Journal of Neuroscience Research 57:255–260 (1999)

Astrocytes Respond to Hypoxia by Increasing Glycolytic Capacity Husnia Marrif1,3 and Bernhard H.J. Juurlink2,3* 1Department

of Pharmacology, College of Medicine, University of Saskatchewan, Saskatoon, SK, Canada of Anatomy and Cell Biology, College of Medicine, University of Saskatchewan, Saskatoon, SK, Canada 3Saskatchewan Stroke Research Centre, College of Medicine, University of Saskatchewan, Saskatoon, SK, Canada 2Department

Astrocytes cope more readily with hypoxic insults than do neurons. We hypothesized that astrocytes can upregulate their glycolytic capacity, allowing anaerobic glycolysis to provide sufficient ATP for cell survival as well as for carrying out critical functions such as taking up glutamate. To test this hypothesis, astrocytes were subjected to hypoxia for 5 hr. Lactate dehydrogenase (LDH) and pyruvate kinase activities increased 3- to 4-fold. Examination of LDH isoenzyme patterns determined that it was the anaerobic isoenzymes that were upregulated. To determine whether increase in enzyme activity translates into increased glycolytic capacity, astrocytes were subjected to varying time periods of hypoxia, and glucose uptake was measured under conditions where astrocytes were forced to consume more ATP. This demonstrated that 8 hr of hypoxia resulted in a doubling of glycolytic capacity. We suggest that how quickly astrocytes upregulate glycolytic capacity may determine whether or not neurons within the stroke penumbra survive. J. Neurosci Res. 57:255–260, 1999. r 1999 Wiley-Liss, Inc. Key words: glia; glycolysis; hypoxia; ischemia; lactate dehydrogenase INTRODUCTION Within the central nervous system, astrocytes are the cells that are more capable of withstanding hypoxic insults than are neurons and oligodendrocytes. This has been most elegantly demonstrated by research done on cultured cells (Yu et al., 1989; Murphy and Horrocks, 1993; Sochocka et al., 1994; Husain and Juurlink, 1995; Lyons and Kettenmann, 1998; Swanson et al., 1997). Astrocytes seem to survive hypoxia as long as there is a supply of glucose for anaerobic glycolysis (Callahan et al., 1990; Kelleher et al., 1993), provided that pH does not drop below 6.5 due to lactacidosis (Swanson and Benington, 1996; Swanson et al., 1997). Although both neurons and oligodendroglia can perform glycolysis, even being in the presence of adequate glucose does not ensure the

r 1999 Wiley-Liss, Inc.

survival of neurons nor oligodendrocytes when subjected to hypoxia (Lyons and Kettenmann, 1998; Murphy and Horrocks, 1993). There are a number of well-defined responses that many cells have to chronic hypoxia. These include downregulation of mechanisms that consume ATP such as inhibition of protein synthesis, protein breakdown, gluconeogenesis, and Na⫹ pump activity (Hochachka et al., 1996). Astrocytes can certainly downregulate their requirements for ATP. We have previously demonstrated that astrocytic ATP demands are decreased 10-fold when astrocytes are deprived of both oxygen and glucose (Yager et al., 1994). We have also demonstrated that astrocytes also downregulate protein synthesis as evidenced by disruption of polyribosomes (Petito et al., 1991). Another response to hypoxia by many cells is an increase in glycolytic capacity (Bunn and Poyton, 1996; Hochachka et al., 1996). This increased ability to perform glycolysis is due to coordinated upregulation of the enzymes of glycolysis (Hance et al., 1980; Ptashne et al., 1985; Robin et al., 1984). Specific hypoxia-inducible anaerobic isoforms of the glycolytic enzymes are upregulated (Ebert et al., 1996). This upregulation is due, in part, to increased transcription (Webster et al., 1990) which is coordinated by activation of the transcription factor, hypoxia-inducible factor-1 (HIF-1; Firth et al., 1994, 1995; Semenza et al., 1994, 1996). To date, no studies on the ability of neural cells to upregulate their capacity to perform anaerobic glycolysis have been performed. This article describes the results of our experiments where we tested the hypothesis that one of astrocytic responses to hypoxia is an upregulation of Contract grant sponsor: Heart and Stroke Foundation of Saskatchewan. *Correspondence to: Bernhard H.J. Juurlink, Department of Anatomy and Cell Biology, College of Medicine, University of Saskatchewan, 107 Wiggins Road, Saskatoon, SK S7N 5E5, Canada. E-mail: [email protected] Received 11 February 1999; Accepted 14 April 1999

256

Marrif and Juurlink

the capacity to perform glycolysis by an increase in the glycolytic enzyme activity. MATERIALS AND METHODS Cell Culture Preparation Cultures highly enriched in astrocytes were prepared from newborn rat neopallia as previously described (Juurlink and Hertz, 1992) except that following cell attachment, the growth medium was replaced with the astrocyte selection medium described by Wiesinger et al. (1991) that consists of Dulbecco’s modified Eagle Medium (DMEM), where sorbitol has replaced glucose and dialyzed horse serum was substituted for horse serum. The cultures were fed with this medium for 2 weeks and thereafter with regular growth medium containing DMEM with glucose and nondialyzed horse serum. Cultures were used at 3–4 weeks of culture. Experimental Studies Establishment of severe hypoxia. Cultures were placed into an anaerobic chamber equipped with an air lock (Forma Scientific Inc.). Within this chamber was a humidified incubator maintained at 37°C. The atmosphere of the anaerobic chamber was 5% CO2, 93% nitrogen. The balance of the atmosphere was hydrogen that in the presence of a catalyst converted the residual oxygen to water. Serum-free medium was made anoxic by gassing with 5% CO2 in nitrogen for 30 min. For the hypoxia experiments, cultures were washed with anoxic medium inside the anaerobic chamber and then fed with anoxic medium. Enzyme activities. Cellular lactate dehydrogenase activity was measured according to Wahlfeld (1983), as we have done previously (Juurlink, 1993). In this method, excess lactic acid is converted to pyruvate with the concomitant production of NADH from NAD. Lactate dehydrogenase activity is expressed as µmoles NADH formed per min per mg protein. Pyruvate kinase activity was measured according to Simon et al. (1977). In this procedure, the rate of pyruvic acid formed from phosphoenolpyruvate was determined by coupling this reaction to the conversion of pyruvate plus NADH to lactate plus NAD by the enzyme lactate dehydrogenase. Pyruvate kinase activity is expressed as units of NADH consumed per min per mg protein. Lactate dehydrogenase isozyme determination. Astrocyte cultures were harvested in 50 mM Tris (pH 7.4) buffer, and 50 µg protein was added to wells in a 1% agarose gel prepared in barbitone buffer and electrophoresed and detected as described by Nissen and Schousboe (1979). 2-Deoxyglucose uptake. For determination of glucose uptake, an adaption of the 2-deoxyglucose uptake

method of Sokoloff et al. (1977) was used as previously described (Peng et al., 1996). Cultures were washed twice with serum-free DMEM and then incubated in serum-free DMEM containing the usual 7.5 mM glucose as well as 2-[1-14C]-deoxyglucose (0.15 µCi/mL) and incubated either under the usual normoxic culture conditions or under anoxic culture conditions. Isotope was obtained from NEN, Du Pont Canada Inc., Mississauga, ON, Canada. After 60 min of incubation in the presence of 2-deoxyglucose, the cultures were washed twice with ice-cold medium and cells digested with 1 M sodium hydroxide. The radioactivity in the digest was measured in a scintillation counter and cell protein calculated using the conventional Lowry technique (Lowry et al., 1951). Rates of glucose metabolism were calculated from the rates of 2-deoxyglucose accumulation multiplied by a factor of 4.55, since glucose phosphorylation proceeds 4.55 times faster than phosphorylation of deoxyglucose (Dienel et al., 1991). To examine whether prior exposure to anoxia increased the ability of cells to take up glucose, some cultures were exposed for 0, 1, 2, 4, 6, or 8 hr to anoxia followed by measuring glucose uptake under anoxic conditions. A cell’s capacity to perform glycolysis may exceed the actual rate of glycolysis and we wished to obtain some measure of the maximal capacity to perform glycolysis. Since glutamate is taken up into cells via a sodium-dependent uptake mechanism resulting in activation of Na⫹-K⫹ ATPase and consumption of ATP, which in turn stimulates glycolysis (Pellerin and Magistretti, 1994), in half of the cultures we examined glucose uptake in the presence of 1.0 mM glutamate in the culture medium. Experiments were performed on three different culture batches. Total dishes analyzed per experimental point were never less than 7. Data are expressed as means ⫾ S.E.M. Statistical analysis of the data was performed with the program Instat and specific tests used are indicated in the figure legends.

RESULTS AND DISCUSSION A 5-hr exposure to anoxia almost triples astrocyte lactate dehydrogenase (LDH) activity (Fig. 1). Since enzymes of glycolysis tend to be coordinately regulated (Hance et al., 1980), the relative capacity to perform glycolysis can be obtained by measuring any of such enzymes. However, we also examined pyruvate kinase activity to verify that more than one glycolytic enzyme was upregulated. As can be seen in Figure 1, a 5-hr exposure to hypoxia more than quadrupled pyruvate kinase activity.

Hypoxic Regulation of Glycolysis in Astroglia

Fig. 1. Lactate dehydrogenase (LDH) and pyruvate kinase activities (mean ⫾ S.E.M.) in astrocyte cultures grown under the usual normoxic conditions or under 5 hr of anoxia. A one-tailed Welch’s t-test demonstrates that the normoxic astrocyte (n ⫽ 15) LDH activity is significantly less (P ⬍ 0.01) than the anoxic astrocyte (n ⫽ 13). Using a one-tailed Welch’s t-test, pyruvate kinase activity was determined to be significantly greater (P ⬍ 0.001) in the cells exposed for 5 hr of anoxia (n ⫽ 10) as compared to control normoxic conditions (n ⫽ 8).

Lactate dehydrogenase is a tetramer that can be comprised of aerobic (B) and anaerobic (A) isoforms. The following tetramer combinations can be formed (Cahn et al., 1962): B4 (LDH 1), B3A1 (LDH 2), B2A2 (LDH 3), B1A3 (LDH 4), and A4 (LDH 5). The activity of LDH 1 is strongly inhibited by low concentrations of pyruvic acid, whereas that of LDH 5 is not (Cahn et al., 1962; Dawson et al., 1964); hence, it is thought that the higher the proportion of the A isoform in the tetramer, the more capable the isoenzyme is of acting under conditions where pyruvate is not rapidly oxidized and thus allowing lactic acid formation. Although the physiological significance of the aerobic and anaerobic isoforms is still not entirely clear, it has been known for a number of years that astrocytes have a high rate of glycolysis and that much of the generated pyruvate is converted to lactate (Hamprecht et al., 1993; Reichenbach et al., 1993; Walz and Mukerji, 1988) which is exported, most likely as a metabolic fuel for neighboring cells such as neurons (Robinson et al., 1998). In the astrocyte cultures, the predominant LDH isoenzymes under normoxic conditions are LDH 5 and 4, with 5 migrating towards the cathodal end of the gel and 4 towards the anodal end (Fig. 2). This is a similar pattern to that seen previously by Nissen and Schousboe (1979)

257

in mouse astrocyte cultures and is in keeping with the aforementioned ability of astrocytes to metabolize glucose to lactate. Exposure of astrocytes to anoxia upregulates the anaerobic isoform, causing an increase in LDH 5, LDH 4, and the appearance of a prominent LDH 3 band. What accounts for the anoxia-induced increases in glycolytic enzyme activity of astrocytes? One possibility is post-translational modifications of enzymes. For example, it is known that phosphorylation of pyruvate kinase inhibits enzyme activity (Oude Weernink et al., 1992). Little is known regarding post-translational modifications of other enzymes of glycolysis. Other possibilities include having more mRNA available for translation through enhanced transcription such as that mediated by HIF-1 (Semenza, 1996) or by post-transcriptional changes that involve increasing the stability of the mRNA (Jungmann et al., 1998). The question we posed was: does the increased glycolytic enzyme activity actually translate into an increased ability to perform glycolysis? To test this, we examined glucose uptake after varying times of culture exposure to anoxia and, as well, when astrocytes were forced to consume extra ATP because of the addition of glutamate to the medium during the 1-hr period when glucose uptake was being measured. Glutamate uptake is sodium-dependent (Worrall and Williams, 1994) and, hence, will stimulate the Na⫹ pump, thus consuming ATP. Figure 3 illustrates glucose utilization during a 1-hr period of anoxia of astrocyte cultures that had either no prior exposure to anoxia or varying periods of prior exposure to anoxia. It can be seen that, in the absence of exogenous glutamate, glucose uptake in the absence of a prior exposure to anoxia had a baseline of 15 nmoles glucose taken up/min/mg cell protein. Glucose uptake did not increase until the astrocytes had been exposed to 4 hr of prior anoxia, and then over the next 4 hr glucose uptake returned to baseline levels. It is noteworthy that only during the last hour of anoxia that glucose uptake is measured and that it is only during this time period that glutamate is added to the culture medium. A major regulator of the rate of glycolysis is the consumption of ATP; thus, when the ADP/ATP ratio increases, glycolytic rate increases (Scrutton and Utter, 1968). Since glucose measurement was carried out under conditions of anoxia, the increase in glucose uptake must reflect increases in ATP consumption that is driving increases in glycolysis. This suggests that there is a decline in cellular ATP between these 2- and 4-hr periods. Our previous results examining the decline in ATP and phosphocreatine during withdrawal of both oxygen and substrate in astrocytes demonstrated that although there was a decline in ATP during the first 3 hr of simulated ischemia, this was

258

Marrif and Juurlink

Fig. 2. Representative gel demonstrating LDH isoenzyme activities in astrocytes subjected to normoxia (lane N) or 5 hr of anoxia (lane A). Cathodal end is at the bottom of the figure.

modest due to a conversion of phosphocreatine to ATP during this time period (Yager et al., 1994). It is likely that in the present experiment that there is little or no decline in ATP over the first 2 hr and that ATP levels only begin to decline between 2 and 4 hr when phosphocreatine has declined substantially, thereby increasing the rate of glycolysis. We have also previously shown that it is during this time period that ATP-consuming mechanisms such as protein synthesis essentially stop (Petito et al., 1991); this decline in ATP-consuming mechanisms perhaps allows the rate of glycolysis to return to baseline levels, as seen in Figure 3. With no prior exposure to anoxia, the addition of glutamate to the medium resulted in a glucose uptake of 18.7 nmoles/min/mg protein (Fig. 3). This is significantly higher (P ⬍ 0.05, Student’s t-test) than the 14.9 nmoles/ min/mg protein in the absence of exogenous glutamate. This must be due to the increased ATP consumption due to glutamate uptake. There is a steady increase in glucose uptake, during the 1-hr measurement period, the longer the cultures are previously exposed to anoxia with a noticeable dip at the 4-hr prior anoxia exposure period. Since ATP demand due to the uptake of glutamate should be the same during this entire period, we interpret this to indicate that the ability to perform glycolysis increases the longer the cells are exposed to a prior period of anoxia. This is in keeping with the increase in glycolytic enzyme activity following exposure to anoxia. What may account for the dip in glucose uptake seen after 4-hr prior exposure to anoxia? It may be that this is related to a downregulation of processes that consume ATP, such as protein synthesis, at this time. Can an hypoxia-induced upregulation of glycolytic capacity be of significance to astrocyte function? It may well be. One important function of astrocytes is the uptake of the potential excitotoxin, glutamate (Rothstein et al., 1996), released by neurons. Under conditions of hypoxemia, the ability of astrocytes to upregulate the

Fig. 3. Glucose uptake (mean ⫾ S.E.M.) in astrocyte cultures exposed to 0, 1, 2, 4, 6, or 8 hr of anoxia prior to measuring glucose uptake. Glucose uptake was measured during a 1-hr period under anoxic conditions; during this measurement, glutamate was added to the medium of half the cultures to a final concentration of 1 mM. An analysis of variance (ANOVA) with a post-hoc Tukey-Kramer test was used for statistical analysis. In the absence of glutamate during the glucose uptake measurement, the 4-hr prior exposure to anoxia value is significantly higher than the 0-hr (P ⬍ 0.01) and the 8-hr (P ⬍ 0.05) value. When glucose uptake is measured in the presence of glutamate, the values from 2-, 6-, and 8-hr prior exposure to anoxia onward, the values are significantly higher (P ⬍ 0.001) than the value with no prior exposure to anoxia. Each analysis point represents the mean of 7–8 cultures.

capacity to perform glycolysis might be the deciding factor in determining whether neurons survive. One such area that experiences hypoxemia is the penumbral region of a stroke lesion where blood flow is not only reduced by 50% or more (Ginsberg and Pulsinelli, 1994; Hakim, 1987) but is characterized by the presence of spreading depression waves (Walz, 1997), allowing massive release of glutamate. Indeed, interfering with the production of ATP by astrocytes results in pathophysiological changes similar to that seen in the ischemic penumbra, leading to the thesis that astrocyte dysfunction is the ultimate cause of neuronal death in the stroke penumbra (Largo et al., 1996). If one considers that a given volume of blood has five times as much glucose in it as there is oxygen for the oxidative metabolism of glucose (Huang et al., 1997), when astrocytes experience hypoxia or hypoxemia, the ability to upregulate the capacity to perform glycolysis might be the deciding factor determining whether astrocytes can perform their neuroprotective functions. It is

Hypoxic Regulation of Glycolysis in Astroglia

clear that under conditions of anoxia, astrocytes still retain their capacity to take up glutamate (Huang et al., 1993, 1997; Swanson, 1992). With a medium glutamate concentration of 0.25 mM, the uptake rate is 40 nmoles/ min/mg protein (Huang et al., 1993). Since glutamate is taken up with two Na⫹ ions, this will ultimately result in the consumption of 40 nmoles ATP/min/mg culture protein. Astrocyte cultures contain approximately 40 nmoles ATP/mg cell protein (Yager et al., 1994). As determined by both medium glucose disappearance (Devon and Juurlink, 1989) and 2-deoxyglucose uptake (Pellerin and Magistretti, 1994; Peng et al., 1996) in astrocyte cultures, the normal rate of glycolysis can be no more than ⬃15 nmoles glucose utilized/min/mg protein; it must be less since glucose taken up is also used in the pentose phosphate shunt. If one assumes that under conditions of anoxia all glucose taken up into the astrocyte is used glycolytically, then there is only about 30 nmoles of ATP being generated/min/mg cell protein; this ATP must be used not only for Na⫹ activity but also other vital cellular processes. It would appear that an increase in the astrocyte glycolytic capacity might be critical to the survival of the tissue. However, an increased capacity to perform glycolysis would be neuronoprotective only if blood flow is sufficient to dissipate the lactic acid that would be produced by the increased glycolytic rate, since severe lactacidosis can result in complete infarction of the tissue (Plum, 1983). In conclusion, upon exposure to hypoxia, astrocytes can double their capacity to perform glycolysis. How rapidly penumbral astrocytes can increase this capacity to perform glycolysis may well decide whether the adjacent neurons will survive.

ACKNOWLEDGMENTS We thank Connie Wong for her technical assistance.

REFERENCES Bunn HF, Poyton RO. 1996. Oxygen sensing and molecular adaptation to hypoxia. Physiol Rev 76:839–885. Cahn RD, Kaplan NO, Zwilling LL. 1962. Nature and development of lactic dehydrogenases. Science 136:962–969. Callahan DJ, Engle MJ, Volpe JJ. 1990. Hypoxic injury to developing glial cells: protective effect of high glucose. Pediatr Res 27:186–190. Dawson DM, Goodfriend TL, Kaplan NO. 1964. Lactate dehydrogenase: functions of the two types. Science 143:929–933. Devon RM, Juurlink BHJ. 1989. Dynamic morphological responses of mouse astrocytes in primary cultures following medium changes. Glia 2:266–272. Dienel GA, Cruz NF, Mori K, Holden JE, Sokoloff L. 1991. Direct measurement of the ␭ of the lumped constant of the deoxyglucose method in rat brain: determination of ␭ and lumped constant from tissue glucose concentration or equilibrium

259

brain/plasma distribution ratio for methylglucose. J Cereb Blood Flow Metab 11:25–34. Ebert BL, Gleadle JM, O’Rourke JF, Bartlett SM, Poulton J, Ratcliffe PJ. 1996. Isoenzyme-specific regulation of genes involved in energy metabolism by hypoxia: similarities with the regulation of erythropoietin. Biochem J 313:809–814. Firth JD, Ebert BL, Pugh CW, Ratcliffe PJ. 1994. Oxygen-regulated control elements in the phosphoglycerate kinase 1 and lactate dehydrogenase A genes: similarities with the erythropoietin 38 enhancer. Proc Natl Acad Sci USA 91:6496–6500. Firth JD, Ebert BL, Ratcliffe PJ. 1995. Hypoxic regulation of lactate dehydrogenase A—interaction between hypoxia-inducible factor 1 and cAMP response elements. J Biol Chem 270:21021– 21027. Ginsberg MD, Pulsinelli WA. 1994. The ischemic penumbra, injury thresholds, and the therapeutic window for acute stroke. Ann Neurol 36:553–554. Hakim AM. 1987. The cerebral ischemic penumbra. Can J Neurol Sci 557–559. Hamprecht B, Dringen R, Pfeiffer B, Kurz G. 1993. The possible roles of astrocytes in energy metabolism of the brain. In: Fedoroff S, Juurlink BHJ, Doucette R, editors. Biology and pathology of astrocyte-neuron interactions. New York: Plenum Press. p 83–91. Hance AJ, Robin ED, Simon LM, Alexander A. 1980. Regulation of glycolytic enzymes activity during chronic hypoxia by changes in rate-limiting enzyme content. J Clin Invest 66:1258–1264. Hochachka PW, Buck LT, Doll CJ, Land SC. 1996. Unifying theory of hypoxia tolerance: molecular/metabolic defense and rescue mechanisms for surviving oxygen lack. Proc Natl Acad Sci USA 93:9493–9498. Huang R, Shuaib A, Hertz L. 1993. Glutamate uptake and glutamate content in primary cultures of mouse astrocytes during anoxia, substrate deprivation and simulated ischemia under normothermic and hypothermic conditions. Brain Res 618:346–351. Huang R, Sochocka E, Hertz L. 1997. Cell culture studies of the role of elevated extracellular glutamate and K⫹ in neuronal cell death during and after anoxia/ischemia. Neurosci Biobehav Rev 21:129–134. Husain J, Juurlink BHJ. 1995. Oligodendroglial precursor cell susceptibility to hypoxia is related to poor ability to cope with reactive oxygen species. Brain Res 698:86–94. Jungmann RA, Huang D, Tian D. 1998. Regulation of LDH-A gene expression by transcriptional and posttranscriptional signal transduction mechanisms. J Exp Zool 282:188–195. Juurlink BHJ. 1993. NADP-linked isozymes are the major forms of isocitrate dehydrogenase in mouse type-1-like astrocytes. Life Sci 52:1087–1090. Juurlink BHJ, Hertz L. 1992. Astrocytes. In: Boulton AA, Baker GB, Walz W, editors. Practical cell culture techniques. Totowa: Humana Press. p. 269–321. Kelleher JA, Chan PH, Chan TY, Gregory GA. 1993. Modification of hypoxia-induced injury in cultured rat astrocytes by high levels of glucose. Stroke 24:855–863. Largo C, Cuevas P, Herreras O. 1996. Is glia disfunction the initial cause of neuronal death in ischemic penumbra? Neurol Res 18:445–448. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275. Lyons SA, Kettenmann H. 1998. Oligodendrocytes and microglia are selectively vulnerable to combined hypoxia and hypoglycemia injury in vitro. J Cereb Blood Flow Metab 18:521–530.

260

Marrif and Juurlink

Murphy EJ, Horrocks LA. 1993. Mechanisms of hypoxic and ischemic injury. Use of cell culture models. Mol Chem Neuropathol 19:95–106. Nissen C, Schousboe A. 1979. Activity and isoenzyme pattern of lactate dehydrogenase in astroblasts cultured from brains of newborn mice. J Neurochem 32:1787–1792. Oude Weernink PA, Rijksen G, Mascini EM, Staal GEJ. 1992. Phosphorylation of pyruvate kinase type K is restricted to the dimeric form. Biochim Biophys Acta 1121:61–68. Pellerin L, Magistretti PJ. 1994. Glutamate uptake into astrocytes stimulates aerobic glycolysis: a mechanism coupling neuronal activity to glucose utilization. Proc Natl Acad Sci USA 91:10625– 10629. Peng L, Juurlink BHJ, Hertz L. 1996. Pharmacological and developmental evidence that potassium-induced stimulation of deoxyglucose uptake in astrocytes is a metabolic manifestation of increase Na⫹-K⫹-ATPase activity. Dev Neurosci 18:353–359. Petito CK, Juurlink BH, Hertz L. 1991. In vitro models differentiating between direct and indirect effects of ischemia on astrocytes. Exp Neurol 113:364–372. Plum F. 1983. What causes infarction in ischemic brain? the Robert Wartenberg Lecture. Neurology 33:222–233. Ptashne KA, Morin ME, Hance A, Robin ED. 1985. Increased biosynthesis of pyruvate kinase under hypoxic conditions in mammalian cells. Biochim Biophys Acta 844:19–23. Reichenbach A, Stolzenburg JU, Eberhardt W, Chao TI, Dettmer D, Hertz L. 1993. What do retinal muller (glial) cells do for their neuronal small siblings. J Chem Neuroanat 6:201–213. Robin ED, Murphy BJ, Theodore J. 1984. Coordinate regulation of glycolysis by hypoxia in mammalian cells. J Cell Physiol 118:287–290. Robinson SR, Scousboe A, Dringen R, Magestretti P, Coles J, Hertz L. 1998. Metabolic trafficking between neurons and glia. In: Laming PR, Sykova´ E, Reichenbach A, Hatton GI, Bauer H, editors. Glial cells: their role in behaviour. Cambridge: Cambridge University Press. p 83–106. Rothstein JD, Dykeshoberg M, Pardo CA, Bristol LA, Jin L, Kuncl RW, Kanai Y, Hediger MA, Wang YF, Schielke JP, Welty DF. 1996. Knockout of glutamate transporters reveals a major role for astroglial transport in excitotoxicity and clearance of glutamate. Neuron 16:675–686. Scrutton MC, Utter MF. 1968. The regulation of glycolysis and gluconeogenesis in animal tissues. Ann Rev Biochem 37:249– 302. Semenza GL. 1996. Transcriptional regulation by hypoxia-inducible factor 1—molecular mechanisms of oxygen homeostasis. Trend Cardiovasc Med 6:151–157. Semenza GL, Roth PH, Fang HM, Wang GL. 1994. Transcriptional regulation of genes encoding glycolytic enzymes by hypoxiainducible factor 1. J Biol Chem 269:23757–23763. Semenza GL, Jiang B-H, Leung SW, Passantino R, Concorde J-P, Maire P, Giallongo A. 1996. Hypoxia response elements in the

aldolase A, enolase 1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia-inducible factor 1. J Biol Chem 271:32529–32537. Simon LM, Robin ED, Philips JR, Acevado J, Axline SG, Theodore J. 1977. Enzymatic basis for bioenergetic differences of alveolar versus peritoneal macrophages and enzyme regulation by molecular O2. J Clin Invest 59:443–448. Sochocka E, Juurlink BH, Code WE, Hertz V, Peng L, Hertz L. 1994. Cell death in primary cultures of mouse neurons and astrocytes during exposure to and ‘recovery’ from hypoxia, substrate deprivation and simulated ischemia. Brain Res 638:21–28. Sokoloff L, Reivich M, Kennedy C, des Rosiers M, Patlak CS, Pettigrew KD, Sakurada O, Shinohara M. 1977. The [14C]deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure, and normal values in the conscious and anesthetized albino rat. J Neurochem 28:897–916. Swanson RA. 1992. Astrocyte glutamate uptake during chemical hypoxia in vitro. Neurosci Lett 147:143–146. Swanson RA, Benington JH. 1996. Astrocyte glucose metabolism under normal and pathological conditions in vitro. Dev Neurosci 18:515–521. Swanson RA, Farrell K, Stein BA. 1997. Astrocyte energetics, function, and death under conditions of incomplete ischemia: a mechanism of glial death in the penumbra. Glia 21:142–153. Wahlfeld AW. 1983. UV-method with L-lactate and NAD. In: Bergmeyer HU, Bergmeyer J, Grabl M, editors. Methods of enzymatic analysis. Vol. III. Enzymes 1: oxyreductases. Deerfield, FL: Verlag Chemie. p 126–133. Walz W. 1997. Role of astrocytes in the spreading depression signal between the ischemic core and penumbra. Neurosci Biobehav Rev 21:135–142. Walz W, Mukerji S. 1988. Lactate production and release in cultured astrocytes. Neurosci Lett 86:296–300. Webster KA, Gunning P, Hardeman E, Wallace DC, Kedes L. 1990. Coordinate reciprocal trends in glycolytic and mitochondrial transcript accumulations during in vitro differentiation of human myoblasts. J Cell Physiol 142:566–573. Wiesinger H, Schuricht B, Hamprecht B. 1991. Replacement of glucose with sorbitol in growth medium causes selection of astroglial cells from heterogenous primary cultures derived from newborn mouse brain. Brain Res 550:69–76. Worrall DM, Williams DC. 1994. Sodium ion dependent transporters for neurotransmitters—a review of recent developments. Biochem J 297:425–436. Yager JY, Kala G, Hertz L, Juurlink BH. 1994. Correlation between content of high-energy phosphates and hypoxic-ischemic damage in immature and mature astrocytes. Brain Res Dev Brain Res 82:62–68. Yu AC, Gregory GA, Chan PH. 1989. Hypoxia-induced dysfunctions and injury of astrocytes in primary cell cultures. J Cereb Blood Flow Metab 9:20–28.