Inhibition of regulated catecholamine secretion ... - Semantic Scholar

2619

Journal of Cell Science 108, 2619-2628 (1995) Printed in Great Britain © The Company of Biologists Limited 1995

Inhibition of regulated catecholamine secretion from PC12 cells by the Ca2+/calmodulin kinase II inhibitor KN-62 Erik S. Schweitzer1,*, Michael J. Sanderson1,† and C. G. Wasterlain2 Departments of 1Anatomy and Cell Biology, and 2Neurology, UCLA School of Medicine, Los Angeles, CA 90024, USA *Author for correspondence †Present address: Department of Physiology, University of Massachusetts Medical Center, 55 Lake Avenue North, Worchester, MA 01655, USA

SUMMARY When stimulated by the cholinergic agonist carbachol, PC12 cells rapidly secrete a large fraction of the intracellular catecholamines by exocytotic release from the large dense-core secretory vesicles in a Ca2+-dependent manner. To investigate whether Ca2+/calmodulin kinase II plays a role in the regulated secretion of catecholamines, we examined the effect of the specific Ca2+/calmodulin kinase II inhibitor KN-62 on the carbachol-induced release of norepinephrine from PC12 cells. Approximately 50% of the regulated release of norepinephrine, stimulated either by carbachol or direct depolarization, was inhibited by pretreatment with KN-62, while the remaining 50% was resistant to KN-62 and therefore independent of Ca2+/calmodulin kinase II. In contrast, H7, an inhibitor of protein kinase C, had no effect on any of the stimulated

release. FURA 2 imaging experiments demonstrated that KN-62 does not act by blocking the stimulation-induced increase in intracellular [Ca2+]. The most likely model consistent with these data is that all the dense-core vesicles fuse with the plasma membrane in a Ca2+-dependent process, but that approximately 50% of the vesicles require an additional step that is dependent on the action of Ca2+/calmodulin kinase II. This step occurs between the influx of Ca2+ and the fusion of vesicle membranes with the plasma membrane, and may be analogous to the Ca2+/calmodulin kinase II phosphorylation of synapsin which mobilizes small, clear synaptic vesicles for exocytosis at the synapse.

INTRODUCTION

second type involves not only small molecules but also neuropeptides or neuromodulator proteins. These two types of regulated secretion differ in a variety of aspects, including the types of vesicles involved (small clear vesicles vs large, densecore vesicles), the site of packaging of the contents into the vesicles (locally from cytoplasmic stores vs in the trans Golgi apparatus), and the kinetics and regulation of the fusion of the vesicles with the plasma membrane (fast vs slow). In both cases, vesicle exocytosis is stimulated by an increase in intracellular Ca2+ (Ca2+i), but the intermediate events between influx of Ca2+ through Ca channels (or in some cases release from internal stores by IP3) and vesicle fusion are still poorly understood. Similarly, the extent of the differences in the molecular machinery involved in secretion from small clear vesicles and large dense-core vesicles has not been clearly defined. Some vesicle components that have been implicated in vesicle fusion, such as synaptophysin and the synapsins, are largely or exclusively associated with small synaptic vesicles and not large dense-core vesicles, and are therefore likely to be involved only in secretion from small clear vesicles (Bennett and Scheller, 1994; Jahn and Südhoff, 1994). Other proteins, including synaptotagmin (Matthew et al., 1981) and SV2 (Buckley and Kelly, 1985; Bajjalieh et al., 1992), are components of both small clear synaptic vesicles and large, dense-core

A central unsolved question in cell biology is what mechanisms regulate cellular secretion. Although considerable progress is being made in identifying the molecular components of secretory vesicles in neuronal and endocrine cells, the detailed sequence of events and the molecular components which play an essential role in regulated secretion remain difficult to define. While Ca2+ has long been recognized as an essential trigger for regulated exocytosis, the specific intermediates through which Ca2+ exerts its effects remain unknown. Calmodulin and the Ca2+/calmodulin-dependent protein kinase II (CaM kinase II) are likely candidates for such intermediates, but it has been difficult to obtain direct evidence for their involvement in secretion. The recent development of highly specific inhibitors of CaM kinase II makes these agents powerful tools for investigating the role that CaM kinase II plays in regulating secretion. Neuronal and endocrine cells exhibit multiple forms of secretion, including constitutive and regulated secretory processes (Schweitzer and Kelly, 1985; Bennett and Scheller, 1994). Moreover, two distinct types of regulated secretion have been distinguished in neuronal and neuroendocrine cells (Kelly, 1993). One type involves the release of small molecule neurotransmitters such as ACh, GABA, and glutamate; the

Key words: CaM kinase II, stimulus-secretion coupling, exocytosis, dense-core vesicle, regulated secretion

2620 E. S. Schweitzer, M. J. Sanderson and C. G. Wasterlain vesicles (Kelly, 1993). Although the specific functions of many of these vesicle proteins are not yet clear, there is increasingly convincing evidence that some are involved in neurotransmitter release (Bennett and Scheller, 1994). Of particular interest in relation to the present study, synapsin I interacts both with vesicle membranes and cytoskeletal elements, and this interaction is inhibited by the phosphorylation of synapsin carried out by CaM kinase II (Torri Tarelli et al., 1992). It appears that this phosphorylation is important for transmitter release to occur, since inhibitors of calmodulin (Schweitzer, 1987) or CaM kinase II (Ishikawa et al., 1990) are effective in blocking the release of small neurotransmitters (ACh and GABA). Moreover, microinjection of dephosphorylated synapsin I into the presynaptic region results in an inhibition of transmitter release, while addition of active CaM kinase II potentiates release (Llinás et al., 1991). Such data support a model in which the phosphorylation of synapsin I by CaM kinase II plays an essential role in transmitter release by freeing a reserve population of synaptic vesicles from the cytoskeleton so that they are competent to fuse with the plasma membrane and release their contents (Benfenati et al., 1993; Greengard et al., 1993). Blocking the phosphorylation of synapsin I would therefore be expected to have no effect on the fusion of vesicles already docked at the active zones, but would prevent the recruitment of the reserve pool of vesicles, thereby decreasing the total amount of transmitter release. Although synapsin I is not associated with dense-core vesicles, other actin-binding proteins, including fodrin (Aunis and Perrin, 1984) and α-actinin (Bader and Aunis, 1983) can bind both dense-core vesicles and actin. No information is currently available about whether these proteins are phosphorylated by CaM kinase II. The secretory pathway involving neuronal dense core vesicles differs in a number of respects from that of small clear synaptic vesicles, and appears to be similar in many respects to the process that releases hormones from the dense core vesicles of endocrine cells (Trifaró et al., 1992). Although the initial fusion event of chromaffin granules is rapid (Chow et al., 1992), the overall kinetics of release are slower, and there is no synapsin I associated with dense core vesicles (Senda et al., 1991; Trifaró et al., 1992). All these secretory processes are stimulated by a rise in [Ca2+]i, although Ca2+-independent processes involving protein kinases C may also play a role (Pozzan et al., 1984; Irons et al., 1992). It appears that protein kinase C (PKC) plays a modulatory, but not an essential role in exocytosis (Terbush and Holz, 1990), and may act through stimulating the autophosphorylation of CaM kinase II (MacNicol et al., 1990). In neuroendocrine PC12 cells, catecholamines, including norepinephrine (NE), are released from large dense-core vesicles along with a variety of secretory proteins and ATP (Greene and Tischler, 1976; Schweitzer and Kelly, 1985; Shoji-Kasai et al., 1992). Recently, several soluble factors have been identified as essential components of the secretory pathway (Walent et al., 1992; Hay and Martin, 1993), but the details of the biochemical events that lead to secretion remain unresolved. PC12 cells contain CaM kinase II (Nose et al., 1985), but its role, if any, in secretion, has never been demonstrated. Nevertheless, the essential role played by Ca2+ in triggering secretion makes CaM kinase II an attractive candidate for an intermediate in triggering exocytosis.

KN-62, a highly-selective inhibitor of CaM kinase II (Tokumitsu et al., 1990), has been shown to have inhibitory effects on the regulated secretion of insulin by isolated pancreatic islets (Wenham et al., 1992) and HIT cells (Guodong et al., 1992). A similar inhibitory effect of KN-62 on cholecystokinin secretion from STC-1 enteroendocrine cells has also been reported (Prpic et al., 1994). In the case of HIT cells and STC-1 cells, the inhibition was shown to be due to a block of Ca2+ influx through L-type Ca2+ channels. In atrial myocytes, KN-62 inhibited a maximum of 50% of the endothelin-stimulated release of atrial natriuretic factor; in these cells, complete inhibition of secretion was accomplished only after treatment with both KN-62 and H7, a PKC inhibitor (Irons et al., 1992). These findings suggest that cardiac endocrine cells utilize two independent mechanisms for controlling secretion - one involving CaM kinase II and the other, PKC. We have examined the effects of KN-62 on the regulated secretion of catecholamines from PC12 cells. KN-62 had an effect similar to that reported for atrial myocytes in that a maximum of 50% of the stimulated secretion is blocked. However, in PC12 cells, unlike atrial myocytes, the remaining 50% of secretion was not blocked by H7. Moreover, the site of action of KN-62 on PC12 cells appeared to be quite different from other endocrine cells studied, in that changes in [Ca2+]i were unaffected. It therefore appears that in PC12 cells there is an obligatory involvement of CaM kinase II in one of the steps between the entry of Ca2+ into the cytoplasm and the fusion of the dense core vesicles with the plasma membrane. MATERIALS AND METHODS Reagents All reagents were obtained from Sigma Chemical Co. (St Louis, MO), except [3H]NE (Amersham), KN-62 and H7 (Seikagaku Corp., Tokyo), FURA 2/AM and calcium ionophore A23187 (Calbiochem, San Diego, CA). NGF was purified from mouse salivary glands by the procedure of Mobley et al. (1976). Trifluoperazine dihydrochloride was a gift from Smith, Kline & French Laboratories (Philadelphia, PA); K-252a was a gift from P. Claude (Univeristy of Wisconsin, Madison). Cell culture PC12 cells (Greene and Tischler, 1976), which were subcloned and screened for normal growth and morphology as well as responsivity to NGF and transfectability (sub-clone A; Schweitzer, 1993) were used at a passage number of less than 20. These cells were screened and found to be free of Mycoplasma by staining with DAPI (Russell et al., 1975). Cells were routinely maintained in Dulbecco’s modified Eagle’s medium, high glucose (Hazelton Research Products, Lexena, KS) with the addition of 25 mM HEPES, 5% supplemented calf serum + 5% horse serum (HyClone Labs, Logan, UT), and penicillin + streptomycin, at 37°C in 9.5% CO2. NGF-treated cells were grown in the presence of 20 ng/ml NGF for 3 days prior to use. Release experiments NE release from PC12 cells was examined essentially as previously described (Schweitzer and Kelly, 1985). PC12 cells were grown in 10 cm tissue culture plates (Falcon), and loaded prior to the experiment by overnight incubation with 1-5 µCi [3H]NE (Amersham). They were then incubated in PC12 medium with specific additions as indicated. Nominally ‘0 Ca2+’ solutions were made by adding MgEGTA (pH 7.4) to a final concentration of 10 mM. For stimulation by direct depolarization, cells were incubated either in ‘Na

Autophosphorylation activity of CaM kinase II In vitro activity of CaM kinase II was measured by quantitating the extent of autophosphorylation of the α and β subunits of CaM kinase II in the presence of 3 units of calmodulin, 10 µM [γ-32P]ATP (0.31.0 Ci/nmole), 10 mM MgCl2, 0.5 mM EGTA, 0.6 mM CaCl2 (final free [Ca2+] was 0.1 mM), and 20 mM Tris-HCl, pH 7.4. Reactions were carried out in a volume of 75 µl at 30°C, and initiated by adding 50 µg of PC12 cell homogenate to the reaction mixture. After 1 minute, reactions were stopped by the addition of sodium dodecyl sulfate and β-mercaptoethanol (4% final concentration of each). Samples were then separated by electrophoresis on 10% sodium dodecyl sulfate-polyacrylamide gels (Laemmli, 1970), which were stained with Commassie Blue, dried, and exposed to DuPont Cronex 6+ X-ray film; the autoradiographs were quantitated by densitometry. Intracellular Ca2+ imaging NGF-treated cells were grown on poly-D-lysine coated coverslips (Schweitzer and Paddock, 1990), loaded by incubation in 5 µM FURA 2/AM in Hanks’ balanced salt solution supplemented with 25 mM HEPES, pH 7.4, and without Phenol Red for 7 minutes at 37°C, followed by incubation in culture medium without dye for 30 minutes at 37°C. Cells were then placed on an inverted microscope at room temperature and observed with a 510 nm filter using fluorescence excitation at 340 nm and 380 nm. The resulting images were acquired by SIT camera to an analog optical disk, and the data later analyzed for Ca2+ levels as described previously (Charles et al., 1991). Grayscale images representing free Ca2+ levels were assigned pseudocolors using the NIH Image software.

RESULTS As shown in Fig. 1, 5 mM carbachol caused a large, rapid increase in the rate of NE release (closed circles) from PC12 cells; in this experiment, this burst of secretion constituted approximately 20% of the total cellular content of NE. The addition of KN-62 by itself did not trigger any secretion of catecholamines (4 minute time point, triangles), but the subsequent addition of carbachol caused a release of only 60% as much NE as from the untreated cells (triangles, 6 minute time point). Under the same conditions, 100% of the carbacholstimulated release was blocked by chelation of extracellular Ca2+ (open circles). Although the PC12 cells used in Fig. 1 were differentiated by growth in the presence of NGF for 3 days prior to measuring release, the same results were obtained with non-NGF treated cells (see, for example, Figs 8, 10). The addition of 1% DMSO (which was used to dissolve the KN62) had no effect on either the basal or stimulated secretion of NE (data not shown). The dose-response relationship between the concentration of KN-62 and the extent to which carbachol-stimulated secretion

15

10

5

0 0

2

4

6

8

Collection Time (minutes)

Fig. 1. Inhibition of carbachol-stimulated release of NE by KN-62. PC12 cells were grown in duplicate dishes for 3 days in the presence of NGF. After loading with [3H]NE, the media bathing the cells were changed at 2 minute intervals. Control cells (d), were stimulated with medium containing 5 mM carbachol at 4 and 6 minutes. Another pair of dishes received medium containing 5 µM KN-62 at 2 minutes, and then 5 mM carbachol + 5 µM KN-62 at 4 and 6 minutes (m). A third pair of dishes was incubated with 0 Ca2+ medium containing 10 mM Mg-EGTA (added at 2 minutes), followed by medium with 5 mM carbachol + 10 mM Mg-EGTA at 4 and 6 minutes (s). The counts plotted represent the amount of [3H]NE released by the cells over the preceding 2 minute period. All data are the means ± ranges of duplicate dishes.

[ 3H]NE Release (10 3 dpm / dish / 2 minutes)

medium’ (150 mM Na, 5 mM K) or ‘K medium’ (100 mM Na, 55 mM K) as described previously (Schweitzer, 1993). Dishes were repeatedly rinsed with control medium to remove extracellular label, and then placed on a metal plate heated to 37°C. Media were changed by tilting the dishes, removing the media, and replacing with fresh media (at 37°) at 2 minute intervals. The media samples were analyzed for [3H]NE by scintillation counting. At the end of the experiments, the total [3H]NE remaining in the cells was determined by adding an extraction solution of 1% Triton X-100 in 150 mM NaCl and 25 mM HEPES, pH 7.4. Total [3H]NE was calculated by adding the amount remaining in the cells to the sum of the amounts released at each 2 minute interval during the experiment.


Inhibition of secretion by KN-62 2621

20

10

0

AAAAAA A 0

0.5

2

5

10

[KN-62] (µM)

Fig. 2. Dose-response relationship between KN-62 and the inhibition of carbachol-stimulated NE release. Duplicate dishes of PC12 cells were stimulated as in Fig. 1, except cells were not previously treated with NGF. At each concentration of KN-62, three sequential time points are shown; each time point represents a 2 minute collection of incubation medium. The first point in each set represents the baseline release in control medium, the second point is the release in the 2 minutes after addition of the KN-62, and the third point represents the release in the 2 minutes after addition of 5 mM carbachol (in the continued presence of KN-62). Data are the means ± ranges.

was inhibited is shown in Fig. 2. A maximum of 50% inhibition of NE release was achieved. Lengthening the time of preincubation longer than 2 minutes did not increase the extent of

50

Fig. 3. Recovery from KN-62 inhibition. Replicate plates of PC12 cells were loaded with [3H]NE, washed, and then incubated with various media at 2 minute intervals. All dishes were subjected to the same number of washes and media changes. Control cells (left-most data set, labeled no KN-62) were incubated in normal medium for 14 minutes and then for 2 minutes with 5 mM carbachol. Remaining dishes were incubated with 5 µM KN62 for 2 minutes at the times indicated by the arrows. This medium was then replaced with medium lacking KN-62, and the cells allowed to recover for various times, as indicated below each data set. At the end of this recovery period, cells were stimulated with medium + 5 mM carbachol for 2 minutes. All data are means ± ranges of duplicate dishes.

= exposure of cells to KN-62 for 2 min

40

30

20

AAAA A AA AAAA AAAA AAAA AAAA AA AAAA A AA AAAA AAAA AAAA AAAA AA

10

no KN-62

0

recovery time after removal of KN-62:

Time (2 minute collection intervals)

0 min

2 min

4 min

6 min


inhibition (data not shown). Moreover, the effect of KN-62 was similar whether the cells were differentiated by prior growth in the presence of NGF or not. Although the fraction of total NE release varied from one experiment to another, ranging from 10-30% of total cellular NE released in the first 2 minutes after treatment with carbachol, the extent of inhibition caused by maximal doses of KN-62 was reproducibly 40-50% of the stimulated release in the absence of KN-62. Because KN-62 is a lipophilic molecule, it was important to establish that KN-62 was exerting its inhibitory effect by specific action on CaM kinase II, rather than by some general toxic effect on the cells. Evidence to support the selective nature of the KN-62 inhibition is illustrated in Fig. 3, in which the reversibility of the KN-62 inhibition is demonstrated. After a 2 minute incubation with KN-62, cells were washed with normal medium without KN-62, and then stimulated at various

10 min

times thereafter. As before, KN-62 inhibited a maximum of 50% of the carbachol-stimulated release. After 10 minutes of washout, the carbachol-stimulated release of NE returned to 85% of control values. It is therefore clear that KN-62 was not exerting its inhibition by any detergent-like effect that irreversibly damaged the cells. Because of the unusual partial inhibition of KN-62 on stimulated release, we sought to analyze the process in more detail. Despite the total dependence of stimulated release on extracellular Ca2+ (Fig. 1), it seemed possible that only half of the stimulated secretion in PC12 cells involves the calmodulin pathway. Fig. 4 demonstrates that this is not the case. In contrast to the effects of KN-62, which specifically inhibits CaM kinase II, treatment of PC12 cells with trifluoperazine, a more general inhibitor of calmodulin (Levin and Weiss, 1976), completely blocked the carbachol-stimulated secretion of NE, suggesting that calmodulin acts on other effectors in addition to CaM kinase II.

30

120 20

10

0

AA A

+ carbachol

0 Ca

2+ +

carbachol

Additions

A

TFP + carbachol

Fig. 4. Inhibition of carbachol-stimulated NE release by TFP. Duplicate dishes of PC12 cells were loaded with [3H]NE and stimulated with 5 mM carbachol. Each bar represents the amount of [3H]NE release in a 2 minute collection interval. For the second set of dishes, cells were exposed to 0 Ca2+ (EGTA-containing) solution for the second 2 minute interval, and to 0 Ca2+ plus 5 mM carbachol for the third 2 minute interval. In the third set, cells were exposed to 100 µM trifluoperazine for the second 2 minute interval, and to 100 µM TFP plus 5 mM carbachol for the third 2 minute interval. Data shown are means of duplicate dishes ± ranges.

Autophosphorylation (% of control)


2622 E. S. Schweitzer, M. J. Sanderson and C. G. Wasterlain

100 80 60 40 20 0 10 -8

10 -7

10 -6

10 -5

[KN-62] (M)

Fig. 5. Inhibition of CaM kinase II by KN-62 in PC12 cell homogenates. Activity of CaM kinase II was assayed by quantitating the extent of autophosphorylation of the α (d) and β (m) subunits of CaM kinase II, as judged by the intensity of bands at molecular masses of 50 kDa and 60 kDa on autoradiographs of SDS gels. Data shown are means ± s.d.

Inhibition of secretion by KN-62 2623 In order to confirm that KN-62 was effective in inhibiting the PC12 cell CaM kinase II, we examined the activity of this enzyme in cell homogenates, as assayed by autophosphorylation of the CaM kinase II subunits. As shown in Fig. 5, KN62 inhibited CaM kinase II activity in vitro with a halfmaximal inhibition at approximately 0.5 µM, a value very similar to the 2 µM that produces half-maximal inhibition of NE release. At 10 µM, at which level the inhibition of secretion has reached its maximum of 50%, the activity of the PC12 CaM kinase has dropped to 12±4% of control values. These data are consistent with the results of Ishii et al. (1991), who reported that 10 µM KN-62 completely inhibited the depolarization-stimulated phosphorylation of tyrosine kinase in intact PC12 cells, which also demonstrated that sufficient KN-62 is able to permeate into intact cells to completely inhibit the activity of CaM kinase II. Taken together, these data strongly

suggest that only a portion of stimulated NE release is dependent on CaM kinase II activity. It was of interest to determine the site of involvement of the CaM kinase II in the 50% of stimulated secretion that was inhibited by KN-62. Since it has been reported that KN-62 blocks the influx of Ca2+ into some endocrine cells (Guodong et al., 1992; Prpic et al., 1994), we examined the effects of carbachol and KN-62 on [Ca2+]i in PC12 cells. Fig. 6 shows the results of such an experiment. FURA 2-loaded cells were illuminated at 340 and 380 nm, and the fluorescence ratios used to calculate [Ca2+]i levels both before and after stimulation with carbachol. Resting [Ca2+]i levels (Fig. 6B,F) were low and stable at 28.6±10.3 nM (n=50); no significant spontaneous fluctuations in this resting level were observed. Either in the presence (top panels) or absence (bottom panels) of KN-62, the cells responded to carbachol with a large increase in [Ca2+]i

Fig. 6. Intracellular Ca2+ levels before and after addition of carbachol ± KN-62. NGF-treated PC12 cells were loaded with FURA 2 and monitored for their levels of intracellular Ca2+. (A-D) Cells incubated with 5 µM KN-62 and then stimulated with carbachol; (E-H) control cells incubated with DMSO and then carbachol. (A and E) Phase contrast images of the experimental and control cells, respectively. (B and F) Cells 60 seconds after the addition of KN-62 (B) or DMSO (F). (C and G) Points near the peak in [Ca2+]i approximately 20 seconds after addition of 5 mM carbachol with (C) or without (G) added KN-62. (D and H) Cells partially recovering in the continued presence of carbachol, approximately 2 minutes after stimulation. The panels shown here correspond to time points of 120 seconds (B and F), 170 seconds (C and G) and 270 seconds (D and H) of Fig. 7. Bars (A and E), 100 µm. The pseudocolor scale bar at the bottom indicates the relationship between hue and [Ca2+], and applies to B-D and F-H.

2624 E. S. Schweitzer, M. J. Sanderson and C. G. Wasterlain 1000

A

[Ca 2+ ] (nM)

800 600 400

carbachol

KN-62 200 0 0

50

100

150

200

250

Time (seconds)

1000

B [Ca 2+ ] (nM)

800 600 400 carbachol

DMSO 200 0 0

50

100

150

200

Time (seconds)

(Fig. 6C,G). This increase in [Ca2+]i was seen both in the cell bodies and the neurites and terminals of the differentiated PC12 cells; both the rate and magnitude of the changes in [Ca2+]i were indistinguishable in these two locations (see, for example, the terminals visible in Fig. 6E-H). There was no significant difference in the peak Ca2+ levels achieved (680±172 nM (n=25) for control cells vs 654±85 nM (n=25) for cells treated with KN-62). The sampling of cells shown in Fig. 6 also demonstrates that this response was seen in virtually all cells, regardless of the presence of the inhibitor. It therefore appears that CaM kinase II is not necessary for the opening of Ca2+ channels in PC12 cells. Despite the continued presence of carbachol, cells demonstrated a recovery toward resting Ca2+ levels (Fig. 6D,H), indicating that either the ACh receptor or the Ca2+ channel, or both, were inactivating. Moreover, a comparison of the kinetics shown in Fig. 6 and Fig. 1 demonstrates that the rate of NE secretion drops off more rapidly than does [Ca2+]i, implying that the secretory mechanism inactivates in the presence of elevated [Ca2+]i. Removal of carbachol led to a complete return to resting Ca2+ levels within 5 minutes, indicating that the cells were still intact and metabolically active (data not shown). Fig. 7 shows the mean [Ca2+]i calculated at one second intervals for 14 different spatial points within a sampling of cells in the microscope fields shown in Fig. 6. There was no significant difference between the response of cells treated with KN-62 and untreated cells. Note that the addition of KN-62 itself did not cause any change in the level of Ca2+ observed in these cells. In all respects, including rise-time, maximum [Ca2+]i, and recovery, the cells treated with KN-62 were indistinguishable from control cells. KN-62 is a highly specific inhibitor of CaM kinase II, and has little or no effect on PKC or other kinases even at con-

250

Fig. 7. Intracellular Ca2+ levels after treatment with KN-62 and carbachol. Same experiment as illustrated in Fig. 6. In these plots, 14 different points within the cells shown in Fig. 6 were chosen, and [Ca2+]i calculated for each 1 second interval during the experiment. The means ± s.d. of the [Ca2+] are indicated for cells treated with KN-62 prior to carbachol stimulation (A) and for control cells treated only with DMSO (B). These pre-treatments were added at the 30 second time points, and carbachol was added at 150 seconds.

centrations 50 times higher than the maximal amounts used here (Tokumitsu et al., 1990). However, since both CaM kinase II and PKC have been implicated in peptide release from endocrine cells, it was important to investigate whether PKC was involved in any of the effects reported here. One possible mode of activation of PKC is the action of carbachol on muscarinic receptors on the surface of PC12 cells. In order to eliminate this possibility, PC12 cells were depolarized directly by exposing them to elevated K levels (55 mM) in the presence of extracellular Ca2+, a treatment which should directly open voltage-gated Ca2+ channels and increase [Ca2+]i. Fig. 8 shows that KN-62 had exactly the same effect of inhibiting depolarization-induced NE release as it had on carbachol-induced release (Fig. 1). As an additional means of bypassing receptor activation, the Ca2+ ionophore A23187 was used to raise [Ca2+]i. Although such treatment resulted in a stimulation of NE release, the kinetics and magnitude of this release differed from that seen with carbachol; nevertheless, treatment of PC12 cells with KN-62 prior to the addition of A231287 caused a partial inhibition of secretion (data not shown), indicating again that the site of action of KN-62 is distal to the increase in [Ca2+]i. An alternative approach to determining whether PKC acts as a mediator of the carbachol-stimulated release is to examine the effect of other inhibitors on release. For example, Irons et al. (1992) have reported that KN-62 inhibits a maximum of 50% of the endothelin-stimulated release of atrial natriuretic factor from atrial myocytes, an effect similar to that reported here. In addition, they reported that the remaining 50% can be inhibited by H7, a selective inhibitor of PKC (Hidaka et al., 1984). In contrast to the results of Irons et al. (1992), Fig. 9 shows that H7 had no effect on PC12 cells over and above that produced by KN-62. The 50% of NE release that is resistant

20

40

30

20

10

0 -4

-2

0

2

4

6



Inhibition of secretion by KN-62 2625

10

0


-4


Fig. 8. Effect of KN-62 on depolarization-stimulated NE release. PC12 cells were grown and loaded with [3H]NE as before, washed repeatedly in high Na medium, and then preincubated with 10 µM KN-62 (m) or DMSO (d) for 2 minutes in Na medium. These media were then removed (0 minute time point) and replaced with K medium either with or without KN-62, respectively, and incubated for another 2 minutes. At the end of this period, the media were removed and replaced with fresh K medium, again with or without KN-62. Data are means ± ranges of duplicate dishes.

-2

0

2

4

6


Fig. 10. Effects of H7 and K-252a on carbachol-stimulated NE release. Duplicate dishes of PC12 cells were grown, loaded with [3H]NE and washed as in Fig. 1. Pairs of dishes were preincubated for 2 minutes with normal medium (s), medium + 10 µM KN-62 (m), 25 µM H7 (d), or 200 nM K-252a (j). These media were then removed, and the cells were incubated for two additional 2 minute intervals in the media containing the same kinase inhibitors as in the preincubation media + 5 mM carbachol. Data shown are the means ± ranges of duplicate dishes.

30

20

10

0 0

2

4

6

8

10

The above experiments indicated that PKC is not an essential component of the machinery that is responsible for the regulated release of NE from PC12 cells. On the other hand, as illustrated in Figs 1 and 4, the carbachol-stimulated release is completely abolished by removal of extracellular Ca2+, and inhibited 50% by KN-62. We therefore conclude that 100% of the catecholamine release is dependent on Ca2+, but only about half of the release requires the action of CaM kinase II. In contrast to a variety of other cell lines, the site of action of CaM kinase II in PC12 cells appears to be ‘downstream’ of the increase in [Ca2+]i, placing it close to the fusion events that release catecholamines from dense-core vesicles.


Fig. 9. Absence of additional effect of the protein kinase C inhibitor H7 on carbachol-induced [3H]NE release from PC12 cells. PC12 cells were loaded with [3H]NE and stimulated with 5 mM carbachol for 2 minutes (between time points 4 and 6 minutes). Control cells (m) received no pre-treatment, while one set of dishes (d) was pretreated with 5 µM KN-62 for 2 minutes (between time points 2 and 4 minutes); another set of dishes (j) was pre-treated with 5 µM KN62 + 25 µM H7 for the same 2 minute period. All data are the means ± ranges of duplicate dishes.

to KN-62 therefore cannot be attributed to a PKC-dependent pathway. Moreover, Fig. 10 shows that H7 by itself has no effect on the carbachol-stimulated release of NE from PC12 cells in the absence of KN-62, demonstrating that the portion of release that depends on CaM kinase II also does not require PKC. In contrast, a potent but less-specific inhibitor of CaM kinase II, K-252a (Knüsel and Hefti, 1992; Hashimoto et al., 1991), has an effect similar to KN-62 in causing a partial inhibition of NE release (Fig. 10).

DISCUSSION The results presented here provide evidence for the essential involvement of CaM kinase II in secretion from the dense core vesicles in PC12 cells. Approximately 50% of the total carbachol-stimulated NE release can be inhibited by preincubating the cells with KN-62. The concentrations of KN-62 used here to inhibit catecholamine release inhibit PC12 cell CaM kinase II completely, but do not affect the activities of PKC, cAMP-dependent kinase II, or myosin light chain kinase (Tokumitsu et al., 1990). No difference was observed in the inhibitory effect of KN-62 whether or not the PC12 cells were previously differentiated by growth in NGF. Treatment of PC12 cells with KN-62 had no effect on the carbachol-stimulated increase in [Ca2+]i. This behavior is in marked contrast to that reported for a variety of other endocrine cell lines, including HIT (pancreatic) cells and STC-1 (enteroendocrine) cells, in which the inhibition of CaM kinase II by KN-62 prevents secretion by blocking the influx of Ca2+ through L-type channels in the plasma membrane

2626 E. S. Schweitzer, M. J. Sanderson and C. G. Wasterlain (Guodong et al., 1992; Prpic, 1994). Since the FURA 2 experiments illustrated here demonstrate that there is no inhibition of the carbachol-induced increase in [Ca2+]i even in the presence of concentrations of KN-62 that exert a maximal inhibitory effect on catecholamine release, the regulated secretion of catecholamines from PC12 cells must involve the action of CaM kinase II at a late step in the process of exocytosis, subsequent to the increase in cytoplasmic Ca2+. It is possible that such a late CaM kinase-dependent step is also involved in regulated secretion in other endocrine cells, but that this effect was masked by the inhibition of the plasma membrane Ca2+ channel seen in the experiments mentioned above; none of the previously published reports examined the effects of KN-62 on secretion under conditions in which the intracellular Ca2+ was directly increased, for example by ionomycin (Guodong et al., 1992). The present report therefore represents the first evidence for the essential involvement of CaM kinase II in a late step in the exocytosis of dense-core secretory vesicles. Significantly, only half the stimulated secretion of NE is dependent on active CaM kinase II. This observation suggests two possibilities: there could be two distinct exocytotic mechanisms, each responsible for the release of roughly 50% of the catecholamine vesicles. Although there are reports that PKC can trigger catecholamine release from PC12 cells (Pozzan et al., 1984), PKC is not required for exocytosis in the closely related adrenal chromaffin cells (Terbush and Holz, 1990). In contrast to the effect on the release of atrial natriuretic factor from atrial myocytes (Irons et al., 1992), the experiments presented here demonstrate that addition of the PKC inhibitor H7 has no effect on any of the release, including the 50% of stimulated release remaining after treatment with KN-62. It seems likely that, if PKC is involved, it acts indirectly by stimulating autophosphorylation of CaM kinase II (MacNicol et al., 1990), thereby bypassing the involvement of calmodulin but not of CaM kinase II. The alternative model is that all exocytosis occurs through a common, Ca2+-dependent mechanism, but that half of the vesicles require an additional step that is dependent on the activation of CaM kinase II in order to fuse with the plasma membrane. Support for a single basic mechanism comes from the observation that, while only 50% of stimulated NE release is inhibitable by KN-62, 100% of the stimulated release is dependent on the presence of extracellular Ca2+, and is blocked by the calmodulin inhibitor trifluoperazine. These data suggest that all vesicles undergo exocytosis by a mechanism that is dependent on both Ca2+ and calmodulin. This model is appealing because it is similar to the situation with the small, clear synaptic vesicles responsible for neurotransmitter release from the presynaptic terminals of neurons, in which release is clearly Ca2+-dependent; calmodulin antagonists have also been reported to inhibit 100% of the regulated release from cholinergic synaptic vesicles in nerve terminals (Schweitzer, 1987). In the case of small clear synaptic vesicles, the vesicles exist in two locations within the presynaptic terminal. One pool of vesicles is located immediately adjacent to the presynaptic release sites, while a reserve pool of vesicles is attached to cytoskeletal elements of the nerve terminal; the phosphorylation of synapsin I by CaM kinase II frees up this second pool of vesicles, permitting their movement to release sites and their subsequent exocytosis. Similarly, two distinct pools of cate-

cholaminergic vesicles have been reported in adrenal chromaffin cells (Neher and Zucker, 1993). Synapsins I and II are specific for neuronal cells, and are absent in endocrine cells (Benfenati et al., 1993; Chin et al., 1994). Moreover, in neurons that release peptides from densecore vesicle, such as in the posterior pituitary, the synapsin I is associated exclusively with the small clear synaptic-like vesicles and not the dense-core vesicles (Senda et al., 1991). Despite the absence of the synapsin proteins in association with dense-core vesicles, the data presented here indicate that CaM kinase II plays an integral role in the secretion of dense-core catecholaminergic vesicles, in a way that is suggestive of the role that synapsin plays in relating synaptic vesicles with the cytoskeleton. It therefore seems reasonable to suggest that a component functionally similar to synapsin is likely to be involved. Other possible sites of CaM kinase II action are synaptophysin (Rubenstein et al., 1993), synaptotagmin (Matthew et al., 1981), the recently described 58 kDa vesicle protein (Takahashi et al., 1991), or the actin-associated proteins αactinin, gelsolin, and scinderin (Rodríguez Del Castillo et al., 1990). The presence of synaptophysin in dense-core vesicles is controversial, but the biochemical evidence that it is a lowabundance component of these vesicles also suggests that it is a component of 100% of the vesicles (Lowe at al., 1988; Obendorf et al., 1988; Blumberg and Schweitzer, 1992); it is therefore difficult to reconcile a model in which CaM kinase II acts to phosphorylate synaptophysin and trigger exocytosis with the observed partial inhibition of catecholamine release by KN-62. Synaptotagmin is a component of both small clear vesicles and large dense-core vesicles (Trifaró et al., 1989), and it not only binds calmodulin but has been reported to be a substrate for CaM kinase II (Bennett et al., 1992; Popoli, 1993). Because synaptotagmin binds Ca2+, it has been suggested to be the actual trigger that causes exocytosis (Brose et al., 1992; Elferink et al., 1993). Experiments on mutants of Drosophila (DiAntonio et al., 1993), C. elegans (Nonet et al., 1993), and PC12 cells (Shoji-Kasai et al., 1992) suggest that synaptotagmin is not essential for release from either synaptic vesicles or dense-core vesicles, but these experiments are complicated by the recent discovery of an additional isoform of synaptotagmin (Hilbush and Morgan, 1994). The 58 kDa protein described by Takahashi et al. (1991) is a component of synaptic vesicles in the brain and is a substrate for CaM kinase II; whether it is found in association with dense-core vesicles or not, and what role, if any, it plays in transmitter release, is not known. Gelsolin and scinderin interact with actin, and scinderin in particular appears to be associated with cytoskeletal rearrangements associated with secretion from dense-core vesicle in adrenal cells (Vitale et al., 1991). Because of its interaction with actin and its presence at secretory sites, scinderin might well be involved in the mobilization of dense-core vesicles in a manner analogous to the role of synapsin I at synapses. Although scinderin has been suggested to be a substrate for PKC, there are no reports of it or the other actin-binding proteins associated with dense core vesicles being phosphorylated by CaM kinase II. A model in which CaM kinase II acts to free a reserve pool of dense-core vesicles by phosphorylating a protein that links vesicles with the cytoskeleton is appealing because it would

Inhibition of secretion by KN-62 2627 explain why KN-62 only partially inhibits the release of catecholamine from PC12 cells. In the present study, the 50% of release that is resistant to KN-62 would represent exocytosis of dense core vesicles that are already docked in release sites. It is known from electrochemical measurements that at least some of the stimulated catecholamine release occurs from such sites with kinetics that are too rapid to require movement of the vesicles from sites within the cell (Chow et al., 1992). Maximal release, however, would require the mobilization of a reserve pool of dense-core vesicles from within the cytoplasm (Trifaró et al., 1992); it seems most likely that this mobilization process is the step that is dependent on the action of CaM kinase II, and is blocked by KN-62. The authors thank Seikagaku Corp. for their gift of KN-62, Diane Voigt for technical assistance, Scott Boitano for assistance with the FURA 2 data analysis, and Chung-Jiuan Jeng, Alexander van der Bliek, Lawrence Kruger, and Cameron Gunderson for helpful comments on the manuscript. This work was supported by NIH grant R01 NS23084 and a grant from the W. M. Keck Foundation (E.S.S), and NIH 49288 (M.T.S.)

REFERENCES Aunis, D. and Perrin, D. (1984). Chromaffin granule membrane-F-actin interactions and spectrin-like protein of subcellular organelles: a possible relationship. J. Neurochem. 42, 1558-1569. Bader, M. F. and Aunis, D. (1983). The 97 kD α-actinin like protein in chromaffin granule membranes from adrenal medulla: evidence for localization of the cytoplasmic surface and for binding to actin filaments. Neuroscience 8, 165-181. Bajjalieh, S. M., Peterson, K., Shinghal, R. and Scheller, R. H. (1992). SV2, a brain synaptic vesicle protein homologous to bacterial transporters. Science 257, 1271-1273. Benfenati, F., Valtorta, F., Rossi, M. C., Onofri, F., Sihra T., Greengard, P. (1993). Interactions of synapsin I with phospholipids: possible role in synaptic vesicle clustering and in the maintenance of bilayer structures. J. Cell Biol. 123, 1845-1855. Bennett, M. K., Calakos, N. and Scheller, R. H. (1992). Syntaxin: a synaptic protein implicated in docking of synaptic vesicles at presynaptic active zones. Science 257, 255-259. Bennett, M. K. and Scheller, R. H. (1994). A molecular description of synaptic vesicle membrane trafficking. Annu. Rev. Biochem. 63, 63-100. Blumberg, D. and Schweitzer, E. S. (1992). Vesamicol binding to subcellular membranes that are distinct from catecholaminergic vesicles in PC12 cells. J. Neurochem. 58, 801-810. Brose, N., Petrenko, A. G., Südhof, T. C. and Jahn, R. (1992). Synaptotagmin: a calcium sensor on the synaptic vesicle surface. Science 256, 1021-1025. Buckley, K. and Kelly, R. B. (1985). Identification of a transmembrane glycoprotein specific for secretory vesicles of neural and endocrine cells. J. Cell Biol. 100, 1284-1294. Charles, A. C., Merrill, J. E., Dirksen, E. R. and Sanderson, M. J. (1991). Intercellular signaling in glial cells: calcium waves and oscillations in response to mechanical stimulation and glutamate. Neuron 6, 983-992. Chin, L. S., Li, L. and Greengard, P. (1994). Neuron-specific expression of the synapsin II gene is directed by a specific core promoter and upstream regulatory elements. J. Biol. Chem. 269, 18507-18513. Chow, R. H., von Ruden, L. and Neher, E. (1992). Delay in vesicle fusion revealed by electrochemical monitoring of single secretory events in adrenal chromaffin cells. Nature 356, 60-63. DiAntonio, A., Parfitt, K. D., Schwarz, T. L. (1993). Synaptic transmission persists in synaptotagmin mutants of Drosophila. Cell 73, 1281-1290. Elferink, L. A., Peterson, M. R. and Scheller, R. H. (1993). A role for synaptotagmin (p65) in regulated exocytosis. Cell 72, 153-159. Greene, L. A. and Tischler, A. S. (1976). Establishment of a noradrenergic clonal cell line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc. Nat. Acad. Sci. USA 73, 2424-2428. Greengard, P., Valtorta, F., Czernik, A. J., Benfenati, F. (1993). Synaptic

vesicle phosphoproteins and regulation of synaptic function. Science 259, 780-785. Guodong, L., Hidaka, H. and Wollheim, C. B. (1992). Inhibition of voltagegated Ca2+ channels and insulin secretion in HIT cells by the Ca2+/calmodulin-dependent protein kinase II inhibitor KN-62: comparison with antagonists of calmodulin and L-type Ca2+ channels. Mol. Pharmacol. 42, 489-498. Hashimoto, Y., Nakayama, T., Teramoto, T., Kato, H., Watanabe, T., Kinoshita, M., Tsukamoto, K., Tokunaga, K., Kurokawa, K., Nakanishi, S., Matsuda, Y. and Nonomura, Y. (1991). Potent and preferential inhibition of Ca2+/calmodulin-dependent protein kinase II by K252a and its derivative, KT5926. Biochem. Biophys. Res. Commun. 181, 423-429. Hay, J. D. and Martin, T. F. (1993). Resolution of regulated secretion into sequential MgATP-dependent and calcium-dependent stages mediated by distinct cytosolic proteins. J. Cell Biol. 119, 139-151. Hidaka, H., Inagaki, M., Kawamoto, S. and Sasaki, Y. (1984). Isoquinolinesulfonamides, novel and potent inhibitors of cyclic nucleotide dependent protein kinase and protein kinase C. Biochemistry 23, 5036-5041. Hilbush, B. S. and Morgan, J. I. (1994). A third synaptotagmin gene, Syt3, in the mouse. Proc. Nat. Acad. Sci. USA 91, 8195-8199. Irons, C. E., Sei, C. A., Hidaka, H. and Glembotski, C. C. (1992). Protein kinase C and calmodulin kinase are required for endothelin-stimulated atrial natriuretic factor secretion from primary atrial myocytes. J. Biol. Chem. 267, 5211-5216. Ishii, A., Kiuchi, K., Kobayashi, R., Sumi, M., Hidaka, H. and Nagatsu, T. (1991). A selective Ca2+/calmodulin-dependent protein kinase II inhibitor, KN-62, inhibits the enhanced phosphorylation and the activation of tyrosine hydroxylase by 56 mM K+ in rat pheochromocytoma PC12h cells. Biochem. Biophys. Res. Commun. 176, 1051-1056. Ishikawa, N., Hashiba, Y. and Hidaka, H. (1990). Effect of a new Ca2+calmodulin-dependent protein kinase II inhibitor on GABA release in cerebrospinal fluid of the rat. J. Pharmacol. Exp. Therapeut. 254, 598-602. Jahn, R. and Südhoff, R. C. (1994). Synaptic vesicles and exocytosis. Annu. Rev. Neurosci. 17, 219-246. Kelly, R. B. (1993). Storage and release of neurotransmitters. Neuron 10, 4353. Knüsel, B. and Hefti, F. (1992). K-252 compounds: modulators of neurotrophin signal transduction. J. Neurochem. 59, 1987-1996. Laemmli, U. K. (1970). Cleavage of structural proteins during the head assembly of bacteriophage T4. Nature 227, 680-685. Levin, R. M. and Weiss, B. (1976). Mechanism by which psychotropic drugs inhibit adenosine cyclic 3′, 5′-monophosphate phosphodiesterase of brain. Mol. Pharmacol. 12, 581-589. Llinás, R., Gruner J. A., Sugimori, M., McGuinness, T. L., Greengard, P. (1991). Regulation by synapsin I and Ca2+-calmodulin-dependent protein kinase II of the transmitter release in squid giant synapse. J. Physiol. 436, 257-282. Lowe, A. W., Madeddu, L. and Kelly, R. B. (1988). Endocrine secretory granules and neuronal synaptic vesicles have three integral membrane proteins in common. J. Cell Biol. 106, 51-59. MacNicol, M., Jefferson, A. B. and Schulman, H. (1990). Ca2+/calmodulin kinase is activated by the phosphatidylinositol signaling pathway and becomes Ca2+-independent in PC12 cells. J. Biol. Chem. 265, 18055-18058. Matthew, W. D., Tsavaler, L. and Reichardt, L. F. (1981). Identification of a synaptic vesicle-specific protein with a wide distribution in neuronal and neurosecretory tissue. J. Cell Biol. 91, 257-269. Mobley, W. C., Schenker, A. and Shooter, E. M. (1976). Characterization and isolation of proteolytically modified nerve growth factor. Biochemistry 15, 5543-5552. Neher, E. and Zucker, R. S. (1993). Multiple calcium-dependent processes related to secretion in bovine chromaffin cells. Neuron 10, 21-30. Nonet, M. L., Grundahl, K., Meyer, B. J. and Rand, J. B. (1993). Synaptic function is impaired but not eliminated in C. elegans mutants lacking synaptotagmin. Cell 73, 1291-1305. Nose, P. S., Griffith, L. C. and Schulman, H. (1985). Ca2+-dependent phosphorylation of tyrosine hydroxylase in PC12 cells. J. Cell Biol. 101, 1182-1190. Obendorf, D., Schwartzbrunner, U., Fischer-Colbrie, R., Laslop, A. and Winkler, H. (1988). In adrenal medulla synaptophysin (protein p38) is present in chromaffin granules and a special vesicle population. J. Neurochem. 51, 1573-1580. Popoli, M. (1993). Synaptotagmin is endogenously phosphorylated by Ca2+/calmodulin protein kinase II in synaptic vesicles. FEBS Lett. 317, 8588.

2628 E. S. Schweitzer, M. J. Sanderson and C. G. Wasterlain Pozzan, T., Gatti, G., Dozio, N., Vicentini, L. M. and Meldolesi, J. (1984). Ca2+-dependent and -independent release of neurotransmitters from PC12 cells: a role for protein kinase C activation?. J. Cell Biol. 99, 628-638. Prpic, V., Basavappa, S., Liddle, R. A. and Mangel, A. W. (1994). Regulation of cholecystokinin secretion by calcium-dependent calmodulin kinase II: differential effects of phenylalanine and cAMP. Biochem. Biophys. Res. Commun. 201, 1483-1489. Rodríguez, Del Castillo, A., Lemair, S., Tcharkarov, L., Jeyapragasan, M., Doucet, J.-P., Vitale, M. L. and Trifaró, J.-M. (1990). Chromaffin cell scinderin: a novel calcium-dependent actin-filament severing protein. EMBO J. 9, 43-52. Rubenstein, J. L., Greengard. P. and Czernik, A. J. (1993). Calciumdependent serine phosphorylation of synaptophysin. Synapse 13, 161-172. Russell, W. C., Newman, C. and Williamson, D. H. (1975). A simple cytochemical technique for demonstration of DNA in cells infected with mycoplasmas and viruses. Nature 253, 461-462. Schweitzer, E. S. and Kelly, R. B. (1985). Selective packaging of human growth hormone into synaptic vesicles in a rat neuronal (PC12) cell line. J. Cell Biol. 101, 667-676. Schweitzer, E. S. (1987). Coordinated release of ATP and ACh from cholinergic synaptosomes and its inhibition by calmodulin antagonists. J. Neurosci. 7, 2949-2956. Schweitzer, E. S. and Paddock, S. (1990). Localization of human growth hormone to a sub-set of cytoplasmic vesicles in transfected PC12 cells. J. Cell Sci. 96, 375-381. Schweitzer, E. S. (1993). Regulated and constitutive secretion of distinct molecular forms of acetylcholinesterase from PC12 cells. J. Cell Sci. 106, 731-740. Senda, T., Ochiai, H., Nakai, Y., Fujita, H. (1991). Immunocytochemical localization of synaptophysin (protein p38) and synapsin I in nerve terminals of rat neurohypophysis. Arch. Histol. Cytol. 54, 233-240. Shoji-Kasai, Y., Yoshida, A., Sato, K., Hoshino, T., Ogura, A., Kondo, S., Fujimoto, Y., Kuwahara, R., Kato, R. and Takahashi, M. (1992). Neurotransmitter release from synaptotagmin-deficient clonal variants of PC12 cells. Science 256, 1820-1823.

Takahashi, M., Arimatsu, Y., Shinobu, F., Fujimoto, Y., Kondo, S., Hama, T. and Miyamoto, E. (1991). Protein kinase C and Ca2+/calmodulindependent protein kinase II phosphorylate a novel 58-kDa protein in synaptic vesicles. Brain Res. 551, 279-292. Terbush, D. R. and Holz, R. W. (1990). Activation of protein kinase C is not required for exocytosis from bovine adrenal chromaffin cells. J. Biol. Chem. 265, 21179-21184. Tokumitsu, H., Chijiwa, T., Hagiwara, M., Mizutani, A., Terasawa, M. and Hidaka, H. (1990). KN-62, 1-[N,O-Bis(5-isoquinolinesulfonyl)-N-methylL-tyrosyl]-4-phenylpiperazine, a specific inhibitor of Ca2+/calmodulindependent protein kinase II. J. Biol. Chem. 265, 4315-4320. Torri Tarelli, F., Bossi, M., Fesce, R., Greengard, P. and Valtorta, F. (1992). Synapsin I partially dissociates from synaptic vesicles during exocytosis induced by electrical stimulation. Neuron 9, 1143-1153. Trifaró, J.-M., Fournier, S. and Novas, M. L. (1989). The p65 protein is a calmodulin-binding protein present in several types of secretory vesicles. Neuroscience 29, 1-8. Trifaró, J.-M., Vitale, M. L. and Rodríguez Del Castillo, A. (1992). Cytoskeleton and molecular mechanisms in neurotransmitter release by neurosecretory cells. Eur. J. Pharmacol. 225, 83-104. Vitale, M. L., Del Castillo, R., Tchakarov, L. and Trifaró, J.-M. (1991). Cortical filamentous actin disassembly and scinderin redistribution during chromaffin cell stimulation precede exocytosis, a phenomenon not exhibited by gelsolin. J. Cell Biol. 113, 1057-1067. Walent, J. H., Porter, B. W. and Martin, T. F. (1992). A novel 145 kD brain cytosolic protein reconstitutes Ca2+-regulated secretion in permeable neuroendocrine cells. Cell 70, 765-775. Wenham, R. M., Landt, M., Walters, S. M., Hidaka, H. and Easom, R. A. (1992). Inhibition of insulin secretion by KN-62, a specific inhibitor of the multifunctional Ca2+/calmodulin-dependent protein kinase II. Biochem. Biophys. Res. Commun. 189, 128-133.

(Received 30 November 1994 - Accepted 23 March 1995)