Mitochondrial transmembrane ion distribution during anoxia

0 downloads 0 Views 2MB Size Report
species may be inhibited during anoxia. The results are dis- cussed in terms of potential regulation of mitochondrial func- tion to provide a quiescent anoxic state ...
Mitochondrial transmembrane during anoxia TAK YEE Department

AW, BO S. ANDERSSON, AND DEAN P. JONES of Biochemistry, Emory University School of Medicine, Atlanta,

Aw, TAK YEE, Bo S. ANDERSSON, AND DEAN P. JONES. Mitochondrial transmembrane ion distribution during anoxia. Am. J. Physiol. 252 (Cell Physiol. 21): C3564361, 1987.-The distribution of pyruvate, phosphate, malate, citrate, K+, aspartate, glutamate, ADP, and ATP between the mitochondrial and cytosolic compartments was studied in isolated rat hepatocytes exposed to 30 min anoxia. The results show that pyruvate and citrate gradients are comparable to aerobic values, indicating that the pH gradient across the membrane under anaerobic conditions is comparable to that under normal aerobic conditions. In contrast, the distribution of phosphate, malate, ATP, ADP, aspartate, and glutamate suggests that transport of these species may be inhibited during anoxia. The results are discussed in terms of potential regulation of mitochondrial function to provide a quiescent anoxic state that is capable of recovering normal function on reoxygenation. pyruvate; malate; citrate; phosphate; aspartate; glutamate; tassium; adenosine triphosphate; adenosine diphosphate

po-

CHEMIOSMOTIC HYPOTHESIS for coupling of oxidative phosphorylation, electron transport is obligately coupled to movement of protons out of the mitochondrial matrix (21). This process establishes the protonmotive force (Ap) which consists of a chemical (ApH) and an electrical component (A$) as related by the equation, Ap = A$ - ZApH, where Z is a factor for conversion of ApH to millivolts. The protonmotive force provides energy for ATP synthesis by obligate coupling of movement of H+ through the membrane (down its electrochemical gradient) with the phosphorylation of ADP. This coupling mechanism requires that the mitochondrial inner membrane be highly impermeable to charged species and that specific solute porter systems be present to maintain osmotic stability and provide for exchange of substrates and products across the membrane (17,Zl). These transport processes are remarkable in that they utilize the components of Ap, i.e., A$J and ApH, to drive the accumulation of precursors and extrusion of the product, ATP (14, 17, 30). For instance, pyruvate and phosphate uptake each occur by electroneutral processes where movement of the anionic species is balanced by either proton symport or hydroxide antiport mechanisms (17, 20). These processes normally result in maintenance of -2- to s-fold higher concentrations of pyruvate (31) and 5- to lo-fold higher concentrations of phosphate in the matrix than in the cytosol (1, 31). The phosphate concentration gradient can provide the driving force for accumulation of dicarboxylates transported by electroneutral antiporters (17, 20). Accumulation of tricarboxIN THE

C356

ion distribution

0363-6143/87

$1.50

Georgia 30322

ylates is dependent on ApH due to a proton-coupled electroneutral exchange for dicarboxylates (l7,20). Thus the transmembranal pH gradient is essential for optimal oxidative phosphorylation not only to drive the ATP synthase, but also indirectly to supply pyruvate and phosphate and to maintain high matrix concentrations of citric acid cycle intermediates. The membrane potential (A$) also can be used directly for phosphorylation as a driving force for the ATP synthase and indirectly in supply of the precursor ADP and elimination of product ATP. Under physiological conditions, ATP is more negatively charged than ADP and the A+ (negative inside) provides a driving force to lower matrix ATP in exchange for accumulation of ADP. Klingenberg (13) concluded that a large portion of the total energy used for oxidative phosphorylation is consumed by this process. During anoxia, electron transport ceases and mitochondria consequently lose the driving force for maintenance of ApH and A$. However, estimates of ApH from the distribution of the weak-acid dimethadione (DMO) and A# from the distribution of triphenylmethylphosphonium ion (TPMP) in hepatocytes exposed to anoxia for 30 min indicate that ApH and A#Jare largely preserved (2, 3). Because many ions are distributed across the inner membrane according to ApH and A#, measurement of ion distribution can provide information on the maintenance of the protonmotive force during anoxia and the mechanisms involved. In addition, since maintenance of precursor pools is essential for mitochondrial function, loss of normal ion distribution may contribute to the irreversible loss of mitochondrial function in prolonged anoxia. We therefore studied the distribution of various ions between the mitochondrial and cytosolic compartments in isolated hepatocytes. The condition chosen for these experiments involved incubation for 30 min under anoxia. This duration of anoxia does not result in cell death and therefore allows study of anoxic changes without complications due to disruption of cells. Cells were fractionated by a method involving digitonin treatment and centrifugation (3, lo), with KCN added to the anaerobic fractionations to prevent a reoxygenation artifact (3). Studies were performed to determine the effect of anoxia on distribution of pyruvate, phosphate, malate, citrate, aspartate, glutamate, K+, ATP, and ADP. Results on pyruvate and citrate confirm that ApH is largely retained during 30 min anoxia, whereas results on phosphate, ATP and ADP indicate that carrier mechanisms for these

Copyright 0 1987 the American Physiological Society

MITOCHONDRIAL

ION

DISTRIBUTION

species are inhibited. The results therefore indicate that inhibition of ion flux occurs under anoxic conditions where All/ and ApH are largely preserved. MATERIALS

AND METHODS

Materials. Collagenase (type IV), digitonin, rotenone, oligomycin, antimycin A, and N-Z-hydroxyethylpiperazine-N’-2ethanesulfonic acid (HEPES) were obtained from Sigma Chemical (St. Louis, MO). Carbonylcyanidep-trifluoromethoxyphenylhydrazone (FCCP) was purchased from Boehringer-Mannheim (Mannheim, FRG). Silicone oil, density 1.05 kg/l from Aldrich Chemical (Milwaukee, WI) and white light paraffin oil, viscosity X25/135, from Fisher Scientific (Pittsburgh, PA) were mixed in a ratio of 6:l. Ultrahigh-purity argon (95%. Incubations ( lo6 cells/ ml) were performed in rotating round-bottom flasks in Krebs-Henseleit buffer containing 25 mM HEPES, pH 7.2 at 37°C under steady-state atmospheres of air or argon as previously described (5). The buffer was preequilibrated for 30 min before addition of hepatocytes and the incubations were carried out for a further 30 min. In some experiments, the incubations also contained FCCP (I PM), antimycin A (1.12 PM), oligomycin (5 pg/ml), or rotenone (50 PM). Digitonin fractionation was performed as previously described with 1 mM KCN included in the fractionation buffer to prevent reoxygenation artifacts (3). Volume measurements (3) showed that very small changes occurred under most conditions used. Thirty minutes of anoxia had no significant effect on cellular or mitochondrial volumes. Addition of FCCP resulted in a small extent of swelling, whereas valinomycin resulted in extensive swelling. Values are expressed as concentrations after correction for volumes of the respective fractions. The adenine nucleotides were measured by high-performance liquid chromatography (11). Aliquots of the perchloric acid extracts were neutralized with 10 M KOH and the supernatants used for adenylate determinations. ATP and ADP were separated on a lo-pm Cl8 reversedphase column and were detected by absorbance at 260 nm. Values are expressed as nanomoles per lo6 cells. Inorganic phosphate was measured in the acid extracts by the calorimetric assay of Fiske and Subbarow at 660 nm (8). To minimize hydrolysis of acid labile phosphates, samples were kept on ice until assay, and all measurements of Pi were performed on the same day of the experiment. Pyruvate, citrate, and malate contents were determined spectrophotometrically by pyridine nucleotidelinked, enzyme-coupled assays. Pyruvate was determined by the method of Lamprecht and Heinz (15), and citrate and malate were measured according to Mollering (24, 25). Glutamate and aspartate were quantitated by highperformance liquid chromatography as previously described by Reed et al. (29). The acid extracts were deriv-

DURING

ANOXIA

c357

atized with l-fluoro-2,4-dinitrobenzene and chromatographed on a 5-pm amine column (Beckmann Instruments). The DNP derivatives of the amino acids were detected by absorbance at 365 nm and were quantitated relative to standards by integration. Potassium and sodium ions were measured using atomic absorption spectrometry. The amounts of K+ and Na+ in the samples were quantitated by comparing them with standard curves that were constructed using different concentrations of the respective ions. Concentrations for all ions and metabolites were estimated using 5 pl/ lo6 cells for total cell volume, 4 p1/106 cells for cytosolic volume, and 1 p1/106 cells for mitochondrial volume. RESULTS

Previous studies have shown that digitonin fractionation provides a useful technique for rapidly separating the mitochondrial compartment of isolated hepatocytes (3, 10). Cytosolic values can be calculated as the difference between digitonin-treated and untreated cells. Inclusion of 1 mM KCN during the fractionation procedure effectively prevents a reoxygenation artifact for anaerobic samples (3) and, thus, allows comparison of mitochondrial and extramitochondrial (cytosolic) ion contents. Values are expressed as concentrations, calculated from independent measurements of volume and ion content (3). Pyruuate. Mitochondrial uptake of pyruvate occurs by a proton-compensated electroneutral mechanism (17, 20). Under aerobic conditions, the concentration gradient is about two- to fivefold (Table 1; Ref. 5). Addition of the protonophore FCCP eliminates the gradient and supports the interpretation that pyruvate uptake in hepatocytes occurs by a H+-coupled mechanism. During anoxia, the total cellular pyruvate decreasesslightly, but the ratio of concentrations in the mitochondrial and cytosolic fractions is maintained at about five (Table 1). Maintenance of this ratio during anoxia could mean that the ApH is retained or that redistribution of pyruvate is inhibited during anoxia. To distinguish between these possibilities, the effect of FCCP on pyruvate distribution during anoxia was measured. Under this condition, the pyruvate gradient was about twofold, i.e., substantially smaller than the fivefold gradient under control aerobic and anaerobic conditions (Table 1). Thus the results indicate that a substantial mitochondrial transmembranal ApH is retained under anaerobic conditions. Phosphate. Mitochondrial uptake of phosphate occurs 1. Transmembranal distribution of pyruvate during anoxia _-TABLE

Pyruvate,

mM

Conditions Total

Air

Mitochondria

Cytosol

Mitochondriall Cytosolic Ratio

1.60k0.29 0.32kO.05 5.0 0.57_to.13 0.46_tO.l0 1.2 1.48k0.25 0.25kO.05 5.9 0.31t0.06 0.53kO.03 0.24t0.04 2.2 are means k SE for 4 cell preparations. Cells (106/ml)

+ FCCP Argon + FCCP

0.57kO.09

0.48t0.10 0.50~0.08

Values were incubated for 30 min under ultrahigh purity argon or air in presence or absence of 1 PM carbonylcyanide-p-trifluoromethoxyphenylhydrazone (FCCP) and fractionated and assayed as described in MATERIALS ANDMETHODS.

C358

MITOCHONDRIAL

ION

DISTRIBUTION

by a proton-compensated electroneutral process (17). The distribution of phosphate among the various ionic species, principally HPOi- and H,PO,, differs between the matrix and cytosol because of the different pH values. In addition, some of the phosphate is bound with metal ions and, therefore, is not a “free” phosphate. These various forms are not distinguished by the Fiske-Subbarow method, so that the values we obtained are the total noncovalently bound phosphate. Measurements of total phosphate in the deproteinated extracts of digitonin-fractionated cells show that the phosphate concentration in the matrix is -4.5 times that in the cytosol (Table 2). Treatment with 1 PM FCCP decreases the ratio of mitochondrial to cytosolic phosphate to about two (Table Z), consistent with the concept that ApH contributes to maintenance of the phosphate gradient. Under anaerobic conditions, utilization of ATP without equivalent rephosphorylation of ADP results in a nearly fourfold increase in cytosolic phosphate (Table 2). However, the mitochondrial phosphate increases only 20% (Table 2). Since phosphate is not being utilized for ATP synthesis, these results indicate that the phosphate uptake system is not functioning under anaerobic conditions. Since respiratory inhibitors, such as rotenone, antimycin A, and oligomycin, also decrease phosphorylation and lead to an accumulation of cellular phosphate, we examined the effects of these inhibitors on the mitochondrial transmembranal phosphate gradient. All three resulted in substantial increases in cytosolic phosphate with only a minor increase in mitochondrial phosphate (Table 2). Since the ApH in the presence of oligomycin is comparable to the aerobic control, these results support the concept that mechanisms exist to limit phosphate uptake during periods of interrupted ATP synthesis. Such inhibition of phosphate transport is consistent with retention of the ApH during anoxia (2) and could function in cells to prevent a massive accumulation of phosphate (e.g., up to 50 mM), which could result in swelling and loss of mitochondrial integrity. 1MaLate. Malate is exchanged across the mitochondrial inner membrane by at least three mechanisms (17). In liver, entry of malate occurs by an electroneutral exchange for phosphate. Under normal aerobic conditions, the phosphate gradient can thereby drive the accumulation of malate. Malate exchanges for other dicarboxylates by another electroneutral system, thus providing a mechanism to concentrate Zoxoglutarate and other citric acid TABLE

~~

DURING

ANOXIA

cycle intermediates in the matrix and allowing efflux of malate. Malate efflux also occurs in exchange for citrate and isocitrate by a proton-compensated electroneutral mechanism. This latter mechanism allows the proton gradient to drive the efflux of malate. Under aerobic conditions, the ratio of malate in the mitochondrial and cytosolic fractions is about three (Table 3). Treatment of cells with 1 PM FCCP decreases the ratio to about one (Table 3), presumably because of the importance of the ApH in maintenance of the malate gradient. After 30 min anoxia, the total cellular malate concentrations decreased over fourfold and the mitochondrial to cytosolic malate ratio fell to 1.7 (Table 3). Because of the loss of the phosphate gradient under this condition (see above), this result is consistent with the importance of the phosphate gradient in maintaining the malate gradient. However, with the large decrease in total malate, other factors may also be involved. Citrate. Uptake occurs by proton-compensated electroneutral exchange with malate as described above. Under aerobic conditions, both the malate gradient and the proton gradient contribute to maintenance of a mitochondrial-to-cytosolic ratio for citrate of X0 (Table 4). Addition of 1 PM FCCP decreases the ratio to less than three, supporting the concept that this accumulation is dependent on ApH. After 30 min anoxia, the ratio for citrate is nearly 10 (Table 4). Since the phosphate gradient (the driving force for malate accumulation) is lost, the malate gradient cannot function over a long term in maintenance of the citrate gradient. Consequently, these results indicate that ApH is the primary driving force for the large ratio of mitochondrial citrate relative to cytosolic citrate. The maintenance of a large citrate ratio therefore is consistent with the conclusion that the ApH is largely retained during 30 min anoxia. Potassium ion. Experiments were performed to determine the effect of anoxia on mitochondrial K+ and Na’. Both are of interest because with the existing mitochondrial membrane potential and cytosolic concentrations of these ions, an increased permeability could lead to massive accumulation with resultant mitochondrial swelling. However, technical problems associated with the removal of extracellular Na+ precluded accurate measurement of mitochondrial Na+ and these results are not included. Mitochondria are typically highly impermeable to K+, and the matrix K+ is maintained at a concentration comparable to the cytosolic value [142 t 6 mM vs. 171

2. Transmembranal distribution of inorganic phosphate during anoxia Inorganic Conditions

Air + FCCP (1 pM) + Oligomycin (5 pg/ml) + Rotenone (50 PM) + Antimycin A (1.12 PM) Argon

Phosphate,

mM Cytosol

Mitochondrial/ Cytosolic Ratio

2.8kO.5 4.6k0.6 7.9zk1.2 7.9t0.5 7.5kO.7 11.2k1.2

4.5 2.0 2.4 1.9 1.8 1.4

n Total

5 3 3 3 3 4

Mitochondria

4.7kO.3 5.5t0.6 9.9t0.7 9.OkO.5 8.5t0.6 11.7t1.0

Values are means k SE; n, no. of cell preparations. Hepatocytes (lo6 cells/ml) and fractionated and assayed as described in MATERIALS AND METHODS. Inhibitor trifluoromethoxyphenylhydrazone.

12.6k0.6 9.0t1.0 18.7k2.2 14.6k2.3 13.3t0.2 15.5k1.5 were incubated concentrations

for 30 min under air or ultrahigh purity argon were as indicated. FCCP, carbonylcyanide-p-

MITOCHONDRIAL

3. Transmembranal during anoxia

distribution

TABLE

Malate,

ION

of malate

mM

Mitochondrial/ Cytosolic Ratio

Conditions Total

Air + FCCP Argon + FCCP

Mitochondria

0.96 0.30 0.21 0.06

Cytosol

0.68 0.29 0.19 0.07

2.10

0.36 0.32 ND

DISTRIBUTION

3.1 1.2 1.7

Values given are average of triplicate measurements from 2 cell preparations. Hepatocytes (lo6 cells/ml) were incubated for 30 min under air or argon in presence or absence of 1 PM carbonylcyanide-ptrifluoromethoxyphenylhydrazone (FCCP). Samples were fractionated and malate contents assayed as described in MATERIALS AND METHODS. ND, not detectable. TABLE 4. Transmembranal distribution of citrate during anoxia Citrate,

mM

Mitochondrial/ Cytosolic Ratio

Conditions Total

Air + FCCP Argon + FCCP

Mitochondria

2.45k0.11 1.15kO.14 0.94kO.26 0.66kO.40

0.65+0.03 0.57t0.02 0.26kO.07 0.32kO.l

Cytosol

0.2t0.01 0.42t0.02 0.10~0.02 0.23kO.02

12.3 2.7 9.9 2.9

Values are means t SE average of triplicate determinations from 3 cell preparations. Cells (106/ml) were incubated for 30 min under air or argon in absence or presence of 1 PM carbonylcyanide-ptrifluoromethoxyphenylhydrazone (FCCP). Digitonin fractionation of samples and citrate measurements were performed as described in MATERIALSAND

METHODS.

TABLE 5. Transmembranal K+ distribution during anoxia K+, mM Mitochondria

Cytosol

Mitochondrial/ Cytosolic Ratio

142t6 122t27

171t4 148t45

0.83 0.82

Conditions

_----

Air Argon

Values are means k SE cells/ml) were incubated for of samples were performed K’ was measured by atomic

for 3 cell preparations. 30 min under air or argon as described in MATERIALS absorption spectrometry.

Hepatocytes (lo6 and fractionation AND

METHODS.

+4mM(n = 3), respectively; Table 51 even though the membrane potential is sufficient to maintain an equilibrium matrix concentration of X0 M. A complete explanation of this phenomenon is not available since the energetics of the system that drives the efflux of K+ (an electroneutral K+/H+ antiporter) are not sufficient to prevent K+ accumulation in the mitochondria. During anoxia, the mitochondrial K+ was 122 t 27 mM compared with the cytoplasmic value of 148 t 45 mM (n = 3; Table 5). Thus 30 min anoxia has little effect on the mitochondrial K+ content. The K+-H+ antiporter is voltage gated and does not function when All/ falls below 140 mV (negative inside; Ref. 6). Since the mitochondrial AI) falls to below this value during 30 min anoxia, it appears likely that this effux mechanism for K+ is inhibited by anoxia. Thus the results indicate that the impermeability of the mitochondrial inner membrane to K+ is preserved during 30 min anoxia. Glutamate-aspartate exchange. Glutamate uptake can occur by a proton-compensated electroneutral process or by an electrophoretic process coupled to aspartate efflux

DURING

c359

ANOXIA

(30). The glutamate-aspartate exchange is an essential component of the malate-aspartate shuttle mechanism for transporting reducing equivalents from cytosolic NADH into the matrix. Because the cytosolic NADHto-NAD+ ratio is less than the mitochondrial ratio, energy input is required for net transfer into the mitochondria. This is provided by coupling of proton movement into the matrix with glutamate in exchange for movement of aspartate out. The coupling makes the net transfer electrophoretic, and this means that energy input from the protonmotive force is available to drive the flux of reducing equivalents into the matrix (26). Under aerobic conditions in hepatocytes, the mitochondrial-to-cytosolic ratio for glutamate is -1.4 (Table 6), a value considerably less than that expected from the ApH and A+. The ratio for aspartate is -1.05, a value much greater than that expected. However, since these are central intermediates in metabolism, their absolute concentrations depend on many interacting factors (16)) and the values obtained here are similar to other reports (31, 34). During anoxia, the ratio for glutamate declines and that for aspartate increases, thus departing from predicted values even more than under aerobic conditions. Three interpretations are consistent with these results, but the correct one cannot be distinguished at present. Firstly, the measured pools in this study may not represent the pools that are interacting with or accessible to the transporter (7). Secondly, under both aerobic and anaerobic conditions, the kinetics of metabolism and transport may be such that metabolism dominates in determining the concentrations in the different compartments. Finally, accessibility of the pools to the transporter, metabolism, and kinetics of the transporter may contribute significantly to relative concentrations under both conditions, but the relative significance of these factors may vary under the different conditions. Thus the greater deviation of the distributions during anoxia from that expected due to ApH and A# may be the result of inhibition of the transport systems. Such inhibition would appear to be necessary to prevent collapse of ApH and A# due to net ion movement in the absence of maintaining energy input from electron transport processes. ATP/ADP. Exchange of ATP and ADP across the 6. Transmembranal distribution of glutamate and aspartate during anoxia

TABLE

Conditions

Glutamate Air Argon Aspartate Air Argon

Glutamate Aspartate,

and mM

Mitochondrial/ Cytosolic Ratio

Mitochondria

Cytosol

9.2&l.. 5.3t0.8

6.4k0.2 4.9t0.5

1.4

2.0-1-0.1 1.8kO.l

1.9kO.l 0.6kO.l

1.1

-~

1.1

3.0

Values are means k SE for 4 cell preparations. Incubations (lo6 cells/ml were performed for 30 min under air or argon as described in MATERIALS AND METHODS. Samples were fractionated and derivatized with iodoacetate and l-fluoro-2,4-fluorodinitrobenzene. The DNP derivatives of glutamate and aspartate were determined by high-performance liquid chromatography (29).

C360 TABLE

MITOCHONDRIAL

7. Mitochondrial

ION

DISTRIBUTION

ADP

ATP

23.Ok3.4 3.6k0.3 are means t SE given

Air Argon

Values were performed Concentrations

as described for respective

ANOXIA

and cytosolic ATP and ADP contents during anoxia Total Cellular

Conditions

DURING

4.8tl.O 6.6t0.4

Mitochondria ATP/ADP

4.8 0.54

ATP

ADP

3.7t0.7 2.1k0.3

2.1kO.2 2.7k0.2

Cytosol ATP/ADP 1.8 0.81

ATP

ADP

19.3k3.3 1.5t0.3

2.7kO.9 3.9t0.3

ATP/ADP

7.1 0.38

in nmol for fractions derived from lo6 cells for 3 preparations. Incubations (lo6 cells/ml) and fractionations in MATERIALS AND METHODS. Adenylates were determined by high-performance liquid chromatography (11). compartments can be estimated assuming 5 ~1 total cell volume and 1 ~1 mitochondrial volume (3).

mitochondrial inner membrane occurs by an electrophoretie mechanism wherein the more negatively charged ATP moves out of the matrix in exchange for ADP (13, 14). Because of the A# (negative inside) in aerobic mitochondria, the cytosolic ATP concentration is high relative to mitochondrial ATP concentration, and cytosolic ADP concentration is low relative to mitochondrial ADP concentration (Table 7). Thus the cytosolic ATP-toADP ratio is maintained considerably higher (-7) than is the mitochondrial ratio (1.8; Table 7). Under anaerobic conditions, the cytosolic ATP/ADP decreases GO-fold to 0.38 (Table 7). In contrast, the mitochondrial ATP/ADP decreases only -50%. These results are consistent with two interpretations: that A# collapses during anoxia or that the mitochondrial ATP is not being as efficiently exchanged for ADP under anoxia as under aerobic conditions. Because TPMP, rhodamine B, and Ca” distributions during anoxia indicate that a substantial A# occurs during anoxia (2, 4), the latter interpretation would appear to be more likely to be correct. By the use of the A$m value during anoxia obtained with the TPMP method, there appears to be sufficient energy to drive the efflux of ATP in exchange for ADP uptake. The adenine nucleotide carrier is known to be a major rate-determining factor in mitochondrial function (32); the current results indicate that during anoxia, the carrier is further inhibited and does not exchange ATP and ADP in accordance with the availability of A#. DISCUSSION

The current studies assessthe effect of 30 min anoxia on mitochondrial transmembranal ion gradients. Two different types of changes in gradients occur as cells become anoxic. For pyruvate and citrate, large gradients are maintained, consistent with maintenance of a ApH that is comparable to normal aerobic conditions. Thus these results confirm studies on DMO distribution, which show that the mitochondrial ApH is retained during 30 min anoxia (2, 3). A second type of response is observed for the critical ions involved in phosphorylation (ATP, ADP, and phosphate) and ions transported by electrophoretic mechanisms. The transmembranal phosphate gradient is lost, indicating that phosphate uptake is inhibited. This may be necessary to prevent swelling due to massive phosphate accumulation resulting from ATP hydrolysis in the cytosol. The transmembranal distributions of ATP and ADP are also dramatically changed from aerobic conditions. Although the cytosolic ratio of ATP to ADP changes nearly ZO-fold on transition to anoxia, the mitochondrial ratio changes less than Z-fold. In spite of the

retention of a substantial membrane potential, the potential does not support continued movement of ATP out and ADP in. This inhibition of ion translocation may in fact be part of the mechanism that allows retention of the mitochondrial A$ for prolonged anaerobic periods. The data for glutamate and aspartate distribution are also consistent with inhibition of the electrophoretie glutamate-aspartate exchange during anoxia. Endogenous inhibitors and inhibitory mechanisms are known for several ion transport systems in mitochondria. For instance, the K+-H+ antiporter is inhibited by Mg2+ (9) and voltage gated (6), and the adenine nucleotide is inhibited by acyl CoAs (27). Although it is not clear that this occurs under physiological conditions, the adenine nucleotide transporter is a major rate-determining factor even during aerobic conditions (32). A decline in mitochondrial ATP and increase in ADP could result in competitive inhibition and limit transport function, but it is not clear how such a large change in cytosolic ATPto-ADP ratio could occur without a similar change in the mitochondrial compartment. However, studies of mitochondria from livers exposed to 3 h of ischemia show inhibited ADP-ATP exchange (22), indicating that other inhibitory mechanisms are involved. Ca2+ efflux is also sensitive to endogenous regulation, potentially by the oxidation-reduction state of pyridine nucleotides (l&19). Also, an endogenous inhibitor of the ATP synthase has been isolated and its association with the synthase has been found to vary in response to changes in respiratory function (28). Inhibition of F,-ATPase (adenosine triphosphatase) activity has also been found in mitochondria from anoxic and ischemic cells (4, 22). Taken together, the evidence that ApH and A# are largely retained during anoxia, that transport of ATP, ADP, malate, phosphate, glutamate, and aspartate appear to be inhibited during anoxia and that several mechanisms are present which regulate transport function, suggests that simultaneous inhibition of ion transport systems occurs. This inhibition particularly affects those systems that allow net-charge movement across the mitochondrial membrane and whose function would thereby lead to collapse of the protonmotive force or to mitochondrial swelling. At present, the ATP synthase is the only system that has been studied in this regard (4). Although proton conductance has not been directly studied, the ATPase activity is inhibited 60% by 30 min anoxia. This inhibition appears to be related to changes in mitochondrial Ca” content, which declines considerably during anoxia, but also may involve other mechanisms. Whether or not a common mechanism exists to regulate various ion transport systems in mitochondria remains open to spec-

MITOCHONDRIAL

ION

DISTRIBUTION

ulation. Extensive regulation of mitochondrial enzymes is known to occur (33), and consequently, physiological regulation of the transport systems, which are also important kinetic determinants, appears reasonable. Moreover, rapid and efficient response to environmental challenges are well known for a variety of systems. Hence, a common “concerted” control mechanism may exist to regulate ionic balance in mitochondria during anoxia. Such regulation could allow transition of the mitochondria to a neahypoxic state which, on reoxygenation, could recover function (2, 12). Thus early changes in mitochondrial ion transport could allow preservation of functional capacity and provide cells with a mechanism to survive episodes of anoxia. We thank Drs. K. F. LaNoue and L. J. Mandel for their helpful comments during manuscript preparation and M. Little for the excellent secretarial assistance. B. S. Andersson thanks the Swedish Medical Research Council for supplemental support. This research was supported by National Institutes of Health Grants HL-30286 and GM-36538. Received

27 May

1986; accepted

in final

form

5 November

1986.

REFERENCES 1. AKERBOOM, T. P. M., H. BOOKELMAN, P. E. ZUURENDONK, R. VAN DER MEER, AND J. M. TAGER. Intramitochondrial and extramitochondrial concentrations of adenine nucleotides and inorganic phosphate in isolated hepatocytes from fasted rats. Eur. J. Biothem. 84: 413-420,1978. 2. ANDERSSON, B. S., T. Y. Aw, AND D. P. JONES. Mitochondrial transmembrane potential and pH gradient during anoxia. Am. J. Physiol. 252 (Cell Physiol. 21): C349-C355, 1987. 3. ANDERSSON, B. S., AND D. P. JONES. Use of digitonin fractionation to determine mitochondrial transmembrane in distribution in cells during anoxia. Anal. Biochem. 146: 164-172, 1985. 4. Aw, T. Y., B. S. ANDERSSON, AND D. P. JONES. Suppression of mitochondrial respiratory function after short-term anoxia. Am. J. Physiol. 252 (Cell Physiol. 21): C362-C368, 1987. bioenergetic hypoxia. 5. Aw, T. Y., AND D. P. JONES. Secondary Inhibition of sulfation and glucuronidation reactions in hepatocytes at low O2 concentrations. J. BioZ. Chem. 257: 8997-9004,1982. 6. BERNARDI, P., AND G. F. AZZORE. Electroneutral H+-K+ exchange in liver mitochondria. Regulation by membrane potential. Biochim. Biophys. Acta 724: 212-223,1983. 7. DUSZYNSKI, J., G. MUELLER, AND K. NANOUE. Microcompartmentation of aspartate in rat liver mitochondria. J. Biol. Chem. 253: 6149-6157, 1978. 8. FISKE, C., AND Y. SUBBAROW. The calorimetric determination of phosphorous. J. Biol. Chem. 66: 375-400, 1925. 9. GARLID, K. D. On the mechanism of regulation of the mitochondrial K’/H’ exchanges. J. BioZ. Chem. 255: 11273-11279, 1980. 10. HOEK, J. B., D. E. NICHOLLS, AND J. R. WILLIAMSON. Determination of the mitochondrial protonmotive force in isolated hepatocytes. J. BioZ. Chem. 255: 1458-1464, 1980. 11. JONES, D. P. Determination of pyridine nucleotides in cell extracts by high-performance liquid chromatography. J. Chromatogr. 225: 446-449,198l. 12. JONES, D. P., F. G. KENNEDY, G. S. ANDERSSON, T. Y. Aw, AND E. WILSON. When is a mammalian cell hypoxic? Insights from studies of cells versus mitochondria. Mol. Physiol. 8: 473-482, 1985. 13. KLINGENBERG, M. The ADP, ATP translocation system of mitochondria. In: Mitochondria and Microsomes, edited by C. P. Lee, G. Schatz, and G. Dallner. Menlo Park, CA: Addison-Wesley, 1981, p. 293-316.

DURING

ANOXIA

C361

14. KLINGENBERG, M., AND H. W. HELDT. The .ADP/ATP translocation in mitochondria and its role in intracellular compartmentation. In: Metabolic Compartmentation, edited by H. Sies. London: Academic, 1982, p. 101-122. 15. LAMPRECHT, W., AND F. HEINZ. Pyruvate. In: Methods of Enzymatic Analysis (3rd ed.), edited by T. Bergmeyer and M. Grabe. Deerfield Beach, FL: VCH Publishers, 1985, vol. 6, p. 570-582. 16. LANOUE, K. F., A. C. SCHOOLWERTH, AND A. J. PEASE. Ammonia formation in isolated rat liver mitochondria. J. Biol. Chem. 258: 1726-1734,1983. 17. LANOUE, K. F., AND A. C. SCHOOLWERTH. Metabolite transport in mammalian mitochondria. In: Bioenergetics, edited by L. Ernsten. Amsterdam: Elsevier, 1984, p. 221-268. 18. LEHNINGER, A. L., A. VERCESI, AND E. A. BABABUNMI. Regulation of Ca2+ release from mitochondria by oxidation-reduction state of pyridine nucleotides. Proc. Natl. Acad. Sci. USA 75: 1690-1694, 1978. 19. L~TSCHER, H. R., K. WINTERHALTER, E. CARAFOLI, AND C. RICHTER. Hydroperoxide-induced loss of pyridine nucleotides and release of calcium from rat liver mitochondria. J. Biol. Chem. 255: 9325-9330,198O. 20. MEIJER, A. J., AND K. VAN DAM. Mitochondrial ion transport. In: Membrane Transport, edited by S. L. Bonting and J. J. H. H. M. de Pont. Amsterdam: Elsevier, 1981, p. 235-254. 21. MITCHELL, P. Keilin’s respiratory chain concept and its chemiosmotic consequences. Science Wash. DC 206: 1148~1149,1979. 22. MITTNACHT, S., S. C. SHERMAN, AND J. L. FARBER. Reversal of ischemic mitochondrial dysfunction. J. BioL. Chem. 254: 9871-9878, 1979. 23. MOLD&JS, P., J. H~GBERG, AND S. ORRENIUS. Isolation and use of liver cells. Methods Enzymol. 51: 60-70, 1978. 24. M~LLERING, H. Citrate. In: Methods of Enzymic Analysis (3rd ed.), edited by J. Bergmeyer and M. Grabe. Deerfield Beach, FL: VCH Publishers, 1985, vol. 7, p. 2-12. H. Malate. Determination with malate dehydrogen25. M~LLERING, ase and aspartate aminotransferase. In: Methods of Enzymic Analysis (3rd ed), edited by J. Bergmeyer and M. Grabe. Deerfield Beach, FL: VCH Publishers, 1985, vol. 7, p. 39-52. 26. NICHOLLS, D. G. Bioenergetics. London: Academic, 1982, p. 170171. AND R. PARVIN. Protective role of 27. PANDE, S. V., T. GOSWAMI, adenine nucleotide translocase in 02-deficient hearts. Am. J. Physiol247 (Heart Circ. Physiol. 16): H25-H34, 1984. 28. PEDERSEN, P. L., K. SCHWARZMANN, AND N. CINTR~N. Regulation of the synthesis and hydrolysis of ATP in biological systems: Role of peptide inhibitors of H’-ATPases. Curr. Top. Bioenerg. 11: 149-199,198l. 29. REED, D. J., J. R. BABSON, P. W. BEATTY, A. E. BRODIE, W. W. ELLIS, AND D. W. POTTER. High-performance liquid chromatography analysis of nanomole levels of glutathione, glutathione disulfide and related thiols and disulfides. Anal. Biochem. 106: 5562, 1980. 30. SCHOOLWERTH, A. C., AND K. F. LANOUE. Transport of metabolic substrates in renal mitochondria. Annu. Reu. Physiol. 47: 143-171, 1985. 31. SIESS, E. A., D. G. BROCK, AND 0. H. WIELAND. Subcellular distribution of adenine nucleotides and of metabolites of tricarboxylate cycle and gluconeogenesis in hepatocytes. In: Metabolic Compartmentation, edited by H. Sies. London: Academic, 1982, p. 235257. 32. TAGER, J. M., R. J. A. WANDERS, A. K. GROEN, W. KUNZ, R. BOHNENSACK, U. KUSTER, G. LETKO, G. BOHME, J. DUSZYNSKI, AND L. WOJTCZAK. Control of mitochondrial respiration. FEBS Lett. 151: l-9, 1983. 33. WILLIAMSON, J. R., AND R. H. COOPER. Regulation of the citric acid cycle in mammalian systems. FEBS Lett. 117: K73-K85, 1980. 34. WILLIAMSON, J. R., AND B. E. CORKEY. Assay of citric acid cycle intermediates and related compounds-update with tissue metabolite levels and intracellular distribution. Methods Enzymol. 55: 200-222,1979.