 = no ischemic neurons, I - - f e w ischemic neurons, 2 = many ischemic neurons, and 3-most neurons are ischemic. Quantitative calculation of neuronal damage in hippocampus was evaluated by direct counting of the neurons in the CA1/2 sections of hippocampus, and the data expressed as the total numbers of the intact neurons and the percentage of necrotic neurons to total neurons. The examiners were blinded to the animal groups.
Data analysis Brain tissue pH, [Mg 2+] and high energy phosphate values were analyzed with a repeated measures analysis of variance. If a significant interaction was detected between groups and time, then two sample t-tests were performed at each time point to compare the group differences. Paired t-tests were performed within each group to compare each time point to the pre-ischemic values. All analyses were corrected for multiple comparisons using Bonferroni correction. Data from the control hypothermic group were analyzed quantita-
193 tively. A Wilcoxon r a n k sum test was used to c o m p a r e the h y p o t h e r m i c a n d n o r m o t h e r m i c ischemic groups o n cortical d a m a g e a n d two sample t-tests were c o n d u c t e d to c o m p a r e intact n e u r o n s b e t w e e n the two groups. A n arc-sine t r a n s f o r m a t i o n was p e r f o r m e d to n o r m a l i z e h i p p o c a m p a l n e u r o n a l c o u n t data, t h e n S t u d e n t ' s ttests were p e r f o r m e d on the p e r c e n t a g e of necrotic n e u r o n s verses total n e u r o n s b e t w e e n n o r m o t h e r m i c a n d h y p o t h e r m i c groups. All data are p r e s e n t e d as m e a n _+ SD.
TABLE 1 MEAN-+SD OF BLOOD GAS AND SERUM GLUCOSE VALUES FOR NORMOTHERMIC ANIMALS AND HYPOTHERMIC ANIMALS pH (ram Hg)
pCO 2 (mm Hg)
pO 2 (mm Hg)
Glucose (mg/dl)
Normothermia (n = 17) Before ischemia 7.36+_0.03 35+5 After ischemia 7.34_+0.02 31 +5
136+24 142_+25 124-+29 -
Hypothermia (n = 6) Before ischemia 7.35-+0.04 39+-7 After ischemia 7.30_+0.01 4 0 + _ 8
139-+26 129_+18 133+_36 -
Results
Physiological p a r a m e t e r s are given in T a b l e 1. A r t e rial blood gases a n d s e r u m glucose values prior to ischemia a n d 30 m i n after ischemia were within n o r m a l physiological ranges for the two groups of animals. No arterial b l o o d gases were t a k e n d u r i n g N M R acquisition of spectra, b e c a u s e o u r previous study had shown that blood gas values were within n o r m a l physiologic r a n g e while a n i m a l s were placed vertically in the N M R p r o b e ( C h o p p et al. 1990a).
Fig. 2 shows 31p N M R spectra from a r e p r e s e n t a t i v e a n i m a l o b t a i n e d before, 24 h, a n d 168 h after 12 m i n f o r e b r a i n ischemia. Relative intensities of spectral peaks were similar at all time points. No i n t e r a c t i o n s for m e t a b o l i c p a r a m e t e r s ( / 3 - A T P / t o t a l p h o s p h a t e , P C r / / 3 - A T P , P C r / P i ) were d e t e c t e d b e t w e e n groups a n d time ( P > 0.4, data not shown), i n d i c a t i n g b r a i n
~ii~i~i~ii:(/
Fig. 1. Glial fibrillary acidic protein (GFAP) immunohistochemical staining of a 6-/zm coronal section from a rat subjected to 12 min forebrain ischemia. Astrocytic reaction with GFAP indicates a sharp demarcation between CA1/2 and CA3 regions of hippocampus (arrow). A 3 x 2 mm2 box indicates the area in one hemisphere in which cortical damage is evaluated. Bars divide CA1/2 into medial, central, and lateral regions for determination of neuronal counts.
194 TABLE 2 MEAN_+SD O F [Mg z+] F O R N O R M O T H E R M I ( " POTHERMIC ISCHEMIC ANIMALS
Baseline 24 h 48 h 72 h 96 h 168 h
A N D HY-
Normothermia (n = 17)
Hypothermia (n = 6)
0.27 + 0.08 0.25 +0.05 0.27_+0.10 (I.22 +_0.04 0.21 ± 0.05 0.25 _+0.07
0.23 ± 0.03 0.39±0.22 0.42±0.18 0.25 ± 0.07 0.21 ± 0,11 0.24 + 0.18
mM mM mM * mM ~ mM mM
mM mM mM mM mM mM
* P = 0.02 compare to hypothermia; + P = 0.006 compare to baseline.
3
12 14 ,5
3O
0
-]0
PPM Fig. 2. 3~p N M R spectra from a representative rat; pre-ischemia (bottom), 24 h after 12 min forebrain ischemia (middle) and 168 h after ischemia (top). Spectral peaks: 1. primarily phosphomonoesters, 2. inorganic phosphate, 3. phosphocreatine, 4. primarily ~,-ATP, 5. primarily a - A T P , 6. ~-ATP.
metabolism has returned to the pre-ischemia value in both normothermic and hypothermic ischemic animals. Fig. 3 shows the time course of N M R pH values for the hypothermic and normothermic ischemic groups. Control hypothermic animals exhibited constant pH values over time (data not shown). No difference was 7.50
7.36
7.22 Q, 7.08
6.94
6.80
J
I
I
J
I
0
24
48
72
96
I
J
I
120
144
168
TIME (I.IOIJRS)
Fig. 3. Mean brain tissue pH + SD m e a s u r e d prior to ischemia (time 0) and at 24, 48, 72, 96 and 168 h after 12 min forebrain ischemia, zx, normothermic animals (n = 17), o, hypothermic animals (n =6).
detected ( P > 0.1) between pre-ischemic pH values in normothermic and hypothermic ischemic animals. An interaction between the normothermic and hypothermic ischemic groups and time was detected ( P = 0.004), indicating that the two groups behave differently over time. Significantly lower pH was detected in the hypothermic ischemic group, at 24 h ( P = 0.0001) and 48 h ( P = 0.018) post-ischemia, compared to the normothermic ischemic group. In the normothermic ischemic animals, alkalosis was exhibited both at 24 h (pH = 7.24 + 0.14, P = 0.0002) and 48 h (pH = 7.20 + 0.14, P = 0.002) post-ischemia, compared to the pre-ischemic values (pH = 7.06 + 0.05). Conversely, hypothermic ischemic animals revealed acidosis (pH = 6.99 __+0.05, P = 0.0002) at 24 h compared to its pre-ischemic values (pH = 7.09 _+ 0.05). Although statistically significant mean alkalosis was not detected in the hypothermic ischemic rats, 5 out of 6 animals exhibited atkalosis either at 72, 96, or 168 h after the ischemic insult (pH from 7.17 to 7.36). Table 2 shows the time course of free [Mg 2+] for the normothermic and hypothermic ischemic animals. [Mg 2+] were constant in the control hypothermic animals (data not shown). No difference was detected ( P > 0.3) between pre-ischemic [Mg 2+] in normothermic and hypothermic ischemic animals. An interaction between groups and time was detected ( P < 0.01). At 48 h post-ischemia, the hypothermic [Mg ÷2 ] was significantly elevated compared to the normothermie [Mg 2+ ]. In the normothermic ischemic animals, [Mg 2+] decreased at 72 h ([Mg z÷] = 0.22 + 0.04 mM, P = 0.006) compared to the pre-ischemic [Mg z+ ] value ([Mg 2+ ] = 0.27 _ 0.08 mM). No neuronal damage was detected in the control hypothermic group. Cerebral cortex revealed few scattered dying neurons in both normothermic and hypotherrnic ischemic animals. No difference in neuronal damage of cerebral cortex was detected between the normothermic and hypothermic ischemic groups (scores =0.81 +__0.66 in normothermia verses 0.67 + 0.52 in hypothermia). Fig. 4 shows the sections of
195 hippocampus for the control hypothermic animals and the normothermic and hypothermic ischemic animals. Control hypothermic animals demonstrated no neuronal damage, while both ischemic groups exhibited severe neuronal injury in the hippocampal region. The numbers of intact neurons and the percentage of necrotic neurons in the medial, central, lateral and overall hippocampus CA1/2 are shown in Table 3. A marginal difference was detected between the two groups only in the medial region (P = 0.056), showing that hypothermic ischemic animals had fewer necrotic neurons (85.8 + 9.7%) compared to the normothermic ischemic animals (93.2 + 5.8%).
Discussion
Our data indicate that lowering the body temperature to 30°C post-ischemia does not significantly protect against or alter the pattern of neuronal injury in hippocampus and cortex after 12 min of forebrain ischemia. However, post-ischemic hypothermia significantly modifies dynamic profiles of change in the corresponding brain tissue pH and [Mg2+]. The data thus suggest that these profiles have little or no predictive value for neuronal damage in this model of forebrain ischemia. Post-ischemic intracellular cerebral alkalosis has been reported both in stroke patients (Syrota et al. 1985; Welch et al. 1989) and in experimental animals (Mabe et al. 1983; Chopp et ah 1990a, b). The profile of cerebral alkalosis has been associated with the duration of ischemia, and alkalosis per se has been thought to mediate ischemic cell damage (Chopp et al. 1990a). Among stroke patients, cerebral alkalosis has been associated with ischemic irreversibility brain infarct (Welch et al. 1989). Our data do not contradict clinical data indicating that cerebral infarction is associated with intracellular alkalosis. Our data, however, suggest that the presence (normothermia) or absence (hypothermia) of post-ischemic intracellular alkalosis is not an indication of mild ischemic cell damage in transient forebrain ischemia. NMR pH measurement is an average of pH values from all cells within the detected volume of the surface coil. In brain tissue, the cells include neurons, glia, endothelial cells and inflammatory cells. The presence of inflammatory cells or reactive glia may contribute to the brain tissue alkalosis. Proliferating cells are activated in an alkalotic environment, and themselves exhibit intracellular alkalosis (McCord and Roy 1982; Schuldiner and Rozengurt 1982; Chesler and Kraig 1989). Possibly, hypothermia alters the time course of reactive response to tissue injury or delays the neuronal injury process. Analysis of the neuronal, glial and
inflammatory cell activity for the normothermic and hypothermic animals at the 24-h or 48-h time points are necessary to resolve the possible role of glia and inflammatory cells in the altering acid/base balance of ischemic brain tissue and delayed neuronal injury by hypothermia. The reason for brain tissue acidosis 24 h after post-ischemia hypothermia is unknown. The effect of intracellular pH on free intracellular [Mg 2+] may play a role in promoting ischemic cell damage. Concomitant reduction of [Mg 2+] and tissue alkalosis has been reported in brain, muscle, and heart (Adam et ah 1989). Mg 2+ may stabilize the plasma membrane, compete with Ca 2÷ to reduce Ca 2+ entry into the cell, and inhibit release of excitatory neurotransmitters (Rothman 1983; Mayer et ah 1984; Kass et al. 1988). Thus, magnesium depletion, paralleling alkalosis, may evoke tissue damage in response to an ischemic insult. In the present study, the level of decrease in [Mg 2÷] observed in normothermic ischemia animals does not evoke severe tissue damage. Likewise maintenance of [Mg 2+] at control levels by hypothermia fails to improve histological outcome. Thus the modifications of [Mg 2+] as measured in the present experiment does not significantly alter neuronal damage. Hypothermic protection in cerebral ischemia has been attributed to a combination of reduction of excitatory neurotransmitter release (Busto et al. 1989b), mitigation of abnormal ion fluxes (Astrup et al. 1981), reduction of edema (Dempsey et al. 1987) and lactate (Schuldiner and Rozengurt 1982), and reduction of leukotrienes (Dempsey et ah 1987). Our previous findings indicated that 34°C post-ischemia hypothermia only protected neuronal damage after 8 min ischemia, but not after 12 min ischemia (Chopp et al. 1991). Other investigators have demonstrated that a lower degree of hypothermia (27°C) during ischemia completely protects against neuronal damage after 20 min of forebrain ischemia, while animals subjected to postischemic hypothermia exhibited variable neuronal damage (Boris-Moiler et al. 1989). Busto et al. found neuronal protection by 30°C post-ischemic hypothermia in animals subjected to 10 min of forebrain ischemia (Busto et al. 1989a). Thus, the neuronal protection by post-ischemic hypothermia depends on both duration of ischemia and depth of hypothermia. The presence of only selective marginal protection by postischemic hypothermia in the present model may be attributed to the complex interplay of a "long" 12-min duration of ischemia coupled with a moderate postischemia 30°C hypothermia. 30°C hypothermia is apparently not low enough to exert strong neuronal protection against 12 min of forebrain ischemia. In summary, our data suggest that, although postischemic hypothermia significantly modifies the profiles of intracellular pH and [Mg 2÷] after a transient
196
Fig. 4. H / E staining of 6 - # m thick paraffin sections of hippocampal C A 1 / 2 region illustrating neuronal damage of (a) control hypothermic animal, (b) normothermic 12-min ischemic animal, and (c) hypothermie 12-rain ischemic animal. Neurons are intact in the control hypothermic animals. Necrotic neurons were exhibited in both ischemic groups of animals (noted by arrows) ~ × 300).
197
Fig. 4. (continued). TABLE 3 MEAN_+ SD OF THE NUMBERS OF INTACT NEURONS AND THE PERCENTAGE OF NECROTIC NEURONS TO TOTAL NEURONS OF HIPPOCAMPUS CA1/2 REGION
Medial Central Lateral Overall
Normothermia (n = 16)
Hypothermia (n = 6)
Numbers
%
Numbers
%
11.7+ 8.1 11.2_+ 8.3 17.3_+ 9.7 40.1 _+24.8
93.2_+5.8 89.6_+5.4 86.8_+8.5 90.0_+6.5
19.6_+14.5 16.9_+12.2 23.9_+14.9 60.4+40.3
85.8_+ 9.7 * 85.8_+10.7 83.3_+11.3 84.2_+11.1
* P = 0.056.
forebrain ischemia, no significant effect of hypothermia on histopathological outcome is detected. Acknowledgements The authors wish to thank Dr. Quan Jiang and Ms. Sue Ann Lee for technical assistance and Ms. Mary L. Rexroad for manuscript preparation. This work was supported by NINDS Grants NS23393 and NS29463.
References Adam, W.R., Craik, D.J., Kneen, M. and Wellard, R.M. (1989) Effect of magnesium depletion and potassium depletion and chlorothiazide on intracellular pH in the rat, studied by 3lp NMR. Clin. Exp. Pharm. Physiol., 16: 33-40.
Astrup, J., Moller-Sorenson, P. and Rahbeck-Sorenson, H. (1981) Inhibition of cerebral oxygen and glucose consumption in the dog by hypothermia, pentobarbital and lidocaine. Anesthesiology, 55: 263-268. Berntman, L., Welsh, F.A. and Harp, J.R. (1981) Cerebral protective effect of low-grade hypothermia. Anesthesiology, 55: 495-498. Boris-Moiler, F., Smith, M.L. and Siesj6, B.K. (1989) Effects of hypothermia on ischemic brain damage: a comparison between preischemic and post-ischemic cooling. Neurosci. Res. Commun., 5: 87-94. Busto, R., Dietrich, W.D., Globus, M.Y.T, Valdes, I., Scheinberg, P. and Ginsberg, M.D. (1987) Small differences in intra-ischemic brain temperature critically determine the extent of ischemic neuronal injury. J. Cereb. Blood Flow Metab., 7: 729-738. Busto, R., Dietrich, W.D., Globus, M.Y. and Ginsberg, M.D. (1989a) Post-ischemic moderate hypothermia inhibits, CA1 hippocampal ischemic neuronal injury. Neurosci. Lett., 101: 299-304. Busto, R., Globus, M.Y.T, Dietrich, D., Martinez, E., Valdes, I. and Ginsberg, M.D. (1989b) Ischemia-induced release of neurontransmitters and free fatty acids in the rat brain: effects of mild intra-ischemic hypothermia. Stroke, 20: 904-910. Chesler, M. and Kraig, R.P. (1989) Intracellular pH transients of mammalian astrocytes. J. Neurosci., 9: 2011-2019. Chopp, M., Knight, R., Tidwell, C.D., Helpern, J.A., Brown, E. and Welch, K.M.A. (1989) The metabolic effects of mild hypothermia on global cerebral ischemia and recirculation in the cat: comparison to normothermia and hyperthermia. J. Cereb. Blood Flow Metab., 9: 141-148. Chopp, M., Chen, H., Vande Linde. A.M.Q, Brown, E. and Welch, K.M.A. (1990a) The time course of post-ischemic intracellular alkalosis reflects the duration of ischemia. J. Cereb. Blood Flow Metab., 10: 860-865.
198 Chopp, M., Vande Linde, A.M.Q, Chen, H., Knight, R., Helpern, J.A. and Welch, K.M.A. (1990b) Chronic intracellular cerebral alkalosis after an 8-minute forebrain ischemic insult in the rat. Stroke, 21(3): 463-466. Chopp, M., Chen, H., Dereski, M.O. and Garcia, J.H. (1991) Mild hypothermie intervention after graded ischemic stress in rats. Stroke, 22: 37-43. Dempsey, A.L., Combs, D.J., Edwards Maley, M., Cowen, D.E., Roy, M.W. and Conaldson, D.L. (1987) Moderate hypothermia reduces post-ischemic edema development and leukotrine production. Surgery, 21: 177-181. Halvorson, H., Vande Linde, A.M.Q, Helpern, J.A. and Welch, K.M.A. (in press) Assessment of magnesium concentrations by P-31 NMR in vivo. NMR in Biomedicine. Kass, I.S., Cottrell, J.E. and Chambers, G. (1988) Magnesium and cobalt, not nimodipine, protect neurons against anoxic damage in the rat hippocampal slice. Anesthsiology, 69: 710-715. Mabe, H., Blomqvist, P. and Siesj6, B.K. (1983) Intracellular pH in the brain following transient ischemia. J, Cereb. Blood Flow Metab., 3: 109-114. Mayer, M.L., Westbrook, G.L. and Cuthrie, P.B. (1984) Voltage-dependent block by magnesium ion of NMDA receptors in spinal cord neurones. Nature, 309: 261-263. McCord, J.M. and Roy, R.S. (1982) The pathophysiology of superoxides: roles in inflammation and ischemia. Can. J. Physiol. Pharmacol., 60: 1346-1352. Minamisawa, H., Nordstrom, C.-H., Smith, M.-L. and Siesj6, B.K. (1990a) The influence of mild body and brain hypothermia on ischemic brain damage. J. Cereb. Blood Flow Metab., 10: 365-374. Minamisawa, H., Smit M.-L. and Siesj6, B.K. (1990b) The effect of mild hyperthermia and hypothermia on brain damage following 5, 10 and 15 minutes of forebrain ischemia. Ann. Neurol., 28: 26-33.
Rothman, S.R. (1983) Synaptic activity mediates death ol hypoxic neurons. Science, 536-537. Schuldiner, S. and Rozengurt, E. (1982) N a / H antiport and swiss 3T3 cells. Mitogenic stimulation leads to cytoplasmic alkatinization. Proc. Natl. Acad. Sci. U.S.A., 79:7778 7782. Smith, M.L., Auer, R.N. and Siesj6, BK. (1984a) The density of distribution of ischemic brain injury in the rat following 2-10 minutes of forebrain ischemia. Acta Neuropathol. (Berl.), 64: 319-332. Smith, M.L., Bendek, G., Dahlgren, N., Rosen, I., Wiek)ch, T. and Siesj6, B.K. (1984b) Models for studying long-term recovery following forebrain ischemia in the rat. 2. A 2-vessel occlusion model. Acta Neurol. Scand., 69: 385-401. Syrota, A., Samson, Y., Boullais, C., Wajnberg, P., Loc'h, C., Crouzwl, C., Maziere, B., Soussaline, F. and Baron, J.C. (1985) Tomographic mapping of brain intracellular pH and extracellular water space on stroke patients. J. Cereb. Blood Flow Metab., 5: 358-368. Vande Linde, A.M.Q., Chopp, M., Chen, H.. Halvorson, tt.R., Helpern, J.A., Brown, E. and Welch, K.M.A. (1990) Decline of cerebral free Mg 2 concentration after forebrain ischemia. Society of Magnetic Resonance in Medicine, Ninth Annual Scientific Meeting and Exhibition, 2: 980. Welsh, F.A., Sims, R.E. and Harris, V.A. (199tl) Mild hypothermia prevents ischemic injury in gerbil hippocampus: J. Cereb. Blood Flow Metab., 10: 557-563. Welch, KM.A., Levine, S.R. and Helpern, J.A. (1989) Pathophysio[ogicla correlates of cerebral ischemia: the significance of cellular acid base shifts. In: Richichi, I. (Ed.), Proceedings of the International Conference of Cardiology and Neurology. Palmi, Italy, Calabro, pp. 205-217.