*:j:ken-ichiro Katsura, * laroslava Folbergrova, *tfinn Bengtsson, *Tibor Kristian, *Gunilla Gid6, and *Bo K. Siesj6

Transcription

1 Journal oj Cerebral Blood Flow and Metabolism 13: The International Society of Cerebral Blood Flow and Metabolism Published by Raven Press, Ltd" New York Recovery of Mitochondrial and Plasma Membrane Function Following Hypoglycemic Coma: Coupling of ATP Synthesis, K + Transport, and Changes in Extra- and Intracellular ph *:j:ken-ichiro Katsura, * laroslava Folbergrova, *tfinn Bengtsson, *Tibor Kristian, *Gunilla Gid6, and *Bo K. Siesj6 *Laboratory fo r Experimental Brain Research. Experimental Research Center, and tdepartment of Clinical Pharmacology, University of Lund, Lund, Sweden; :f:second Department of Internal Medicine, Nippon Medical School, Tokyo, Japan; and Institute of Physiology, Czech Academy of Sciences, Prague, Czech Republic Summary: The primary objective of the present study was to evaluate the recovery of plasma and mitochondrial membrane functions after 3 min of hypoglycemic coma and to establish whether a lingering accumulation of free fatty acids (FFAs) delays the recovery. A secondary objective was to study whether production of metabolic acids following glucose infusion leads to a fall in intracellular ph (phj). Phosphocreatine, creatine, ATP, ADP, and AMP, as well as glycogen, glucose, lactate, pyruvate, and FF As of rat brain cortex and caudoputamen were measured, and "free" ADP was calculated from the creatine kinase equilibrium. Extracellular ph (ph e) and K + concentration (K + e) were measured with ion-sensitive microelectrodes, and phj was derived by the CO2 method. Glucose injection was followed by resumption of oxidative phosphorylation within 2 min and by an equally rapid restoration of normal K + e levels. These functions recovered although tissue FF As remained elevated for at least 7-8 min. Tissue lactate content increased only moderately and production of metabolic acids did not lead to intracellular acidosis. After 15 min of recovery, phj was moderately increased, although phe fell toward 7 It is speculated that the dissociation between intra- and extracellular ph is compatible with an up-regulation of an Na + ih + antiporter, e.g., by phosphorylation. Key Words: Energy metabolites-extracellular ph-free fatty acids-hypoglycemic coma-intracellular ph-k + transport-recovery. The brain damage incurred after transient ischemia or hypoglycemic coma is probably secondary to energy failure (Siesj6, 198 1, 1984; Auer and Siesj6, 1988). As discussed in a recent review (Siesj6, 1992a,b), the molecular mechanisms of cell damage are triggered by the decrease in phosphorylation potential (ATP. ADP-I. Pi -I). The secondary consequences encompass loss of cellular ion homeostasis, with release of K + from cells, and uptake of Na+, Cl-, and Ca2+, as well as extensive Received December 16, 1992; final revision received February 22, 1993; accepted March 25, Address correspondence and reprint requests to Dr. K. -l. Katsura at Laboratory for Experimental Brain Research, Experimental Research Center, Lund University Hospital, S Lund, Sweden. Abbreviations used: Cr, creatine; DC, direct current; FFA, free fatty acid; PCr, phosphocreatine. lipolysis, causing accumulation of lysophospholipids, free fatty acids (FF As), thromboxane A 2, prostaglandins, leukotrienes, and platelet-activating factor (see Bazan et al., 199 1; Lindsberg et al., 199 1). At least following transient ischemia, a significant part of the final damage is probably incurred in the immediate postinsult period, i.e., at the time when oxygen supply is restored (Kontos, 1989; Siesj6 et al., 1989; Watson and Ginsberg, 1989). This is because oxidation of reduced compounds, of xanthine/hypoxanthine, and of FF As such as arachidonic acid leads to enhanced production of free radicals. It is much less obvious that such postinsult mechanisms operate following hypoglycemic coma since acidosis is not present and since an anaerobic/ aerobic transition, with a sudden oxidation of reduced compounds, does not occur. However, at least in the caudoputamen, postinsult aggravation 82

2 BRAIN RECOVER Y FOLLOWING HYPOGLYCEMIA 821 of histopathologically verified cell damage has been observed (Kalimo et ai., 1985). In view of this, it is justified to analyze recovery events following ischemia and hypoglycemia. The present article concerns hypoglycemic coma. Tissue was analyzed after 3 min of hypoglycemic coma, as well as, 2, 5, and 15 min following repolarization of cell membranes (2, 4, 7, and 17 min following intravenous injection of glucose). The main objective of the experiments was to correlate resumption of mitochondrial metabolism to restoration of normal K + e levels. The main question posed was whether recovery of mitochondrial and plasma membrane function is faster following hypoglycemic coma than fo\1owing ischemia. The results given novel information on the coupling of mitochondrial and plasma membrane function during recovery following hypoglycemic coma. Furthermore, the results on extra- and intrace\1ular ph have a bearing on mechanisms regulating intracellular ph. MATERIAL AND METHODS Male Wistar rats of an SPF strain (Mpllegaard Breeding Center, Copenhagen, Denmark), weighing 3-35 g, were starved overnight but had access to tap water ad libitum. One-half hour before anesthesia, they were injected intraperitoneally with insulin 2 IU. kg-i (Actrapid; Novo Industri A/S, Denmark). Anesthesia was induced with 3% halothane (Halothane; ISC Chemicals, Bristol, U.K.) and maintained during operation with 1.5% halothane in a 1:2 mixture of OzINzO. After the operative procedure was finished, the halothane concentration was reduced and kept at.5% throughout the experiments. The rats were tracheotomized and artificially ventilated so as to give an arterial Pco2 of 35-4 mm Hg and a P2 of 1 mm Hg. Catheters were introduced into one femoral or tail artery for blood pressure monitoring, as well as P2, Pco2 ph, and glucose measurements, into one femoral vein or tail vein for drug administration, and into one jugular vein for blood pressure control by withdrawal or injection of blood through a fixed syringe attached to the catheter. The rats were heparinized (heparin 18 IU. kg-i) and paralyzed with d-tubocurarine (.5 mg. kg-i). A frontoparietal craniotomy (3 x 3 mm), with a well attached to the bone surrounding the craniotomy filled with mock CSF bubbled with 5% CO2 at 37"C, gave access to the brain surface for microelectrode measurements of the direct current (DC) potential and either phe or K + e (Mutch and Hansen, 1984; Bengtsson et ai., 199; see also Katsura et ai., 1992). The operated rats were placed on a plastic table in a Faraday cage, measuring.6 x.7 x 1.1 m, before the introduction of the microelectrode. The dura was excised and the microe1ectrode was advanced 5 ± 2 ILm into the neocortex by a micromanipulator. A ground electrode of Agi AgCl in 2% agar and 15 mm NaCI was inserted into the neck musculature (for methodological details, see Bengtson et ai., 199). Atropine (2 mg. kg-i) was administered to avoid bradycardia and reduce airway secretion. EEG was continuously recorded by bipolar needle electrodes inserted in the muscles lateral to the skull bone. The rats were maintained at a rectal temperature of 37 C with the help of a thermistorcontrolled heating blanket. Control rats underwent the same operative procedure but either were not injected with insulin or were injected with insulin and simultaneously given a continuous infusion of a 25% glucose solution. Blood (or plasma) glucose levels were monitored by repeated measurements so as to maintain glucose concentrations of 6--1 mm, requiring infusion rates of 1-2 ml. h - I. After 3 min of hypoglycemic coma, recovery was induced by administration of an intravenous bolus dose of glucose (.5 ml of 5% glucose solution); simultaneously, a continuous glucose infusion (1% glucose solution) was started and maintained so as to give plasma glucose levels of 6--1 mm. Experimental design Apart from the control animals, all rats were allowed a 3-min period of hypoglycemic coma counted from the time of depolarization as judged from the DC potential shift. Depolarization occurred within 2-3 h from the time of insulin injection. The rats were then either killed by freezing of the brain in situ or given a glucose injection that repolarized membranes within 2 min. Glucoseinjected animals were allowed recovery periods of, 2, 5, and 15 min following the start of repolarization or 2, 4, 7, and 17 min following glucose injection. All rats were killed by freezing of the brain in situ with liquid nitrogen according to Ponten et al. (1973). Brains were then chiseled out under intermittent irrigation with liquid nitrogen and stored at - 8 C. Subsequently these brains were transferred to a cold chamber ( - 2 C) in which cortical tissue and, in some experiments, caudoputamen were cut out. The brain specimens were then stored at - 8 C for subsequent chemical analyses, all performed within 1 month after the experiments. The tissue was analyzed for labile phosphates, glycolytic metabolites, total COz (Tco2), and FFAs as described in previous publications (Ponten and Siejso, 1964; Folbergrova et ai., 1972a,b; Agardh et ai., 1981). There were two separate series. In one of these, encompassing animals with 3 min of hypoglycemic coma as well as with 2, 5, and 15 min of recovery (following the start of membrane repolarization), the tissues analyzed were neocortex and caudoputamen. The second series comprised animals without electrodes in which neocortical tissue was sampled at the time of repolarization (2 min following glucose injection). In the tabulated data, the two control groups for neocortex were pooled. Measurements of phe and K+ e' and calculation of phi and free ADP Extracellular ph (ph e) and K + e concentration (K + e) were measured with ion-sensitive microelectrodes, as described in previous publications (Ekholm et ai., 1993; Katsura et ai., 1992). Intracellular ph (ph) was calculated from Tc2, tissue Pcoz (PtCoz), and phe (Siesjo et ai., 1972). Ptco2 was assumed to exceed the arterial COz tension (P ac2) by 6 mm Hg (Ponten and Siesjo, 1966) except during hypoglycemic coma when PtCOZ was obtained by adding 4 mm Hg to PaCOZ' Extracellular space volume was assumed to be 2% in control and recovery animals and 1% during hypoglycemic coma (Pelligrino et ai., 1981). Calculation of phi was based upon the CO2 method (Siesjo et ai., 1972; J Cereb Blood Flow Metab, Vol. 13, No.5, 1993

3 822 K.-I. KATSURA ET AL. Katsura et al., 1992). For the group studied 2 min following glucose infusion, we calculated phi by using a phe derived from separate experiments and a Tco2 that was the average of those measured in the 3-min coma and 7-min recovery groups. Utilizing phi values thus calculated, and assuming free intracellular magnesium concentrations of.33 mm in the control and recovery periods and.52 mm during hypoglycemic coma (Brooks and Bachelard, 1989), we calculated the free ADP concentrations (ADPr) from the equilibrium equations for the creatine kinase reaction (Lawson and Veech, 1979; Ekholm et al., 1992a). We also calculated ADP r on the assumption of no change in Mg 2 + concentration during and after hypoglycemic coma. As shown in a previous publication, small to moderate deviations of phi or of Mg 2 + concentration have little influence on the calculated ADPr (Ekholm et al., 1992). Statistics Values presented are means ± SD. The data for metabolites were subjected to one-way analysis of variance followed by Scheffe's test. RESULTS Physiological variables and Tco2 contents As shown in Table 1, the physiological variables and Tco 2 values were similar in all groups. Thus, the animals had normal blood pressure, blood oxygen, and carbon dioxide tensions and plasma ph. Blood (or plasma) glucose concentrations were normal or moderately raised in all groups except the "coma" groups. Body temperature was kept constant during the experiments around 37 C. Definition of groups Figure 1 shows a typical record of the DC potential and K + e in hypoglycemic coma. The onset of coma was defined in terms of a sudden negative shift of the DC potential. This was, or was not, followed by a shortlasting, abortive recovery (see Pelligrino et ai., 1981; Harris et ai., 1984). After 3 min, glucose was injected intravenously. After -2 min (13 ± 48 s; n 22), the DC potential started to change toward normal. Recovery was initially defined from that point, i.e., as 2, 5, and 15 min following the start of change of DC potential. Since the 2-min group showed such extensive recovery of energy state, a new group was added with 2-min recovery counting from glucose injection. In the following, the recovery is defined from the point of glucose injection (2, 4, 7, and 17 min). Energy and glycolytic metabolites The results are shown in Table 2. As expected (see Lewis et ai., 1974; Agardh et ai., 1978; Behar et ai., 1985), 3 min of hypoglycemic coma was associated with an extensive breakdown of phosphocreatine (PCr) and ATP and with accumulation of creatine (Cr), ADP, and AMP in both neocortex and caudoputamen. There was also an extensive reduction of the sum of the adenine nucleotides and a substantial decrease in adenylate energy charge (data not shown). The results of Table 2 demonstrate that PCr and Cr concentrations had partly normalized after 2 min and were close to normal within 4 min, suggesting extensive restoration of mitochondrial metabolism. This contention is supported by the extensive recovery of the adenylate energy charge, which was >95% of control after 4 min, and by the normalization of overall ADP and AMP concentrations. To evaluate further the recovery of the cellular energy state, we calculated PCr and ATP concentrations as percentage of control and derived ADPf from the creatine kinase equilibrium, using data on changes in phi (see below). As Fig. 2 shows, the PCr concentration had normalized already after 4 min. Furthermore, the data make it clear that the ATP/ADPf ratio had reached control values, or was above normal, after 4 min. Figure 2 (bottom) illustrates how differences in the assumed Mg2 + concentration influence the results. Even if the Mg2+ concentration increased during and after hypoglycemic coma, the analysis predicts that the ATPI ADPfratio had normalized within 4 min. Hence, the results suggest that the phosphorylation potential of TABLE 1. Physiological variables and total tissue CO2 content (Teo2) in control and hypoglycemic groups MABP Temperature Pc2 P2 Blood glucose Tc2 (mm Hg) (C) (mm Hg) (mm Hg) ph (mm) (mmol. kg-i) Control (n 4) 135 ± ± ± ± ± ± ±.41 Control-I-G (n 8) 137 ± ± ± ± ± ± ±.86 3-min coma (n 7) 129 ± ± ± ± ±.1 NM ll.41 ± min recovery (n 5) 143 ± ± ± ± ± ± ±.63 7-min recovery (n 4) 148 ± ± ± ± ± ± ± min recovery (n 6) 14 ± ± ± ± ± ± ± 2.39 Control (n 6) 13 ± ± ± ± ±. 5.4 ±.3a NM 2-min recovery (n 6) 156 ± ± ± ± ± ± Loa NM Values are means ± SD, n a Plasma glucose. no. of experiments. Control-I-G, controls injected with insulin and glucose; NM, not measured. J Cereb Blood Flow Metab. Vol. 13, No.5, 1993

4 BRAIN RECOVERY FOLLOWING HYPOGLYCEMIA 823 (mv) o DC potential (mm) 5 c:; :l c. u -2-4 J I I Insulin, -6, I' II I II 3 min depolarization Glucose (min) FIG. 1. Typical records of direct current (DC) potential and K+ e measurements during hypoglycemic coma. The duration of the coma period was defined as the time from the sudden DC potential shift until glucose injection. The lag seen in the figure is artificial and due to the different positions of the pens. the (remaining) adenine nucleotide pool is quickly restored following glucose injection. As expected, hypoglycemic coma was associated with a drastic fall in tissue glucose and glycogen concentrations, as well as in the lactate and pyruvate concentrations (Table 2). Infusion of glucose led to a prompt increase in the tissue glucose pool (4 min), but there was no significant resynthesis of glycogen within the first 17 min of recovery. Lactate concentration was predictably reduced during hypoglycemia (see Lewis et ai., 1974) and increased above normal upon infusion of glucose, as reported by Behar et al. (1985). Pyruvate concentration was reduced during hypoglycemia and restored to normal in the recovery groups. FFAs Changes in individual FF A concentrations are illustrated in Fig. 3. As predicted, hypoglycemic coma was associated with a massive increase in FFA concentration (see Agardh et ai., 1981). The present results demonstrate that the increase in TABLE 2. Labile metabolites of rat brain cortex and caudoputamen Control Control-I-G 3-min coma 2-min recovery 4-min recovery 7 -min recovery 17 -min recovery (1) (8) (7) (6) (5) (4) (6) Phosphocreatine Neocortex 4.44 ± ± ± ±.49a 4.49 ± ± ±.29 Caudoputamen 4.93 ± ± ±.43a 5.2 ± ± ±.46 Creatine Neocortex 6.2 ± ± ±.63" 7.33 ± ± ± ±.45 Caudoputamen ± ± ±.98a 5.82 ± ± ±.6 ATP Neocortex 2.84 ± ±.9.76 ± ±.7a 1.55 ±.22a 1.68 ±.3a 1.84 ±.7a Caudoputamen 2.88 ± ± ± ±.29a 1.91 ±.9a ±.14a ADP Neocortex.28 ±.3.26 ± ±.6.37 ±.5a.27 ±.5.23 ±.2.25 ±. 1 Caudoputamen.26 ±.6.24 ±.3.62 ±.6.27 ±.4.25 ±.2.27 ±.3 AMP Neocortex.6 ±..6 ±. 1.4 ± O.lOa. 14 ±.4.6 ±.2.6 ±. 1.7 ±.4 Caudoputamen.6 ±. 1.5 ± ±.14a.7 ±.2.6 ±. 1.6 ±. 1 Glucose Neocortex 1.97 ± ± ±.15a.24 ± O.l1a 1.23 ± o.na ±.73a 3.64 ± 1.2 Caudoputamen 2.67 ± ±.32.3 ± ± ± ±.73 Glycogen Neocortex 2.39 ± ±.57.7 ±.7".12 ±.8a. 1 ±.4a.23 ±.15a Caudoputamen ± ± ±.4a. 11 ± ±.5a. 18 ±.4a Lactate Neocortex 1.48 ± ± ± ± ± ± ± 1.63a Caudoputamen 1.57 ± ± ± ± ± ±.79 Pyruvate Neocortex.9 ±. 1.1 ±. 1.3 ± O.Ola.6 ±.2a. 12 ± ±.2.1 ±. 1 Caudoputamen.7 ±. 1.8 ±.2.3 ± ±.2. 1 ± ±.3 Values are means ± SD. Numbers of rats comeocortex are within parentheses; numbers for measurements in caudoputamen are all 4. Control-I-G, controls injected with insulin and glucose. p <.5 compared with individual control-i-g group with analysis of variance followed by Scheffe's test. J Cereb Blood Flow Metab, Vol. 13, No.5, 1993

5 824 K.-I. KATSURA ET AL. :a 12 Y ---- PCr -. - ATP '-' 1 8. U." 6 <II '-' <II. :: < Co? ';' C e U os " e " " '" " u '" '" '" c c c c c S S S S S, N.!:; FIG. 2. Top: Changes of ATP and phosphocreatine (PCr) concentrations expressed as % of control. Bottom: Changes in ATP/free AOP (AOP,) ratio. (--), constant Mg2+ concentration (.33 mm); (. -e), Mg2+ increased to.52 mm during and after hypoglycemic coma (see text). Control-I-G, controls injected with insulin and glucose; rec, recovery. Values are expressed as means ± SO. FF A concentrations persisted for at least 7 min into the recovery period and that full recovery was not observed in the period of 7-17 min. K+ e concentrations At the onset of hypoglycemic coma, K + e rose to values of 5 mm (see Pelligrino et ai., 1982; Harris et ai., 1984). As illustrated in Fig. 1, K + e started to decrease 2-3 min after injection of glucose; i.e., just after that repolarization was observed (as judged by the positive shift in DC potential) and the phosphorylation potential was rising. After 5 min, K + e val- ues decreased to the precoma value (3.5 mm). A further reduction of K + e (undershoot) was sometimes observed, with a gradual return. Acid-base changes The present results confirm previous ones, demonstrating that phi either remains unchanged or increases somewhat during hypoglycemic coma (Pelligrino et ai., 198 1; Behar et ai., 1985) and that phe decreases during recovery (Bengtsson et ai., 199). As demonstrated, this secondary decrease in phe was associated with a phi that was either normal or even increased (Fig. 4). Thus, phi remained unchanged or increased at a time when tissue lactate content increased (see Table 2). DISCUSSION The present results should be viewed against those previously published. Agardh et al. (1978, 1982) described recovery following 3 or 6 min of hypoglycemic coma, but the observations were restricted to postinsult periods of 3 min. Behar et al. (1985) used IH and 31p nuclear magnetic resonance to study changes in lactate, labile phosphates, and phi' Their results describe recovery events similar to those reported in the present study. By necessity, though, their data fail to give information on changes occurring during the first 2-3 min of recovery and cannot be used to correlate changes in lactate, labile phosphates, and phi to ion fluxes or phe. Mitochondrial phosphorylation, as this can be studied under optimal conditions in vitro, is affected only moderately by 3 min of hypoglycemic coma (Agardh et ai., 1982). The present results 16: 5 1 D 18:.;: ' 4 ], '" 3 -< 2 1 os Co? ';' e " " u '" C C S.5.5 e e e e U U,. r.a 18:1 IiI1 2:4 22:6 FIG. 3. Changes in the individual free fatty acids (FFAs). Note that the 2-min recovery group is missing. Values are expressed as means ± SO. Control-I-G, controls injected with insulin and glucose; rec, recovery. J Cereb Blood Flow Metab, Vol. 13, No.5, 1993

6 BRAIN RECOVER Y FOLLOWING HYPOGLYCEMIA S 7.4 -ph i - - ph. 7.3 FIG. 4. Changes of intra- and extracellular ph (ph; and phe' respectively) during and after 3 min of hypoglycemic coma. Note that the 2-min recovery (rec) group is missing. 'Significantly higher than the control-i-g group (p <.5, analysis of variance followed by Scheffe's test). Control-I-G, controls injected with insulin and glucose. Values are expressed as means ± SO * 6.8 '". '" '" ' 1: 1: '5.5 's e 's. <:> u,.. demonstrate that the mitochondria can resume phosphorylation also under the less favorable conditions in vivo, e.g., when FF As have accumulated. It is clear from the present results on changes in K + e that also Na +, K + -ATPase activity is quickly resumed. Thus, neither the mitochondrial nor the plasma membrane function suffers serious damage unless the period of hypoglycemic coma is extended to 6 min (Agardh et ai., 1982). These conclusions should be drawn with the caveat that a minority of cells, such as the granule cells in the tip of the dentate gyrus, may develop rapid and extensive mitochondrial failure (Auer et ai., 1984, 1985). We compared the present results with those obtained in an ischemia recovery study (Ekholm et ai., 1993). Although there is no acidosis and no reduction of blood flow during hypoglycemia, the lag before the DC potential and before the K + e started to return to basal level was very similar, suggesting that following both hypoglycemia and ischemia, the normalization of the ATP/ADPf ratio, i.e., the phosphorylation potential, is relatively rapid. In both conditions, the tissue glucose concentration is virtually nil at the end of the insult. Hence, any lag in recovery related to the relatively slow transport of glucose from plasma to tissue is similar in the two conditions. The results demonstrate that neither the reestablishment of capillary circulation nor the lingering intracellular acidosis following ischemia introduces a lag in the normalization of the cellular phosphorylation potential. In both conditions, ATP synthesis occurs promptly after the initial delay, and Na +, K + -ATPase activity is quickly resumed. The present results confirm previous studies showing that although lactate content rises moderately during recovery, phj does not fall (Behar et ai., 1985). The results suggest that phj is rising at the time when phe falls. Such an apparent dissociation between phj and phe has previously been observed following status epilepticus [Siesj6 et al. (1985); see also increase in phj following ischemia described by Mabe et al. (1983)]. It is tempting to speculate that these insults lead to a sustained stimulation of extrusion of H+ from cells, explaining why phj can increase or remain constant when phe falls (Siesj6 et ai., 1993). If confirmed, such results hint that hypoglycemia and other insults could cause sustained alterations of membrane function, e.g., by phosphorylation of the ubiquitous Na + IH+ exchanger. Acknowledgment: This study was financially supported by continuous grants from the Swedish Medical Research Council (no. 14X-263), the U.S. Public Health Service via the NIH (no. 5 ROI NS7838), and the Medical Faculty, Lund University, Lund, Sweden. REFERENCES Agardh C-D, Folbergrova J, Siesj6 BK (1978) Cerebral metabolic changes in profound insulin-induced hypoglycemia and in the recovery period following glucose administration. J Neuroc hem 31:1l Agardh C-D, Chapman AG, Nilsson B, Siesj6 BK (1981) Endogenous substrates utilized by rat brain in severe insulininduced hypoglycemia. J Neurochem 36:49-5 Agardh C-D, Chapman AG, Pelligrino D, Siesj6 BK (1982) Influence of severe hypoglycemia on mitochondrial and plasma membrane function in rat brain. J Neurochem 38: Auer RN, Siesj6 BK (1988) Biological differences between ischemia, hypoglycemia and epilepsy. Ann NeuroI24: J Cereb Blood Flow Metab, Vol. ]3, No. 5, 1993

7 826 K.-I. KATSURA ET AL. Auer RN, Wieloch T, Olsson Y, Siesj6 BK (1984) Distribution of hypoglycemic brain damage. Acta Neuropathol (Ber/) 64: Auer RN, Kalimo H, Olsson Y, Wieloch T (1985) The dentate gyrus in hypoglycemia. Pathology implicating excitotoxin mediated neuronal necrosis. Acta Neuropathol (Ber/) 67: Bazan NG, Squinto SP, Braquet P, Panetta T, Marcheselli VL (1991) Platelet-activating factor and polyunsaturated fatty acids in cerebral ischemia or convulsions: intracellular PAFbinding sites and activation of a FOS/JUN/AP-l transcriptional signaling system. Lipids 26: Behar KL, den Hollander JA, Petroff OAC, Hetherington HP, Prichard JW, Shulman RG (1985) Effect of hypoglycemic encephalopathy upon amino acids, high-energy phosphates, and phi in the rat brain in vivo: detection by sequential IH and 31p NMR spectroscopy. J Neurochem 44: Bengtsson F, Boris-M6l1er F, Hansen AJ, Siesj6 BK (199) Extracellular ph in the rat brain during hypoglycemic coma and recovery. J Cereb Blood Flow Metab 1: Brooks KJ, Bachelard HS (1989) Changes in intracellular free magnesium during hypoglycemia and hypoxia in cerebral tissue as calculated from 31P-nuc1ear magnetic resonance spectra. J Neurochem 53: Ekholm A, Asplund B, Siesj6 BK (1992) Perturbation of cellular energy state in complete ischemia: relationship to dissipative ion fluxes. Exp Brain Res 9: Ekholm A, Katsura K, Kristian T, Liu M, Folbergrova J, Siesj6 BK (1993) Coupling of cellular energy state and ion homeostasis during recovery following brain ischemia in normoglycemic rats. Brain Res 64: Folbergrova J, MacMillan V, Siesj6 BK (I972a) The effect of hypercapnic acidosis upon some glycolytic and Krebs cyc1eassociated intermediates in the rat brain. J Neurochem 19: Folbergrova J, MacMillan V, Siesj6 BK (1972b) The effect of moderate and marked hypercapnia upon the energy state and upon the cytoplasmic NADH/NAD+ ratio of the rat brain. J Neurochem 19: Harris R, Wieloch T, Symon L, Siesj6 BK (1984) Cerebral extracellular calcium activity in severe hypoglycemia: relation to extracellular potassium activity and energy state. J Cereb Blood Flow Metab 4: Kalimo H, Auer RN, Siesj6 BK (1985) The temporal evolution of hypoglycemic brain damage. III. Light and electron microscopic findings in the rat caudoputamen. Acta Neuropathol (Bed) 67: 37-5 Katsura K, Asplund B, Ekholm A, Siesj6 BK (1992) Extra- and intracellular ph in the brain during ischemia, related to tissue lactate content in normo- and hypercapnic rats. Eur J Neurosci 4: Kontos HA (1989) Oxygen radicals in cerebral ischemia. In: Cerebrovascular Diseases (Ginsberg M, Dietrich W, eds), New York, Raven Press, pp Lawson JW, Veech RL (1979) Effects of ph and free Mg 2 + on the Keq of the creatine kinase reaction and other phosphate hydrolyses and phosphate transfer reactions. J Bioi Chem 254: Lewis LD, Ljunggren B, Ratcheson RA, Siesj6 BK (1974) Ce- rebral energy state in insulin-induced hypoglycemia, related to blood glucose and to EEG. J Neurochem 23: Lindsberg PJ, Hallenbeck JM, Feuerstein G (1991) Plateletactivating factor in stroke and brain injury. Ann Neurol 3: Mabe H, Blomqvist P, Siesj6 BK (1983) Intercellular ph in the brain following transient ischemia. J Cereb Blood Flow Metab 3: Mutch WA, Hansen AJ (1984) Extracellular ph changes during spreading depression and cerebral ischemia: mechanisms of brain ph regulation. J Cereb Blood Flow Metab 4: Pelligrino D, Almqvist L-O, Siesj6 BK (1981) Effects of insulininduced hypoglycemia on intracellular ph and impedance in the cerebral cortex of the rat. Brain Res 221: Pelligrino D, Yokoyama H, Ingvar M, Siesj6 BK (1982) Moderate arterial hypotension reduces cerebral cortical blood flow and enhances cellular release of potassium in severe hypoglycemia. Acta Physiol Scand 115: Ponten U, Siesj6 BK (1964) A method for the determination of the total carbon dioxide content of frozen tissues. Acta Physiol Scand 6: Ponten U, Siesj6 BK (1966) Gradients of CO2 tension in the brain. Acta Physiol Scand 67: Ponten U, Ratcheson RA, Salford LG, Siesj6 BK (1973) Optimal freezing conditions for cerebral metabolites in rats. J Neurochem 21: Siesj6 BK (1981) Cell damage in the brain: a speculative synthesis. J Cereb Blood Flow Metab 1: Siesj6 BK (1984) Cerebral circulation and metabolism. J Neurosurg 6: Siesj6 BK (1988) Hypoglycemia, brain metabolism, and brain damage. Diabetes Metab Rev 4: Siesj6 BK (1991) The role of calcium in cell death. In: Neurodoegenerative Disorders: Mechanisms and Prospects for Therapy (Price D, Aguayo A, Thoenen H, eds). Chichester, Wiley, pp Siesj6 BK (l992a) Pathophysiology and treatment of focal cerebral ischemia. l. Pathophysiology. J Neurosurg 77: Siesj6 BK (l992b) Pathophysiology and treatment of focal cerebral ischemia. II. Mechanisms of damage and treatment. J Neurosurg 77: Siesj6 BK, Deshpande JK (1987) Electrolyte shifts between brain and plasma in hypoglycemic coma. J Cereb Blood Flow Metab 7: Siesj6 BK, Folbergrova J, MacMillan V (1972) The effect of hypercapnia upon intracellular ph in the brain, evaluated by the bicarbonate-carbonic acid method and from the creatine phosphokinase equilibrium. J Neurochem 19: Siesj6 BK, von Hanwehr R, Nergelius G, Nevander G, Ingvar M (1985) Extra- and intracellular ph in the brain during seizures and in the recovery period following the arrest of seizure activity. J Cereb Blood Flow Metab 5:47-57 Siesj6 BK, Agardh C-D, Bengtsson F (1989) Free radicals and brain damage. Cerebrovasc Brain Metab Rev 1: Siesj6 BK, Katsura K, Mellergiird P, Ekholm A, Lundgren J, Smith M-L (1993) Acidosis-related brain damage. Prog Brain Res 96:23-48 Watson BD, Ginsberg MD (1989). Ischemic injury in brain: role of oxygen radical-mediated processes. Ann NY Acad Sci 559: J Cereb Blood Flow Me/ab, Vol. 13, No. 5, 1993