View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1366 (1998) 97^112

Mitochondria and neuronal glutamate excitotoxicity

David G. Nicholls *, Samantha L. Budd 1

Neurosciences Institute, Department of Pharmacology and Neuroscience, University of Dundee, Dundee DD1 9SY, UK

Received 5 January 1998; accepted 17 February 1998

Abstract

The role of mitochondria in the control of glutamate excitotoxicity is investigated. The response of cultured cerebellar granule cells to continuous glutamate exposure is characterised by a transient elevation in cytoplasmic free calcium concentration followed by decay to a plateau as NMDA receptors partially inactivate. After a variable latent period, a secondary, irreversible increase in calcium occurs (delayed calcium deregulation, DCD) which precedes and predicts subsequent cell death. DCD is not controlled by mitochondrial ATP synthesis since it is unchanged in the presence of the ATP synthase inhibitor oligomycin in cells with active glycolysis. However, mitochondrial depolarisation (and hence inhibition of mitochondrial calcium accumulation) without parallel ATP depletion (oligomycin plus either rotenone or antimycin A) strongly protects the cells against DCD. Glutamate exposure is associated with an increase in the generation of superoxide anion by the cells, but superoxide generation in the absence of mitochondrial calcium accumulation is not neurotoxic. While it is concluded that mitochondrial calcium accumulation plays a critical role in the induction of DCD we can find no evidence for the involvement of the mitochondrial permeability transition. ß 1998 Elsevier Science B.V. All rights reserved.

Keywords: Mitochondrion; Glutamate; NMDA; Excitotoxicity; Calcium; Neuron

1. Introduction crotic or apoptotic characteristics and can occur with a delay of a few minutes to 1^2 days. Furthermore, a Even in neurones, where the ATP requirement for less severe, but chronic, restriction in neuronal ATP ionic homeostasis is particularly high, the safety mar- generation capacity may underlie a range of neuro- gin between cellular ATP supply capacity and de- degenerative disorders, including Alzheimer's, Hun- mand is considerable as long as glucose and oxygen tington's and Parkinson's diseases, amyotrophic lat- are available in excess. However even a brief inter- eral sclerosis and AIDs-related dementia as well as ruption in ATP generation, as in transient global encephalopathies associated with mitochondrial mu- ischaemia, can initiate neurodegeneration, which de- tations (reviewed in [1]). The excitatory neurotrans- pending on the severity of the insult can show ne- mitter glutamate plays a central role in neuronal cell death associated with these neurodegenerative disor- ders, and the interaction between the NMDA-selec- * Corresponding author. Fax: +44 (1382) 667120; tive glutamate receptor and the mitochondrion forms E-mail: [email protected] 1 Present address: CNS Research Institute, LMRC First the basis of this review, which will focus on necrotic Floor, 221 Longwood Avenue, Brigham and Women's Hospital, cell death in neuronal culture. Boston, MA 02115, USA. The sequence of events culminating in excitotoxic

0005-2728 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved. PII: S0005-2728(98)00123-6

BBABIO 44667 30-7-98 98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 cell death is initiated by the entry of Ca2‡ through analysis will focus on neuronal bioenergetics and will predominantly NMDA-selective glutamate receptors attempt to deconvolute the interactions between mi- activated by elevated extracellular glutamate. Gluta- tochondrial ATP synthesis, Ca2‡ accumulation and mate is not excitotoxic to anoxic cells as long as the generation of reactive oxygen species in cultured NMDA receptors are inactivated prior to the resto- neurones exposed to excitotoxic concentrations of ration of oxygen [2]. The two conditions of greatest glutamate. risk are ¢rstly the period when circulation is restored following transient global ischaemia and before the highly active glutamate transporters can re-accumu- 2. The initial glutamate exposure late the amino acid into neurones and glia, and sec- ondly in the penumbra surrounding a focal ischaemia In primary neuronal culture, the excitotoxic cas- where respiring neurones are exposed to high gluta- cade can be initiated by as little as 5 min exposure mate concentrations di¡using from the ischaemic to high glutamate concentrations in the absence of core. This latter region is of particular signi¢cance, Mg2‡ (to prevent voltage-dependent block of the since it demonstrates that glutamate exposure can NMDA receptor). There is overwhelming evidence destroy the bioenergetic integrity of neurones contin- that the entry of Ca2‡ into the cells predominantly uously supplied with glucose and oxygen. through NMDA-selective glutamate receptors during It is convenient to divide the excitotoxic process this period triggers the subsequent neurodegenera- into two stages: the initial exposure to glutamate; tion; however Ca2‡ loading via voltage-activated and the subsequent `latent' period, where the contin- Ca2‡ channels during KCl-depolarisation is much ued presence of glutamate is not obligatory and less excitotoxic and the reason for this selectivity is which culminates in the failure of cytoplasmic Ca2‡ the subject of current controversy. The possibilities homeostasis, termed by Tymianski `delayed Ca2‡ de- are that the absolute amount of Ca2‡ entering regulation' [3] and subsequent cell death (Fig. 1). Our through the NMDA receptor may greatly exceed that through voltage-activated Ca2‡ channels [4,5], or that Ca2‡ entering through the NMDA receptor is focussed onto a vulnerable excitotoxic locus within the cell [3,6]. When determined with high-a¤nity Ca2‡ indica- tors, such as fura-2, little di¡erence is seen in the 2‡ apparent elevation in [Ca ]c evoked with glutamate or KCl [7^9]; however the use of low a¤nity indica- tors of free Ca2‡ capable of reporting far higher concentrations before saturating has indicated that 2‡ glutamate-evoked [Ca ]c elevations may be consid- erably underestimated and may exceed 5 WM, con- siderably in excess of levels achieved with KCl-depol- arisation [7,10].

2.1. Neuronal mitochondria sequester cytoplasmic 2+ [Ca ]c transients

Fig. 1. Phases of acute glutamate excitotoxicity. Neurones ex- At the termination of a transient glutamate expo- posed to glutamate show a transient elevation in cytoplasmic 2‡ sure, [Ca ]c tends to return to baseline as the cation 2‡ 2‡ free Ca , [Ca ]c. Following this initial response, the signal de- is sequestered into internal organelles or is extruded cays to a plateau as NMDA receptors partially inactivate. The plateau is maintained for a variable time (`latent period') until across the plasma membrane. Our early studies with 2‡ isolated mitochondria from liver and brain demon- a secondary, irreversible, increase in [Ca ]c occurs (delayed Ca2‡ deregulation). strated that the activity of the mitochondrial Ca2‡

BBABIO 44667 30-7-98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 99 uniporter was highly dependent on the extramito- ganglion cells exposed to a range of metabolic in- 2‡ 2‡ chondrial Ca concentration, [Ca ]o [11] and ex- hibitors, including protonophores, cyanide and glu- ceeded the activity of the independent mitochondrial cose removal, but noted that these e¡ects could be Ca2‡ e¥ux pathway (coupled to H‡ and Na‡ in liver a consequence of impaired Ca2‡ extrusion from 2‡ and brain, respectively [12]) when [Ca ]o rose above the cells as well as inhibited mitochondrial sequestra- the `set-point' at which uptake and e¥ux balanced tion. [13]. Mitochondria would therefore be predicted au- In a detailed series of papers, White and Reynolds 2‡ 2‡ tomatically to bu¡er [Ca ]c above the set-point and [9,22,23] have analysed the pathways of Ca remov- 2‡ to release it back into the cytoplasm when [Ca ]c al from the cytoplasm following acute, non-toxic, 2‡ recovered to below this value (reviewed in [12]). With glutamate exposure. Restoration of baseline [Ca ]c a predicted set-point of 0.3^0.5 WM [13] even the following as little as 15 s exposure to 3 WM glutamate 2‡ Ca transients measured with high-a¤nity indica- was highly dependent upon both vim and extracel- tors during KCl stimulation rise well above these lular Na‡ ; thus no recovery occurred during wash- values and should therefore be in£uenced by mito- out of glutamate by a Na‡- free medium containing chondrial Ca2‡ sequestration. In 1981, we obtained protonophore. Interestingly, when the same Ca2‡ direct evidence of mitochondrial sequestration of de- transient was generated by the combination of high polarisation-evoked Ca2‡ loads by measuring mito- KCl and veratridine (activating voltage-activated chondrial 45Ca2‡ pools within synaptosomes: Ca2‡ Ca2‡ channels and allowing Na‡ entry via voltage- ‡ 2‡ loading of these intact isolated terminals by KCl-ac- activated Na channels), recovery of [Ca ]c to tivation of voltage-activated Ca2‡ channels resulted baseline could still occur. This was interpreted as in the large majority of the accumulated Ca2‡ being indicating a fundamental di¡erence in the handling further transported into the matrix [14,15]. of glutamate-evoked versus KCl/veratridine-evoked In 1990, Thayer and Miller [16] showed that brief Ca2‡ loads. While the reason for this di¡erence KCl-mediated depolarisation of dorsal root ganglion was not directly investigated, a likely explanation 2‡ 2‡ cells was followed by a partial recovery in [Ca ]c to was considered to be a greater uptake of Ca into a plateau which varied from 200 to 600 nM, but was the cell (and hence the mitochondrion) in the former absent in cells depolarised in the presence of proton- condition. ophore. This plateau was interpreted to be a conse- quence of the slow release from the mitochondrion of 2.2. In situ mitochondrial membrane potential (vim) 2‡ 2‡ 2+ Ca temporarily accumulated at the peak [Ca ]c. during acute Ca loading of neurones Addition of the protonophore during the plateau 2‡ produced a large transient elevation in [Ca ]c con- The key parameter determining the energetic status sistent with a rapid release of the mitochondrial pool. of mitochondria is the proton-motive force, vp, A similar plateau was observed in these cells follow- across the inner membrane (reviewed in [24]). In ing trains of s 25 action potentials [17]; again the the presence of physiological concentrations of Pi 2‡ 2‡ level of the [Ca ]c plateau during recovery (450^550 and Mg , isolated mitochondria from liver, heart nM) correlated well with the set-point predicted for or brain maintain a total vp of some 220 mV of isolated brain mitochondria [13]. A comparable pro- which the membrane potential vim comprises 150^ 2‡ tonophore enhancement of the KCl-evoked [Ca ]c 180 mV with a vpH of 30.5 to 31 pH unit contrib- transient and abolition of the subsequent plateau has uting the remaining 30^60 mV [25]. The determina- also been reported for bullfrog sympathetic neurones tion of both components of vp for in situ mitochon- [18]. These results were consistent with earlier mod- dria is exceedingly complex; in collaboration with elling with isolated mitochondria [19] and have re- Hoek and Williamson [26], we quanti¢ed vp for in cently been reproduced in whole-cell patch-clamped situ hepatocyte mitochondria: vim was determined chroma¤n cells following depolarisation-evoked ac- from the Nernstian distribution between the cyto- tivation of voltage-activated Ca2‡ channels [20]. plasm and matrix of the lipophilic cation TPMP‡, In 1990, the group of Duchen [21] also reported while vpH was measured by weak acid distribution; enhanced cytoplasmic Ca2‡ transients in dorsal root a value for vp of s 200 mV was consistent with

BBABIO 44667 30-7-98 100 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112

isolated mitochondria. Subsequently, we quanti¢ed endogenous Pi by an ADP/glucose/hexokinase trap the vim component for mitochondria within isolated [11] the capacity of isolated mitochondria to accumu- nerve terminals arriving at an estimate of 150 mV for late Ca2‡ is greatly restricted. Thus, while the same glucose-maintained synaptosomes [27]. reversible drop in vp occurs during Ca2‡ accumula- Direct determination of the Nernstian gradient be- tion, the vpH generated by protons extruded by the tween matrix and cytoplasm by confocal microscopy respiratory chain is no longer neutralised by Pi accu- is extremely di¤cult due to the small size of the mulation and a progressive build-up of vpH occurs. mitochondria and the enormous dynamic range re- This can ultimately result in an equipartition between quired to quantify a 300-fold gradient. Instead, de- the two components: a vp of 220 mV being com- terminations at the population or single-cell level of prised of 110 mV of vim and almost 2 pH units of resolution membrane-permeant cationic £uorescent transmembrane gradient [42,43]. Finally, induction indicators report their concentration within the mi- of the mitochondrial permeability transition (MPT, tochondrial matrix either by their £uorescent see below) will lead to a collapse in both vim and quenching (e.g. rhodamine-123 [28^34] or tetrameth- vp. It is not yet clear which of the above mechanisms ylrhodamine esters [35^37]) or by a change in their underlies any decrease in vim in glutamate-exposed emission spectra (e.g. JC-1 [9]). An essential control, neurones. which is not always performed, is to establish that the signal is not a¡ected by a change in plasma mem- 2.3. Deconvolution of vim e¡ects on mitochondrial brane potential, for example high [K‡] or glutamate Ca2+ accumulation and ATP synthesis in the absence of external Ca2‡, see [9,28]. With this important proviso in mind, £uorescent changes indi- In intact cells with maintained in situ metabolism, cative of mitochondrial depolarisation have been re- Ca2‡ homeostasis cannot be separated from cellular ported following acute glutamate exposure of cul- metabolism and bioenergetics and should be consid- tured neurones [9,28,30,31,33,34,37], although ered as part of an integrated picture (Fig. 2). Neuro- particularly with JC-1 the pattern of response can nes are almost entirely dependent on exogenous glu- be complex, with both hyperpolarizing and depola- cose, and the consequent supply of pyruvate to the rizing responses being detected [9]. The ATP syn- mitochondrion and its subsequent oxidation via the thase inhibitor oligomycin did not prevent the KCl- tricarboxylic acid cycle drives both ATP synthesis 2‡ evoked vim decrease [28]. and Ca accumulation. Since the two processes The in£uence of mitochondrial Ca2‡ loading on compete for the proton circuit they should be con- vim is complex and has been analysed in detail sidered together in an analysis of the e¡ects of cel- only for isolated mitochondria: in the presence of lular Ca2‡ loading. While Ca2‡ accumulation by the 2‡ excess Pi a single addition of Ca causes a transient mitochondria may a¡ect ATP synthesis, alterations depression in both the vim and vpH components of in ATP synthesis will in turn a¡ect the activity of ion vp, the extent of which is dependent on the Ca2‡ pumps responsible for removing Ca2‡ from the cy- concentration [38^40]. When net accumulation of toplasm. An analysis of these interactions is one of Ca2‡ is complete both components of vp are restored the main goals of our laboratory [31,41]. to their initial values. The transient depression in vp While protonophore addition has been widely em- can be su¤cient to interrupt ATP synthesis or even ployed to investigate mitochondrial Ca2‡ pools in to causes reversal of the ATP synthase and hydrol- neurones (see above), the collapse in vim will not ysis of ATP ^ consistent with early observations that only inhibit mitochondrial Ca2‡ accumulation and Ca2‡ accumulation could take precedence over ATP release the cation into the cytoplasm, but should synthesis [41]. The signi¢cance of phosphate is fre- also instantaneously reverse the mitochondrial ATP quently overlooked: however, the anion is co-accu- synthase resulting in rapid hydrolysis of cytoplasmic mulated into the matrix in parallel with Ca2‡ where ATP [41]. Unless the maximal rate of glycolytic ATP it forms an osmotically inactive, but rapidly dissoci- synthesis comfortably exceeds the cellular ATP de- able, calcium phosphate complex [12]. In the absence mand during Ca2‡ loading plus the rate of ATP hy- of added Pi, and particularly following depletion of drolysis by the uncoupled mitochondria, the cells will

BBABIO 44667 30-7-98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 101 become depleted of ATP and ATP-dependent proc- longed cytoplasmic Ca2‡ homeostasis [31,41]. As in esses will be compromised. other cells [27,44], the addition of oligomycin causes We have employed rat cerebellar granule cells as a slight hyperpolarisation of the in situ mitochondria our experimental model; these cells have a high den- [45] consistent with a `state 3^state 4' transition [24]. sity of mitochondria both along their neurites and Since the entire ATP generation by the cell is glyco- packed around the nucleus in the relatively thin an- lytic in the presence of oligomycin experiments can nulus of cytoplasm [31]. We initially determined be designed in which vim can be collapsed, for ex- whether the granule cells showed the same response ample by the further addition of a mitochondrial to protonophores during KCl-depolarisation as the respiratory chain inhibitor, such as rotenone dorsal root ganglion cells investigated by Thayer and [31,41], without a priori a¡ecting cellular ATP gen- Miller [16]. Cerebellar granule cells maintain a eration (Fig. 2). 2‡ 2‡ [Ca ]c of 6 100 nM, well below the estimated set- If accumulation of Ca into endoplasmic reticu- point for rat brain mitochondria (0.3^0.5 WM [13]) lum is ignored, the change in free Ca2‡ concentration and would thus be predicted to be largely depleted of reported by a £uorescent probe is the net resultant of Ca2‡. This was con¢rmed by the failure of protono- cytoplasmic Ca2‡ chelation, accumulation and re- phore addition to initiate a transient release of Ca2‡ lease from mitochondria and uptake and e¥ux into the cytoplasm. In contrast, addition of protono- across the plasma membrane, which in neurones is 2‡ 2‡ phore during the plateau of [Ca ]c following KCl- due to both the plasma membrane Ca -ATPase and depolarisation resulted in a transient spike in the Na‡/Ca2‡ exchanger [21,46,47]. In view of the evi- fura-2 signal, consistent with release from the mito- dence discussed above that mitochondria sequester chondrial compartment. When protonophore was much of the Ca2‡ load imposed by KCl depolarisa- added prior to 50 mM KCl, the size of the depolar- tion, it would be predicted that abolition of the mi- 2‡ isation-evoked [Ca ]c spike was greatly enhanced, tochondrial pool would enhance the cytoplasmic consistent with earlier studies [18]. While this could Ca2‡ transient. This is not, however, what we ob- be interpreted as a direct consequence of a failure of serve: KCl-depolarisation of granule cells in the pres- mitochondrial sequestration, a rapid secondary ele- ence of rotenone plus oligomycin actually produces a 2‡ 2‡ vation in [Ca ]c led us to suspect that the protono- smaller peak [Ca ]c elevation than in the presence of phore was causing a collapse in ATP levels, prevent- oligomycin alone [41]. Thus, mitochondrial depolar- ing Ca2‡ extrusion from the cells. This depletion was isation by protonophores enhances the KCl-evoked 2‡ con¢rmed by analysis of ATP/ADP ratios within the [Ca ]c transient, while mitochondrial depolarisation cells: within 5 min, 2 WM CCCP decreased the ratio by rotenone plus oligomycin decreases the transient. from a control value of s 7 to 6 3 [41]. There is thus The trivial explanation that rotenone might directly ambiguity as to whether the enhanced cytoplasmic inhibit voltage-activated Ca2‡ channels was elimi- 2‡ [Ca ]c response is due a failure of mitochondrial nated since when both rotenone and protonophore 2‡ Ca accumulation due to the collapsed vim, or a were present before KCl-depolarisation, the peak failure of Ca2‡ extrusion from the cell in response to height was enhanced to the same extent as with a lowered ATP/ADP ratio. FCCP alone [41]. There are two explanations for Seven-day in vitro granule cells incubated in the this counter-intuitive result: mitochondrial depolar- presence of 15 mM glucose do not have a su¤ciently isation must either reduce Ca2‡ entry via the VACCs active glycolytic pathway to be able to maintain a responsible for the initial transient, or Ca2‡ extrusion high ATP/ADP ratio and low basal cytoplasmic from the cell must be enhanced. Ca2‡ in the presence of protonophore, when glycol- Application of the same protocol to granule cells ysis is required to maintain the cell's normal func- exposed to glutamate showed that while oligomycin tions in the face of the freely reversing mitochondrial did not signi¢cantly a¡ect the peak and initial pla- ATP synthase. However, when the ATP synthase is teau following maximal stimulation of the NMDA inhibited by oligomycin, preventing both synthesis receptor, the combination of rotenone plus oligomy- and hydrolysis of ATP by the mitochondria, glycol- cin to depolarise the mitochondria prior to receptor 2‡ ysis can maintain high ATP/ADP ratios and pro- activation decreased both the initial [Ca ]c transient

BBABIO 44667 30-7-98 102 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112

and the subsequent plateau [31]. In further experi- glutamate-evoked cytoplasmic Ca2‡ transients, while ments (Budd and Nicholls, unpublished), we have removal of the mitochondrial matrix as a Ca2‡ sink 2‡ con¢rmed that metabolic restriction by the complex without depleting ATP result in a decrease in [Ca ]c. III inhibitor antimycin A, protonophore addition Furthermore, the total 45Ca2‡ accumulated during and inhibition of succinate dehydrogenase by malo- exposure to glutamate was also decreased by rote- nate each enhanced Ca2‡ deregulation following none plus oligomycin [31]; thus mitochondrial depol- NMDA receptor activation, while the same inhibi- arisation must restrict Ca2‡ entry into the cell and/or tors delayed deregulation in the presence of oligomy- enhance e¥ux of the cation. The mechanism of this cin. Thus, as in the case of KCl depolarisation, in- control is currently being investigated. hibitors which will decrease ATP levels enhance One hypothesis is that mitochondria control the

BBABIO 44667 30-7-98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 103

Fig. 2. Schematic inter-relationships between the NMDA receptor, mitochondrial Ca2‡ transport and cellular bioenergetics. Mitochon- dria in granule cells utilize pyruvate to generate vp, synthesize ATP and accumulate Ca2‡. Ca2‡ extrusion pathways driven by ATP 2‡ 2‡ (e.g. the Ca -ATPase) maintain a low [Ca ]c. In each ¢gure, the predicted activity of pathways is represented by the line thickness. In each scheme, the initial response to NMDA receptor activation is depicted. (A) In the absence of inhibitors, ATP synthesis is pre- dominantly via the ATP synthase. Ca2‡ entering the cell during NMDA receptor activation is largely accumulated into the mitochon- dria. Chronic receptor activation can lead to mitochondrial Ca2‡ overload. (B) In the presence of oligomycin, glycolysis supports the total cellular ATP demand. Both ATP synthesis and hydrolysis by the mitochondria are inhibited. The mitochondria still load with Ca2‡ and despite the presence of the inhibitor cellular ATP falls and Ca2‡ homeostasis is lost. Thus Ca2‡-loading of the mitochondri- on is excitotoxic by a process independent of the oligomycin-sensitive ATP synthase. (C) The complex I inhibitor rotenone prevents the respiratory chain from pumping protons. Glycolysis now supplies ATP both for the cell and for the ATP synthase which reverses 2‡ to generate a reduced proton gradient. Even in the absence of glutamate, the cell can only maintain a low [Ca ]c for a limited peri- od. When glutamate is added the reversal of the ATP synthase accelerates as it attempts to compensate for the inward £ux of Ca2‡ into the matrix and the resulting ATP depletion leads to an immediate loss of cytoplasmic Ca2‡ homeostasis. (D) The combination of rotenone plus oligomycin allows the proton gradient to decay preventing mitochondrial Ca2‡ loading. Cytoplasmic ATP is main- tained by glycolysis; the decreased 45Ca2‡ uptake into the cells, together with the maintained low cytoplasmic Ca2‡, each demonstrate that Ca2‡ in£ux through the NMDA receptor is balanced by e¥ux across the plasma membrane when mitochondria can no longer se- quester Ca2‡, due either to a feed-back inhibition of the NMDA receptor or an enhanced Ca2‡ e¥ux. (E) The protonophore FCCP collapses cellular ATP by reversal of the ATP synthase. 6 net Ca2‡ £ux across the plasma membrane by rapid 3. Delayed Ca2+ deregulation changes in ATP availability. ATP depletion of gran- 2‡ ule cells by simple protonophore addition enhances The initial peak in [Ca ]c following glutamate the KCl-evoked peak Ca2‡ elevation and prevents addition and the subsequent recovery to a plateau the maintenance of a subsequent plateau [21,41]. largely re£ects the activation and subsequent partial 2‡ Glucose depletion greatly enhances the [Ca ]c tran- desensitisation of the NMDA receptors. Once the sient in dorsal root ganglion cells depolarised with plateau is established, the cytoplasmic free Ca2‡ of KCl [21]. Lowered glucose availability, both in the glutamate-exposed cells can remain relatively low presence and absence of oligomycin, enhances both and stable [49] until a sudden, essentially irreversible, the peak and subsequent plateau fura-2 signal of increase occurs (Fig. 3A). Individual cells survive for granule cells during glutamate exposure (Budd, varying periods before this ¢nal Ca2‡ deregulation Ward and Nicholls, unpublished). However, since and survival ¢ts a single exponential curve, consistent the rationale of our experiments in the presence of with a stochastic process [50]. Continued NMDA oligomycin has been to eliminate e¡ects of mitochon- receptor activation is not required during this period, drial ATP synthesis, how could mitochondria with since transient exposure of neurones to glutamate for inhibited ATP synthase activity still in£uence cyto- as short a period as 5 min can lead to subsequent cell plasmic ATP/ADP ratios? The experimental observa- death, although prolongation of the glutamate expo- tion is that glutamate causes an equivalent decrease sure does increase toxicity [3]. Plasma membrane in- in granule cell ATP/ADP ratios in the presence and tegrity is initially maintained, since £uorescent absence of oligomycin [31]. One factor which is fre- probes are still retained in the cytoplasm. While the quently overlooked is the availability of phosphate ATP/ADP ratio falls in populations of deregulating (Pi), which is co-accumulated into the mitochondrial cells [31,51] it has not been possible to determine the matrix together with Ca2‡ [12] and the rapid accu- ratio in an individual cell during the process. Thus it mulation of the cation into the mitochondrion fol- is not easy to determine whether there is an upstream lowing NMDA receptor activation could temporarily failure of ATP synthesis or a downstream inhibition deplete the cytoplasm of Pi, which would in turn of ATP-dependent ion pumps; for example, the plas- decrease the ATP/ADP ratio which could be main- ma membrane Ca2‡-ATPase or Na‡/K‡-ATPase, tained by glycolysis. Thus it has been observed that due to proteolytic or oxidative damage. 2‡ low extracellular Pi decreases ATP levels in cultured Deregulation of [Ca ]c in our preparation appears neurones and sensitises cells to glutamate-evoked to be a result of failed e¥ux rather than enhanced 2‡ death [48]. in£ux, since the rise in [Ca ]c is not blocked by a

BBABIO 44667 30-7-98 104 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112

Fig. 3. Delayed Ca2‡ deregulation following the addition of glutamate/glycine to rat cerebellar granule cells. (A) Addition of 100 WM glutamate plus 10 WM glycine to cells results in a stochastic Ca2‡ deregulation. (B) In the presence of antimycin A to inhibit the respi- ratory chain, glutamate/glycine addition leads to an immediate deregulation which can be accounted for by insu¤cient ATP synthesis (see Fig. 1C). (C) In the presence of oligomycin plus antimycin A to depolarize the mitochondria without collapsing ATP, glutamate/ glycine-evoked Ca2‡ transients are decreased and no Ca2‡ deregulation occurs within the timespan of the experiment. Each trace rep- 2‡ resents [Ca ]c in an individual cell soma determined with fura-2. For further details see [31]. cocktail of channel inhibitors (Budd and Nicholls, ergy demand upon a neurone. The increased cyto- unpublished), while the rate of Mn2‡ quenching of plasmic Ca2‡ and Na‡ concentrations will result in fura-2 £uorescence, indicative of Ca2‡ entry, does activation of the plasma membrane Ca2‡-ATPase not increase as cells deregulate [52]. There may how- and Na‡/K‡-ATPase, respectively. The latter may ever be a contribution to deregulation from mito- be quantitatively more important, since even though chondrial dumping of Ca2‡ into the cytoplasm as a NMDA receptors display some selectivity for Ca2‡ result of mitochondrial depolarisation and/or the over Na‡, the much higher concentration of the lat- permeability transition. ter ion in the extracellular medium means that the in£ux of Na‡ may exceed that of Ca2‡ [67]. How- 3.1. The latent period prior to deregulation ever, glutamate is still excitotoxic in Na‡-free media [3]. In addition to these plasma membrane e¡ects, the 2‡ The processes which occur during this intermediate peak values of [Ca ]c in excess of 5 WM during `latent' period culminate in cell death which may glutamate exposure reported in recent experiments have apoptotic [53,54] necrotic [55^57] or mixed with low a¤nity Ca2‡ indicators [7,10] would be pre- [51,58,59] characteristics depending on the severity dicted from isolated mitochondrial studies [40] to of the initial insult. Thus, within a culture of cerebel- result in such a rapid uptake into the mitochondrial lar granule cells transiently exposed to glutamate, a matrix that vp could be transiently lowered below subpopulation were found to depolarise their mito- the value required for ATP synthesis. chondria and undergo necrosis, while the surviving Depletion of cytoplasmic ATP will lead to a cata- 2‡ cells subsequently repolarised their mitochondria and strophic deregulation of [Ca ]c by inhibiting the regenerated ATP, but underwent later apoptosis [51]. plasma membrane Ca2‡-ATPase and also limiting Apoptosis is out of the scope of this review, however, the activity of the two ATP-requiring enzymes in while activation of the caspase cascade is a common the glycolytic pathway, hexokinase and phosphofruc- ¢nal pathway for a range of factors inducing granule tokinase [68]. In our studies, cultured granule cells cell apoptosis [60^62], caspase inhibitors are unable treated with FCCP, rotenone or antimycin A (Fig. to prevent necrosis [63^66]. 3B) show massive Ca2‡ deregulation upon exposure to glutamate [31]. In this condition, glycolysis is re- 3.2. Restricted bioenergetics facilitates delayed Ca2+ quired not only to generate ATP for the activated deregulation plasma membrane ion pumps, but also to supply the reversed mitochondrial ATP synthase, which at- NMDA receptor activation places a multiple en- tempts to regenerate vp which is lowered by the up-

BBABIO 44667 30-7-98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 105 take of Ca2‡ into the matrix. It is evident that these cells. A further implication of this ¢nding is that conditions create an instant ATP de¢cit and imme- glycolysis can supply su¤cient ATP to supply the diate failure of cytoplasmic Ca2‡ homeostasis. cell's energy requirements even when these are en- There is an extensive literature demonstrating the hanced during NMDA receptor activation. synergistic e¡ect of metabolic restriction upon gluta- mate excitotoxicity both in vivo, e.g. [69^72], and in 3.4. Delayed Ca2+ deregulation is not causally vitro [37,59,73^75]. This synergism can also be ob- dependent upon a failure of glycolysis served in in vivo models of neurodegenerative dis- eases: thus the behavioural and morphological ef- The requirement of glycolysis for ATP at two steps fects of Huntington's disease following injection of (hexokinase and phosphofructokinase) implies that a or 3-nitropropionate into the striatum can be dimin- cell whose ATP levels are declining, even transiently, ished by NMDA receptor antagonists [76^78] as can could su¡er an irreversible collapse as glycolysis be- the motor e¡ects of 1-methyl-4-phenylpyridinium comes limiting, exacerbating a further decrease in (MPP‡) inhibition of mitochondrial complex I in ATP. In 1986, we found that synaptosomes showed animal models of Parkinson's disease [71]. an initial enhancement of glycolysis, consistent with a Pasteur e¡ect, following respiratory inhibition by 3.3. Delayed Ca2+ deregulation is not causally rotenone, but that this was followed by a progressive dependent upon a failure of oxidative failure [79]. This glycolytic failure has been subse- phosphorylation quently analysed in more detail [68] under conditions where the bioenergetic safety margin was eroded by While Ca2‡-deregulation of glutamate-exposed increased energy demand (e.g. ionophore addition) cultured cerebellar granule cells is an imperfect mod- and inhibited ATP generation (including rotenone el for glutamate excitotoxicity, it does allow some addition). It was concluded that the ATP require- simple hypotheses to be tested. The ¢rst is that a ment for hexokinase was the limiting factor during failure of mitochondrial ATP synthesis is necessary glycolytic failure [68]. Interestingly, this raises the and su¤cient for deregulation. This might be ex- question as to how the neurone restarts glycolysis pected in view of the synergistic e¡ects of energy on recovery from severe ATP depletion. In this con- restriction and NMDA receptor activation discussed text, it has been demonstrated that lactate, which above, as well as the possibility that mitochondria accumulates during anoxia, can be utilised by se- might be damaged by excessive Ca2‡ accumulation verely ATP-depleted cells, generating pyruvate and (discussed below). Since granule cells have su¤- allowing mitochondrial oxidative phosphorylation ciently active glycolysis to maintain Ca2‡ homeosta- to regenerate ATP allowing glycolysis to occur [80]. sis when mitochondrial oxidative phosphorylation is If delayed Ca2‡ deregulation occurs in a neurone inhibited from the outset in the presence of oligomy- at a stage where the capacity of the cell to generate cin [41], this hypothesis would predict that glutamate ATP is insu¤cient to meet its energy demands, it would not be excitotoxic in the presence of the in- might be predicted that an additional substrate sup- hibitor. However, no statistically signi¢cant di¡er- ply might be able, at least temporarily, to rescue ence is observed in the time or extent of Ca2‡ dereg- cells. Lactate and pyruvate are e¡ective neuronal ulation in granule cells exposed to glutamate in the substrates [81,82] and Eimerl and Schramm [83] presence or absence of the inhibitor [31]. In other have reported that pyruvate plus phosphate added words, delayed Ca2‡ deregulation occurs in cells after glutamate exposure decreases cell death. In re- whose mitochondria are polarised, transporting elec- cent experiments (Budd, Ward and Nicholls, unpub- trons and accumulating Ca2‡, but are not required to lished), we have found that granule cells incubated in synthesise (or hydrolyze) ATP. A decline in ATP/ glucose-free media and maintained by lactate or pyr- ADP ratio is seen in granule cell populations exposed uvate undergo glutamate-evoked delayed Ca2‡ de- to glutamate for 60 min in the presence of oligomy- regulation after a similar delay as cells maintained cin [31], but it is currently unclear whether this is the by glycolysis. Thus, a time-dependent inhibition of cause or e¡ect of the Ca2‡ deregulation in individual the glycolytic pathway following glutamate exposure

BBABIO 44667 30-7-98 106 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 is not an inherent component of the delayed Ca2‡ glutamate has been considered to be a reliable pre- deregulation observed in our preparation. dictor of the extent of subsequent cell death. How- ever, this has recently been challenged by the report 3.5. In situ mitochondria undergo progressive that NMDA-evoked uptake is still much more toxic bioenergetic deterioration in glutamate-exposed than an equivalent extent of KCl-evoked 45Ca up- cells take [6], suggesting that the extreme toxicity of glu- tamate-evoked Ca2‡ uptake may be due to the selec- A number of key catabolic mitochondrial enzymes tive direction of the Ca2‡ onto an excitotoxic locus are readily inhibited by oxidative damage, including in the close vicinity of the intracellular face of the pyruvate dehydrogenase [84] and aconitase [85]. The NMDA receptor [3]. However, since the mitochon- mitochondrial electron transport chain itself is a fur- drial matrix is the only compartment capable of se- ther potential locus for inhibition since complexes I, questering this Ca2‡, it is of central importance to III and IV each exert some control over the rate of establish the role which matrix Ca2‡ accumulation oxygen consumption of isolated brain mitochondria plays a role in excitotoxicity. [86]. There is some disagreement as to the bioener- An additional factor, which has not generally been getic consequences of partial inhibition of these com- considered, is that the initial rate of mitochondrial plexes: by comparing the titration of respiratory Ca2‡ loading, rather than the absolute amount accu- chain inhibitors with intact mitochondria and single mulated may be critical. Thus, low a¤nity Ca2‡ 2‡ complexes Davey and Clark [86] concluded that the probes indicate very large [Ca ]c transients in re- individual complexes had to be inhibited by at least sponse to glutamate [7,10] which would be predicted 60% before signi¢cant e¡ects were observed on mi- to divert all mitochondrial proton pumping to Ca2‡ tochondrial respiration and ATP synthesis. However accumulation [12]. Indeed, the net accumulation of repopulation of mitochondrially depleted rho-0 cells 45Ca2‡ by granule cells is largely complete within with mitochondria from patients with either Parkin- 5 min (Budd and Nicholls, unpublished). Addition son's disease displaying 26% de¢ciency in complex I of NMDA or glutamate to granule cells in low K‡ activity [87] or Alzheimer's disease displaying 52% medium in the absence of Mg2‡ (the usual experi- de¢ciency in complex IV activity [88] both resulted mental protocol) depolarises the plasma membrane 2‡ 2‡ in a slowed recovery of [Ca ]c following Ca loads. [90]; however, glutamate addition to granule cells A relatively direct demonstration that in situ mi- predepolarised in high KCl medium has been re- tochondria undergo a progressive deterioration of ported to show much decreased excitotoxicity [34]. bioenergetic function following glutamate exposure Since NMDA receptors partially desensitise during is the ¢nding by Atlante et al. [89] that mitochondria prolonged depolarisation, this implies that the initial isolated from glutamate-exposed cerebellar granule Ca2‡ entry in polarised cells may be critical in trig- cells display a decrease in state 3 respiration which gering excitotoxicity. It is notable that no lag can be 2‡ 2‡ was proportional to the time for which the cells had detected between an increase in [Ca ]c and [Ca ]m been exposed to glutamate. After a 5-h exposure, the following NMDA receptor activation, whereas some subsequently isolated mitochondria displayed no sig- lag is apparent for KCl-evoked Ca2‡ loading [91]. ni¢cant respiratory control. Inhibited state 3 respira- tion implies a metabolic or respiratory chain limita- 3.7. Glutamate is not excitotoxic to cells maintaining tion rather than a loss of respiratory control, a high ATP with depolarised mitochondria although the locus of inhibition was not established in this study. In the presence of oligomycin, in situ mitochondria maintain a high vim and thus retain the ability to 3.6. Delayed Ca2+ deregulation may be triggered by accumulate Ca2‡. Consistent with this, 45Ca2‡ is ac- mitochondrial Ca2+ accumulation cumulated in the 20 min following addition of gluta- mate to the same extent as in the absence of the The total 45Ca2‡ accumulated by cortical neurones inhibitor [31]. As in the case of depolarisation- [4] or cerebellar granule cells [5] in the presence of evoked Ca2‡ loading discussed above, inhibition of

BBABIO 44667 30-7-98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 107 the respiratory chain of oligomycin-treated cells control of glutamate excitotoxicity, it is important to should collapse vim without a¡ecting ATP synthesis distinguish between a relatively stable and reversible 2‡ by glycolysis. Respiratory chain inhibitors under elevation in [Ca ]c, which can be a result of en- these conditions will, however, collapse vim and hanced in£ux and/or restricted e¥ux from the cell abolish mitochondrial Ca2‡ accumulation [31] in ad- and the catastrophic and irreversible rise signalling dition to blocking electron transport at de¢ned sites the onset of Ca2‡ deregulation. Thus, Khodorov et and thus potentially a¡ecting the generation of reac- al. [92] exposed 7^8 DIV cerebellar granule cell cul- tive oxygen species. tures to the combination of antimycin A and oligo- The total accumulation of 45Ca2‡ within cultured mycin to depolarise the in situ mitochondria by a cerebellar granule cells exposed to glutamate can ap- similar mechanism to our earlier experiments [41]. proach 20 nmol/Wl of cell volume (equivalent to 20 These authors found that the combination of inhib- 2‡ mM!) before the cells die [5]. The mitochondrion is itors resulted in a delayed secondary rise in [Ca ]c the only organelle capable of accumulating such after glutamate exposure, in contrast to our ¢ndings amounts of Ca2‡, and loss of the mitochondrial that the same inhibitors decreased both peak and sink for Ca2‡ entering via the NMDA receptor plateau responses to continuous glutamate [31], lead- should therefore logically result in a rapid saturation ing them to conclude that `mitochondria play a dom- of cytoplasmic Ca2‡ chelators and a massive eleva- inant role in the protection against neuronal Ca2‡ 2‡ tion in [Ca ]c. It was therefore remarkable that overload induced by excitatory amino acids' [92]. granule cells exposed to glutamate in the presence Apparently opposite ¢ndings with the same prepara- oligomycin plus rotenone [31] or antimycin A (Fig. tion require some comment: the initial recovery fol- 2‡ 2‡ 3C) show a relatively small peak Ca elevation, lowed immediately by a secondary rise in [Ca ]c maintain a low subsequent Ca2‡ plateau and can after glutamate found by Khodorov et al. is very 2‡ 2‡ survive for up to 5 h in a Mg -free medium in the similar to the time course of [Ca ]c in our cells in continuous presence of 100 WM glutamate, 10 WM the presence of protonophore, and suggests an en- glycine. Any cell death which occurred in the pres- ergy limitation; these authors worked at room tem- ence of these two inhibitors was independent of the perature and employed 5 mM glucose (in contrast to presence of glutamate. Thus, mitochondrial Ca2‡ ac- our 37³C and 15 mM glucose). Furthermore, no con- cumulation may be an essential stage in the chain of trols were performed with glutamate in the presence events associated with glutamate excitotoxicity. of oligomycin alone to control for ATP depletion Such a surprising result requires careful controls, under these conditions where the maximal rate of the ¢rst is to investigate the possibility that rotenone glycolytic ATP synthesis and Ca2‡ extrusion will be or oligomycin have previously unrecognised direct restricted, but NMDA receptor activity will be vir- inhibitory actions at the NMDA receptor. Oligomy- tually unchanged. Since di¡erent experimental or cul- cin alone is not neuroprotective [31], while as dis- ture conditions will a¡ect the complex balance be- cussed above, rotenone addition in the absence of tween NMDA receptor activity, Ca2‡ extrusion oligomycin results in immediate glutamate-evoked capacity, glycolysis and the capacity for mitochon- Ca2‡ deregulation [31]. This would actually be pre- drial ATP synthesis, it will clearly not be possible dicted on bioenergetic grounds: inhibition of the res- to separate e¡ects of mitochondrial ATP synthesis piratory chain places a severe demand upon the gly- and Ca2‡ sequestration by using conditions which colytic capacity of the cell: in addition to the normal deplete ATP as well as depolarizing the mitochon- cellular house-keeping functions which require ATP, dria. respiratory chain inhibition leads to a reversal of the The experiments we performed with rotenone/oli- mitochondrial ATP synthase as glycolytic ATP is gomycin [31] decreased both the mitochondrial Ca2‡ 2‡ utilised to regenerate vim (Fig. 2). accumulation and the cytoplasmic [Ca ]c elevations in response to glutamate. While we interpreted the 2+ 3.8. Elevated [Ca ]c is not itself neurotoxic neuroprotection a¡orded by this combination to the abolition of the matrix accumulation, it was possible 2‡ In this analysis of the role of mitochondria in the that the decreased cytoplasmic [Ca ]c could account

BBABIO 44667 30-7-98 108 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 for the neuroprotection. It is therefore of consider- oxidation by non-synaptic rat brain mitochondria, 2‡ able signi¢cance that the increased [Ca ]c and de- but did detect the reactive oxygen species with sub- creased matrix Ca2‡ produced by protonophores in strates utilizing complex 1. No decrease in peroxide cultured hippocampal neurones [8], cortical neurones production was found following a state 4^state 3 [93] and an immortalised hippocampal neuronal cell transition; thus it was proposed that complex I could line [94] is not only not neurotoxic, but a¡ords pro- be a source of reactive oxygen species in state 3. c3 tection against glutamate excitotoxicity [8,93,94]. Continued production of O2 † in state 3 by isolated Putting our results together with these, one can con- brain mitochondria is also indicated by the study of clude that mitochondrial Ca2‡ loading is potentially Dykens [99], where Ca2‡ loading of malate/gluta- excitotoxic, but that elevated cytoplasmic Ca2‡, while mate supplemented mitochondria in the presence of controlling the extent of mitochondrial Ca2‡ loading ADP increased production of highly reactive hydrox- and signalling delayed Ca2‡ deregulation, is not re- yl radicals. sponsible for the intermediate events leading to a Con£icting consequences of in situ mitochondrial progressive deterioration in cell function. depolarisation have been reported on the generation of reactive oxygen species in cultured neurones: the 3.9. The role of mitochondrially initiated oxidative production of hydrogen peroxide detected by di- damage chloro£uorescin is inhibited by protonophores [100], in accordance with results with isolated mito- Glutamate excitotoxicity requires oxygen; thus chondria [97,101], although the pH sensitivity of the hippocampal neurones exposed to glutamate under signal from the oxidised product, dichloro£uorescein, hypoxic conditions show no more cell death than creates complications during the large cytoplasmic due to hypoxia alone [2]. Additionally, NMDA an- acidi¢cation produced both by protonophores [102] tagonists, which protect against excitotoxic damage and by NMDA receptor activation [102,103]. to cortical neurones following chemical ischaemia, The non-£uorescent membrane permeant hydro- need only be present during reperfusion [95]; in other ethidine (HEt) is oxidised within neurones to the c3 words, the return of oxidative metabolism triggers a £uorescent ethidium cation by O2 † and has been critical period of toxic NMDA receptor activation. widely used to detect the formation of the radical Reoxygenation is associated both with mitochondrial in a large number of neuronal and non-neuronal Ca2‡ accumulation and the generation of reactive studies [104^108]. Similar studies have been per- oxygen species (ROS) and a central question con- formed with dihydrorhodamine-123 [109], which is cerns the causal relationships between these two pa- oxidised to cationic rhodamine-123 predominantly rameters and the ultimate death of the cell. We wish by peroxynitrite [110]. Since both cationic products to focus on just one aspect of a complex topic, are membrane permeant, they will accumulate in the namely the role of the in situ mitochondrial mem- matrix of polarised mitochondria, even if they were brane potential in the generation of potentially toxic the product of non-mitochondrial oxidation; thus c3 superoxide O2 † radicals. they do not automatically signal the site of free rad- c3 The predominant source of O2 † in most tissues is ical production in imaging studies [111]. In addition, the mitochondrial respiratory chain (reviewed in generated ethidium undergoes £uorescent quenching [96]). Complex III, and in particular the level of re- when accumulated within the mitochondria matrix duction of ubisemiquinone at the Qp site, has gener- and conversely shows a £uorescent dequenching ally been considered to be the main source of reactive upon mitochondrial depolarisation [45]. This can be oxygen species in mitochondria. Thus, with succinate di¤cult to separate from the inherent radical-induced as substrate, the generation of reactive oxygen spe- oxidation, particularly in experiments involving glu- cies by isolated heart mitochondria is relatively high tamate- or protonophore-mediated mitochondrial de- in state 4, but virtually abolished when ADP is polarisation. Thus, a number of studies have re- added and vim decreases to state 3 levels [97,98]. ported that protonophore treatment of neurones to However, in a recent paper, Herrero and Barja [98] depolarise in situ mitochondria causes an enhanced failed to detect peroxide generated during succinate generation of ROS detected by dihydrorhodamine-

BBABIO 44667 30-7-98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 109

123 [109] or hydroethidine [104,108], in contrast to been reported to be much less e¡ective against brain results with isolated mitochondria. We have re-exam- mitochondria [119]. The inhibitor has been employed ined the behaviour of the latter probe in cultured in an attempt to identify a role for the MPT in intact cerebellar granule cells and found that the £uorescent neurones, but while some measure of protection yield of the ethidium generated by hydroethidine ox- against glutamate is generally observed [9,36,37] idation is enhanced by mitochondrial depolarisation, but see [30], it is likely that it is the ability of the consistent with £uorescent dequenching as the mito- agent to inhibit calcineurin, rather than the MPT, chondria release ethidium and compounded by fur- which underlies most of the observed e¡ects. Thus, ther £uorescent enhancement as the released ethidi- while Ankarcrona et al. [120] found that cyclosporin um binds to non-mitochondrial nucleic acids [45]. A protected granule cells against both early necrosis This vim sensitivity could be minimised by the use and delayed apoptosis induced by glutamate, a sim- of very low concentrations of HEt, such that the ilar protection was a¡orded by the more selective ethidium remains bound within the mitochondria calcineurin inhibitor FK-506, which does not interact and is not released on depolarisation, or by extrac- with the mitochondrial permeability transition pore. tion and quanti¢cation of the generated ethidium at Calcineurin inhibition by cyclosporin A has multiple the end of the experiment. e¡ects, including an increase in spontaneous action With these precautions, we con¢rm reports that potential ¢ring [121] and an inhibition of NMDA NMDA receptor activation enhances the generation receptors [122], while chronic cyclosporin A induces c3 of O2 † by granule cells, but ¢nd no enhancement neuronal apoptosis in cortical cultures [123,124]. In upon protonophore addition, in agreement with the our experiments, cyclosporin A, the more selective above studies using dichloro£uorescin but in contra- methylvaline-4-cyclosporin [125] and bongkrekic diction to those using dihydrorhodamine-123 or acid [126] each only a¡orded a slight delay before HEt. It may, therefore, also be necessary to re-exam- the onset of delayed Ca2‡ deregulation (Castilho, ine the sequence of events discussed in the context on Budd and Nicholls, unpublished), although a more apoptotic signalling, where a decrease in vim is pro- e¡ective protection has been reported for hippocam- posed to initiate an increase in the generation of re- pal neurones [34]. active oxygen species, since this has been, in part, Even if the MPT is an essential stage in both apop- based upon experiments using HEt [112]. totic and necrotic cell death, there are many ques- tions which remain to be answered. Since increasing 3.10. The role of the mitochondrial permeability excitotoxic stress causes a shift from apoptosis to transition necrosis in a single cell preparation [51], this would suggest that the proportion of mitochondria within a While a consensus appears to be emerging that the neurone undergoing the transition might de¢ne the MPT, which can be readily observed with isolated fate of the cell: a small proportion swelling and re- mitochondria, occurs in intact cells as an obligatory leasing the proposed pro-apoptotic factors cyto- component of apoptotic and necrotic neuronal cell chrome c [127,128] and apoptosis-inducing factor death, we would wish to counsel caution. While there [112] might initiate apoptosis with the residual mito- is evidence in support of an activation of the MPT in chondria maintaining su¤cient ATP levels for apop- myocardial reperfusion injury [113^115] and during tosis to occur, while a more powerful insult would oxidative [116] and anoxic [117] stress of hepatocytes, disrupt the majority of mitochondria and lead to a role in glutamate-evoked neuronal excitotoxicity is death by ATP depletion and subsequent necrosis. currently more speculative. However, delayed Ca2‡ deregulation still occurs in Isolated brain mitochondria incubated in the pres- cells where mitochondria are prevented from synthe- ence of adenine nucleotides and Pi are able to accu- sizing ATP by oligomycin [31]. Secondly, with iso- mulate large amounts of Ca2‡ and to maintain a lated mitochondria the MPT occurs within seconds stable set-point [13]. Cyclosporin A is an e¡ective of Ca2‡ overload; in contrast delayed Ca2‡ deregu- inhibitor of the MPT in isolated liver mitochondria lation can occur hours after a transient glutamate 2‡ exposed to Ca and Pi (reviewed in [118]) but has exposure, when the mitochondria would have un-

BBABIO 44667 30-7-98 110 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 loaded their Ca2‡. Finally, the MPT as normally [8] J.M. Dubinsky, S.M. Rothman, J. Neurosci. 11 (1991) 2545^ investigated in isolated mitochondria is inhibited by 2551. Mg2‡, adenine nucleotides and low pH and is di¤- [9] R.J. White, I.J. Reynolds, J. Neurosci. 16 (1996) 5688^5697. [10] S. Rajdev, I.J. Reynolds, Neurosci. Lett. 162 (1993) 149^152. cult to see with NAD-linked substrates [129]. The [11] F. Zoccarato, D.G. Nicholls, Eur. J. Biochem. 127 (1982) conditions within the cytoplasm of a glutamate ex- 333^338. posed neurone would on each of these counts be [12] D.G. Nicholls, K.E.O. Aî kerman, Biochim. Biophys. Acta predicted to be resistant to the MPT and it is cur- 683 (1982) 57^88. rently unclear what factor would induce the transi- [13] D.G. Nicholls, I.D. Scott, Biochem. J. 186 (1980) 833^839. [14] I.D. Scott, K.E.O. Aî kerman, D.G. Nicholls, Biochem. J. 192 tion in situ. (1980) 873^880. [15] K.E.O. Aî kerman, D.G. Nicholls, Biochim. Biophys. Acta 645 (1981) 41^48. 4. Conclusions [16] S.A. Thayer, R.J. Miller, J. Physiol. (Lond.) 425 (1990) 85^ 115. The mechanism of glutamate-induced neuronal ne- [17] J.L. Werth, S.A. Thayer, J. Neurosci. 14 (1994) 346^356. [18] D.D. Friel, R.W. Tsien, J. Neurosci. 14 (1994) 4007^4024. crosis is far from understood. The mitochondrion [19] K.E.O. Aî kerman, D.G. Nicholls, Rev. Physiol. Biochem. occupies centre stage in three roles: as the prime Pharmacol. 95 (1983) 149^201. generator (and possible dissipator) of cellular ATP; [20] J. Herrington, Y.B. Park, D.F. Babcock, B. Hille, Neuron 16 as the main sink for the Ca2‡ accumulated via the (1996) 219^228. NMDA receptor and as the major source of poten- [21] M.R. Duchen, M. Valdeolmillos, S.C. O'Neill, D.A. Eisner, J. Physiol. (Lond.) 424 (1990) 411^426. tially excitotoxic reactive oxygen species. These three [22] R.J. White, I.J. Reynolds, J. Neurosci. 15 (1995) 1318^1328. roles are all interconnected, and we may have made a [23] R.J. White, I.J. Reynolds, J. Physiol. (Lond.) 498 (1997) 31^ contribution to unravelling the interactions by high- 47. lighting the role of mitochondrial Ca2‡ accumula- [24] D.G. Nicholls, S.J. Ferguson, Bioenergetics, Vol. 2, Academ- tion. ic Press, London, 1992. [25] D.G. Nicholls, Eur. J. Biochem. 50 (1974) 305^315. [26] J.B. Hoek, D.G. Nicholls, J.R. Williamson, J. Biol. Chem. 255 (1980) 1458^1464. Acknowledgements [27] I.D. Scott, D.G. Nicholls, Biochem. J. 186 (1980) 21^33. [28] M.R. Duchen, Biochem. J. 283 (1992) 41^50. SLB was supported by a Medical Research Coun- [29] V.P. Bindokas, R.J. Miller, J. Neurosci. 15 (1995) 6999^ cil studentship. The work was supported by a grant 7011. [30] N.K. Isaev, D.B. Zorov, E.V. Stelmashook, R.E. Uzbekov, from the Wellcome Trust. M.B. Kozhemyakin, I.V. Victorov, FEBS Lett. 392 (1996) 143^147. [31] S.L. Budd, D.G. Nicholls, J. Neurochem. 67 (1996) 2282^ References 2291. [32] J.H. Prehn, V.P. Bindokas, J. Jordan, M.F. Galindo, G.D. [1] M.F. Beal, N. Howell, I. Bodis-Wollner, Mitochondria and Ghadge, R.P. Roos, L.H. Boise, C.B. Thompson, S. Krajew- Free Radicals in Neurodegenerative Disease, Wiley-Liss, ski, J.C. Reed, R.J. Miller, Mol. Pharmacol. 49 (1996) 319^ New York, 1997. 328. [2] J.M. Dubinsky, B.S. Kristal, M. Elizondo-Fournier, J. Neu- [33] B.I. Khodorov, V. Pinelis, O. Vergun, T. Storozhevykh, N. rosci. 15 (1995) 7071^7078. Vinskaya, FEBS Lett. 397 (1996) 230^234. [3] M. Tymianski, M.P. Charlton, P.L. Carlen, C.H. Tator, [34] J. Keelan, O. Vergun, L. Patterson, M.R. Duchen, Soc. Neu- J. Neurosci. 13 (1993) 2085^2104. rosci. Abstr. 23 (1997) 895.1. [4] D.M. Hartley, M.C. Kurth, L. Bjerkness, J.H. Weiss, D.W. [35] L.M. Loew, W. Carrington, R.A. Tuft, F.S. Fay, Proc. Natl. Choi, J. Neurosci. 13 (1993) 1993^2000. Acad. Sci. U.S.A. 91 (1994) 12579^12583. [5] S. Eimerl, M. Schramm, J. Neurochem. 62 (1994) 1223^ [36] A.L. Nieminen, T.G. Petrie, J.J. Lemasters, W.R. Selman, 1226. Neuroscience 75 (1996) 993^997. [6] R. Sattler, M.P. Charlton, M. Hafner, M. Tymianski, Soc. [37] A.F. Schinder, E.C. Olson, N.C. Spitzer, M. Montal, J. Neu- Neurosci. Abstr. 23 (1997) 894.4. rosci. 16 (1996) 6125^6133. î [7] K. Hyrc, S.D. Handran, S.M. Rothman, M.P. Goldberg, [38] K.E.O. Akerman, Biochim. Biophys. Acta 502 (1978) 359^ J. Neurosci. 17 (1997) 6669^6677. 366.

BBABIO 44667 30-7-98 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112 111

[39] G.M. Heaton, D.G. Nicholls, Biochem. J. 156 (1976) 635^ [64] S.A. Loddick, A. MacKenzie, N.J. Rothwell, Neuroreport 7 646. (1996) 1465^1468. [40] D.G. Nicholls, Biochem. J. 170 (1978) 511^522. [65] F.J. Gottron, H.S. Ying, D.W. Choi, Mol. Cell. Neurosci. 9 [41] C.S. Rossi, A.L. Lehninger, J. Biol. Chem. 239 (1964) 3971^ (1997) 159^169. 3980. [66] R.C. Armstrong, T.J. Aja, K.D. Hoang, S. Gaur, X. Bai, [42] D.G. Nicholls, Biochem. J. 176 (1978) 463^474. E.S. Alnemri, G. Litwack, D.S. Karanewsky, L.C. Fritz, [43] D.G. Nicholls, M. Crompton, FEBS Lett. 111 (1980) 261^ K.J. Tomaselli, J. Neurosci. 17 (1997) 553^562. 268. [67] L. Kiedrowski, G. Brooker, E. Costa, J.T. Wroblewski, Neu- [44] M.R. Duchen, T.J. Biscoe, J. Physiol. (Lond.) 450 (1992) 33^ ron 12 (1994) 295^300. 61. [68] M. Erecinska, D. Nelson, J. Deas, I.A. Silver, Brain Res. 726 [45] S.L. Budd, R.F. Castilho, D.G. Nicholls, FEBS Lett. 415 (1996) 153^159. (1997) 21^24. [69] J.G. Greene, J.T. Greenamyre, Prog. Neurobiol. 48 (1996) [46] C.D. Benham, M.L. Evans, C.J. McBain, J. Physiol. (Lond.) 613^621. 455 (1992) 567^583. [70] R.C. Henneberry, in: M.F. Beal, N. Howell, I. Bodis-Woll- [47] J.L. Werth, Y.M. Usachev, S.A. Thayer, J. Neurosci. 16 ner (Eds.), Mitochondria and Free Radicals in Neurodegen- (1996) 1008^1015. erative Disease, Wiley-Liss, New York, 1997, pp. 111^143. [48] M. Glinn, B. Ni, R.P. Irwin, S.W. Kelley, S.-Z. Lin, S.M. [71] F. Blandini, R.H.P. Porter, J.T. Greenamyre, Mol. Neuro- Paul, Soc. Neurosci. Abstr. 23 (1997) 895.11. biol. 12 (1996) 73^94. [49] J.M. Dubinsky, J. Neurosci. 13 (1993) 623^631. [72] E. Brouillet, P. Hantraye, R.J. Ferrante, R. Dolan, A. Leroy [50] J.M. Dubinsky, B.S. Kristal, M. Elizondo-Fournier, Neuro- Willig, N.W. Kowall, M.F. Beal, Proc. Natl. Acad. Sci. pharmacology 34 (1995) 701^711. U.S.A. 92 (1995) 7105^7109. [51] M. Ankarcrona, J.M. Dypbukt, E. Bonfoco, B. Zhivotov- [73] A. Novelli, J.A. Reilly, P.G. Lysko, R.C. Henneberry, Brain sky, S. Orrenius, S.A. Lipton, P. Nicotera, Neuron 15 (1995) Res. 451 (1988) 205^212. 961^973. [74] A.M. Marini, J.P. Schwartz, I.J. Kopin, J. Neurosci. 9 [52] B.I. Khodorov, D.A. Fayuk, S.G. Koshelev, O.V. Vergun, (1989) 3665^3672. V.G. Pinelis, N.P. Vinskaya, T.P. Storozhevykh, E.N. Ar- [75] P.G. Lysko, J.A. Cox, M.A. Vigano, R.C. Henneberry, senyeva, L.G. Khaspekov, A.P. Lyzhin, N. Isaev, I.V. Vic- Brain Res. 499 (1989) 258^266. torov, J.M. Dubinsky, Int. J. Neurosci. 88 (1996) 215^241. [76] J.G. Greene, J.T. Greenamyre, J. Neurochem. 64 (1995) [53] T. Nitatori, N. Sato, S. Waguri, Y. Karasawa, H. Araki, K. 2332^2338. Shibanai, E. Kominami, Y. Uchiyama, J. Neurosci. 15 [77] R. Henshaw, B.G. Jenkins, J.B. Schulz, R.J. Ferrante, N.W. (1995) 1001^1011. Kowall, B.R. Rosen, M.F. Beal, Brain Res. 647 (1994) 161^ [54] M. Ankarcrona, B. Zhivotovsky, T. Holmstro«m, A. Diana, 166. J.E. Eriksson, S. Orrenius, P. Nicotera, Neuroreport 7 (1996) [78] M.F. Beal, E. Brouillet, B. Jenkins, R. Henshaw, B. Rosen, 2659^2664. B.T. Hyman, J. Neurochem. 61 (1993) 1147^1150. [55] F. Dessi, C. Charriaut-Marlangue, M. Khrestchatisky, Y. [79] R.A. Kauppinen, D.G. Nicholls, J. Neurochem. 47 (1986) Ben-Ari, J. Neurochem. 60 (1993) 1953^1955. 1864^1869. [56] J. Ikeda, S. Terakawa, S. Murota, I. Morita, K. Hirakawa, [80] A. Schurr, R.S. Payne, J.J. Miller, B.M. Rigor, Brain Res. J. Neurosci. Res. 43 (1996) 613^622. 744 (1997) 105^111. [57] B.J. Gwag, J.Y. Koh, J.A. Demaro, H.S. Ying, M. Jacquin, [81] M. Villalba, A. Mart|¨nez-Serrano, P. Go¨mez-Puertas, P. D.W. Choi, Neuroscience 77 (1997) 393^401. Blanco, C. Bo«rner, A. Villa, M. Casado, C. Gime¨nez, R. [58] E. Bonfoco, D. Krainc, M. Ankarcrona, P. Nicotera, S.A. Pereira, E. Bogo¨nez, T. Pozzan, J. Satrustegui, J. Biol. Lipton, Proc. Natl. Acad. Sci. U.S.A. 92 (1995) 7162^7166. Chem. 269 (1994) 2468^2476. [59] Z. Pang, J.W. Geddes, J. Neurosci. 17 (1997) 3064^3073. [82] C. Vicario, C. Arizmendi, G. Malloch, J.B. Clark, J.M. [60] B.H. Ni, X. Wu, Y.S. Du, Y. Su, E. Hamilton-Byrd, P.K. Medina, J. Neurochem. 57 (1991) 1700^1707. Rockey, P. Rosteck Jr., G.G. Poirier, S.M. Paul, J. Neuro- [83] S. Eimerl, M. Schramm, J. Neurochem. 65 (1995) 739^743. sci. 17 (1997) 1561^1569. [84] T. Tabatabaie, J.D. Potts, R.A. Floyd, Arch. Biochem. Bio- [61] R.P. Xiao, H.H. Valdivia, K. Bogdanov, C. Valdivia, E.G. phys. 336 (1996) 290^296. Lakatta, H.P. Cheng, J. Physiol. (Lond.) 500 (1997) 343^ [85] M. Patel, B.J. Day, J.D. Crapo, I. Fridovich, J.O. McNa- 354. mara, Neuron 16 (1996) 345^355. [62] J.B. Schulz, M. Weller, T. Klockgether, J. Neurosci. 16 [86] G.P. Davey, J.B. Clark, J. Neurochem. 66 (1996) 1617^1624. (1996) 4696^4706. [87] J.P. Sheehan, R.H. Swerdlow, W.D. Parker, S.W. Miller, [63] H. Hara, R.M. Friedlander, V. Gagliardini, C. Ayata, K. R.E. Davis, J.B. Tuttle, J. Neurochem. 68 (1997) 1221^1233. Fink, Z.H. Huang, M. Shimizu-Sasamata, J.Y. Yuan, [88] J.P. Sheehan, R.H. Swerdlow, S.W. Miller, R.E. Davis, J.K. M.A. Moskowitz, Proc. Natl. Acad. Sci. U.S.A. 94 (1997) Parks, W.D. Parker, J.B. Tuttle, J. Neurosci. 17 (1997) 2007^2012. 4612^4622.

BBABIO 44667 30-7-98 112 D.G. Nicholls, S.L. Budd / Biochimica et Biophysica Acta 1366 (1998) 97^112

[89] A. Atlante, S. Gagliardi, G.M. Minervini, E. Marra, S. [110] N.W. Kooy, J.A. Royall, H. Ischiropoulos, J.S. Beckman, Passarella, P. Calissano, Neuroreport 7 (1996) 2519^2523. Free Radic. Biol. Med. 16 (1994) 149^156. [90] M.J. Courtney, J.J. Lambert, D.G. Nicholls, J. Neurosci. [111] L.M. Henderson, J.B. Chappell, Eur. J. Biochem. 217 10 (1990) 3873^3879. (1993) 973^980. [91] T.-I. Peng, J.T. Greenamyre, Soc. Neurosci. Abstr. 23 [112] P.X. Petit, S.A. Susin, N. Zamzami, B. Mignotte, G. (1997) 685.19. Kroemer, FEBS Lett. 396 (1996) 7^13. [92] B.I. Khodorov, V.G. Pinelis, T. Storozhevykh, O.V. Ver- [113] R.A. Altschuld, C.M. Hohl, L.C. Castillo, A.A. Garleb, gun, N.P. Vinskaya, FEBS Lett. 393 (1996) 135^138. R.C. Starling, G.P. Brierley, Am. J. Physiol. 262 (1992) [93] A.K. Stout, H.M. Raphael, J.M. Scanlon, I.J. Reynolds, H1699^H1704. Soc. Neurosci. Abstr. 23 (1997) 650.6. [114] E.J. Gri¤ths, A.P. Halestrap, J. Mol. Cell Cardiol. 25 [94] S. Tan, Y. Sagara, D. Schubert, Soc. Neurosci. Abstr. 23 (1993) 1461^1469. (1997) 542.1. [115] E.J. Gri¤ths, A.P. Halestrap, Biochem. J. 307 (1995) 93^ [95] J.J. Vornov, J. Neurochem. 65 (1995) 1681^1691. 98. [96] V.P. Skulachev, Q. Rev. Biophys. 29 (1996) 169^202. [116] G.E. Kass, M.J. Juedes, S. Orrenius, Biochem. Pharmacol. [97] A. Boveris, N. Oshino, B. Chance, Biochem. J. 128 (1972) 44 (1992) 1995^2003. 617^630. [117] J.G. Pastorino, J.W. Snyder, J.B. Hoek, J.L. Farber, Am. [98] A. Herrero, G. Barja, J. Bioenerg. Biomembr. 29 (1997) J. Physiol. 268 (1995) C676^85. 241^249. [118] P. Bernardi, V. Petronilli, J. Bioenerg. Biomembr. 28 (1996) [99] J.A. Dykens, J. Neurochem. 63 (1994) 584^591. 131^138. [100] I.J. Reynolds, T.G. Hastings, J. Neurosci. 15 (1995) 3318^ [119] B.S. Kristal, J.M. Dubinsky, J. Neurochem. 69 (1997) 524^ 3327. 538. [101] M. Cino, R.F. Del Maestro, Arch. Biochem. Biophys. 269 [120] M. Ankarcrona, J.M. Dypbukt, S. Orrenius, P. Nicotera, (1989) 623^638. FEBS Lett. 394 (1996) 321^324. [102] G.J. Wang, R.D. Randall, S.A. Thayer, J. Neurophysiol. 72 [121] R.G. Victor, G.D. Thomas, E. Marban, B. O'Rourke, (1994) 2563^2569. Proc. Natl. Acad. Sci. U.S.A. 92 (1995) 6269^6273. [103] Z. Hartley, J.M. Dubinsky, J. Neurosci. 13 (1993) 4690^ [122] Y.F. Lu, K. Tomizawa, A. Moriwaki, Y. Hayashi, M. To- 4699. kuda, T. Itano, O. Hatase, H. Matsui, Brain Res. 729 [104] J.M. Scanlon, E. Aizenman, I.J. Reynolds, Eur. J. Pharma- (1996) 142^146. col. 326 (1997) 67^74. [123] J.W. McDonald, M.I. Behrens, C. Chung, T. Bhattachar- [105] N. Zamzami, P. Marchetti, M. Castedo, D. Decaudin, A. yya, D.W. Choi, Brain Res. 759 (1997) 228^232. Macho, T. Hirsch, S.A. Susin, P.X. Petit, B. Mignotte, G. [124] J.W. McDonald, M.P. Goldberg, B.J. Gwag, S.I. Chi, Kroemer, J. Exp. Med. 182 (1995) 367^377. D.W. Choi, Ann. Neurol. 40 (1996) 750^758. [106] M. Castedo, A. Macho, N. Zamzami, T. Hirsch, P. Mar- [125] A. Nicolli, E. Basso, V. Petronilli, R.M. Wenger, P. Ber- chetti, J. Uriel, G. Kroemer, Eur. J. Immunol. 25 (1995) nardi, J. Biol. Chem. 271 (1996) 2185^2192. 3277^3284. [126] N. Brustovetsky, M. Klingenberg, Biochemistry 35 (1996) [107] A. Macho, M. Castedo, P. Marchetti, J.J. Aguilar, D. De- 8483^8488. caudin, N. Zamzami, P.M. Girard, J. Uriel, G. Kroemer, [127] J. Yang, X.S. Liu, K. Bhalla, C.N. Kim, A.M. Ibrado, J.Y. Blood 86 (1995) 2481^2487. Cai, T.I. Peng, D.P. Jones, X.D. Wang, Science 275 (1997) [108] V.P. Bindokas, J. Jordan, C.C. Lee, R.J. Miller, J. Neuro- 1129^1132. sci. 16 (1996) 1324^1336. [128] R.M. Kluck, E. Bossy-Wetzel, D.R. Green, D.D. New- [109] L.L. Dugan, S.L. Sensi, L.M.T. Canzoniero, S.D. Handran, meyer, Science 275 (1997) 1132^1136. S.M. Rothman, T.S. Lin, M.P. Goldberg, D.W. Choi, [129] P. Bernardi, K.M. Broekemeier, D.R. Pfei¡er, J. Bioenerg. J. Neurosci. 15 (1995) 6377^6388. Biomembr. 26 (1994) 509^518.

BBABIO 44667 30-7-98