ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 222, No. 1, April 1, pp. 321-325, 1983

Net Adenine Nucleotide Transport in Rat Liver Mitochondria Is Affected by Both the Matrix and the External ATP/ADP Ratios

JUDITH AUSTIN AND JUNE R. APRILLE’

Department of Biology, TV&S University, Me&rd, Massachusetts 02155

Received August 23, 1982

The adenine nucleotide pool size of rat liver mitochondria may be regulated by a transport mechanism which allows net movement of ATP or ADP across the inner mitochondrial membrane (J. R. Aprille and J. Austin (1981) Arch. Biochem. Biophvs. 212, 689-699). Since transport can occur in either direction, and ATP is the preferred substrate, variations in the matrix or external ATP/ADP ratio were examined for an effect on the direction and rate of net adenine nucleotide flux. Fastest rates of net uptake were seen when the external ATP/ADP ratio was high and the internal ratio was low. Conversely, if the external ratio was low and the internal ratio was high, net loss was observed. The rate and direction of net transport were also changed by simply varying the external ATP concentration alone between 0 and 4 mM. Concentrations of less than 1 mM produced net loss of matrix adenines, whereas higher concentrations caused a net increase. The results suggest that changes in relative concentrations of ATP in the external and internal compartments can result in changes in the adenine nucleotide pool size. Another important result was clarification of earlier ambiguous findings with carboxyatractyloside. If care was taken to avoid changes in the matrix ATP/ADP ratio, the addition of carboxyatractyloside had no effect on either net uptake or net efflux of adenine nucleotides.

Rat liver mitochondria are capable of net transport of adenine nucleotides by a mechanism different from the one-for-one exchange which characterizes the adenine nucleotide translocase (l-4).

Up to this point our efforts to describe this transport have involved the incuba- tion of isolated mitochondria in media lacking any oxidizable substrates. Under these conditions the only sources of energy for the mitochondria are endogenous sub- strates and ATP hydrolysis (when ATP was added to the incubation). As the work progressed it was shown that the ability of the mitochondria to accumulate adenine nucleotides was dependent on the energi- zation of the inner membrane. This sug-

‘Author to whom correspondence should be ad- dressed.

gested that the addition of a respiratory substrate should stimulate adenine nu- cleotide uptake. Surprisingly, all the sub- strates tested produced some degree of in- hibition. Further experiments with suc- cinate showed that this inhibition was related to substrate oxidation (1).

One purpose of the work presented here was to determine whether this apparent inhibition might be a secondary effect of the increase in the matrix ATP/ADP ratio which was observed upon the addition of an oxidizable substrate. We have previ- ously proposed that net adenine nucleotide transport occurs continuously in both di- rections across the inner mitochondrial membrane (1). If ATP is the preferred substrate for efflux, as it appears to be for influx, then a rise in the matrix ATP/ADP ratio would be expected to increase the rate

321 0003-9861/83 $3.00 Copyright 0 1983 by Academic Press, Inc. All rights of reproduction in any form reserved.

322 AUSTIN AND APRILLE

TABLE I

EFFECT OF MATRIX AND EXTERNAL ATP/ADP RATIOS ON THE NET FLUX OF ADENINE NUCLEOTIDES

Net adenine nucleotide flux

(nmol/mg/min)

Matrix

ATP/ADP

External

ATP/ADP

A. 1 mM ATP 0.04 r 0.02 5.3 + 1.2 93 ?I 39 1 mM ATP, 1 pg/ml oligomycin 0.63 + 0.02 0.19 f 0.01 24f5

B. 1 mM ADP, 5 PM carboxyatractyloside -0.23 f 0.02 5.8 + 1.0 0.06 + 0.01 1 mM ADP, 1 pgg/ml oligomycin 0.14 * 0.07 0.02 f 0.01 0.09 f 0.01

Note. Incubations were carried out as described under Methods with the additions noted. P’,P5-di(adenosine- 5’)pentaphosphate (0.575 mM) was added in the experiments under part B to inhibit adenylate kinase (9). All incubations contained 5 mM glutamate + 5 mM malate. ATP/ADP ratios reported here were determined

at 10 min but similar values were observed throughout the 15-min incubation period.

of efflux relative to the rate of influx, pro- ducing a decrease in the observed rate of net uptake.

A second purpose of the study was to determine whether, by similar logic, the inhibition of net adenine nucleotide up- take by translocase inhibitors observed previously (1) is also secondary to alter- ations in the matrix ATP/ADP ratio. The resolution of the effects of this class of inhibitors is particularly important in de- termining whether net adenine nucleotide transport is distinct from adenine nucleo- tide translocase activity.

METHODS

Mitochondria were isolated by differential centrif- ugation from the livers of adult male Sprague-Daw- ley rats as described previously (1, 5). The isolation

medium was 250 mM sucrose, 1 mM Tris-Cl, 0.1 mM EDTA, pH 7.4. EDTA was omitted from the medium

for the last wash and final suspension. Protein de- terminations were by the biuret method (6).

To study changes in the matrix adenine nucleotide pool size, mitochondria (final cone 1 mg/ml) were incubated at 30°C in a medium containing 225 mM sucrose, 2 mM KzHPOI-KHzPOI, 5 mM MgClz, 10 mM

KCI, 10 mM Tris-Cl, pH 7.4. All incubations also con- tained 5 mM glutamate and 5 mM malate. Mitochon-

dria were preincubated in the medium for 15 s before the reaction was started by the addition of either ATP or ADP. At 2, 5, 10, and 15 min, l-ml samples of the incubation were extracted by centrifugation through a layer of silicone oil into 12% perchloric acid (1). The acid extracts were neutralized and as- sayed enzymatically for total adenine nucleotides

(ATP + ADP + AMP) (7,8). Rates of net adenine nu- cleotide uptake or loss from the matrix were calcu-

lated by linear regression. When it was necessary to determine the ATP/ADP

ratio in the external media, 0.2 ml of the incubation suspension, including the mitochondria, was ex-

tracted with an equal volume of 12% perchloric acid, neutralized, and assayed for adenine nucleotides as

above. Since the mitochondrial adenine nucleotides make up only a small portion of the total adenine nucleotides in the suspension (-1% when the exter-

nal adenine nucleotide concentration is 1 mrd), in- cluding them does not introduce a significant source

of error. The experimental conditions were designed to produce external ATP/ADP ratios that were ei-

ther *l or 81. The exact external ATP/ADP was not an important consideration as long as that ob-

jective was attained. Where absolute matrix adenine nucleotide con-

tents are given in the results, parallel incubations with [“Cl sucrose were used to determine the volume of medium carried through the layer of silicone oil

with the mitochondria. A correction was made for the adenine nucleotides which would be contained in this volume, based on the assumption that it is in

equilibrium with the bulk medium. When mitochondria are extracted by the silicone

oil method, unless the acid-precipitated pellet is re- homogenized, the extraction of adenine nucleotides

is not complete. More than 95% of the protein in the sample layered on the silicone oil was recovered in the pellet, but if the pellet was rehomogenized, the

adenine nucleotide values increased by an average of 21 f 4.1%. This did not vary as a function of incu- bation time, nor did it appear to vary significantly as a function of different experimental conditions. Since we were examining relative differences, values reported here were not corrected for this constant fractional loss.

ADENINE NUCLEOTIDE TRANSPORT IN MITOCHONDRIA

.- - /‘- 4.0mM I

323

0’ I 0 5 IO 15

Time (mid

FIG. 1. Total matrix adenine nucleotide content as a function of time during incubation with

different concentrations of ATP. Incubation conditions were as described under Methods, with the

ATP concentrations indicated. Under all conditions the matrix ATP/ADP ratio was $1 and did not vary during the incubations.

Unless otherwise noted, all values are the averages

of three separate experiments f SE.

RESULTS

When mitochondria are incubated in media containing magnesium and phos- phate in the presence of 1 mM ATP, the matrix adenine nucleotide pool size in- creases at a rate of 0.44 f 0.02 nmol/mg/ min (1). Under these conditions, the ma- trix ATP/ADP ratio is low, approximately 0.44. If a respiratory substrate, such as glutamate + malate, is included, this rate of adenine nucleotide uptake is decreased by 89% and the matrix ATP/ADP ratio increases to 5.3 (1) (also see Table IA).

To test whether the inhibition of uptake might be a secondary effect of this change in the ATP/ADP ratio, oligomycin was in- cluded in the incubation medium. In the presence of oligomycin, any ADP present remained unphosphorylated and, due to the specificity of the adenine nucleotide trans- locase, was concentrated in the matrix ad- enine nucleotide pool, resulting in a low matrix ATP/ADP ratio (Table IA). Under these conditions the apparent inhibition of uptake by glutamate + malate was abol- ished.

We were curious as to what effect shift- ing the matrix ATP/ADP ratio would have in the presence of external ADP. If the mitochondria are incubated with 1 mM

ADP and oligomycin, both the matrix and external pools are composed predomi- nantly of ADP. In these circumstances a slow rate of net adenine nucleotide uptake was seen (Table IB). If instead the mito- chondria are incubated with carboxy- atractyloside and ADP, the matrix ade- nine nucleotide pool will be high in ATP while the external adenine nucleotide pool remains high in ADP. Under these con- ditions a net loss of adenine nucleotides was observed (Table IB).

The results so far suggested that the lack of net change in the matrix adenine nu- cleotide pool size when both the matrix and external pools are predominantly ATP is due to a dynamic equilibrium between the two pools. That is, it is possible that movement is occurring in both directions at equal rates. If this is the case and if the rate of influx is concentration dependent, then changing the external ATP concen- tration should create a nonequilibrium sit- uation. We found that it was possible to vary both the rate and direction of net ad-

324 AUSTIN AND APRILLE

TABLE II

MATRIXADENINENUCLEOTIDE CONTENT AFTER INCUBATIONWITHVARIOUSCONCENTRATIONSOFATP

ATP concentration Matrix adenine during incubation nucleotide content

(mM) (nmol/mg)

0.25 7.36

1.0 13.18 2.0 18.99 4.0 23.68

Note. Mitochondria were incubated for 15 min with 5 mM glutamate + 5 mM malate and the indicated

concentrations of ATP as described under Methods. At the end of the incubation, the mitochondria were

pelleted by centrifugation, washed once in 250 mM sucrose, 1 mM Tris-Cl, pH 7.4, and extracted for ad-

enine nucleotides. Values reported are the averages of two experiments. The average adenine nucleotide

content of the freshly isolated mitochondria in these experiments was 12.65 nmol/mg.

enine nucleotide transport by simply vary- ing the external ATP concentration be- tween 0 and 4 mM (Fig. 1). External ATP concentrations of less than 1 mM produced a net loss of adenine nucleotides while con- centrations greater than 1 mM led to a net increase. (With 1 mM ATP, either no change or a slight increase was observed.) Similar results were obtained if the mi- tochondria were separated from the in- cubation medium by centrifugation and subsequently extracted for adenine nucleo- tides instead of being extracted through silicone oil (Table II). Regardless of the external ATP concentration, under the ex- perimental conditions in Fig. 1 and Table II, the matrix ATP/ADP ratio was 91.

In the presence of carboxyatractyloside and other inhibitors of the adenine nu- cleotide translocase, we had previously ob- served a strong inhibition of net adenine nucleotide uptake when the mitochondria were incubated with 1 mM ATP (no respi- ratory substrate present) (1). The addition of translocase inhibitors produced an in- crease in the matrix ATP/ADP ratio sim- ilar to that observed upon the addition of a respiratory substrate (l), suggesting that this apparent inhibition was also indirect rather than at the site of transport. To test

this possibility, mitochondria were incu- bated with glutamate + malate either with 4 mM ATP or without external adenine nu- cleotides. In either case the matrix ATP/ ADP ratio did not change appreciably if carboxyatractyloside was added (Table III). Under these conditions no inhibition by carboxyatractyloside of net loss or net up- take of matrix adenine nucleotides was ob- served.

DISCUSSION

The work reported here shows that the net rate of change in the size of the matrix adenine nucleotide pool is affected by both the matrix and external ATP/ADP ratios. The greatest rate of net uptake is seen when the matrix pool is predominantly ADP and the external pool ATP. Net loss of adenine nucleotides occurs when the matrix is predominantly ATP and the ex- ternal pool ADP. These results are most simply explained by assuming that ATP is the preferred substrate for transport in either direction, and that net changes in the matrix pool size occur mainly as a function of perturbations in the relative concentration of ATP in the external and internal compartments. However, changes in the ATP concentration usually occur via alterations in the ATP/ADP ratio, and the potential importance of the reciprocal

TABLE III

NETADENINENUCLEOTIDEFLUXINTHEPRESENCE ANDABSENCEOFCARBOXYATRACT~LOSIDE

Net adenine

nucleotide flux (nmol/ Matrix

mg/min) ATP/ADP

0 mM ATP -0.32 + 0.01 5.2 + 1.0 0 mM ATP, 5 PM

carboxyatractyloside -0.32 + 0.02 4.8 k 0.6 4 mM ATP 0.26 +- 0.02 6.1 + 0.3 4 rnhl ATP, 5 PM

carboxyatractyloside 0.30 + 0.12 5.2 -c 0.2

Note. Incubation conditions were as described un- der Methods with additions as indicated. All incu- bations included 5 mM glutamate t 5 mM malate.

ADENINE NUCLEOTIDE TRANSPORT IN MITOCHONDRIA 325

change in the ADP concentration must also be considered. ADP is transported at a much slower rate (l), and it therefore may act as a competitive inhibitor for ATP transport.

It is important to remember that the incubations included 5 mM MgCla, so vir- tually all the adenine nucleotides in the matrix and medium are chelated. Previous work has shown that it is probably the Mp-chelated species of ATP and ADP that are the actual substrates for this transport process (1). An exact free con- centration for ATP * Mg in the matrix can- not be deduced, because of possible bind- ing to proteins, but in our experiments there was a much higher content of ATP - Mg per microliter of water space in the matrix as compared to the external medium. Therefore, it is reasonable to sup- pose that there was a concentration gra- dient for ATP * Mg across the inner mem- brane. For example, when mitochondria were incubated with 1 mM ATP, at the end of the 15-min incubation the matrix total adenine concentration was -16 mM (using an experimentally determined value of 0.77 pl/mg protein for the matrix Hz0 space). Similar differences in the adenine nucleo- tide content of the matrix versus the cy- tosol have been described in situ (10, 11). This indicates that the matrix adenine nu- cleotide pool can be sustained against a concentration gradient. The energy for maintaining this concentration difference is probably derived from the respiration- dependent proton gradient (1).

Previous attempts to determine the ef- fect of adenine nucleotide translocase in- hibitors on net adenine nucleotide trans- port have produced ambiguous results (l- 4). Our work with carboxyatractyloside shows that it does not inhibit net adenine nucleotide transport in either direction and that the apparent inhibition observed un- der some conditions may be a secondary effect of changes in the matrix ATP/ADP ratio. This suggests that in further work, changes in the ATP/ADP ratio must be considered as a possible complication in testing other inhibitors or variations in in- cubation conditions. Most importantly, this result clearly distinguishes the observed transport process from the only other

known adenine transporter, adenine nu- cleotide translocase.

We found that both net uptake and net loss of adenine nucleotides could be pro- duced under the same incubation condi- tions simply by changing the external ATP concentration. This suggests that move- ment occurs simultaneously in both direc- tions and that when no movement is ob- served it might be because the rate of in- flux equals the rate of efflux. This idea has been verified in additional studies (to be reported) in which separate unidirectional influx and efflux rates for this transport process were measured directly.

ACKNOWLEDGMENTS

This study was submitted by J. Austin in partial

fulfillment of the requirements for the M.S. in bioi- ogy. The work was supported by NIH Grant NS14936.

We thank Lisa Carroll for preparing the manuscript.

REFERENCES

1. APRILLE, J. R., AND AUSTIN, J. (1981) Arch.

B&hem. Biophys. 212, 689-699. 2. MEISNER, H., AND KLINGENBERG, M. (1968) J. Biol.

Chem. 243, 3631-3639. 3. POLLAK, J. K., AND SUTTON, R. (1980) Biochw J.

192, 75-83. 4. ZOCCARATO, F., RUGOLO, M., SILIPRANDI, D., AND

SILIPRANDI, N. (1981) Eur. J. B&hem. 114,194-

199. 5. ASIMAKIS, G. K., AND APRILLE, J. R. (1980) FEBS

L&t. 117, 157-160.

6. JACOBS, E. E., JACOB, M., SANADI, D. R., AND BRADLEY, L. B. (1956) J. Biol Chem. 223, 147 156.

7. LAMPRECHT, W., AND TRAUTSCHOLD, I. (1974) in

Methods of Enzymatic Analysis (Bergmeyer, H. U., ed.), Vol. 4, pp. 2101-2110, Verlag-Chemie, Weinheim/Academic Press, New York.

8. JAWOREK, D., GRUBER, W., AND BERGMEYER, H. U. (1974) in Methods of Enzymatic Analysis

(Bergmeyer, H. U., ed.), Vol. 4, pp. 212’7-2131, Verlag-Chemie, Weinheim/Academic Press, New York.

9. MELNICK, R. L., RUBENSTEIN, C. P., AND MOTZKIN, S. M. (1979) Anal B&hem. 96, 7-11.

10. SOBOLL, S., SCHOLZ, R., AND HELDT, W. (1978) Eur. J. B&hem. 87, 377-390.

11. ZUURENDONK, P. F., TISCHLER, M. E., AKERBOOM, T. P. M., VAN DER MEER, R., WILLIAMSON, J. R., AND TAGER, J. M. (1979) in Methods in Enzymology (Fleischer, S., and Packer, L.,

Eds.), Vol. 56, pp. 207-223, Academic Press, New York.