Inhibitory Action of Chlorpromazine, Dibucaine, and Other Phospholipid-interacting Drugs on Calcium-activated, Phospholipid-dependent Protein Kinase*

Ca2+-activated, phospholipid-dependent protein ki- nase recently found in mammalian tissues (Takai, Y., Kishimoto, A., Iwasa, Y., Y., T., Y. J. Biol. Chem 254, 3692-3695) is inhibited by various phospholipid-interacting drugs such as chlorpromazine, imipramine, phentolamine, dibucaine, verapamil, and tetracaine. This effect is attrib- uted to the inhibition of the activation process but not to the interaction with the active site of enzyme. This is supported by the fact that a catalytic fragment of this enzyme, which is obtained by limited proteolysis with Ca2+-dependent neutral protease, is fully active without Ca” and phospholipid and is not susceptible to any of these drugs. Kinetic analysis suggests that these drugs cause such inhibition competitively with phospholipid. None of these drugs appears to compete with Ca2’ or to counteract the unique effect of unsaturated diacylglycerol. Unsaturated diacylglycerol has been shown previously to increase markedly the affinity of enzyme for Ca2+ as well as for phospholipid and thereby serve as an initiator for the activation of this protein kinase. Neither cyclic AMP-dependent nor cyclic GMP-dependent protein kinase is susceptible to these phospholipid-interacting

pounds can cause inhibition of cell fusion (3), cell spreading and motility ( 4 4 , exocytosis (7), thrombin-induced platelet aggregation (a), and phytohemagglutinin-induced lymphocyte transformation (9). Ca'+ is essential for all of these cellular processes (for reviews, see Refs. 10 and 11). Recent analysis in this laboratory has found in mammalian tissues a new species of multifunctional protein kinase which is specifically activated by Ca2+ and membrane phospholipid but not by cyclic nucleotides (12,13). This protein kinase is activated by association with membrane phospholipid in the presence of Ca'+, and a small amount of unsaturated diacylglycerol greatly increases the apparent affinity of enzyme for phospholipid and sharply decreases the Ca2+ concentration giving rise to maximum enzyme activation (14, 15). Since a variety of extracellular messengers, including thrombin and plant mitogen, induce rapid turnover of phosphatidylinositol in the respective target cell membranes to provide such an active unsaturated diacylglycerol (for reviews, see Refs. 16-18), it is possible that this protein kinase may play roles in transmembrane control of various cellular processes mentioned above. The results to be described here will indicate that various drugs such as chlorpromazine and dibucaine appear to interact with phospholipid and thereby strongly inhibit the activation of this protein kinase. The Ca2+-activated, phospholipid-dependent protein kinase, cyclic AMP-dependent protein kinase, and cyclic GMP-dependent protein kinase will be referred to as protein kinases C, A, and G, respectively.

EXPERIMENTAL PROCEDURES
Materials and Chemicals-Protein kinase C was partially purified from rat brain soluble fraction as previously described (19). The catalytic fragment of this kinase was prepared from the native enzyme by limited proteolysis with rat brain Ca2+-dependent neutral protease and was isolated by gel filtration procedure under the conditions specified earlier (19). Protein kinases A and G were prepared as described previously (20,21). These enzyme preparations were free of each other and of endogenous phosphate acceptor proteins. Histone used as phosphate acceptor was prepared from calf thymus as specified earlier (22). Phospholipid was extracted from bovine brain by the method of Folch et al. (23) and fractionated on a silicic acid column as described by Rouser et al, (24). Diolein was purchased from Nakarai Chemicals. [y-"PIATP was prepared by the method of Glynn and ChappeU (25). Chlorpromazine hydrochloride, imipramine hydrochloride, phentolamine hydrochloride, dibucaine hydrochloride, verapamil hydrochloride, and tetracaine hydrochloride were donated by Shionogi Research Laboratory, CIBA-GEIGY Japan, CIBA-GEIGY, Teikoku Chemical Industry, Eisai Co., and Kyorin Pharmaceutical Co., respectively.
Enzyme Assay-Unless otherwise noted, protein kinase C was routinely assayed at 2 X M Ca2+ in the presence of phospholipid and diolein under the conditions given in the legend to Fig. 1. Various phospholipid-interacting drugs were added as indicated in each experiment. For the assay of catalytic fragment, Ca2+, phospholipid, and diolein were omitted. Protein kinases A and G were assayed with calf thymus whole histone as phosphate acceptor (20,21).
Determinations-Phospholipid phosphorus and protein were determined by the methods of Bartlett (28) and Lowry et al. (29), respectively. Radioactivity of J2P incorporated into histone was determined using a Nuclear-Chicago Geiger-Muller gas flow counter, model 4338.

Protein Kinase and Phospholipid-interacting
Drugs 8379

RESULTS AND DISCUSSION
Among a large number of local anesthetics, tranquilizers, and other drugs which are known as phospholipid-interacting compounds, the following six drugs were chosen simply because these were often used to modify various cellular processes (1, 2): chlorpromazine, imipramine, phentolamine, dibucaine, verapamil, and tetracaine. Fig. 1 shows that protein kinase C was inhibited to variable extents by these drugs when the enzyme was assayed in the presence of Ca", phospholipid, and diolein as an unsaturated diacylglycerol. Chlorpromazine appeared to be the most potent inhibitor. In preceding reports from this laboratory (19,30), protein kinase C Phospholipid was first mixed with diolein in a small volume of chloroform. After the chloroform was removed in uacuo, the mixture was suspended in 20 m~ Tris/HCl at pH 7.5 by sonication as described previously (14) and employed for the assay. The incubation was carried out for 3 min at 30°C. The reactions were stopped by the addition of 25% trichloroacetic acid, and acid-precipitable materials were collected on a Toyo-Roshi membrane filter (pore size, 0.45 pm). The radioactivity was determined as described previously (27)   has been shown to be alternatively activated in an irreversible manner by limited proteolysis with Ca'+-dependent neutral protease. In this process, a catalytically fully active fragment is produced which is entirely independent of Ca", phospholipid, and unsaturated diacylglycerol. It was noted that such a catalytic fragment was not susceptible to any of these drugs, as shown in Table I. It is unlikely, therefore, that these drugs interact with the active center of enzyme or with the substrate protein and thereby inhibit the enzymatic reaction. In addition, Ca2+-dependent neutral protease was inhibited by none of these drugs. Thus, these drugs specifically blocked the reversible activation process which may be elicited by association with membrane phospholipid but did not interfere with the irreversible activation process which may be mediated by Ca2'-dependent neutral protease. Inversely, the irreversible proteolytic activation process was inhibited in purified system by various thiol protease inhibitors such as leupeptin and E-64. These protease inhibitors did not affect the reversible activation process nor the already proteolytically activated protein kinase C. It was also noted that protein kinases A and G were not susceptible to any of the phospholipid-interacting drugs nor to protease inhibitors.
Next, experiments were performed to explore the mode of inhibitory action of these drugs. Since the reversible activation of protein kinase C requires the simultaneous presence of Ca2+, phospholipid, and unsaturated diacylglycerol, each of the three factors was varied in the presence and absence of the phospholipid-interacting drugs mentioned above. As shown in Fig. 2, these drugs appeared to slightly modify the K , value for Ca", the concentration needed for the halfmaximum velocity, but the results seemed to be difficult to interpret in consideration of the preserved activation in the absence of diolein. Nevertheless, it is unlikely that these drugs compete with this divalent cation and thereby inhibit the reaction. Kinetic analysis has previously shown that unsaturated diacylglycerol greatly increases the affinity of enzyme for Ca2+ as well as for phospholipid, thus rendering the enzyme more active, particularly at lower concentrations of Ca2' (14, 15). The inhibitory action of these drugs did not counteract such a unique action of unsaturated diacylglycerol, nor was it cancelled by the addition of increasing amounts of the unsaturated neutral lipid. Instead, the results given in Fig. 3 suggested that the inhibitory action of all the drugs mentioned above was competitive with phospholipid. The kinetic analysis described above does not clarify the mechanism of activation or inhibition of protein kinase C but seems to provide an additional indication that some phospholipid bilayer structure may seriously influence the activation of enzyme. A preliminary analysis indicated that, in the presence of various phospholipid-interacting drugs, the enzyme could still be associated with membranes but did not reveal its catalytic activity. Presumably, these drugs interact with phospholipid and thereby confer profound modification of the specific hydrophobic lipid-protein interaction. It seems to be pharmacologically accepted that most of the drugs employed for the present studies interact with excitable membranes and block excitatory processes (1). Although the inhibition of protein kinase C may not be necessarily the sole target of these drug actions, several drugs mentioned above may serve as a useful tool for differentiating roles of the three sets of closely similar enzyme systems, namely protein kinases A, G, and C.