Glucose deprivation depolarizes plasma membrane of cultured astrocytes and collapses transmembrane potassium and glutamate gradients

Neuroscience Vol. 26, No.

I, pp. 283-289,

1988

0306-4522/88

$3.00 + 0.00

Pergamon Press plc 0 1988 IBRO

Printed in Great Britain

GLUCOSE DEPRIVATION DEPOLARIZES PLASMA MEMBRANE OF CULTURED ASTROCYTES AND COLLAPSES TRANSMEMBRANE POTASSIUM AND GLUTAMATE GRADIENTS R. A. KAUPPINEN,*$ K. ENKVIST,~ I. HOLOPAINEN~$ and K. E. 0. WKERMANt *Department of Clinical Neurophysiology, University Central Hospital of Kuopio, SF-70210 Kuopio, Finland; YDepartment of Biochemistry and Pharmacy, Academy of Abe, SF-20500 Turku, Finland; and JInstitute of Biomedical Sciences, University of Tampere, SF-33101 Tampere, Finland Abstract-Primary cultures of astrocytes were used to investigate the effects of glucose deprivation on plasma membrane potential, on the respiration and on the eaergy status of these cells. Plasma membrane potential, as monitored with a cyanine dye, 3,3’-diethylthiadicarbocyanine, hyperpolarized by about 100% when glucose was added to substrate-deprived cells. The effect of glucose was prevented by iodoacetate or ouabain. In the absence of glucose, cellular adenosine triphosphate/adenosine diphosphate ratio was extensively reduced and pyruvate was unable either to restore energy status or to hyperpolarize the plasma membrane of astrocytes, although it was the preferential substrate for mitochondria within the cells. Glucose deprivation and inhibition of glycolysis or respiration in the presence of glucose caused dramatic decrease in transmembrane potassium ion and L-glutamate gradients. The gradients were not restored in the presence of pyruvate. Thus, aerobic glycolysis, rather than oxidation of pyruvate, is required to maintain maximal plasma membrane potential, adenosine triphosphate/adenosine diphosphate ratios as well as K+ and L-glutamate gradients. This evidence, together with the unresponsiveness of astrocyte respiration to ouabain, indicates a functional dissociation between energy dissipation at the plasma membrane and mitochondrial synthesis of adenosine triphosphate. The results are discussed with regard to the vulnerability of glia at low levels of blood glucose and the contribution of glial dysfunction to development of hypoglycaemic encephalopathy.

Despite numerous studies on hypoglycaemic encephthe mechanisms underlying neuronal alopathy, ‘,5.30,44 vulnerability are still not understood. Recently, microdialysis of the extracellular space in the rat hippocampus and striatum’ demonstrated elevation of the excitatory amino acid neurotransmitters, glutamate and aspartate, in insulin-induced hypoglycaemia. These changes take place concomitantly as cortical electroencephalogram (EEG)-activity levels off.’ This is accompanied by a steep drop in the cellular concentration of adenosine triphosphate (ATP).‘.” An application of a glutamate N-methylD-aspartate (NMDA)-receptor antagonist, 2-amino7-phosphoheptanoic acid, into the caudate nucleus of rats significantly prevented loss of neurons.43 These results would support the operation of an excitotoxic

§To whom correspondence should be addressed. Present address: Department of Physics in Relation to Surgery, The Royal College of Surgeons of England, 35-43 Lincoln’s Inn Fields, London WC2A 3PN, U.K. Abbreviations: ADP, adenosine diphosphate; ATP, adenosine triphosphate; Cl-CCP, carbonylcyanide-mchlorophenylhydraione; DiS-C,-(S), 3,3’-diethylthiadicarbocyanine: EDTA, ethvlenediamine tetra-acetic acid; EEG, &ctroen&phal&graphy; NMDA, N-methyl-Daspartate; PBS, phosphate-buffered saline; TES, 2(2-hydroxy-l,l-bis(hydroxymethyl)ethyl amino) ethanesulphonate.

mechanism in the development of hypoglycaemic cell damage as recently reviewed for neuronal cell death of various aetiologies.” The origin of increased extracellular glutamate in the hypoglycaemic brain is not known. Nerve large terminalsz4 and glia” contain two separate, pools of glutamate. Thus, the question arises as to which of these structures is more vulnerable to lack of glucose. We recently showed that isolated nerve i.e. synaptosomes, can oxidize nonterminals, glycolytic endogenous substrates in the absence of external glucose and maintain submaximal ATP adenosine diphosphate (ADP) ratios to support a plasma membrane potential at nearly physiological conditions.“,” Much less is known about the effects of experimental hypoglycaemia on the glial cells, which have an important role in maintaining extracellular homeostasis of inorganic ions and neurotransmitters.‘2,29,38The aim of this study was to determine the responses of plasma membrane potential, K+ and L-glutamate gradients as well as energy status and respiration using cultured mature rat astrocytes to glucose deprivation. EXPERIMENTAL PROCEDURES Cell culture and incubations Cultures of primary astrocytes were prepared by Holopainen.

283

I3 Briefly , cerebral

hemispheres

as described from new-

284

R. A.

KAUPPINM

born Wistar rats (3-6 h old) were removed, carefully freed from the meninges and mechanically dissociated into culture medium (Dulbecco’s minimum essential medium, Gibco Biocult Lab. Ltd, Scotland), which was supplemented with 10% heat-inactivated serum from fetal calf, 2 mM L-glutamate and antibiotics. Flasks (Nunc A/S, COpenhagen, Denmark) were kept at 37°C in CO, (5%) and air ventilated and humidified incubator. The culture medium was changed twice a week. After 46 weeks in culture, when the confluent cell layer comprised mainly polygonal glial cells, the cells were used for experiments. It has been demonstrated that by 34 weeks in culture the enzyme activities of glycolysis and mitochondrial oxidative metabolism have developed to the levels found in cells in .sit~.~ Over 95% of the cells stained positive for ghal

a,

AA 0.005

b 5 min

et ul.

fibrillary acidic protein. The cells grown in culture flasks were removed bv incubation in the uresence of 0.2% EDTA in phosphate-buffered saline (PBS) (in g/l, 8 N&l, 0.2 KCl. I.15 Na,HPO.. 0.2 KH,PO,) for 30min at 37°C followed by gentle &aping with y rubber policeman. After this treatment, over 90% of the cells excluded Trypan Blue. The cells were kept at 4C in the experimentat medium containing 137m~ NaCI, 5.4 mM KCI, 4.2 mM NaHCO,, I .2 mM MgCl,, 20 ITIM TES buffer 2-(2-hydroxy-l,l-bis(hydroxymethyl)ethyl amino) ethanesulphonate (pH 7.4). The viability of the cells did not change in 6 h under these conditions. Re~~~rut~on,&col_vsis and adenylutes

The rate of oxygen uptake was measured in Hansa-Tech oxygen electrode chambers at 37°C.‘” The rate of glycolysis was estimated from the production of ‘H,O from D-3[‘HIglucose (5 mM, 2 pCi/ml)16 as previously described.‘s ATP and ADP were measured from neutralized perchloric extracts of the cells with a luciferin assay as described by Kauppinen and Nicholls.” The potential-sensitive cationic cyanine dye, 3,3’-diethylthiadicarbocyanine iodide (DiS-C,-(5)), was used as an indicator of plasma membrane potential.‘,” Astrocytes were incubated in a thermostatted and stirred cuvette at 37°C in a Shimadzu UV 3000 dual-wavelength spectrophotometer. The incubation medium contained ~PM D&C,-(5) and absorbance was monitored at a wavelength pair of 646590 nm.” Determination of a6Ruhidium and L-[-‘H]glutamate Rradients

IOA

Astrocytes were incubated in the presence of 4 P’M*“RbCl, (4 &i/ml) and of recently lyophilized L-[3H]glutamate (0.3 PM, 10 nCi/ml). The cells were separated from incubation medium by rapid centrifugation through an oil layer of butylphtalate and dinonylpbtalate into 5% perchloric acid. For radioactivity counting, the pellets were treated as described previously.36 For each s6Rb+ and t$H]glutamate study, the volume of the intracellular water was measured in a parallel experiment by using “H,O (8 pCi/ml) and f’4C]insulin (4 yCi/ml) as markers for total and extracellular water spaces. respectively. Protein was measured by the method of Lowry et al.“’

Y

Maierials

s6RbCl, u-3-(‘HIglucose and t,-[3H]glutamate were obtained from the Radiochemical Centre (Amersham, Bucks, U.K.). Dis-C,-(5) was from Molecular Probes Inc. (Oregon, U.S.A.) and CarbonyIcyanide-m-chloropheny~hydrazone (Cl-CCP) from Boehringer (Mannheim, West Germany). buabain’ and gramicidin were purchased from Sigma (Poole, Dorset, U.K.). ATP-monitoring reagents were purchased from LKB Wallac Ltd, Turku, Finland. Standard laboratory reagents were at least analytical grade and were obtained from Merck (Darmstad, West Germany) or from Sigma (Poole. Dorset, U.K.).

Fig. 1. Optical monitoring of astrocyte plasma membrane potential by DiS-C,-(5). Influence of substrate deprivation, external substrates and metabolic inhibitors. Cultures of primary astrocytes (0.18 mg of protein/ml) were incubated in the thermostatted cuvette of a Shimadzu UV 3000 dualwavelength spectrophotometer in the presence of 2/1h1 DiS-C,-(5) monitoring the absorbance at 646590 nM and in the absence of external substrates. Downward deflection indicates a decrease of absorbance at 646nm. Additions were made as indicated by arrows. Symbols are as follows: 2 PM gramicidin (GRA); 10 mM o-glucose (GL); 5 mhi Napyruvate (PYR); 5 mM iodoacetate (IOA); 0.1 mhr ouabain (OUA).

RFZXJi>TS Glucose-deprivation

depolarizes

astrocyte

plasma

membrane

Cationic cyanine dyes respond to changes in membrane potential of organelles or cells by shifts in their absorbance or fluorescence spectra.4.39,41 The optical changes observed are usually linearly dependent on the magnitude of membrane potential up to - 100 mV. When astrocytes were added to substratefree incubation medium containing the cyanine dye

Responses of primary astrocytes to hypoglycaemia

285

brane polarization of astrocytes seems to require glycolytic activity rather than glycolytic end-product, pyruvate. Inhibition of Nat-K+-ATPase by ouabain (Fig. Id) completely blocks glucose-induced hyperpolarization, indicating that this effect requires the activity of a Na+-pump and cannot be ascribed to glucose transport alone. There is a correlation between the changes in the absorbance of cyanine dye and the calculated K+ equilibrium potential in the presence of the cation conducting ionophore, valinomycin (Fig. 2a). Since K+ permeability is the major determinant of membrane potential under these conditions4 it is fair to assume that the cyanine dye responses are linearly dependent on membrane potential. Hyperpolarization is dependent on glucose concentration (Fig. 2b). The maximal effect is obtained at 6 mM external glucose. The concentration of glucose, which hyperpolarizes the astrocyte plasma membrane Glucose (mw potential to half of its maximum, was 2.2m~ (Fig. 2b). This is close to the K,,, of plasma membrane Fig. 2. Relationships of the relative membrane potentialtransport of glucose in astrocytoma cells,26 but is dependent absorbance of DiS-C,-(5) and apparent K+ diffusion potential (a) and external glucose concentration substantially greater than the reported K, of gly(b). Experimental conditions were as in Fig. I, except that colysis for glucose in astrocytes (0.7 mM).‘* There is protein concentration was 0. I mg/ml. In (a), the relative sum a steep decrease in membrane polarization below of the increase in absorbance on stepwise additions of KC1 in the presence of 4 PM valinomycin are plotted against the 3 mM. Assuming that the astrocytes consist of a metabolically homogeneous cell population, depolarapparent K+ diffusion potential (reduced Goldman equation), assuming an internal K+ concentration of 120 mM. In ization of the plasma membrane in the absence of (b), the increase of absorbance above substrate-deprived external glucose amounts to some 25mV. 12

A

level is plotted against external D-glucose concentration added stepwise in the absence of ionophores.

DiS-C,-(5) a decrease in the absorbance (Fig. la) monitored at 646-590 nm is seen indicating that an equilibration of the probe with the membrane potential occurs.3.39Gramicidin, a cation-selective channelforming ionophore, reverses the decrease in absorbance (Fig. la-d) by dissipating ion gradients across the plasma membrane. The level of absorbance in the presence of 2 PM gramicidin was thus taken as the zero value of membrane potential.3,1’ The effects of glucose and pyruvate on the absorbance of DiS-C,-(5) of astrocytes are illustrated by the experiments in Fig. lc and d. Addition of IO mM glucose to substrate-deprived cells induces a rapid decrease in absorbance, indicating hyperpolarization of the plasma membrane. Subsequent addition of 5 IIIMpyruvate does not influence the kinetics of this effect (compare Fig. 1b and c). Glucose always causes a 50% decrease in the signal as compared with that seen in the absence of substrate. This indicates that the oxidation of endogenous substrates46 is able to support only submaximal plasma membrane potential (Fig. la vs b and c). If pyruvate is added to substrate-depleted cells, the plasma membrane potential is not changed (Fig. Ic); but glucose added after pyruvate hyperpolarizes (Fig. lb). Inhibition of glycolysis by iodoacetate causes a 70% reversal of the action of glucose (Fig. Ic). Thus, the plasma memNSC26,1--J

Pyruvate is the preferential substrate for astrocyte mitochondria

Because glucose and pyruvate differ significantly in their ability to support astrocyte plasma membrane potential, we studied the utilization of external substrates for mitochondrial respiration in the cells. The basal respiration of substrate-deprived cells was 5.4 k 0.2 nmol of O/mg protein/min (n = 4) and 10m~ glucose caused a 30% stimulation of oxygen consumption to 7.5 f 2.0 nmol of O/mg protein/min (n = 6) (p < 0.05, paired t-test). Addition of pyruvate to cellular suspension in the presence of 10m~ glucose caused further 35% activation of oxygen uptake (Fig. 3a). The stimulating effect of pyruvate was not abolished by malonate (not shown), which indicates that the increase in oxygen consumption in mitochondria occurs within intact cells, since this inhibitor of the Krebs cycle does not penetrate the plasma membrane.42 The rate of glycolysis in the presence of external glucose proceeds at 2.1 k 0.6 nmol/mg protein/min (n = 4). Because oxidation of 1 mol of glucose requires 12 mol of oxygen, the glycolytic activity is three times as great as expected from oxygen consumption (7.5 f 2.0 nmol of O/mg proteimmin). Assuming, that in the presence of 10m~ glucose, oxidation of pyruvate is responsible for the oxygen uptake; only 30% of glucose consumed is completely oxidized.

286

R. A. KAUPPINENet ul. Table 1. Cellular adenosine triphosphate and adenosine diphosphate concentrations. Influence of substrate deprivation, glucose and pyruvate ATP ADP @mol/mg protein)

Addition None Glucose (10rnM) Pyruvate (5 mM)

Fig. 3. Oxygen uptake of cultured primary astrocytes. Influence of external substrates and metabolic inhibitors. Astrocytes (0.22 mg of protein/ml) were incubated in a Hansa-Tech oxygen electrode chamber in the absence (a) and presence (b) of 10 mM D-glucose; additions were made as indicated by arrows. Symbols are as in Fig. 1 plus 1 PM carbonylcyanide-m-clorophenylhydrazone (CCP) and 2 mM Na-arsenite (ARS). Values given are rates of respiration in nmol O/min/mg protein.

The uncoupler of oxidative phosphorylation, ClCCP, which depolarizes mitochondria within cells thus increasing energy dissipation at the mitochondrial membranes,2,‘7 stimulates oxygen consumption only in the presence of glycolytic substrate (Fig. 3a, b). Iodoacetate inhibits endogenous respiration after 10min and, in the presence of this inhibitor, Cl-CCP is unable to stimulate oxygen uptake (Fig. 3b). This is consistent with exhaustion of endogenous oxidizable substrates. Pyruvate can partially restore the uncoupler-stimulated respiration in iodoacetate-inhibited astrocytes (Fig. 3b). A further addition of arsenate, which blocks pyruvate dehydrogenase together with lipoate-dependent enzymes, completely inhibits the respiration in the cells (Fig. 3b). At a concentration sufficient to block the Na+-pump (Fig. Id), ouabain does not influence oxygen consumption (Fig. 3a). Glucose is required for maintenance of cellular energy levels

Table 1 summarizes cellular concentrations of ATP and ADP in the absence and presence of external substrates. An ATP/ADP ratio of about 1 is maintained in the presence of 10 mM glucose, but glucose deprivation or pyruvate as only an external substrate led to an extensive decrease of the ratio. Pyruvate elevated significantly ATP as compared to substratedeprived cells, but the ATP/ADP ratio remained unchanged. Thus, the depolarization of plasma membrane under these two conditions with respect to glucose-treated cells correlates with a decay of energy status.

78.3 + 4.7

ATP/ADP

383.1 k 24.5

463.7 k 34.6* 386.6 5 18.6 164.7 + 8.4;

602.4 + 89.0

0.20 + 0.01 1.2 + 0.04* 0.29 & 0.04

Astrocytes (3.5 mg of protein/ml) were incubated either in the absence of external substrates or in the presence of 10 mM glucose or 5 mM pyruvate for 20 min. The samples were quenched with 7% (v/v) perchloric acid in 25 mM Na-EDTA on ice for 3 min, centrifuged and supernatants were neutralized with 3 M KOH in 1 M Tris to pH 7 and re-centrifuged. Extracts were immediately assayed for ATP and ADP as described under Experimental Procedures and each assay was standardized by additing 0.13 PM ATP at the end. Values are mean fS.E. for four experiments. Level of significance with respect to glucose-deprived cells. ‘p < 0.001.

Glucose dependency of 86Rb’ and L-[ 3H] gradients

One can anticipate that failure to maintain membrane potential will invariably affect the distribution of ions, including K+, and neurotransmitters, such as L-glutamate, since uptake mechanisms of these compounds are directly linked to the Nat-electrochemical potential.‘2,34 The permeability of Rb+ across the plasma membrane is nearly identical to that of K+,45 hence @Rb+ can be used as a probe of the Ki gradient. 35.36The effects of external glucose, pyruvate and some metabolic inhibitors on the distribution of 86Rb+ and L-[‘HIglutamate are summarized in Table 2. Hyperpolarization of the astrocyte plasma membrane (Figs 1 and 2) upon addition of external glucose is accompanied by an extensive accumulation

Table 2. Effects of glucose, pyruvate and metabolic inhibitors on s6Rb+ and Q3H]glutamate gradients across the astrocyte plasma membrane Addition ____ None 10 mM glucose 10 mM glucose 5 mM iodoacetate IO mM glucose 5 mM arsenate 5 mM pyruvate 5 mM pyruvate 5 mM arsenate

*6Rb+ in/out

r_-[3H]glutamate

5.8 i 0.7 7.8 I 1.3* 5.5 f.o.5

8.7 + 0.5 19.4 + 1.8t 6.1 50.1

6.2 k 0.5

8.7 + 1.0

5.9 * 0.2 4.7 f 0.3

9.2 + 0.4 3.9 + 0.6t

Astrocytes (0.3mg of protein/ml) were incubated in the presence of 86Rb+ (4 PM, 4 pCi/ml) and L-[‘HIglutamate (O.~PM, IOpCi/ml) for IOmin, and substrates and/or inhibitors were added. Samples were taken 10 min after the additions were made. The intracellular concentrations of both tracers were calculated using the intracellular volume determined in a parallel experiment for each assay. Values are mean + S.D. for 3-6 experiments. *p < 0.05, tp i 0.01 with respect to glucose-free cells.

Responses of primary astrocytes to hypoglycaemia

of both *6Rb+ and r$H]glutamate, and the effect of glucose is reversed by subsequent addition of iodoacetate. Pyruvate in the absence of glucose did not induce uptake of either tracer, as could be expected from data in Fig. lc. Importantly, anaerobic glycolysis in arsenate-treated cells supports submaximal gradients of both 86Rbt and L-[3H]glutamate, but only when external glucose is present. The data illustrated in Table 2 are plotted in Fig. 4, which shows the significant positive correlation between “Rb+ and L-[3H]glutamate accumulation. Thus, the gradients of these two compounds are supported by a common energetic mechanism.

DISCUSSION

The results of the present study demonstrate an ultimate dependence of functioning glycolysis in maintenance of energy status and functions of mature cultured astrocytes. The importance of aerobic glycolysis in maintaining cellular energy status is emphasized by the fact that the mitochondrial ATP synthesis during oxidation of pyruvate in the absence of glucose did not restore cellular ATP/ADP ratio. In the respiration-inhibited cells, however, anaerobic glycolysis was unable to maintain 86Rb+ and L-glutamate gradients. Thus, the physiological cytosolic phosphorylation potential of ATP to drive plasma membrane energy-dependent reactions is sustained only when glycolysis and respiration coexist. The oxidation of endogenous substrates, e.g. glycogen26 or glutamate,27,46 inevitably led to an extensive depolarization of the plasma membrane and loss of

201 I

-1

I ‘O -

l

A--

I

0

I

K+ and L-glutamate gradients. Glycogen content of astroglial cells is shown to be dependent on the concentration of glucose in the incubation.26 Inhibition of respiration of glucose-deprived primary astrocytes led to a disappearance of cellular ATP.** Thus, anaerobic glycolysis does not proceed in the absence of external glucose probably due to low glycogen stores. As a support for this claim, glucose stimulates oxygen uptake of substrate-deprived astrocytes’* (Fig. 3, trace a). Thus, glycogen does not serve as a significant energy source in hypoglycaemic astrocytes, and non-glycolytic substrates27,45 are oxidized in glucose-deprived cells. Intact nerve terminals respond differently to lack of glucose,‘8*‘9 Basal respiration and ATP/ADP ratio are unchanged, and, hence, plasma membrane has a high K+ gradient. Mitochondrial membrane potential is decreased by lO-20mV, and this drop is reversed either by external glucose or pyruvate.‘81’9 In the present work, the mitochondrial membrane potential of astrocytes was not measured. However, a hyperpolarization of membrane potential as indicated by a decrease of DiS-C,-(5) absorbance, was accompanied by a steepening of 86Rb+ gradients upon addition of glucose to the substrate-deprived cells. Pyruvate had no effect on the dye absorbance or on *‘Rb+ gradients. These results suggest that the contribution of mitochondrial membrane potential to the absorbance of DiS-C,-(5) is insignificant. Thus, a possible hyperpolarization of mitochondria within astrocytes did not interfere with our results. Aerobic glucose “waste” by cultured transformed glial cells** and astrocytes*’ has been demonstrated. In our cell culture, aerobic lactate production in stoichiometry was three times as great as oxygen consumption. Assuming that the cell preparation consisted of a respiring and a non-respiring subpopulation, one would expect much higher amounts of excess glycolysis. In anaerobic cells, glycolysis would be the only energy level supporting pathways with a low efficacy of ATP synthesis. However, it has been shown recently that primary astrocytes release pyruvate into the culture medium sufficient to support growth of neurons in the same culture.37 Therefore, mitochondrial oxidation of cytoplasmic reduced nicotinamide adenine dinucleotide is required in order to produce pyruvate, since in the absence of mitochondrial respiration glucose ends up as a lactate. Thus, the high rate of aerobic glycolysis is peculiar to astrocytes and not an indication of defective mitochondrial oxidation. The mechanisms responsible for high aerobic glycolytic activity in glial cells22~28 is not known. It has been suggested that the reason for excess glycolytic activity in rapidly growing tumour cells is due to the binding of hexokinase to the outer membrane of the mitochondria.6 An immunohistochemical study on hexokinase localization within brain cells has revealed large amounts of the enzyme coupled to the outer membrane of the mitochondria of astrocytes in

1+-

v

b

01

I/

I

5

10 fabibn

0rlhut)

Fig. 4. Correlation between MRb+ and t_-[3H]glutamate gradients across astrocyte plasma membrane. Standard deviations are indicated by two-dimensional error bars. Symbols as follows: (0) no additions; (D) 10 mM D-ghCOSe, (0) 10mM D-ghOSe and 5 mM iodoacetate, (0) 10 mM o-glucose and 5 mM arsenate, (A) 5 mM pyruvate, (A) 5 mM pyruvate and 5 mM arsenate. Correlation coefficient 0.979.

287

288

R. A.

KAUPPINEN

situ.15 In cultured astrocytes, mitochondrial hexokinase makes up about 20% of the total activity.“: We found that 6 mM external glucose is required to support maximal plasma membrane potential in astrocytes, the same concentration coincidentally that is the saturating level for hexokinase in astrocytoma cells.26 If this type of activation of glycolysis prevails in astrocytes, the stimulation of respiration by glucose (Fig. 3a) could be the result of increased ATP turnover at the mitochondrial membrane during the hexokinase step, due to an improved supply of substrate for the enzyme rather than to better availability of pyruvate for mitochondria. This could also explain the unresponsiveness of astrocyte respiration to ouabain. It seems likely that there is a functional uncoupling between energy dissipation by the Na+-pump and ATP synthesis by oxidative phosphorylation in astrocytes. However, Olson and Holzman” were able to find a response to ouabain, at a concentration 10 times higher than that used here, indicating that an interrelationship does exist. Puthophysiological

releoance

Although caution must be exercised in drawing conclusions from studies in vitro about the conditions prevailing in uivo, the present results can be discussed in light of the pathophysiology of brain hypoglycaemia. In the rat pathological changes in EEG activity are known to appear at a blood glucose concentration of 3.5 mM and isoelectricity is reached at 0.5-l mM.10.30Field potentials evoked from granule cells of the dentate gyrus are severely attenuated in the at 2 mM external glucose and, importantly,

rt al.

absence of glucose, pyruvate and malate cannot resume lost potential amplitude.’ We observed dramatic depolarization of the astrocyte plasma membrane followed by a decrease in steady-state Rb I and L-glutamate gradients when external glucose was lowered to the same range. Electrophysiological studies have revealed the resting plasma membrane potential of astrocytes to be about 76 mV.‘.*’ Thus, lack of glucose can be expected to cause substantial depolarization of these cells (Figs 1 and 2). Lack of glucose however, does not significantly lessen the ability of isolated nerve terminals to sustain the plasma membrane potential in the first place.18 One can assume. therefore, that intact nerve endings continue to function in the early stage of hypoglycaemic encephalopathy. Extrapolating our results to the astrocyte in situ, the extensive and rapid depolarization of astrocyte plasma membrane with decreased blood glucose would lead to accumulation of both K’ and L-glutamate in the extracellular space of the brain. High extracellular K’ launches an amplified cascade to increase extracellular glutamate’,” by depolarizing the presynaptic terminal, which induces leakage of the transmitter, and by reducing the capacity for glutamate uptake to nerve endings.“,“’ The enhanced excitation would cause the collapse of ionic gradients across the postsynaptic neuronal membrane, which would cause further potentiating cell damage in hypoglycaemia.” Acknowledgements--Supported by Grants from The Finnish Academy of Sciences and The Sigrid Juselius Foundation. Expert secretarial help by MS Eija Hokkanen is greatly acknowledged.

REFERENCES

substrates utilized by rat brain in Severe 1. Aeardh C.-D.. ChaDman A. G.. Nilsson B. and Siesiij B. K. (1981) Endogenous insulin induceh hypoglycaemia. J. Neurochem. 36,-49(tSOO. K. E. 0. and JHrvisalo J. 0. (1980) Effects of ionophores and metabolic inhibitors on mitochondrial 2. Akerman membrane potential within isolated hepatocytes as measured wiih the safranine method. Biochem. J. 192, 183-190. 3. Akerman K. E. O., Scott I. G., Heikkill J. E. and Heinonen E. (1987) Ionic dependence of membrane potential and glutamate receptor-linked responses in synaptoneurosomes as measured with a cyanine dye, Dis-C>-(5). /. Neurochem. 48, 552-559. of membrane potential using optical indicators. Biosci. Rep. 1, 183-196. 4. Bashford C. L. (1981) The measurement H. P., Prichard J. W. and Shulman R. G. (1985) 5. Behar K. L.. den Hollander J. A,. PetrofT 0. A. C., Hetherington Effect of hypoglycemic encephalopathy upon amino acids, high-energy phosphates, and pH, in the rat brain in Detection by sequential ‘H and 31P NMR spectroscopy. J. Neurochem. 44, 1045-1055. E., Morris H. P. and Pedersen P. L. (1981) Energy metabolism of tumor cells: requirement for a form 01 6. Bustamante hexokinase with a propensity for mitochondrial binding. J. biol. Chem. 256, 8699-8704. A. (1987) Cellular origins of endogenous amino acids released 7. Butcher S. P., Sandberg M., Hagberg H. and Hamberger into extracellular fluid of the rat striatum during severe insulin-induced hypoglycemia. J. Neurochem. 48, 722--728. pH from glial cells of rat brain in I%~o. J. cerehr. 8. Chesler M. and Kraig R. P. (1987) Direct recording of intracellular BIood Flow Metab. 7-(Suppl. l), Sl14. of evoked potentials from dentate granule cells by low glucose. 9. Cox D. W. G. and Bachelard H. S. (1982) Attenuation pyruvate + malate, and sodium fludride. Brain Res. 239, 527-534. in the CNS. Relationships between net uptake and 10 Erecinska M. and Nelson D. (1987) Amino acid neurotransmitters exchange in rat brain synaptosomes. FEBS Letf. 213, 61--66. of cytoplasmic calcium transients on 11. Heinonen E., Akerman K. E. O., Kaila K. and Scott I. G. (1985) Dependence membrane potential in isolated nerve endings of guinea-pig. Eiochim. biophys. Acta 815, 203-208. 12 Hertz L. (1982) Astrocytes. In Handbook of Neurochemistry (ed. Lajtha A.), Vol. 1, pp. 319-355. Plenum Press, New York. I. (1984) Modification of taurine and hypotaurine uptake systems in cultured primary astrocytes by serum 13 Holopainen free medium and dibutyryl-cyclic-AMP treatment. Int. J. deal Neurosci. 2, 529-534. uptake mechanism in synaptic neurotransmission. Br. J. Pharmoc. 41, 571~ 591. 14 Iversen L. L. (1971) Role of transmitter

oiw:

Responses

of primary

astrocytes

to hypoglycaemia

289

15. Kao-Jen J. and Wilson J. E. (1980) Localization of hexokinase in neural tissue: Electromicroscopic studies of rat cerebellar cortex. J. Neurochem. 35, 667-678. 16. Katz J. and Ronstad R. (1966) The metabolism of tritiated glucose by rat adipose tissue. J. biol. Chem. 241,360&3610. 17. Kauppinen R. A. and Hassinen I. E. (1984) Monitoring of mitochondrial membrane potential in isolated perfused rat heart. Am. J. Physiol. 247, H508-H516. 18. Kauppinen R. A. and Nicholls D. G. (1986) Synaptosomal bioenergetics. Role of glycolysis, pyruvate oxidation and responses to hypoglycaemia. Eur. J. Biochem. 158, 1599165. 19. Kauppinen R. A. and Nicholls D. G. (1986) Pyruvate utilization by synaptosomes is independent of calcium. FEBS Left. 199, 222-226. 20. Lewis L. D., Liunggren B., Norberg K. and Siesjii B. K. (1974) Cerebral energy state in insulin-induced hypoglycaemia, related to blood &cose and EEG J. Neurochem. 23, 673-679. 21. Lowrv 0. H.. Rosebrouuh N. J.. Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. bill. Chem. 153, 2655275. 22. Lust W. D., Schwartz J. P. and Passonneau J. V. (1975) Glycolytic metabolism in cultured cells of nervous system I. Glucose transport and metabolism in the C-6 glioma cell line. Molec. cell. Biochem. 8, 1699176. 23. MacVicar B. A., Crichton S. A., Burnard D. M. and Tse F. W. Y. (1987) Membrane conductance oscillations in astrocytes induced by phorbol esters. Nature 329, 242-243. 24. Nicholls D. G. and Sihra T. S. (1986) Synaptosomes possess an exocytotic pool of glutamate. Nature 321, 772-773. 25. Olson J. and Holzman D. (1982) Respiration and cell volume of primary cultured cerebral astrocytes in media of various osmolarities. Brain Res. 246, 273-279. 26. Passonneau J. V. and Crites S. K. (1976) Regulation of II glvcogen metabolism in astrocvtoma and neuroblastoma cells. I J. biol. Chem. 251, 2015-2022. ’ metabolism in glioma and neuroblastoma cells. Proc. 27 Pate] M. S., Russell J. J. and Gershman H. (1981) Ketone-body natn. Acad. Sci. U.S.A. 78, 72167218. F. R. and Trouet A. (1985) Effects of antimycin, glucose deprivation and serum on cultures 28 Pauwels P. J., Opperdoes of neurons, astrocytes and neuroblastoma cells. J. Neurochem. 44, 143-148. between neurons and glia. Trend.7 Neurosci. 5, 29 Pentreath V. W. (1982) Potassium signalling of metabolic interactions 339-345. R. A., Blank A. C. and Ferrendelli J. A. (1981) Regionally selective metabolic effects of hypoglycemia in 30 Ratcheson the brain. J. Neurochem. 36, 1952-1958. S. M. and Olney J. W. (1987) Excitotoxicity and the NMDA receptor. Trends. Neurosci. 10, 299-302. 31. Rothman B. F., Winterith M., Tholey G., Dierich A. and Mandel P. (1979) Effect of pentobarbital on 32. Roth-Schecheter mitochondrial synthesis in cultured glial cells. J. Neurochem. 33, 669-676. overflow of neuroactive amino acids during severe 33. Sandberg M., Butcher S. P. and Hagberg H. (1986) Extracellular insulin-induced hypoglycemia: in uiuo dialysis of the rat hippocampus. J. Neurochem. 47, 1788184. of glutamate in astrocytes cultured from 34. Schousboe A., Svenneby G. and Hertz L. (1977) Uptake and metabolism dissociated mouse brain hemispheres. J. Neurochem. 29, 999-1005. in intact synaptosomes. Biochem. J. 186, 21-33. 35. Scott I. D. and Nicholls D. G. (1980) Energy transduction 36. Scott A. G., Akerman K. E. O., Heikkila J. E., Kaila K. and Andersson L. C. (1986) Development of a neural phenotype in differentiating ganglion cell-derived human neuroblastoma cells. J. cell. Physiol. 128, 285-292. in the low molecular weight tropic activity for central 37. Selak I., Skapcr S. D. and Varon S. (1985) Pyruvate participation nervous system neurons in glial-conditioned media. J. Neurosci. 5, 23-28. 38. Shank R. P. and Aprison M. H. (1981) Present status and significance of the glutamine cycle in neural tissue. Life Sci. 28, 837-842. 39. Sims P. J., Waggoner A. S., Wang C.-H. and Hoffman J. F. (1974) Studies on the mechanisms by which cyanine dyes measure membrane potential in red blood cells and phosphatidylcholine vesicles. Biochemistry 13, 33153330. 40. Tholey G., Roth-Schechter B. F. and Mandel P. (1980) Development of glial cells in primary cultures: Energy metabolism and lactate dehydrogenase isoenzymes. Neurochem. Res. 5, 847-854. 41. Waggoner A. S. (1979) Dye indicators of membrane potential. Ann. Rev. Biophys. Bioenerg. 8, 47-68. 42. Webb J. L. (1966) Enzyme and Metabolic Inhibitors, Vol. 2, pp. l-244. Academic Press, New York. 43. Wieloch T. (1985) Hypoglycemia-induced neuronal damage prevented by an N-methyl-D-aspartate antagonist. Science 230, 681683. 44. Wieloch T., Harris R. J., Symon L. and Siesjo B. K. (1984) Influence of severe hypoglycemia on brain extracellular calcium and potassium activities, energy, and phospholipid metabolism. J. Neurochem. 43, 160-168. 45. Yellen G. (1987) Permeation in potassium channels: Implications for channel structure. Ann. Reo. Biophys. Chem. 16, 227-246. 46. Yu A. C., Schousboe A. and Hertz L. (1982) Metabolic fate of “C-1abelled glutamate in astrocytes in primary cultures. J. Neurochem. 39, 954960. (Accepted

9 December

1987)