Complex regulation of store-operated Ca2+ entry pathway by PKC-ε in vascular SMCs

Tarik Smani, Tina Patel, Victoria M. Bolotina

Abstract

The role of PKC in the regulation of store-operated Ca2+ entry (SOCE) is rather controversial. Here, we used Ca2+-imaging, biochemical, pharmacological, and molecular techniques to test if Ca2+-independent PLA2β (iPLA2β), one of the transducers of the signal from depleted stores to plasma membrane channels, may be a target for the complex regulation of SOCE by PKC and diacylglycerol (DAG) in rabbit aortic smooth muscle cells (SMCs). We found that the inhibition of PKC with chelerythrine resulted in significant inhibition of thapsigargin (TG)-induced SOCE in proliferating SMCs. Activation of PKC by the diacylglycerol analog 1-oleoyl-2-acetyl-sn-glycerol (OAG) caused a significant depletion of intracellular Ca2+ stores and triggered Ca2+ influx that was similar to TG-induced SOCE. OAG and TG both produced a PKC-dependent activation of iPLA2β and Ca2+ entry that were absent in SMCs in which iPLA2β was inhibited by a specific chiral enantiomer of bromoenol lactone (S-BEL). Moreover, we found that PKC regulates TG- and OAG-induced Ca2+ entry only in proliferating SMCs, which correlates with the expression of the specific PKC-ε isoform. Molecular downregulation of PKC-ε impaired TG- and OAG-induced Ca2+ influx in proliferating SMCs but had no effect in confluent SMCs. Our results demonstrate that DAG (or OAG) can affect SOCE via multiple mechanisms, which may involve the depletion of Ca2+ stores as well as direct PKC-ε-dependent activation of iPLA2β, resulting in a complex regulation of SOCE in proliferating and confluent SMCs.

  • protein kinase C-ε
  • Ca2+-independent phospholipase A2
  • diacylglycerol
  • smooth muscle cells

physiologically agonist stimulation of a G protein-coupled receptor leads to the activation of PLC (5, 6), which cleaves phosphatidylinositol 4,5-bisphosphate into 1,2-diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP3). IP3 causes Ca2+ release from sarco(endo)plasmic reticulum stores, which leads to their depletion and the activation of store-operated Ca2+ (SOC) entry (SOCE) in a variety of cell types, including vascular smooth muscle cells (SMCs) (13, 26, 28, 29). At the same time, DAG can activate PKC, which may regulate and fine tune different signaling cascades (24, 39), including cell proliferation (for a review, see Ref. 23). Among many other targets, PKC has been shown to activate Ca2+-independent PLA2β (iPLA2β) in different cell types (1, 34). Jenkins and others (21) demonstrated that exogenous stimulation of PKC with vasopressin, PMA, and DAG induced iPLA2β activation and the release of arachidonic acid and induced the translocation of PKC-ε and PKC-δ from the cytosol to plasma membrane in the A10 SMC cell line. In ventricular myocytes, a novel PKC isoform (found in the membrane fraction) has been shown to modulate iPLA2 activity (34).

Recently, we (15, 31, 32) found and others (1619) confirmed that iPLA2β is a major determinant for the activation of SOCE. We (9) demonstrated that upon depletion of intracellular Ca2+ stores, a diffusible Ca2+ influx factor (CIF) is produced that displaces inhibitory calmodulin (CaM) from iPLA2β, leading to its activation and the production of lysophospholipids, which, in turn, activate plasma membrane Ca2+ influx channels and SOCE in SMCs. Upon refilling of the stores and the termination of CIF production, CaM binds back to iPLA2β, inhibiting its activity and terminating SOCE. Thus, iPLA2β emerged as a crucial transducer of the signal from depleted stores to plasma membrane channels and a potential target for the regulation and fine tuning of SOCE by other signaling cascades, including those that involve PKC. In previous studies, Albert and Large (3, 4) demonstrated 1-oleoyl-2-acetyl-sn-glycerol (OAG) and phorbol 12,13-dibutyrate can cause the activation of Ca2+-conducting channels that can be also activated by norepinephrine in SMCs dispersed from the portal vein (3, 4). Also, Su and others (35) presented evidence that OAG can activate cation current in Jurkat cells that may be similar to that activated by store depletion. In addition, in multiple studies, DAG has been shown to activate some transient receptor potential C channels, which is thought to be a direct effect on the channel itself (for a review, see Ref. 18). Thus, DAG (and OAG) may have multiple effects, and the mechanism of DAG and PKC involvement in the activation of the specific SOCE pathway remains unclear.

This study was designed to test which steps in the SOCE pathway may be affected by DAG and whether iPLA2β may be involved in the regulation of SOCE by PKC in SMCs. Our results demonstrate that DAG (or OAG) can affect SOCE via multiple mechanisms, which may involve the depletion of Ca2+ stores as well as direct PKC-ε-dependent activation of iPLA2β, resulting in a complex regulation of SOCE in proliferating and confluent SMCs.

MATERIALS AND METHODS

Cells.

Primary cultured SMCs from rabbit aortas were prepared following the same protocol as previously described (31, 32). All animal experiments were carried out in accordance with protocols and guidelines established by the United States National Institutes of Health.

Intracellular Ca2+ measurements.

SMCs were loaded with fura-2 AM (2 μM fura-2 AM for 30 min at 37°C, followed by a 15-min wash), and changes in intracellular Ca2+ [fluorescence at 340/380 nm (F340/F380)] were monitored as previously described (15, 3133). A dual-excitation fluorescence-imaging system (InCyt Im2, Intracellular Imaging, Cincinnati, OH) was used for experiments using individual cells. Experiments were done in 0 mM Ca2+ solution [containing (in mM) 120 NaCl, 4.7 KCl, 4 MgCl2, 0.2 EGTA, and 10 HEPES], and the Ca2+ influx was determined from changes in fura-2 fluorescence after the readdition of Ca2+ (2 mM). Changes in intracellular Ca2+ were expressed as a change in ratio (ΔRatio), which was calculated as the difference between the peak F340/F380 ratio after extracellular Ca2+ was added and its level right before Ca2+ addition. The basal Ca2+ influx in the primary culture of SMCs was ΔRatio = 0.29 ± 0.10 (n = 130), which was subtracted from summary data shown by the bar graphs but not from original traces. Data were summarized from the large number of individual cells (20–40 cells in a single run, with 3–9 identical experiments done in at least 3 cell preparations).

SMC transfection.

Transfection was done using Lipofectamine plus (Invitrogen) and following standard protocols. The oligonucleotides for antisense and scrambled PKC-ε were as specified in Ref. 36. Briefly, SMCs on days 6 and 7 (60–70% confluence) and 95% confluent SMCs were transfected with antisense and scrambled oligonucleotides [the same as successfully used in a previous study (36)] directed against PKC-ε: antisense oligonucleotide 5′-CATGAGGGCCGATGTGACCT-3′ and scrambled oligonucleotide 5′-TACGCATAACGCGCTGGTGG-3′. Experiments were conducted 60–72 h after transfection. The oligonucleotides were labeled with fluorescein, and its presence in the cells was verified by imaging (excitation at 480 nm and emission at 515 nm) before the experiments.

Western blot analysis.

Protein samples were incubated with Laemmli sample buffer at 95°C for 2 min, and 30 μg of total protein were loaded on a 7.5% SDS-polyacrylamide gel for electrophoresis. Proteins were transferred to nitrocellulose membranes in transfer buffer (25 mM Tris, 192 mM glycine, and 20% methanol) at 40 V overnight. The membrane was then blocked in PBS containing 0.05% Tween 20 (PBST) with 3% milk for 1 h and then incubated with primary (anti-PKC-ε) antibody (Santa Cruz Biotechnology, Santa Cruz, CA) for 2 h at room temperature. The primary antibody was removed, and blots were washed three times for 10 min with milk-PBST. Blots were then incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG antibody diluted 1:2,000 in milk-PBST, washed three times in PBST, and treated with enhanced chemiluminescence reagents (Super ECL, Pierce) for 1 min. Blots were then exposed to photographic films, and the optical density was determined using Un-scan-it analysis software (Silk Scientific).

iPLA2 assay.

iPLA2 Activity was performed using a modified commercial assay kit originally designed for the cytosolic PLA2 (cPLA2; Cayman Chemical). To detect the activity of iPLA2 instead of cPLA2, the assay buffers were modified to contain no Ca2+ (Ca2+ is needed for cPLA2 but not for iPLA2 activity). Phospholipase activity was assayed by incubating the samples with the substrate (arachidonoyl thio-PC) for 1 h at room temperature in a modified Ca2+-free assay buffer of the following composition: 300 mM NaCl, 10 mM HEPES, 8 mM Triton X-100, 4 mM EGTA, 60% glycerol, and 2 mg/ml BSA (pH 7.4). The generated free thiols were visualized by the addition of DTNB for 5 min, and the absorbance was determined at 405 nm using a standard microplate reader. The background iPLA2-independent component of basal lipase activity was determined in control samples when all specific iPLA2 activity had been inhibited with bromenol lactone (BEL; 10 μM for 5 min) and was subtracted from all the readings. The specific activity of iPLA2 was expressed in absorbance units per milligram of protein.

Immunostaining.

Cells grown in the Lab-Tek Chamber glass-bottomed slide system (Nalge Nunc) were fixed for 5 min in methanol at −20°C and air dried at room temperature. SMCs were then treated for 1 h with 1% heat-inactivated goat serum containing 0.1% Triton X-100 in PBS (for blocking) and incubated in a humidified chamber overnight a 4°C with primary antibody solution (rabbit anti-PKC-ε, Santa Cruz Biotechnology, 1:25 dilution). After washes with PBS, slides were incubated for 1 h in the dark with FITC-conjugated goat anti-rabbit antibody (1:200 dilution, Jackson Laboratories, West Grove, PA). Slides were mounted with Vectashield medium with Texas red (Vector Labs), and cell samples were analyzed using a high-resolution imaging system with the Openlab software package from Improvision (Coventry, UK). Raw images of SMCs (showing the localization of FITC conjugated with an anti-PKC-ε antibody within the whole cell) were visualized at 488-nm excitation and 510-nm emission. The nucleus (labeled by Texas red) was visualized at 544-nm excitation and 620-nm emission. As negative controls, cells were incubated with secondary antibodies only.

Drugs and treatments.

Drugs were purchased from Sigma. Fura-2 and fura-2 AM were from Invitrogen. Chiral enantiomers of BEL (S-BEL and R-BEL) were separated by HPLC utilizing a Chirex column of 3,5-dinitrobenzoyl-R-phenylglycine attached to a silica matrix (Phenomenex) as previously described (21). It is important to notice that BEL and its enantiomers are suicidal substrates for iPLA2; inhibition is irreversible, requires basal activity of this enzyme, and strongly depends on the temperature, duration of treatment, and concentration used (8, 15, 21, 3133). In experiments with intact cells, the optimal conditions for BEL treatment (to ensure complete inhibition of iPLA2) are as follows: intact cells need to be pretreated (in bath solution not containing BSA or serum) with 10–25 μM BEL for 30 min at 37°C, and BEL can then be washed away before the beginning of the experiments. In cases where iPLA2 is already active, significantly shorter (1–5 min) treatments may fully inhibit the enzyme.

Statistical analysis.

Group data are presented as means ± SE. A single or paired Student's t-test was used to determine the statistical significance of the obtained data. The significance between multiple groups was evaluated using ANOVA followed by Turkey's test. Data were considered significant at P < 0.05.

RESULTS

To test the potential effects of DAG and the role of PKC in SOCE, we used a variety of pharmacological and molecular tools and examined the effects of PKC activation and inhibition on Ca2+ influx in proliferating (or confluent) rabbit aortic SMCs. Ca2+ influx was triggered either by depletion of the stores with thapsigargin (TG) or via different shortcuts in the iPLA2β and lysophospholipid-mediated SOCE pathway (32), which is shown in Fig. 1. By consecutive activation/inhibition of different components of this pathway (with the tools shown in Fig. 1), we started to look at which specific components and step(s) in the signal transduction pathway from the stores to ion channels may be regulated by PKC (and activated by OAG).

Fig. 1.

Scheme of the store-operated Ca2+ (SOC) entry (SOCE) pathway mechanism. The scheme summarizes the experimental procedures used to highlight the different facets and regulation of the SOCE pathway in smooth muscle cells (SMCs). 2-APB, 2-aminoethoxydiphenyl borate; DES, diethylstilbestrol; BEL, bromenol lactone; PM, plasma membrane; LPC, lysophosphatidylcholine; iPLA2, Ca2+-independent PLA2; CaM, calmodulin; CIF, Ca2+ influx factor; OAG, 1-oleoyl-2-acetyl-sn-glycerol; IP3R, inositol 1,4,5-trisphosphate receptor; ER, endoplasmic reticulum; SR, sarcoplasmic reticulum; SERCA, sarco(endo)plasmic reticulum Ca2+-ATPase; TG, thapsigargin.

Similarities and PKC dependence of TG- and OAG-induced Ca2+ entry in proliferating SMCs.

First, we found that TG (2 μM)-induced Ca2+ influx in SMCs can be significantly impaired by chelerythrine (10 μM), a widely used generic inhibitor of PKC (17, 19), as shown in Fig. 2A. In the same SMC, the application of OAG (50 μM), a membrane-permeable analog of DAG (a known activator of PKC), induced Ca2+ entry (Fig. 2B) that looked very similar to TG-induced SOCE: the amplitudes of the TG- and OAG-induced Ca2+ entry were not significantly different, and both were suppressed by chelerythrine, although the inhibitory effect of chelerythrine was significantly stronger for OAG-induced Ca2+ entry (81 ± 10% inhibition of the OAG effect vs. 63 ± 6% inhibition of the TG effect). Importantly, the application of OAG after TG did not result in any additional Ca2+ entry (Fig. 2A), suggesting that they both may be activating the same Ca2+ entry pathway.

Fig. 2.

PKC inhibition impairs TG- and OAG-induced Ca2+ influx. A, left: representative traces showing the changes in intracellular Ca2+ concentration [presented as the ratio of fluorescence at 340 to 380 nm (F340/F380)] in fura-2-loaded aortic SMCs. TG (2 μM) was applied in the absence of extracellular Ca2+. To test for Ca2+ influx, Ca2+ (2 mM) was added at the time shown (arrow). TG-induced responses are shown in a representative control cell and the cell to which PKC inhibitor [chelerythrine (Cheler); 10 μM] was applied 2 min before Ca2+, as shown (*). The bar graph on the right shows the average amplitude of TG-induced Ca2+ influx (Δratios ± SE) from control and Cheler-treated SMCs and from SMCs where OAG was applied after the addition of extracellular Ca2+ to TG-treated SMCs (TG + OAG). Each bar summarizes the result from 70–105 cells (numbers shown above each bar). B: similar to A, but OAG (50 μM) was applied to SMCs instead of TG. Bar graphs represent means ± SE of Ca2+ influx from 83–93 cells.

Role of SOC channels and iPLA2β in OAG-induced Ca2+ entry.

To determine if SOC channels can be directly involved in Ca2+ influx induced not only by TG but also by OAG, we tested the effects of 2-aminoethoxydiphenyl borate (2-APB; 100 μM) and diethylstilbestrol (DES; 10 μM), two widely used inhibitors of SOC channels (10, 11, 27, 40). Figure 3A shows that OAG-induced Ca2+ influx was indeed impaired when 2-APB or DES were added to the cells. Similar to OAG, we found that 2-APB-sensitive Ca2+ influx in SMCs could be also triggered by PMA (Fig. 3B), which is another potent activator of PKC. PMA-induced Ca2+ influx was also inhibited by chelerythrine, confirming that PKC activation may indeed lead to the activation of SOC channels and SOCE in SMCs.

Fig. 3.

OAG and PMA-induced Ca2+ influx are prevented by SOC channel inhibitors. A, left: representative traces showing the changes in intracellular Ca2+ concentration in fura-2-loaded SMCs. OAG (50 μM) was applied in the absence of extracellular Ca2+, and 2 mM Ca2+ was then added as indicated (arrow). Traces are for SMCs treated with OAG (control) and when 2-APB (100 μM) or DES (10 μM) were added 2 min before Ca2+, as shown (*). Bar graph on the right shows the average amplitudes of OAG-induced Ca2+ influx (Δratios ± SE) from the experiments as show on the left. Each bar summarizes the result from 32–121 cells (numbers shown above each bar). B, left: representative traces showing PMA-induced changes in intracellular Ca2+ concentration in fura-2-loaded SMCs. PMA (1 μM) was applied in the absence of extracellular Ca2+, and 2 mM Ca2+ was then added as shown (arrow). Traces are for SMCs treated with PMA (control) and when 2-APB (100 μM) was added 1 min before Ca2+, as shown (*). Bar graph on the right shows average amplitudes of PMA-induced Ca2+ influx (Δratios ± SE) from control, 2-APB-treated, and Cheler (10 μM)-treated SMCs. n = 23–40 cells, as indicated above each bar.

To determine which specific step(s) in the signal transduction pathway from the stores to ion channels (Fig. 1) may be regulated by PKC (and activated by OAG), we started to test the effects of PKC inhibition on SOCE induced by different shortcuts in the iPLA2-dependent SOCE pathway (32), starting from plasma membrane channels and moving upstream. To test the possibility of a direct PKC-dependent modification of SOC channels, we tested if PKC inhibition could affect channel (and SOCE) activation by lysophospholipids, a signaling step in the SOCE pathway downstream of iPLA2β that leads to SOC activation (9, 30, 32, 33). Figure 4 shows that lysophosphatidylcholine-activated SOCE in SMCs was inhibited by 2-APB (consistent with SOC channel activation) but was totally insensitive to chelerythrine. This result showed that PKC does not affect SOC channels directly and does not play any significant role in membrane-delimited lysophospholipid-mediated signal transduction downstream from iPLA2β.

Fig. 4.

LPC evoked a Cheler-insensitive Ca2+ influx. Left: representative traces show intracellular Ca2+ changes in fura-2-loaded SMCs treated with LPC (300 nM). LPC was added in Ca2+-free solution for 4–5 min before Ca2+ (2 mM) administration. Traces are for cells exposed to LPC alone (−Cheler), cells treated for 3 min with 10 μM Cheler before Ca2+ addition (+Cheler), and cells treated with LPC and 2-APB (100 μM) added as shown (*). The bar graph on the right shows average amplitudes of LPC-induced Ca2+ influx (Δratios ± SE) from control, 2-APB-treated, and Cheler (10 μM)-treated SMCs. n = 20–27 cells, as indicated above each bar.

Next, we tested if iPLA2β itself may be a target for PKC regulation and may be involved in OAG-induced Ca2+ entry. To assess the role of iPLA2β in OAG-induced Ca2+ influx, we used a chiral-specific mechanism-based suicidal substrate for iPLA2β as a new advanced pharmacological tool. Gross and others (21) demonstrated that two chiral enantiomers of BEL have different specificity to two different isoforms of iPLA2: S-BEL is specific to iPLA2β, whereas R-BEL is more specific to iPLA2γ. Recently, we (15) have shown that S-BEL but not R-BEL produced a dose-dependent inhibition of SOCE in rat basophilic leukemia (RBL) cells. Figure 5A demonstrates that in SMCs, S-BEL also produced a dose-dependent inhibition of TG-induced SOCE with an IC50 of ∼3 μM. In contrast, R-BEL had very little effect. Figure 5B shows that, similar to TG-induced SOCE, OAG-induced Ca2+ influx was also inhibited by S-BEL but not R-BEL, thus confirming that iPLA2β (but not iPLA2γ) is important not only for TG-induced Ca2+ entry but also for OAG-induced Ca2+ entry.

Fig. 5.

OAG-induced Ca2+ influx is inhibited by S-BEL but not by R-BEL. A, left: representative traces showing the changes in intracellular Ca2+ in SMCs induced by TG (2 μM) in control cells and in cells pretreated with BEL enantiomers. S-BEL was used to inhibit iPLA2β or R-BEL to block iPLA2γ (25 μM for 30 min at 37°C). Extracellular Ca2+ (2 mM) was added at the time shown (arrow). The bar graph on the right shows the dose dependence of the effects of S-BEL and R-BEL on TG-induced Ca2+ influx in SMCs. The best fit was generated with a Hill equation with an IC50 for S-BEl of ∼3 μM. n = 60–140 cells. B, left: representative traces of OAG-induced Ca2+ influx in control cells and in SMCs pretreated with S-BEL to block iPLA2β (25 μM for 30 min at 37°C) or R-BEL to block iPLA2γ (25 μM for 30 min at 37°C). The bar graph on the right summarizes data from 40–50 cells of the experiments on the left.

To test whether OAG could activate iPLA2β in SMCs, we determined iPLA2 activity in intact and homogenized SMCs. Figure 6A shows that in intact SMCs, TG and OAG both produced a significant activation of iPLA2. Inhibition of either PKC (with chelerythrine) or iPLA2β with S-BEL impaired OAG-induced activation of iPLA2. Interestingly, OAG activated iPLA2 not only when applied to intact cells but also in homogenized cells (Fig. 6B). This was in contrast to TG, which was unable to activate iPLA2 in cell homogenates in which Ca2+ stores were disintegrated, as we also showed in RBL cells (15). OAG-induced iPLA2β activation in cell homogenates was PKC dependent (chelerythrine sensitive) and could be a result of the previously described direct PKC-dependent phosphorylation of iPLA2β (21).

Fig. 6.

OAG-induced a PKC-dependent activation of iPLA2 in intact and homogenized SMCs. A: summary data showing the activity of iPLA2 in samples from intact untreated SMCs (basal), from cells treated for 5 min with 5 μM TG, from SMCs treated for 3 min with 50 μM OAG, from cells pretreated for 5 min with 10 μM Cheler and then 50 μM OAG, and from SMCs pretreated with 25 μM BEL (30 min at 37°C) and then 50 μM OAG. Data are from 12–15 measurements from 3 different cultures. B: average activity of iPLA2 in samples from untreated SMCs that were first homogenized and then treated as in A. Data are from 9–12 measurements from 4 different cultures. *Significant increases in activity compared with basal levels (P < 0.05).

OAG-induced store depletion and SOCE activation.

To examine if any earlier events in the SOCE pathway (preceding iPLA2β) may be also affected by OAG, we focused on the process of store depletion. As described above, we found a lot of similarities in TG- and OAG-induced SOCE in proliferating SMCs. Moreover, we noticed that a significant Ca2+ rise could be observed in a large number of SMCs when OAG was applied in Ca2+-free solution (as shown, for example, in Fig. 2B), which resembled a passive Ca2+ release upon the application of TG (see Fig. 2A). To test whether OAG may indeed cause Ca2+ release and significant depletion of Ca2+ stores, we compared the amount of Ca2+ that could be released from stores by ionomycin (IM; 100 nM) (37) in the presence and absence of OAG. Figure 7A shows that IM-induced Ca2+ release (which was observed in the absence of extracellular Ca2+) was significantly lower in cells pretreated with 50 μM OAG (ΔRatio = 0.24 ± 0.02, n = 85) compared with untreated cells (ΔRatio = 0.55 ± 0.01, n = 85). Importantly, the readdition of extracellular Ca2+ after IM induced a similar Ca2+ influx in the presence or absence of OAG. These important results suggest that the significant part of OAG-induced Ca2+ entry may be a direct result of store depletion and SOCE activation, which could explain the high similarity of OAG- and TG-induced Ca2+ influx patterns. Interestingly, we found that, in contrast to TG-induced irreversible store depletion and SOCE activation, OAG-induced SOCE was reversible and time dependent, similar to SOCE activated in other cells by agonists that involve DAG production (14, 37). Figure 7B shows that OAG-induced SOCE reached a maximum within the first minute after OAG application and within 5 min declined to ∼25% of its maximum level. These results strongly suggest that the depletion of Ca2+ stores and activation of SOCE could be one of the many effects of DAG that is naturally produced upon cell stimulation. These findings easily explain the high level of similarity found between TG- and OAG-induced SOCE and emphasize the fact that some OAG effects may be simply related to its ability to mimic the effects of TG.

Fig. 7.

OAG induced a partial depletion of intracellular Ca2+ stores and Ca2+ influx that declined within 5–10 min. A, left: representative traces showing the changes in intracellular Ca2+ concentration (F340/F380) in control SMCs or those pretreated with 50 μM OAG in Ca2+-free conditions. Ionomycin (IM; 100 nM) was applied to release Ca2+ from the stores, and Ca2+ (2 mM) was then added at the time indicated (arrow) to test for Ca2+ influx. Summary data are shown on the right. The bar graph presents the average Ca2+ release and Ca2+ influx tested in the absence (n = 85 cells) and presence of OAG (n = 140 cells). B, left: overlay of representative traces showing Ca2+ responses in SMCs when OAG (50 μM) was applied in the absence of extracellular Ca2+ and Ca2+ (2 mM) was then added at different times after OAG application, as shown (* and arrows). Summary data on the right show the time course of Ca2+ influx induced by OAG. Each bar is an average from 42–80 cells.

Role of PKC-ε in the activation of SOCE in proliferating SMCs.

While observing the high level of similarity between TG- and OAG-induced Ca2+ influx, and the similar roles that SOC channels and iPLA2β play in both of them, we found profound differences in the role for PKC in SOCE in actively proliferating SMCs and in SMCs that had reached a high degree of confluence and are known to undergo a transition from a differentiating to contractile (nonproliferating) phenotype (7, 16, 25). In all the experiments described above, actively proliferating SMCs were used (40–50% confluence), and, as shown in Fig. 2, both TG- and OAG-induced Ca2+ influx had similar amplitudes, were nonadditive, and could be impaired by the inhibition of SOC, iPLA2β, or PKC. Meanwhile, Fig. 8 shows that when SMCs reach full confluence (and stop proliferating), TG- and OAG-induced Ca2+ entry also have similar amplitudes, are nonadditive, and were impaired by 2-APB and S-BEL (but not by R-BEL). However, contrary to actively proliferating SMCs (Fig. 2), nonproliferative cells appeared to be insensitive to chelerythrine (Fig. 8, B and C). Calphostin C (a different PKC inhibitor) and KN-62 (an inhibitor of CaM kinase) also failed to affect TG-induced SOCE in confluent SMCs (Fig. 8B). Thus, despite the fact that PKC was involved in TG- and OAG-induced SOCE in proliferating SMCs, they appeared to lose SOCE dependence on PKC when they reach confluence and stop proliferating. The expression and activity of some PKC isoforms may change during cell proliferation, and Assender and co-workers (2) showed that the expression levels of specific PKC-ε increased with the state of dedifferentiation of SMCs. PKC-ε is a novel PKC isoform that fits the profile that may be related to the SOCE pathway in SMCs: it is Ca2+ independent (and so can work in the presence of a strong Ca2+ chelator, which is a signature condition for SOC activation), can be activated by DAG, and is upregulated in proliferating SMCs. As a first step in testing the role of PKC-ε in SOCE in SMCs, we compared the expression of PKC-ε in proliferating SMCs (in which SOCE was highly sensitive to PKC inhibition) with confluent SMCs (in which SOCE was insensitive to PKC inhibition). Immunostaining (Fig. 9A) and Western blot analysis (Fig. 9B) showed that, indeed, PKC-ε is highly expressed in proliferating SMCs, whereas it was hardly detectable in cells after they reached 100% confluence.

Fig. 8.

OAG and TG induced Ca2+ influx in confluent SMCs. A, left: representative traces showing the changes in intracellular Ca2+ concentration in fura-2-loaded aortic SMCs. OAG (50 μM) was applied in the absence of extracellular Ca2+, and 2 mM Ca2+ was then added at the time shown (arrow). TG (2 μM) was added at the end of the experiments. The bar graph on the right shows average amplitudes (Δratios ± SE) of TG-induced Ca2+ influx, OAG-treated SMCs, and from SMCs where TG was applied after the addition of extracellular Ca2+ in OAG-treated SMC, as shown on the left. Summary data are from 42–60 cells, as shown above each bar. B: bar graph showing the changes in intracellular Ca2+ concentration (Δratios ± SE) in confluent SMCs. Bars are for SMCs treated with TG (2 μM) for 4–5 min in the absence of extracellular Ca2+, and Ca2+ (2 mM) was then added (control); for SMCs treated with TG when 2-APB (100 μM), Cheler (10 μM), calphostin C (1 μM), or KN-62 (10 μM) were added 2 min before Ca2+; and for cells pretreated with 25 μM of either R-BEL or S-BEL for 30 min. Numbers of cells are indicated above each bar. C: same as in B except that OAG was applied to induce Ca2+ influx. Numbers of cells are indicated above each bar.

Fig. 9.

Distribution of PKC-ε in proliferating and confluent SMCs. A: immunostaining of proliferating (left) and confluent (right) SMC with PKC-ε antibody conjugated to FITC. B: Western blot (top) and densitometry analysis (bottom) showing PKC-ε expression in samples from proliferating and confluent SMCs. Average data were from 3 different Western blots.

To test the role of PKC-ε in the activation of SOCE, actively proliferating SMCs were transfected with antisense or sense oligonucleotides specific to PKC-ε [as described in a previous study (36)]. Figure 10A shows that TG-induced Ca2+ influx was significantly impaired in SMCs transfected with PKC-ε-specific antisense oligonucleotides, whereas sense oligonucleotides did not produce any changes, compared with control SMCs. Similarly, SMC transfection with antisense but not sense oligonucleotides for PKC-ε inhibited OAG-induced Ca2+ influx (Fig. 10B). Transfection of 95% confluent SMCs with antisense oligonucleotides for PKC-ε was without any effect on OAG- or TG-induced Ca2+ influx (Fig. 10C). Thus, PKC-ε appeared to be a specific PKC isoform that regulates SOCE and may play a crucial role in proliferating SMCs.

Fig. 10.

Effect of SMC transfection with antisense nucleotides for PKC-ε on TG- and OAG-induced Ca2+ influx. A, left: representative traces showing the changes in intracellular Ca2+ concentration in proliferating SMCs (40–60%). Cells were treated with TG (2 μM for 5 min in the absence of extracellular Ca2+), and Ca2+ then was added at the time shown (arrow). Traces are for nontransfected SMCs (control) and SMCs transfected with PKC-ε sense (s) or antisense (a/s) oligonucleotides 60–72 h before the experiments. Insert, representative Western blot of PKC-ε expression from SMCs transfected with sense and antisense oligonucleotides. Summary data on the right show TG-induced Ca2+ influx from the experiments on the left. The average was from a total of 75–150 cells. B: same as in A except that OAG (50 μM) was applied to induce Ca2+ influx. The average was from a total of 50–60 cells. C: bar graphs showing summary data of TG-induced (left) and OAG-induced (right) Ca2+ influx in SMCs confluent at 95% under same conditions as in A and B. The average was from 108–135 cells for TG-treated SMCs and 30–35 cells for OAG-treated SMCs.

DISCUSSION

In summary, our new findings demonstrate which parts of the SOCE pathway may be affected by OAG/DAG and which specific isoform of PKC may be involved in a complex regulation of SOCE in vascular SMCs. Striking similarities between Ca2+ influxes triggered by TG and OAG, the nonadditive nature of TG- and OAG-induced responses, and the crucial roles played by SOC and iPLA2β in both processes indicate that TG- and OAG-induced Ca2+ influx is likely to be mediated by the same SOCE pathway, which may be triggered by the depletion of Ca2+ stores.

Several lines of evidence suggest that OAG may produce the depletion of Ca2+ stores in SMCs. Indeed, OAG application resulted in significant (but not complete) Ca2+ loss from stores, which may be due to the OAG-induced uncoupling of sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA) function that was demonstrated earlier by Cardoso and colleagues (12) in muscle cells and platelets. Transient inhibition of SERCA by OAG may easily explain the transient activation of Ca2+ entry that we found to peak within the first minute and then fade away, leaving 25% of the maximum Ca2+ entry after 5 min. The transient nature of the OAG-induced effect was similar to what we earlier observed in platelets upon their activation with thrombin, which may involve naturally produced DAG (37). Some of the controversies in earlier reports related to the ability of DAG (or OAG) to activate SOCE may be explained by the time dependence of its effect on store depletion. One will expect that the time window when OAG effects could be observed and the level of the residual SOCE may significantly vary, being longer lasting and more pronounced in some cell types, whereas being hardly detectible in other cell types and experimental conditions. Importantly, OAG-induced activation of single SOC channels (3) could be a result of OAG-induced store depletion (in the case of cell-attached membrane patches in intact SMCs) or direct PKC-dependent activation of iPLA2β when SOC channels were studied in outside-out membrane patches in which iPLA2β (and PKC) may remain present and fully functional (32).

The fact that OAG was able to activate SOCE in both proliferating and confluent SMCs, which had different levels of expression of PKC-ε and different sensitivities of SOCE to general PKC inhibitors, strongly suggests that OAG may mimic the effects of TG and activate SOCE via depletion of stores, which does not seem to involve PKC. However, the important role of PKC-ε in both TG- and OAG-induced SOCE and their equal sensitivity to PKC inhibitors in proliferating SMCs strongly suggest that PKC-ε can regulate some other step(s) in the SOCE pathway. We found that lysophospholipid-induced SOCE activation (a part of the pathway downstream of iPLA2) is not sensitive to PKC inhibition, whereas a PKC inhibitor can significantly impair iPLA2 activation in cell homogenates. This suggests that iPLA2β itself can be a target for PKC regulation in proliferating SMCs. In vitro activation of iPLA2β by PKC-dependent phosphorylation was reported earlier in the membrane fraction of ventricular myocytes and macrophage-like P388D1 cells (1, 34). Such PKC-induced activation of iPLA2β may either originate a “shortcut” signal in the SOCE pathway or may amplify the signal originating in depleted stores, in both cases resulting in SOCE activation. Our new finding that PKC plays a role only in proliferating but not confluent SMC suggests that the contribution of the PKC-dependent mechanism may vary in different cell types (or even in the same cells, but in different stages of their growth and proliferation). It also suggests that the major effect of OAG on SOCE may be via direct PKC-independent inhibition of SERCA function, as suggested by Cardoso et al. (12).

Thus, the role of DAG (OAG) and PKC in SOCE regulation may be a rather complex and flexible phenomenon that may physiologically adjust to the needs and expression patterns of different types of cells and different levels of their proliferation and differentiation. We do not believe that PKC is crucial for SOCE, because chelerythrine, which is known to block the majority of PKC isoforms, did not have any effect in confluent SMCs, where SOCE works just fine with or without it. The fact that chelerythrine blocks significant (but noticeably not all) SOCE in proliferating cells makes us believe that PKC can be an important modulator of SOCE, and our study provides evidence that iPLA2β can be a target for such PKC-dependent (and OAG/DAG-dependent) modulation. When one thinks about PKC and Ca2+ regulation, many additional processes within the cell should be taken into consideration. As one of many examples, SERCA inhibitors (which are widely used as a specific method to trigger the activation of SOCE) may produce a significant Ca2+ rise in the cytosol (due to passive Ca2+ release), which, in the absence of efficient intracellular Ca2+ buffering, may lead to the Ca2+-induced activation of PLC (20). This will result in the production of DAG, which can trigger the activation and reported translocation of PKC to the plasma membrane (23, 24, 34, 39), where it can modulate iPLA2β activity and SOCE. This nonspecific pathway can cross talk with the SOCE pathway, and its ability to modulate SOCE will greatly depend on the ability of the cell to effectively buffer the passive Ca2+ leak from the stores (which will translate into whether PLC can get activated or not and if DAG will be produced) and the availability of DAG-dependent PKC that can translocate to the plasma membrane and regulate iPLA2β activity. All these can vary in different cell types as well as in the same cells at different stages of their proliferation and may be one of the reasons why PKC can play an important role in proliferating but not confluent SMCs. Further studies are needed to assess the potential ability of some other protein kinases to modulate SOCE through iPLA2β and to discover the whole net of complex interactions between different signaling cascades that may be involved in the regulation of the SOCE pathway.

GRANTS

This work was supported by American Heart Association Grant 04252860; Fondos de Investigación Sanitaria del Gobierno de España Grant PI050396 (to T. Smani), and National Heart, Lung, and Blood Institute Grants HL-54150 and HL-71793 (to V. M. Bolotina). T. Smani is an investigator of the “Ramon y Cajal” program supported by the Spanish Ministry of Education.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract