Non-protein-mediated transfer of phosphatidic acid between microsomal and mitochondrial membranes

ARCHIVESOF BIOCHEMISTRYAND BIOPHYSICS Vol. 260, No. 1, January, pp. 301-308,1988

Non-Protein-Mediated Microsomal

Transfer of Phosphatidic Acid between and Mitochondrial Membranes

JOLANTA BARANSKA AND LECH WOJTCZAKl Department

of Cellular

Biochemistry, Nencki Institute of Experimental Pasteura 3, 02-093 Warsaw, Poland

Biology,

Received March 30,1987, and in revised form August 18, 1987

The transfer of phosphatidic acid between rat liver microsomes loaded with [““PIphosphatidic acid and rat liver mitochondria was studied in the absence of added lipid transfer proteins. It was found that during 1 h at 37°C in the medium containing 100 mM KCl, 20-30s of phosphatidic acid but only 2.5% of phosphatidylcholine were transferred. This spontaneous transfer of phosphatidic acid remained the same after pretreatment of microsomes and mitochondria with 125 mM KC1 or microsomes alone with 1 mM Tris, pH 8.6, procedures reported to remove adsorbed lipid transfer proteins. This transfer was insensitive to thiol-blocking reagents. The initial rate of this non-proteinmediated transfer of phosphatidic acid was virtually independent of the concentration of the acceptor membranes (mitochondria), thus indicating that it occurs by diffusion of the phospholipid through the aqueous phase rather than by membrane collision. About 80% of phosphatidic acid synthesized in the outer mitochondrial membrane was recovered in the inner membrane after a l-h incubation, pointing to a high rate of the intermembrane transfer of this phospholipid within intact mitochondrion. o 19% Academic Press, Inc.

Phospholipid transfer proteins of varying specificity that mediate the intermembraneous exchange of phospholipids have been found in cytosol fractions from a variety of animal and plant tissues (for reviews, see Refs. (l-3)). However, even in the absence of these proteins a transfer of phospholipids between membranes does occur, though usually at a much lower rate. Two mechanisms of such non-protein-mediated transfer have been proposed, one of which involves collision of membranes (4, 5), and the other one a transfer through the aqueous phase in form of monomers or micelles (6-13). Our previous studies (14) showed the presence of phosphatidic acid transfer activity in the rat liver cytoplasmic fraction. However, we also found that, in an ionic medium, the spontaneous transport of 1To whom correspondence should be addressed.

phosphatidic acid occurred at a considerable rate, so that the cytoplasmic fraction had little additional effect. We have suggested therefore that, under ionic conditions of the cell cytoplasm, phosphatidic acid is transported between the endoplasmic reticulum and mitochondria at least partly without participation of phospholipid transfer proteins. Since no phosphatidic acid transfer protein could be detected in the mitochondrial intermembrane compartment, we have speculated that the transfer of this phospholipid from the outer to the inner mitochondrial membranes proceeds entirely by a nonprotein-mediated mechanism. Recently Megli et al. (15) have reported that the “spontaneous” transfer of phosphatidylcholine and phosphatidylethanolamine from liposomes to mitochondria is due to the nonspecific phospholipid transfer protein that is adsorbed on iso301

0003-9861/88 $3.00 Copyright All rights

0 1988 by Academic Press, Inc. of reproduction in any form reserved.

302

BARANSKA

AND WOJTCZAK

Microsomes labeled with i4C in phosphatidylcholine were prepared according to Bjerve (19) by incubating the particles with [i4C]choline chloride (10 &i/ml, sp act 6.8 mCi/mmol). After 30 min at 37°C microsomes were sedimented by centrifugation and washed with 250 mM sucrose containing 1 mM EDTA and unlabeled choline. If microsomes previously labeled with ‘H in proteins were used, doubly labeled particles resulted. Microsomes or mitochondria were loaded with [32P]phosphatidic acid by incubating the particles with glycerol 3-[s’P]phosphate according to the procedure of Zborowski and Wojtczak (20) as described previously (14). After 60 min at 3’7°C for microsomes and 30 min at 20°C for mitochondria the particles were sedimented by centrifugation and washed twice with 250 mM sucrose containing 1 mM EDTA and 1 mM glycerol 3-phosphate. When microsomes previously labeled with i4C in proteins were used in this procedure, doubly labeled particles were obtained. For phospholipid transfer experiments microsomal pellets obtained in these procedures were suspended in 250 mM sucrose, sonified for 3 X 30 s, and nondispersed clumps were removed by centrifugation at 10,OOOg for 10 min. Incubations. All incubations for phospholipid transfer were carried out under constant shaking with air as the gas phase. The transfer of phosphatidic acid and phosphatidylcholine was studied using doubly labeled microsomes, with 14Cin proteins and 32Pin phosphatidic acid or with 3H in proteins and 14Cin phosphatidylcholine as donor membranes and mitochondria as the MATERIALS AND METHODS acceptor. At various periods of time 0.3- to 0.5-ml Biological material. Liver mitochondria and microaliquots of the incubation mixture were pipetted into somes from Wistar rats were obtained by conven- centrifugation tubes containing 5 ml cold 250 mM tional procedures (16). Where indicated, mitochonsucrose plus 1 mM EDTA and rapidly centrifuged at dria and microsomes were additionally washed with 10,OOOg.The resulting mitochondrial pellet was 125 mM KC1 plus 10 mM Tris-HCI (pH 7.4) plus 1 mM washed with the same solution, solubilized with conEDTA (15), or microsomes were washed with 1 mM centrated formic acid, and counted for radioactivity. Tris-HCl, pH 8.6 (17). Mitochondria depleted of the In parallel samples both the mitoehondrial pellets outer membrane (mitoplasts) were prepared accord- and the supernatants containing microsomes were ing to Schnaitman and Greenawalt (18). After sedi- extracted with acidic butanol according to Bjerve et mentation of mitoplasts at 10,OOOgfor 10 min, the al. (21), and the content of phospholipids was anaresulting supernatant was centrifuged at 100,OOOg lysed by thin-layer chromatography. In some experifor 60 min to collect fragments of the outer mito- ments with [32P]phosphatidic acid parallel samples chondrial membranes. Microsomes containing la- were pipetted into 5% trichloroacetic acid and the beled proteins were obtained from animals injected amount of water-soluble 32P, indicative for the hywith 75 pCi 14C-labeled amino acid mixture (sp act 54 drolysis of phosphatidic acid, was measured. mCi/mAtom carbon) or 800 pCi L-[G-aH]leueine (sp Lipid extraction and chromatography. Lipids were act 250 mCi/mmol), or 400 &i L-[U-‘*C]leucine (sp extracted from aqueous suspensions of the memact 200 mCi/mmol) dissolved in O.S%NaCl, 60-90 min branes according to Bligh and Dyer (22) and from prior to decapitation. membrane pellets using methanol/chloroform (l/2, v/v) (23). When quantitative recovery of lysophospholipids was essential extraction with acidic bu2 Abbreviations used: CCCP, carbonyl cyanide m- tanol as described by Bjerve et al. (21) was used. chlorophenyl-hydrazone; EGTA, ethylene glycol Phospholipids were separated and identified by bis(@aminoethyl)-N,N’-tetraacetic acid; NEM, N- thin-layer chromatography on silica gel G or H ethylmaleimide. (Merck A. G., Darmstadt, FRG) using the following

lated rat liver mitochondria. This was documented by the observation that the “spontaneous” transfer was diminished when mitochondria were prewashed with a saline solution and that the transfer was inhibited by thiol-blocking reagents like mersalyl or NEM.’ These authors also suggested that the spontaneous transfer of phosphatidic acid described by ourselves (14) might be due to the adsorbed phospholipid transfer protein as well. The first aim of the present investigation was, therefore, to check whether this suggestion were true. This was accomplished by subjecting microsomes and mitochondria to treatment known to remove adsorbed lipid transfer proteins and by looking for the effect of SH-blocking reagents. Having established that the spontaneous transfer was in fact a non-protein-mediated one, our second aim was to differentiate between its two possible mechanisms, namely membrane collision and free diffusion throught the water phase. In addition, some further insight into the phosphatidic acid transfer between the outer and the inner mitochondrial membranes was also obtained.

NON-PROTEIN-MEDIATED

TRANSFER

solvent systems: (A) chloroform/methanol/water (65/25/4, v/v/v) (24); (B) chloroform/methanol/acetic acid/water (65/25/8/4, v/v/v/v) (21); or (C) chloroform/methanol/28% ammonia/water (65/35/5/ 2.5, v/v/v/v) (25). Spots were visualized using iodine vapor and autoradiography. Enzyme o.ssaya Rotenone-insensitive NADH-cytochrome c reductase (EC 1.6.2.1), NADPH-cytochrome c reductase (EC 1.6.2.4), and succinate-cytochrome c reductase were measured as described by Sottocasa et al. (26). Chemicals and other procedures. rat-Glycerol 3[32P]phosphate was prepared by heating inorganic [“PIphosphate with glycerol (27). W-labeled amino acid mixture and L-[G-3H]leucine were from Amersham International plc (Amersham, UK); [‘4C]choline chloride and [3”P]phosphate were from the Institute of Nuclear Research (Swierk, Poland) and L-[U-‘4C]leucine was from UVVVR (Prague, Czechoslovakia). Radioactivity was measured in a scintillation spectrometer making use of the Cerenkov effect for 32P (28) or using liquid scintillation cocktail for 3H, i4C, and for the mixture of 3H, i4C, and “P. Protein was determined by the biuret method (29). RESULTS

Transfer between Microsmes and Mitochondria The non-protein-mediated transfer of phosphatidic acid and phosphatidylcholine was studied in experiments in which doubly labeled microsomes were used as donor membranes and mitochondria as the acceptor. Labeling of proteins in microsomes permitted determination of microsomal contamination in mitochondrial pellets after incubation. The amount of microsomes sedimented with mitochondria was usually about 3% in zero time and lo-13% after 5-20 min incubation. Values for phospholipid transfer were routinely corrected for this contamination. The labeling of microsomal phosphatidic acid or phosphatidylcholine (see Materials and Methods) always resulted in the formation of a certain proportion of the lyso forms. The lysophosphatidic acid amounted to about 10% of glycerol 3-r2P]phosphate incorporated into microsomal phospholipids and the lysophosphatidylcholine, to as much as 25% of [14C]choline incorporated. The amount of the lyso forms did not essentially change during

OF PHOSPHATIDIC

ACID

303

subsequent incubation for 60 min in the presence of EDTA. As shown in Fig. lA, about 5% of the radioactivity present in phosphatidylcholine plus lysophosphatidylcholine was transferred from microsomes to mito-

FIG. 1. Transfer of phosphatidylcholine and lysophosphatidylcholine (A) and phosphatidic and lysophosphatidic acids (B) from microsomes to mitochondria. The transfer of labeled phospholipids was determined as described under Materials and Methods. The concentrations of mitochondria were 8.6 and 6.3 mg protein/ml and those of microsomes 2.7 and 3.9 mg protein/ml in (A) and (B), respectively. In (A), microsomes were labeled with 3H in proteins (180,000 cpm/ml) and with i4C in phosphatidylcholine and lysophosphatidylcholine (total, 150,000 cpm/ml). In (B), microsomes were labeled with 14Cin proteins (30,000 cpm/ml) and loaded with 32P-labeled phosphatidic and lysophosphatidic acids (100,000 cpm/ml). Incubation temperature was 37°C and the medium contained 100 mM KCI, 10 mM TrisHCl (pH 7.4), and 2 mM EDTA. In (B) it was supplemented with 1 mM K-phosphate and 2 FM CCCP. The reaction was started by the addition of microsomes. Phospholipids were extracted with acidic butanol (21), and the chromatographic systems A and B were used for the separation of diacyl- and monoacyl-labeled phospholipids in (A) and (B), respectively. The points represent radioactivity in the mitochondrial pellet, corrected for microsomal contamination, as a percentage of the added radioactivity of the respective phospholipid; values for the zero time were subtracted. In (B) correction was also made for partial hydrolysis of phosphatidic acid, which was estimated from the amount of water-soluble 32P and amounted to 3,8, and 21% after 5,20, and 60 min of incubation, respectively. (0) Phosphatidylcholine or phosphatidic acid; (0) lysophosphatidylcholine or lysophosphatidic acid; (A) both diacyl and monoacyl forms counted together.

304

BARANSKA

AND WOJTCZAK

chondria during 60 min incubation. The analysis of transferred phospholipids revealed that lysophosphatidylcholine was transferred in 10% and phosphatidylcholine in 2.5%. In contrast, the transfer of phosphatidic acid plus lysophosphatidic acid amounted to over 20% of the total radioactivity present in these phospholipids. Also in contrast to phosphatidylcholine, phosphatidic acid and its lyso form were transferred at the same rate as expressed in a percentage of the particular phospholipid present in donor membranes (Fig. 1B). It has to be mentioned that the transfer of phosphatidic acid was routinely measured in the medium supplemented with an uncoupler of oxidative phosphorylation (CCCP) and inorganic phosphate (see legend to Fig. 1) in order to minimize obscuring the results by accumulation in mitochondria of inorganic [32P]phosphate resulting from hydrolysis of phosphatidic acid during incubation. However, these additions had no effect on the transfer of either phosphatidic acid or phosphatidylcholine. The transfer was also corrected for a decrease of the amount of phosphatidic acid during incubation due to hydrolysis, presumably by the action of phosphatidate phosphohydrolase (EC 3.1.3.4). A possible involvment in the “spontaneous” transfer of phosphatidic acid of adsorbed transfer proteins was investigated in experiments in which donor and acceptor membranes were pretreated to minimize the content of adsorbed soluble proteins. Figure 2 illustrates an experiment in which mierosomes prewashed with 125 mM KC1 prior to loading with phosphatidic acid were used as donor membranes. It shows that the transfer was the same to mitochondria isolated in the usual way and to those prewashed with 125 mM KCl. In addition, it can also be seen that NEM had no effect on the rate of phosphatidic acid transfer in either case. It should be also mentioned that in control experiments the spontaneous transfer of phosphatidylcholine was not affected by prewashing the mitochondria with KC1 solution or preincubation with NEM. Apparently, our mitochondrial preparations,

FIG. 2. Effect of saline washing on the spontaneous transfer of phosphatidic acid from microsomes to mitochondria. Before loading with [32P]phosphatidic acid, microsomes were suspended in 125 mM KC1 plus 10 mM Tris-HCl (pH 7.4) plus 1 mM EDTA and sedimented by the usual procedure. Mitochondrial suspension was divided into two parts, one of which was additionally washed with 250 mM sucrose plus 10 mM Tris-HCl (pH 7.4) and the other one with the KCITris-HCI-EDTA solution as for microsomes. The medium and conditions for phosphatidic acid transfer were as described in Fig. 1. The concentration of mitochondria was 3.2 mg protein/ml and that of microsomes 2.9 mg protein/ml. Where indicated, the incubation mixture also contained NEM, 80 nmol/mg mitochondrial protein, added to the otherwise complete medium 5 min prior to starting the reaction with microsomes. Values for zero time were subtracted. (0) Mitochondria washed with sucrose; (0) mitochondria washed with sucrose, incubated with NEM; (A) mitochondria washed with KCl; (A) mitochondria washed with KCl, incubated with NEM.

in contrast to those of Megli et al. (15), contained only negligible amounts of adsorbed nonspecific lipid transfer protein(s). It was also found that washing of microsomes with 1 mM Tris buffer, pH 8.6, had no effect on the rate of phosphatidic acid transfer and that removing of the outer mitochondrial membrane did not diminish the ability of the resulting mitoplasts to accept phosphatidic acid from microsomes (not shown). In fact, mitoplasts were usually better acceptors than intact mitochondria. The dependence of the transfer of phosphatidic acid on the concentration of ac-

NON-PROTEIN-MEDIATED

TRANSFER

ceptor membranes is illustrated in Fig. 3. As shown, the rate of phosphatidic acid transfer is by no means proportional to the concentration of the acceptor membranes. The increase of mitochondrial concentration two and four times had practically no effect on the initial rates of phosphatidic acid transfer and potentiated it only by factors of 1.3 and 1.5, respectively, after 60 min incubation. A similar result was obtained with mitochondria and microsomes prewashed with 125 mM KC1 (not shown). In contrast to this, increasing the amount of donor membranes (microsomes) up to 4.5 mg protein/ml increased proportionally the initial rate of phosphatidic acid transfer, whereas a further doubling of this amount to 9 mg protein/mg increased the transfer rate by a factor of 1.4 (not shown). Transfer between Outer and Inner Mitochondrial Membranes

OF PHOSPHATIDIC

ACID

305

FIG. 3. The dependence of the transfer of phosphatidic acid from microsomes to mitochondria on the concentration of mitochondria. Mitochondria at various concentrations were incubated with microsomes (2.6 mg protein/ml) labeled with 14C in proteins (180,000 cpm/ml) and loaded with [32P]phosphatidic acid (55,000 cpm/ml) at 37°C in the medium as in Fig. 1B. Values were corrected for microsomal contamination of the mitochondrial pellets and for partial hydrolysis of phosphatidic acid, which was estimated from the amount of water-soluble 32P and amounted to 4,17, and 33% after 5,20, and 60 min, respectively. Values for zero time were subtracted. The concentrations of mitochondria (acceptor membranes) were 2.2 (A), 4.4 (0), and 8.8. (0) mg protein/ml.

When mitochondria were incubated for 30 min with glycerol 3-[32P]phosphate to promote the synthesis of [32P]phosphatidic acid, followed by the incubation for 60 min with cofactors for the synthesis of cardiolipin, and then subjected to the membrane separation procedure, it was found that a 78% of total phosphatidic acid recovered large proportion of lipid 32P, of which in both fractions, represented phosphaabout 97% was in form of phosphatidic tidic acid associated with the inner mitoand lysophosphatidic acids, was present chondrial membrane. In contrast, small in mitoplasts (mitochondria stripped amounts of cardiolipin and phosphatioff their outer membrane) (Table I). dylglycerol which were formed were inAlthough the mitoplast fraction was trinsic to mitoplasts only, and minute strongly contaminated with outer mem- amounts found in the outer membrane branes (which has often been the case fraction, could be entirely ascribed to after a prolonged incubation preceding inner membrane contamination, as indimembrane fractionation), it is clearly visi- cated by succinate-cytochrome c reducble that the ratio of lipid 32Pto rotenone- tase (not shown). That the redistribution insensitive NADH-cytochrome c reduc- of newly synthesized phosphatidic acid in tase (the marker for the outer membrane) mitochondria is time dependent can be was six times higher in the mitoplast deduced by comparing the present result fraction than in outer membranes. This with our previous observation that only indicates that mitoplasts contained much one-fourth of phosphatidic acid was recovmore phosphatidic acid than could be ac- ered in the inner membrane after 40 min counted for by outer membrane contami- of incubation (see Table VI in Ref. (18)). In a separate experiment it was found nation. From data presented in Table I it that, under identical conditions, only 3% can be calculated that this contamination contributed to only 42,000 cpm, whereas of 32Ppresent in the phospholipid fraction the remaining 219,000 cpm, equivalent to was incorporated to phosphatidylglycerol

306

BARANSKA

AND WOJTCZAK TABLE I

SUBMITOCHONDRIALDISTRIBUTIONOF

32P cm (4

Fraction Mitochondria (unfractionated) Outer membranes Mitoplasts Recovery

294,000 20,270 261,080 281,350

PHOSPHATIDIC ACID

NADH-cytochrome

c reductase

% of total

Arbitrary units W

% of total

(A)/(B) ratio

100 7 89 96

6,083 1,893 3,947 5,840

100 31 65 96

48.3 10.7 66.1

Note. Mitochondria were incubated with glycerol 3-[32P]phosphate under conditions promoting the synthesis of phosphatidic acid (see Materials and Methods). Thereafter, mitochondria were separated by centrifugation, washed, and incubated at 30°C for 60 min in the medium containing 100 mM KCI, 10 mM Tris-HCl (pH 7.4), 10 mM 2-mercaptoethanol, 10 mM MgCls (added last), 2 mM EGTA, 1 mM CTP, and 1 mM glycerol 3-phosphate. Subsequently, they were subjected to digitonin treatment (18) and separated into outer membrane and mitoplast fractions by fractionated centrifugation. Lipid 32Pand NADH-cytochrome c reductase were measured in both fractions. The last column shows the ratio between the radioactivity of phosphatidic acid (column A) and the activity of NADH-cytochrome c reductase (column B).

(2.5%) and cardiolipin incubation.

(0.5%) after a l-h

DISCUSSION

A substantial spontaneous transfer of phosphatidic acid between guinea pig liver mitochondria and microsomes has already been observed by Stuhne-Sekalec and Stanacev (30). In our previous study (14) it has been suggested that the transfer of this phospholipid between intracellular membranes can proceed, to a considerable extent, without participation of transfer proteins. The present investigation confirms the observations of Stuhne-Sekalec and Stanacev (30) and further substantiates our previous findings. To our rough estimation, microsomes loaded with phosphatidic acid by the procedure described (14) contain about 1 nmol of this phospholipid per milligram of protein, which constitutes less than 1% of total microsomal phospholipids. Therefore, the spontaneous transfer of phosphatidic acid from microsomes to mitochondria, expressed in absolute terms, is much lower than the transfer of phosphatidylcholine. However, the ability of a phospholipid to be transported or ex-

changed between membranes must be related to its abundance in the donor membrane, which approximately reflects its relative accessibility at the surface. In this respect phosphatidic acid was several times more readily transported than phosphatidylcholine (Fig. 1). Also in this sense, lysophosphatidylcholine was relatively better transported spontaneously than phosphatidylcholine. Since in model systems lysophosphatidylcholine is known to be more readily transferred through the water phase than phosphatidylcholine (13), it was important to note that the transfer of 32Pfrom microsomes loaded with [32P]phosphatidic acid to mitochondria was equally due to the transport of phosphatidic acid and of its lyso derivative (Fig. 1B). Because under our experimental conditions acylation of lysophosphatidic acid in acceptor membranes was unlikely, our results clearly demonstrate that the spontaneous transfer of phosphatidic acid, and not only of its lyso derivative, between microsomes and mitochondria was quite substantial. To establish whether the spontaneous transfer of phosphatidic acid really occurs by a non-protein-mediated mechanism, it was necessary to show that it was not due

NON-PROTEIN-MEDIATED

TRANSFER

to transfer protein(s) adsorbed on microsomes and/or mitochondria used in these experiments as donor and acceptor membranes. Association of high molecular weight lipid transfer protein with liver microsomes (17) and the presence of a nonspecific transfer protein bound to rat liver mitochondria (15) have been reported. In the first part of the present investigation it was shown that neither washing of microsomes and mitochondria with 125 mM KC1 nor preincubation of mitochondria with NEM, procedures which remove or inactivate the adsorbed nonspecific transfer protein (15), affected the spontaneous transfer of phosphatidic acid appreciably (Fig. 2). The concentration of NEM in this experiment was even higher than that used by Megli et al. (31) to block the spontaneous transfer of phosphatidylcholine (50 nmol/mg protein). In addition, washing of microsomes with 1 mM Tris buffer, pH 8.6, reported to release the high molecular weight lipid transfer protein (1’7), had no effect on phosphatidic acid transfer either. Thus, the non-proteinmediated transfer of phosphatidic acid between microsomes and mitochondria, proceeding at a relatively high rate in a saline medium, seems to be well documented. To differentiate between the two mechanisms of the non-protein-mediated transfer of phospholipids, viz. by membrane collision or diffusion through the water phase, it has been proposed (8, 10) to examine the effect of changing the concentration of acceptor membranes. Our results in this line are in favor of the diffusion, since initial rates of the transfer of phosphatidic plus lysophosphatidic acids were independent of the concentration of the acceptor membrane (Fig. 3). A similar picture was also observed for the transfer of phospholipids labeled with [‘4C]choline. However, as already mentioned, in that case mostly lysophosphatidylcholine was transferred. Some increase of the transfer of phosphatidic acid with increasing concentration of the acceptor membranes after a longer time (Fig. 3) is not incompatible with the mechanism of free diffusion, because the picture is disturbed by the back transfer which becomes more

OF PHOSPHATIDIC

ACID

307

pronounced when the concentration of the acceptor membrane is low. The fact that the transfer depended, within a certain range, on the concentration of donor membranes may be due to the fact that the “saturation” of the water phase with monomeric or micellar phosphatidic acid was not attained at low concentration of microsomes. These results are compatible with the results of De Cuyper and Joniau (32) who have demonstrated in an elegant way, using free flow electrophoresis, that the transfer of phosphatidic acid proceeds between phospholipid vesicles by diffusion and not by membrane collision. Moreover, these authors have also shown that phosphatidic acid can be more readily exchanged between membranes than other phospholipids, which is most likely related to its negative charge. The surface charge of the membranes plays, most likely, an important role in the spontaneous transfer of phosphatidic acid. This is indicated by our observation (14) that this transfer was substantial in the KC1 medium but did not occur in slightly buffered sucrose. Probably, screening of the acceptor membrane is necessary to make it accessible for strongly negatively charged phosphatidic acid in monomeric or micellar form. It has to be stressed that phosphatidic acid which is spontaneously transferred from microsomes to mitochondria is metabolically active, since it is a precursor of 32P-labeled cardiolipin and phosphatidylglycerol, as demonstrated previously (14). The present study helps to understand the mechanism by which phosphatidic acid is transported from the outer to the inner mitochondrial membrane. In a previous publication (14) we were unable to show the presence of an exchange protein for this phospholipid in the intermembrane compartment. Moreover, Blok et al. (33) could not find transport proteins for nitrogen-containing phospholipids in this compartment either and have proposed that, inside the mitochondrion, the transport may proceed due to a close contact of the two membranes. Our results (Table I)

308

BARANSKA

AND

have shown that the transfer of phosphatidic acid between the two membranes, possibly by diffusion through the water phase or otherwise facilitated by the close contact of the two membranes, is fast enough not to be the limiting factor for the synthesis of phosphatidylglycerol and cardiolipin. In conclusion, the present study has shown that phosphatidic acid can be detached from microsomes (and presumably other natural and model phospholipid membranes) and diffuse through the water phase under ionic conditions much easier than other phospholipids. This property may make the transport of phosphatidic acid within a living cell relatively independent of phospholipid transfer proteins. ACKNOWLEDGMENTS The authors thank Dr. Jozef Zborowski for helpful discussions. The skillful technical assistance of Irena Makowska and Maria Buszkowska is gratefully acknowledged. This study was supported by a grant from the Polish Academy of Sciences under Project CPBP 4.01. REFERENCES 1. WIRTZ, K. W. A. (1974) B&him. Biophys. Acta 344,95-117. 2. ZILVERSMIT, D. B., AND HUGHES, M. E. (1976) in Methods in Membrane Biology (Korn, E. D., Ed.), Vol. 7, pp. 211-259, Plenum, New York. 3. KADER, J. C., DOUADY, D., AND MAZLIAK, P. (1982) in Phospholipids (Hawthorne, J. N., and Ansel, G. B., Eds.), New Comprehensive Biochemistry, Vol. 4, pp. 279-311, Elsevier, Amsterdam. 4. KREMER, J. M. H., KOPS-WERKHOVEN, M. M., PATHMAMANOHARAN, C., GIJZEMAN, 0. L. J., AND WIERSEMA, P. H. (1977) B&him. Biophys. Acta 471,177-188. 5. BARSUKOV, L. I., VICTOROV, A. V., VASILENKO, I. A., EVSTIGNEEVA, R. P., AND BERGELSON, L. D. (1980) Biochim. Biophys. Acta 598, 153-168. 6. MARTIN, F. J., AND MACDONALD, R. C. (1976) Bie chemistry 15,321-327. 7. PAPAHADJOPOULOS, D., Hur, S., VAIL, W. J., AND POSTE, G. (1976) Biochim. Biophys. Acta 448, 245-264.

WOJTCZAK 8. ROSEMAN, M. A., AND THOMPSON, T. E. (1980) Bie chemistry 19,439-444. 9. NICHOLS, J. W., AND PAGANO, R. E. (1981) Bio-

chemistry 20,2783-2789. 10. MCLEAN, L. R., AND PHILLIPS, M. C. (1981) Bio-

chemistry 20,2893-2900. 11. DE CUYPER, M., JONIAU, M., AND DANCREAU, H. (1983) Biochemistry 22,415-420. 12. DE CUYPER, M., JONIAU, M., ENGBERTS, J. B. F. N., AND SUDHOLTER, E. J. R. (1984) Colloids Surfaces 10,313-319. 13. MCLEAN, L. R., AND PHILLIPS, M. C. (1984) Bie

chemistry 23,4624-4630. 14. BARAASKA, J., AND WOJTCZAK, L. (1984) B&him.

Biophys. Acta 773,23-31. 15. MEGLI, F. M., DE LISI, A., VAN AMERONGEN, A., WIRTZ, K. W. A., AND QUAGLIARIELLO, E. (1986) Biochim. Biophys. Acta 861,463-470. 16. HOGEBOOM, G. H. (1955) in Methods in Enzymology (Colowick, S. P., and Kaplan, N. O., Eds.), Vol. 1, pp. 16-19, Academic Press, New York. 17. WETTERAU, J. R., AND ZILVERSMIT, D. B. (1984) J.

Biol. Chem. 259,10863-10866. 18. SCHNAITMAN, C., AND GREENAWALT, J. W. (1969) J. Cell Biol. 38, 158-175. 19. BJERVE, K. S. (1973) Biochim. Biophys. Acta 296,

549-562. 20. ZBOROWSKI, J., AND WOJTCZAK, L. (1969) Biochim. Biophys. Acta 187,73-84. 21. BJERVE, K. S., DAAE, L. N. W., AND BREMER, J. (1974) And Biochem. 58,238-245. 22. BLIGH, E. G., AND DYER, W. J. (1959) Canad. J.

B&hem. Physiol. 37,911-917. 23. WOJTCZAK, L., WLODAWER, P., AND ZBOROWSKI, J. (1963) Biochim. Biophys. Acta 70,290-305. 24. WAGNER, H., HURHAMMER, L., AND WOLFF, P. (1961) B&hem. 2. 334,175-184. 25. BARA~~SKA, J. (1980) Biochim. Biophys. Acta 619, 258-266. 26. SOTTOCASA, G. L., KUYLENSTIERNA, B., ERNSTER, L., AND BERGSTRAND, A. (1967) J. Cell Biol. 32, 415-438. 27. KENNEDY, E. P. (1953) J. Biol. Chem. 201,399-412. 28. GOULD, J. M., CARTHER, R., AND WINGET, G. D. (1972) Anal. Biochem 50,540-548. 29. GORNALL, A. G., BARDAWILL, C. J., AND DAVID, M. M. (1949) J. Biol. Chem. 177.751-766. 30. STUHNE-SEKALEC, L., AND STANACEV, N. Z. (1979) Canad. J. Biochem. 58,1082-1090. 31. MEGLI, F. M., LANDRISCINA, C., AND QUAGLIARIELLO, E. (1981) Biochim. Biophys. Acta 640, 274-284. 32. DE CUYPER, M., AND JONIAU, M. (1985) B&him. Biophys. Acta 814,374-380. 33. BLOK, M. C., WIRTZ, K. W. A., AND SCHERPHOF, G. L. (1971) B&him Biophys. Acta 233,61-75.