Stimulation of Na’-Ca” Exchange in Cardiac Sarcolemmal Vesicles by Phospholipase D*

Treatment of canine cardiac sarcolemmal vesicles with phospholipase D resulted in a large stimulation (up to 400%) of Na+-Ca2+ exchange activity. The phospholipase D treatment decreased the apparent Km (Ca2+) for the initial rate of Nai+-dependent Ca2+ uptake from 18.2 +/- 2.6 to 6.3 +/- 0.3 microM. The Vmax increased from 18.0 +/- 3.6 to 31.5 +/- 3.6 nmol of Ca2+/mg of protein/s. The effect was specific for Na+-Ca2+ exchange; other sarcolemmal transport enzymes ((Na+, K+)-ATPase; ATP-dependent Ca2+ transport) are inhibited by incubation with phospholipase D. Phospholipase D had little effect on the passive Ca2+ permeability of the sarcolemmal vesicles. After treatment with 0.4 unit/ml of phospholipase D (20 min, 37 degrees C), the sarcolemmal content of phosphatidic acid rose from 0.9 +/- 0.2 to 8.9 +/- 0.4%; simultaneously, Na+-Ca2+ exchange activity increased 327 +/- 87%. It is probable that the elevated phosphatidic acid level is responsible for the enhanced Na+-Ca2+ exchange activity. In a previous study (Philipson, K. D., Frank, J. S., and Nishimoto, A. Y. (1983) J. Biol. Chem. 258, 5905-5910), we hypothesized that negatively charged phospholipids were important in Na+-Ca2+ exchange, and the present results are consistent with this hypothesis. Stimulation of Na+-Ca2+ exchange by phosphatidic acid may be important in explaining the Ca2+ influx which accompanies the phosphatidylinositol turnover response which occurs in a wide variety of tissues.

regulating the trans-sarcolemmal fluxes of Ca2+ which accompany each heart beat. Sarcolemmal Na+-Ca'+ exchange can be stimulated by membrane potential (2-4), high pH ( 5 ) , proteinase treatment (6), or calmodulin-dependent phosphorylation (7). In a recent study (8), we found that phospholipase C treatment caused mild stimulation of Na+-Ca2+ exchange. We speculated that the negatively charged phospholipids phosphatidylinositol and phosphatidylserine may be espe- cially important in the Na+-Ca2+ exchange process. In the present study, we further pursue the relationship between Na+-Ca'+ exchange and negatively charged phospholipids through the use of phospholipase D which converts phospholipids to phosphatidic acid. Striking stimulation of Na+-Ca2+ exchange is observed after phospholipase D treatment.
Breakdown of phosphatidylinositol has been correlated with the influx of Ca'+ which accompanies stimulus-response coupling in many tissues (for reviews, see . It has been speculated that phosphatidic acid, a breakdown product of phosphatidylinositol, may be involved in the Ca2+ influx response. Our results provide evidence of a stimulatory role for phosphatidic acid in a well defined Caz+ transport pathway.

MATERIALS AND METHODS
Sarcolemmal vesicles were prepared from canine ventricles as described previously (g).' The vesicles used for most of the experiments reported here had the following sarcolemmal marker enzyme activities: K+-dependentp-nitrophenyl phosphatase activity was 35.8 k 2.7 pmol/mg of protein/h and was purified 76.0 + 8.0-fold as compared with the initial tissue homogenate. The yield of this enzyme was 18.3 f 2.7%. (Na+,K')-ATPase activity was 51.0 2 5.6 and 140.3 f 6.2 pmol/mg of protein/h in the absence and presence of alamethicin (12.5 pg/ml; donated by Dr. J. E. Grady, The Upjohn Co.), respectively. n = 6 for all data. Further characterization is given in Ref. 9.
Sarcolemmal vesicles (1-3 mg of protein/ml) were treated with phospholipase D by mixing 0.014 ml of Na'-loaded vesicles (140 mM NaCl, 10 mM Mops' (pH 7.4, 37 "C)) with an equal volume of phospholipase D dissolved in the identical medium. Aliquots (0.005 ml) were used directly in Ca'+-transport experiments after 20 f 1 min. The phospholipase D (Streptomyces chromofuscus, Calbiochem, La Jolla, CA) activity was 33 IU/mg and the sarcolemmal protein/ phospholipase D ratio in most experiments was about 80. Proteinase contamination in the phospholipase D was assessed using a Boehringer Mannheim kit (catalogue no. 582433) which determines the ability of a sample to solubilize a fibrin film; 1.0 IU of phospholipase D had less than 0.004 IU of proteinase activity using trypsin as a standard.
Na+-Ca'+ exchange was measured as the initial rate of Nai+dependent Ca'+ uptake as described in detail previously (5, 6, 8, 9). Briefly, 0.005 ml of Na+-loaded (140 mM NaCl, 10 mM Mops (pH 7.4, 37 "C)) sarcolemmal vesicles was rapidly diluted into 0.25 ml of Ca2+ uptake medium containing 140 mM KC1, variable CaC12, 1.25 pCi of ''CaC12, 0.4 p~ valinomycin, 10 mM Mops (pH 7.4, 37 "C). After 1.0 or 1.5 s (unless otherwise noted), the CaZ+ uptake was automatically quenched by addition of 0.03 ml of stopping solution. Our usual stopping solution was 140 mM KCI, 1 mM LaCls, but we found this stopping solution resulted in unusually high blank values in experiments using phospholipase D-treated sarcolemmal vesicles. Therefore, many of the Ca2+ uptake experiments were quenched using 0.03 ml of 140 mM KCI, 10 mM EGTA followed by the immediate addition of 1.0 ml of ice-cold 140 mM KCl, 1 mM EGTA. The vesicles were then harvested by Millipore filtration (0.45 pm) and washed with two 3-ml aliquots of cold 140 mM KCI, 1 mM EGTA. Qualitatively similar results were obtained with either stopping technique. Blank values were obtained by using Ca'+ uptake medium which contained 140 mM NaCl instead of KCl. Blank values were subtracted for all data points to correct for superficially bound Ca'+ and Na+ gradient-independent Ca'+ uptake.
The endogenous Ca2+ level in the Ca2+ uptake solutions was about 2 p~ as determined by a Ca2+-selective electrode (Orion). The data shown in Fig. 2 have been corrected for the additional Ca2+.
Data are presented as mean ? S.E.

Fig
. 1 demonstrates the stimulation of the initial rate of Na+-Ca'+ exchange activity in canine cardiac sarcolemmal vesicles after phospholipase D pretreatment. In all experiments, Na+-Ca2+ exchange is measured as N&+-dependent Ca2+ uptake. At the higher phospholipase D concentrations Na+-Ca'+ exchange is 4-fold higher than the initial level. The stimulation by phospholipase D is specific for Na+-Ca'+ exchange. Two other sarcolemmal transport activities, (Na+,K+)-ATPase and the ATP-dependent Ca2+ pump, show moderate or large inhibitions, respectively, after phospholipase D treatment (Fig. 1). The source of the phospholipase D was S. chrornofuscus; this enzyme (15) does not require the low pH or high Ca2+ concentration required of the more commonly used phospholipase D from peanut or cabbage.
The dependence of Na+-Ca2+ exchange activity on [Ca'+] is changed after exposure of the sarcolemmal vesicles to phos-pholipase D. Fig. 2 shows a representative experiment in which both the capacity and apparent affinity for Ca'+ are increased after enzyme treatment. In a series of five experiments, the V,,, of Na+-Ca2+ exchange was elevated from 18.0 f 3.6 to 31.5 f 3.6 nmol of Ca2+/mg of protein/s while the apparent K,,, for Ca2+ decreased from 18.2 k 2.6 to 6.3 f 0.3

P "
The enhanced Na,+-dependent Ca'+ uptake reflects Na+-Ca2+ exchange activity and is not due to effects of phospholipase D on other transport pathways. The Ca'+ taken up by phospholipase D-treated vesicles could be readily released by the addition of the Ca2+ ionophore A23187 (0.75 PM). In 1.0 min, 92% of the Ca2+ was released by ionophore addition. This indicates that Ca'+ had been accumulated against a concentration gradient (using the energy of the Na gradient) and that phosphatidic acid was not acting primarily as a Ca" ionophore to enhance passive Ca2+ influx. Likewise, if monensin (2 PM) was used to dissipate the Na+ gradient obtained by diluting Na+-loaded vesicles into KC1 medium, the Nai+dependent Ca'+ uptake activity could be eliminated (in both control and phospholipase D-treated vesicles). This demonstrates that the observed Ca2+ transport activity was dependent upon the presence of an outwardly directed Na' gradient. In all Na+-Ca2+ exchange experiments, valinomycin (0.4 PM) was included in the Ca2+ uptake medium to maintain an inside-positive membrane potential for maximal Nai+-dependent Ca'+ uptake activity (4). Thus, the effects of phospholipase D were due to direct stimulation of Na+-Ca2+ exchange and were not due to effects on sarcolemmal membrane potential. In fact, the stimulatory effects of phospholipase D on Na+-Ca2+ exchange were evident even in the absence of valinomycin or if sucrose or choline chloride were used in the Ca2+ uptake medium instead of KC1.
In the experiments described above, sarcolemmal vesicles were preincubated with phospholipase D and this mixture was then used directly in Na+-Ca'+ exchange experiments (see "Materials and Methods"). Thus, the phospholipase D and any soluble hydrolysis products were present during the brief period of the Ca2+ uptake reaction and may have affected the measurement. Since the vesicles are diluted 50-fold to initiate

Nai-Ca2+
Exchange and Phospholipase D the Ca'+ transport, the concentrations of these substances would be very low in the final uptake medium. Control experiments were run in which control andphospholipase D-treated (0.4 unit/ml) vesicles were diluted and pelleted by centrifugation to remove the phospholipase D. After resuspension, the vesicles which had been exposed to phospholipase D had Na+-Ca2+ exchange activity which was 315% of control levels. This demonstrates that the stimulated exchange was due to a membrane-bound product of phospholipase D hydrolysis.
The preparation of cardiac sarcolemmal vesicles contains a mixture of inside-out and right-side-out vesicles (9).' We tested whether the phospholipase D treatment was preferentially stimulating the Na+-Ca2+ exchange of one type of sarcolemmal vesicle using our published technique (9) for examining the exchange of inside-out vesicles only. We found (not shown) that the inside-out vesicles and the total population (inside-out plus right-side-out vesicles) were stimulated to equal extents by phospholipase D treatment. Apparently, both the inside-out and right-side-out sarcolemmal vesicles are susceptible to and responsive to phospholipase D treatment.
Since phosphatidic acid has been reported to act as a Ca2+ ionophore (16-23), we examined the effects of phospholipase D treatment on the passive Ca2+ permeability of the sarcolemmal vesicles. Control or phospholipase D-treated vesicles were first loaded with Ca2+ by Nai+-dependent Ca2+ uptake. EGTA was added after 30 s to inhibit further Ca'+ uptake and the slow loss of Ca'+ from the vesicles was a measure of the passive Ca2+ permeability. As shown in Fig. 3, there was little change in passive Caz+ flux after phospholipase D treatment.
We observe a large stimulation of Na+-Ca2+ exchange activity when the initial rate of Na,+-dependent Ca'+ uptake is measured over 1.0 or 1.5 s (Figs. 1 and 2). The magnitude of the stimulation decreases with time and after 30 s of Ca2+ uptake only a moderate stimulation is noted. This explains why the Ca'+ content of the phospholipase D-treated vesicles (loaded with Ca'+ by 30 s of Na+-Ca2+ exchange) in Fig. 3 is We analyzed the sarcolemmal membrane by thin layer chromatography using a solvent system which separated the phospholipids into three spots representing phosphatidic acid, phosphatidylethanolamine, and a combination of the other sarcolemmal phospholipids (phosphatidylcholine, phosphatidylserine, phosphatidylinositol, and sphingomyelin). After phospholipase D treatment (0.4 unit/ml), the vesicle phosphatidic acid level increased from 0.9 k 0.2 to 8.9 2 0.4% ( n = 4) of the total phospholipid content. In these same sarcolemmal samples, the initial rate of Na+-Ca2+ exchange rose 327 k 78% after enzyme treatmwt. The phospholipid contents of both of the two other phospholipid spots decreased by about equal percentages. The normal phospholipid composition of the vesicles is given in Ref. 8. One incidental finding was that the sarcolemmal membranes apparently possess a phosphatidic acid phosphatase activity. After phospholipase D treatment, the total phospholipid content of vesicles was slightly diminished. This decrease could be largely accounted for by the appearance of inorganic phosphate. Evidently, some of the phosphatidic acid produced by the phospholipase D was being hydrolyzed by the sarcolemma to diacylglycerol. If exogenous phosphatidic acid was added to sarcolemmal vesicles, inorganic phosphate would rapidly appear in the medium consistent with this interpretation. Plasma membrane phosphatidic acid phosphatase has been described in other systems (24) and will be the object of a separate study.
Thin section electron microscopy (courtesy of Dr. J. S. Frank, Department of Medicine, UCLA) of phospholipase D (0.4 unit/ml)-treated vesicles did not reveal morphological damage such as that seen after phospholipase C treatment (8).

DISCUSSION
We have demonstrated that phospholipase D pretreatment causes a large stimulation in the initial rate of Na+-Ca2+ exchange in cardiac sarcolemmal vesicles. The stimulation is specific for Na+-Ca2+ exchange; we observe no stimulation of other sarcolemmal ion transport enzymes (Fig. 1). The phospholipase D increases both the capacity and Ca2+ affinity of the exchange mechanism (Fig. 2). The enhanced Ca2+ uptake is Na+ gradient-dependent and results in the formation of an outwardly directed Ca2+ gradient. This eliminates the possibility that the phospholipase D-induced increase in Ca" uptake is due to an ionophoretic action (16-23) of phosphatidic acid. Although a small increase in passive Ca2+ flux may be occurring (Fig. 3), this is not the dominant effect of phospholipase D in our system.
It is most likely that the phosphatidic acid produced by phospholipase D is responsible for the stimulation of the Na+-Ca2+ exchange. Stimulation persists after soluble hydrolysis products are removed by centrifugation. Large stimulation of Na+-Ca2+ exchange is apparent when about 9% of the sarcolemmal phospholipid is phosphatidic acid.
In a recent study (8), we found that phospholipase C (which produces diacylglycerol) caused moderate stimulation (20-80%) of sarcolemmal Na+-Ca2+ exchange. Since the negatively charged phospholipids phosphatidylserine and phosphatidylinositol were little effected by this enzyme, their relative sarcolemmal content (per cent of total phospholipid) increased after phospholipase C treatment. We speculated that negatively charged phospholipids could enhance Na+-Ca2+ exchange activity. The present results demonstrate in a much more direct manner the stimulation of Na+-Ca2+ exchange by a negatively charged phospholipid. Perhaps phosphatidic acid associates with the Na+-Ca2+ exchange protein and increases Na+-Ca2+ Exchange and Phospholipase D 19 the availability of Ca2+ for exchange. Increased turnover of plasma membrane phosphatidylinositol is associated with a large number of stimulus-response coupling mechanisms involving receptors which control Ca2+ influx (for reviews, see Refs. 10-12). A prevalent finding is that the phosphatidylinositol is hydrolyzed to diacylglycerol which is phosphorylated to phosphatidic acid. It has been hypothesized that it is the phosphatidic acid which is responsible for the Ca'+ influx. The hypothesis is primarily based on results which show that phosphatidic acid can act as a Caz+ ionophore in model systems (16, 22, 23). Evidence has also accrued that phosphatidate can increase Ca2+ influx into intact cells from several tissues (18-21). The implication in the cellular systems is that, by analogy, the phosphatidic acid is again acting as a Cap+ ionophore. There is no strong evidence, however, to support this assertion. An alternative explanation is that phosphatidic acid mobilizes cellular Cap+ flux by activation of plasma membrane Na+-Ca'+ exchange. Na+-Ca2+ exchange activity has been reported in plasma membrane vesicles from an increasing number of tissues and may be widespread (e.g. . Electrogenic Na+-Ca2+ exchange can transport Ca2+ in both directions across the plasma membrane. Whether stimulation of Na+-Ca'+ exchange will result in a net influx or efflux of Ca2+ will be a function of the intra-and extracellular Na+ and Ca'+ levels and of the membrane potential. Thus, another possibility is that phosphatidate-stimulated Na+-Ca2+ exchange may be important in returning elevated cellular Ca2+ to its normal level following the "phosphatidylinositol response."