Modulation of PKCδ tyrosine phosphorylation and activity in salivary and PC-12 cells by Src kinases

Cyril Benes, Stephen P. Soltoff

Abstract

Protein kinase C (PKC) δ becomes tyrosine phosphorylated in rat parotid acinar cells exposed to muscarinic and substance P receptor agonists, which initiate fluid secretion in this salivary cell. Here we examine the signaling components of PKCδ tyrosine phosphorylation and effects of phosphorylation on PKCδ activity. Carbachol- and substance P-promoted increases in PKCδ tyrosine phosphorylation were blocked by inhibiting phospholipase C (PLC) but not by blocking intracellular Ca2+ concentration elevation, suggesting that diacylglycerol, rather than d-myo-inositol 1,4,5-trisphosphate production, positively modulated this phosphorylation. Stimuli-dependent increases in PKCδ activity in parotid and PC-12 cells were blocked in vivo by inhibitors of Src tyrosine kinases. Dephosphorylation of tyrosine residues by PTP1B, a protein tyrosine phosphatase, reduced the enhanced PKCδ activity. Lipid cofactors modified the tyrosine phosphorylation-dependent PKCδ activation. Two PKCδ regulatory sites (Thr-505 and Ser-662) were constitutively phosphorylated in unstimulated parotid cells, and these phosphorylations were not altered by stimuli that increased PKCδ tyrosine phosphorylation. These results demonstrate that PKCδ activity is positively modulated by tyrosine phosphorylation in parotid and PC-12 cells and suggest that PLC-dependent effects of secretagogues on salivary cells involve Src-related kinases.

  • serine/threonine phosphorylation
  • parotid acinar
  • protein kinase C

the involvement of tyrosine phosphorylation in heterotrimeric GTP-dependent protein (G protein)-coupled receptor-mediated events has been documented in many cellular systems, including freshly isolated cells and cultured cell lines. These events may involve the activation of nonreceptor tyrosine kinases such as Src and/or the transactivation of growth factor receptors (1, 6-8, 30, 43). Previously, we observed that carbachol and substance P promoted rapid increases in the tyrosine phosphorylation of protein kinase C (PKC) δ in salivary gland epithelial cells (rat parotid acinar cells) (46). In these cells, the muscarinic and substance P receptors are linked to phospholipase C (PLC)-β via a Gq-type G protein. The activation of these G protein-coupled receptors (GPCRs) produces rapid increases in diacylglycerol (DAG) andd-myo-inositol 1,4,5-trisphosphate production, which promote the activation of PKC and increases in intracellular Ca2+ concentration ([Ca2+]i), resulting in the initiation of fluid secretion (saliva formation) by these cells. The rapid (within seconds) nature of tyrosine phosphorylation of PKCδ after activation of these receptors suggests that PKCδ may play a role in fluid secretion or other neurotransmitter-promoted events in these cells. Phorbol esters, which activate many PKCs, including PKCδ, also produce many regulatory effects on salivary cells, including the modulation of early response genes (55), Ca2+-dependent K+ channels (33), RNA synthesis (54), cAMP response to β-adrenergic stimuli (53), and exocytosis (56). The present study had several goals, including the following: 1) to determine the factors involved in tyrosine phosphorylation of PKCδ in freshly isolated salivary acinar cells; 2) to determine whether secretagogues alter PKCδ enzyme activity and whether such alterations are dependent on tyrosine phosphorylation; and 3) to determine whether Src-related tyrosine kinases contribute to PKCδ enzyme activity and tyrosine phosphorylation in salivary and other cells.

PKCδ is a member of the PKC family of proteins, which is subdivided into three classes: Ca2+-dependent classic PKCs (cPKCs), consisting of α, βI, βII, and γ; novel PKCs (nPKCs), consisting of δ, ε, η, and θ; and atypical PKCs (aPKCs), consisting of ζ and ι/λ. The cPKCs and nPKCs bind and are activated by DAG, for which phorbol esters can substitute. PKCδ becomes tyrosine phosphorylated in cells exposed to various stimuli, including growth factors (10, 26, 27) and other receptor ligands (44,46, 47, 50), phorbol esters (10, 19, 46), and H2O2 (21, 49), and in cells transformed with rasHa or v-src (9, 58). A number of groups have reported that PKCδ can be tyrosine phosphorylated in vitro by c-Abl, Src, Lyn, and other members of the Src family of tyrosine kinases (2, 10, 14, 19, 47, 49, 50,58). However, there are conflicting reports concerning the contribution of tyrosine phosphorylation to the enzymatic activity of PKCδ (13). There was an increase in the tyrosine phosphorylation of PKCδ in cells in which the oncogenes v-src (58) or rasHa (9) were overexpressed, and this was associated with a decrease in PKCδ enzymatic activity. In other cells, there was a positive association between tyrosine phosphorylation and PKCδ enzyme activation in response to various stimuli (27, 28, 37), and other studies failed to find any contribution of tyrosine phosphorylation in PKCδ activity (26).

PKCδ has multiple tyrosine residues, and sites of tyrosine phosphorylation have been localized to the regulatory domain at the NH2-terminal side [Tyr-52 (50) and Tyr-187 (26)], the catalytic domain at the COOH-terminal side of PKCδ [Tyr-512 and Tyr-523 (21, 49)], and at a site between the regulatory and catalytic domains [Tyr-311 (2)]. Recent studies demonstrated that PKCδ is phosphorylated on serine and threonine sites (23, 39, 40) similar to those on cPKC family members (4, 20). In serum-starved cells, serum promoted the phosphorylation of a COOH-terminal hydrophobic site (Ser-662) on PKCδ via a mammalian target of rapamycin (mTOR)-dependent pathway, and a site (Thr-505) in the activation loop of the kinase domain was phosphorylated by 3-phosphoinositide-dependent protein kinase-1 (PDK1), which itself is activated by the lipid kinase phosphatidylinositol (PtdIns) 3-kinase. These sites and another one (Ser-643) were critical for the functional activity of PKCδ. To put PKCδ tyrosine phosphorylation in the context of Thr-505 and Ser-662 phosphorylation, we used phosphospecific antibodies to examine the phosphorylation status of PKCδ on Thr-505 and Ser-662 in freshly isolated parotid cells exposed to stimuli that increase the tyrosine phosphorylation of PKCδ. We also examined the active involvement of mTOR and PtdIns 3-kinase in these phosphorylations in these cells.

On the basis of our previous findings, we examined various aspects concerning the mechanism of tyrosine phosphorylation of PKCδ in freshly isolated rat parotid acinar cells. Little information is available concerning the role of tyrosine phosphorylation as a signaling mechanism by exposure of these salivary gland cells to GPCR ligands that promote fluid secretion. We investigated postreceptor events that could contribute to the tyrosine phosphorylation of PKCδ, including PLC activation, [Ca2+]i elevation, and the involvement of protein tyrosine kinases and phosphatases. A major part of the present study was to determine whether tyrosine phosphorylation produced any alterations on the enzymatic activity of PKCδ in parotid cells in response to secretory stimuli. To extend correlative observations that indicated that increases in PKCδ tyrosine phosphorylation produced increases in PKCδ activity, the reversibility of changes in enzymatic activity was examined by dephosphorylating PKCδ in vitro using PTP1B, a protein tyrosine phosphatase. We also examined the in vitro and in vivo contributions of Src and Src-related kinases to PKCδ activation and tyrosine phosphorylation. As a comparison for evaluating the contribution of tyrosine phosphorylation to PKCδ activity in parotid acinar epithelial cells, we also examined this in PC-12 cells, a nonepithelial cell type.

MATERIALS AND METHODS

Chemicals.

All chemicals were reagent grade or better. Carbamylcholine (carbachol), orthovanadate, and H2O2 were purchased from Sigma (St. Louis, MO). Pervanadate was formed by combining H2O2 and vanadate in a 1:1 molar ratio and was made fresh on the day of the experiment. Substance P was obtained from Peninsula Laboratories (Belmont, CA). PMA (phorbol 12-myristate 13-acetate) was obtained from Life Technologies (Grand Island, NY). U-73122 was purchased from BioMol (Plymouth Meeting, PA). 4-Amino-5- (4-methylphenyl)-7-(t-butyl) pyrazolo[3,4-d]pyrimidine (PP1) was purchased from Calbiochem (La Jolla, CA) and BioMol. Anti-phosphotyrosine antibody was a generous gift of Dr. Thomas Roberts (Dana Farber, Boston, MA). Anti-PKCδ polyclonal antibody (SC-213) and SRC2 antibody (SC-18; reactive toward p60Src and other selected Src-related kinases, including p62Yes and p59Fyn) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-PKCδ monoclonal antibody (MAb; P36520) was purchased from Transduction Laboratories (Lexington, KY). Phospho-PKCδ (Thr-505) and phospho-PKC (pan) antibodies, which recognize the Thr-505 and Ser-662 sites, respectively, on PKCδ were obtained from Cell Signaling (New England Biolabs). Anti-Src MAb (327) was a generous gift from Dr. Joan Brugge (Harvard Medical School, Boston, MA). Protein A-Sepharose beads and protein G-Sepharose beads were bought from Amersham Pharmacia. 1,2-Bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid acetoxymethyl ester (BAPTA-AM) was purchased from Molecular Probes (Eugene, OR). Igepal was purchased from ICN Biomedicals (Costa Mesa, CA). Recombinant PTP1B (14) was purchased from Upstate Biotechnology (Lake Placid, NY) and was also a generous gift from Dr. Benjamin Neel (Beth Israel Deaconess Medical Center, Boston, MA). DAG and phosphatidylserine (PS) were purchased from Avanti Polar Lipids (Alabaster, AL). Dulbecco's modified Eagle's medium (DMEM) was purchased from Bio Whittaker (Walkersville, MD).

Cell preparation and solutions.

Parotid acinar cells were prepared from male Sprague-Dawley rats (200–250 g; Charles River Laboratories, Kingston, NY, or Taconic, Germantown, NY) using previously established techniques (45). Briefly, rat parotid glands were removed and treated with trypsin and collagenase to get a suspension of single cells and small groups of cells. Cells were suspended at 1–1.5 mg of protein/ml in a medium composed of the following: 116.4 mM NaCl, 5.4 mM KCl, 1 mM NaH2PO4, 25 mM sodium HEPES, 1.8 mM CaCl2, 0.8 mM MgCl2, 5 mM sodium butyrate, and 5.6 mM glucose, pH 7.4. Cells were maintained on ice before use. Aliquots (1.5 ml) of the cell suspension were equilibrated at 37°C for at least 10 min before exposure to stimuli. In experiments designed to examine the contribution of [Ca2+]i to the tyrosine phosphorylation of PKCδ, cells were exposed to 25 μM BAPTA-AM or vehicle (DMSO) for 30 min at 37°C before they were exposed to stimuli. In some experiments, 5 mM EGTA was also added to the BAPTA-loaded cells before the addition of stimuli.

PC-12 cells were cultured in DMEM containing heat-inactivated 10% calf serum, 5% heat-inactivated fetal calf serum, and 1% penicillin/streptomycin. Cells were grown in 100-mm dishes, used at ∼75% confluence, and were serum starved in DMEM containing 0.1% serum overnight before exposure to stimuli.

Immunoprecipitations and Western blotting.

Parotid cells were exposed to various agents or vehicle (water or DMSO) and were then collected by a brief spin in a microcentrifuge (Brinkmann 5414). The supernatant was removed, and cells were lysed in 1 ml of ice-cold lysis buffer [137 mM NaCl, 20 mM Tris base, pH 7.5, 1 mM EGTA, 1 mM EDTA, 10% (vol/vol) glycerol, and 1% vol/vol Nonidet P-40 (NP-40) or Igepal] containing the following phosphatase and protease inhibitors: 1 mM vanadate, 1 mM ZnCl2, 4.5 mM sodium pyrophosphate, 47.6 mM NaF, 9.26 mM β-glycerophosphate, 0.5 mM dithiothreitol (DTT), 2 μg/ml leupeptin, 2 μg/ml pepstatin, 2 μg/ml aprotinin, and 2 μg/ml 4-(2-aminoethyl)-benzenesulfonyl fluoride hydrochloride. The lysates were vortexed and then sedimented at 16,000 g for 15 min at 4°C. In experiments conducted using PC-12 cells, the cells were exposed to stimuli in DMEM containing 0.1% serum for the designated times, washed twice with ice-cold phosphate-buffered saline (PBS) solution (136.9 mM NaCl, 2.68 mM KCl, 1.47 mM KH2PO4, and 15.65 mM NaH2PO4, pH 7.4), lysed in 1 ml of lysis buffer, vortexed, and then sedimented at 16,000 g for 15 min at 4°C. For parotid and PC-12 cell lysates, the cleared supernatants were added to fresh tubes, and proteins were immunoprecipitated using various antibodies in the following amounts: ∼0.5 μg/ml MAb PKCδ; 6 μg/ml anti-P-Tyr; 2 μg/ml MAb Src; and ∼1 μg/ml polyclonal PKCδ. Proteins were collected using protein A-Sepharose beads (4 mg/ml of lysate) except for the use of protein G-Sepharose (4 mg/ml) in experiments in which Src and PKCδ were immunoprecipitated individually or simultaneously with antibodies to both proteins. The composition of the buffers and the washing protocols for the immunoprecipitations were performed as described previously, as were the Western blotting conditions (46). For Western blotting, the nitrocellulose filters were exposed to the following dilutions of antibodies: anti-PKCδ (polyclonal; SC-213), 0.2 μg/ml; phospho-PKCδ (Thr-505), 1:1,000 dilution; phospho-PKC (pan), 1:1,000 dilution; anti-P-Tyr, 1 μg/ml; anti-Src (MAb 327), 1 μg/ml; and SRC2, 0.2 μg/ml.

In Western blotting experiments to determine whether PKCδ and Src or Src-related kinases coimmunoprecipitated with each other, proteins that immunoprecipitated with polyclonal and monoclonal antibodies to PKCδ were immunoblotted using polyclonal (SRC2) and MAb 327 Src antibodies. In addition, proteins that immunoprecipitated with anti-Src antibodies (MAb 327 and SRC2) were immunoblotted using polyclonal and monoclonal anti-PKCδ antibodies. Proteins were visualized with a chemiluminescence system (NEN) and X-ray film (Kodak).

PKCδ activity assay.

Cells were exposed to various agents and then lysed in 1 ml of ice-cold lysis buffer. After the lysates were cleared by centrifugation for 15 min at 16,000 g, anti-PKCδ MAb (∼0.5 μg/ml) was added to the lysate for ∼3 h at 4°C, and protein A-Sepharose (4 mg/ml) was added for 1 h to collect PKCδ. The immunoprecipitates were washed twice in PBS/NP-40 (1%) or PBS/Igepal (1%), once in 0.1 M Tris (pH 7.5)/LiCl (0.5 M), and twice in 25 mM Tris (pH 7.5)/0.5 mM EGTA/5 mM MgCl2/0.5 mM DTT. All wash solutions were used ice cold. The beads were resuspended in a final 100-μl reaction buffer [5 mM MgCl2, 0.5 mM EGTA, 10 μM PKCδ synthetic substrate peptide (AKRKRKGSFFYGG), 1 mM DTT, and 25 mM Tris · HCl (pH 7.5)]. As noted, in some experiments the assays were conducted in the additional presence of 20 μg/ml PS or the combination of 10 μM DAG plus 20 μg/ml PS in the assay buffer. The assays were initiated with the addition of ATP {50 μM ATP and 10 μCi [γ-32P]ATP (10 μCi/μl specific activity)}. The samples were incubated for 30 min at 30°C with intermittent mixing, and then duplicate 10-μl aliquots of each sample were spotted onto p81 phosphocellulose paper. Background activity was measured using lysates to which protein A-Sepharose was added without antibody. The p81 papers were washed five times in 0.425% phosphoric acid, and the amount of 32P was determined by liquid scintillation counting. The duplicate values from each immunoprecipitate were averaged and treated as one sample. Duplicate samples (2 separate immunoprecipitates) were usually collected for each of the various conditions in each experiment. The duplicate samples were averaged and treated as the results from one cell preparation (n = 1). Statistics were performed on data from three or more separate preparations/experiments. These assay conditions were similar to those in a previous study in which the sequence of the substrate peptide, which is based on the pseudosubstrate region of PKCδ, was determined to be optimal for PKCδ (35). The basal and stimulated PKCδ activities were completely blocked in vitro by the presence of various PKC inhibitors (1 μM RO31-8220 or 1 μM GF-109203X) in the assay mixture (not shown). Typical basal values ranged between 5,000 and 15,000 counts per minute, depending on the cell preparation.

Dephosphorylation of PKCδ using PTP1B.

In some experiments, immunoprecipitates were exposed to recombinant PTP1B (0.25 μg, specific activity 67.8 nmol · min−1 · μg−1 usingp-nitrophenylphosphate). The immunoprecipitates were washed twice in PBS/Igepal (1%), once in 0.1 M Tris (pH 7.5)/LiCl (0.5 M), and twice in 25 mM Tris (pH 7.5)/0.5 mM EGTA/5 mM MgCl2/5 mM DTT. Dephosphorylations or mock dephosphorylations (no added PTP1B) were performed in 100 μl of this same buffer for 30 min at 30°C with intermittent mixing. After dephosphorylation, the immunoprecipitates were washed twice in PBS/1% NP-40 or Igepal, once in 0.1 M Tris (pH 7.5)/LiCl (0.5 M), and twice in 25 mM Tris (pH 7.5)/0.5 mM EGTA/5 mM MgCl2/0.5 mM DTT. PKCδ activity assays were performed as described above.

Joint immunoprecipitation of Src and PKCδ.

In experiments in which the effect of Src on PKCδ activity was measured, cells were lysed in RIPA buffer (0.1% SDS, 1% NP-40 or Igepal, 0.5% deoxycholate, 158 mM NaCl, and 20 mM Tris, pH 8.0), and PKCδ was immunoprecipitated using anti-PKCδ MAb alone or a combination of anti-PKCδ antibody and anti-Src (MAb 327) antibodies. In some experiments, Src was immunoprecipitated in the absence of PKCδ using anti-Src antibody (MAb 327). The immunoprecipitates were collected with protein G-Sepharose beads and were washed three times in RIPA and twice in 20 mM HEPES (pH 7.4)/5 mM MgCl2/5 mM MnCl2. Some immunoprecipitates were subjected to in vitro phosphorylation reactions in the same buffer (± 100 μM ATP) for 45 min at 30°C with intermittent mixing. The immunoprecipitates were washed twice in PBS/1% Igepal, once in 0.1 M Tris (pH 7.5)/LiCl (0.5 M), and twice in 25 mM Tris (pH 7.5)/0.5 mM EGTA/5 mM MgCl2/0.5 mM DTT. PKCδ activity assays were performed as described above. Some immunoprecipitates used for PKCδ activity assays were subsequently used for Western blotting. For these samples, 2× sample buffer was added to the immunoprecipitates after the PKCδ assays were complete, and the samples were subjected to SDS-PAGE.

Data.

The mean value ± SE of n number of different experiments are as indicated. The percent inhibition of PKCδ activity by various conditions was calculated as the difference between stimulated and basal activities in the inhibitory condition (PP1 in vivo or PTP1B in vitro) compared with the difference between stimulated and basal under control conditions. Statistics were performed using a paired t-test. Western blots are representative of at least three different experiments.

RESULTS

Tyrosine phosphorylation of PKCδ is blocked by U-73122 in parotid cells exposed to carbachol or substance P but not PMA or pervanadate.

In rat parotid acinar cells, the tyrosine phosphorylation of PKCδ is increased in cells exposed to carbachol and substance P, which are agonists to PLC-linked GPCRs, as well as in cells exposed to PMA (46). The GPCR-dependent changes in tyrosine phosphorylation occur very rapidly and are maximal within the first minute of exposure to receptor ligands (46). To investigate whether PLC activation was involved in promoting the tyrosine phosphorylation of PKCδ, parotid cells were exposed to the PLC inhibitor (3, 12, 17, 57) U-73122 (10 μM), which blocked the elevation of [Ca2+]i in parotid acinar cells (not shown). The effects of carbachol and substance P were blocked in U-73122-treated cells (Fig.1 A), suggesting that the receptor-mediated tyrosine phosphorylation of PKCδ was initiated by the production of DAG or the elevation of [Ca2+]i. The tyrosine phosphorylation of PKCδ was also increased by pervanadate, an inhibitor of PTP1B (18), and other protein tyrosine phosphatases. Pervanadate (1 min) increased the tyrosine phosphorylation to a much greater level than carbachol (1 min) or PMA (5 min) (Fig. 1, B andC).1 The effects of PMA and pervanadate on the tyrosine phosphorylation of PKCδ were not blocked by U-73122 (Fig. 1, B and C), indicating that the effects of PMA and pervanadate are independent of PLC activity. The responses to pervanadate, which also increased the tyrosine phosphorylation of other proteins (not shown), suggest that the inhibition of a tyrosine phosphatase is sufficient to promote the tyrosine phosphorylation of PKCδ in these cells. Unlike the effects of pervanadate, equivalent exposures (100 μM, 1 min) of H2O2 or vanadate were ineffective in producing large increases in the tyrosine phosphorylation of PKCδ or other proteins (not shown). The effect of PMA on U-73122-treated cells suggests that the production of DAG is also sufficient to promote an increase in PKCδ tyrosine phosphorylation. In addition, either the constitutive activity of a tyrosine kinase or a kinase that is activated by carbachol, substance P, PMA, and pervanadate is able to promote the tyrosine phosphorylation of PKCδ in these epithelial cells. Additional experiments suggested that Src or a related protein kinase is the tyrosine kinase that phosphorylates PKCδ in both parotid and PC-12 acinar cells.

Fig. 1.

Receptor-initiated, but not phorbol 12-myristate 13-acetate (PMA) or pervanadate-initiated, increases in protein kinase C (PKC)δ tyrosine (Tyr) phosphorylation are dependent on phospholipase C. Parotid acinar cells were treated with 0.1% DMSO (−) or 10 μM U-73122 (+) for 20 min and were then exposed to various agents. A: cells were exposed to vehicle (−), carbachol (Carb; 20 μM, 1 min), or substance P (Sub P; 10−7 M, 0.2 min). Proteins were immunoprecipitated (IP) with polyclonal anti-PKCδ antibody and immunoblotted (IB) with anti-P-Tyr antibody (top). Left and right panels were from different experiments. One blot was stripped and reprobed with PKCδ antibody (bottom) to show that there were similar levels of this protein in all samples. B: cells were exposed to vehicle (−), carbachol (20 μM, 0.2 min), PMA (200 nM, 5 min), or pervanadate (Pervan; 100 μM, 1 min). Proteins were immunoprecipitated with anti-P-Tyr antibody and immunoblotted with anti-PKCδ antibody.C: cells were exposed to vehicle (−), pervanadate (100 μM, 1 min), or carbachol (10−4 M, 1 min). Proteins were immunoprecipitated with polyclonal anti-PKCδ and immunoblotted with anti-P-Tyr antibody. Immunoblots were exposed to X-ray film for two different lengths of time to show that treatment of cells with U-73122 did not reduce their response to pervanadate.

Elevation of [Ca2+]i does not contribute to PKCδ tyrosine phosphorylation.

Although we previously found that exposure of salivary gland epithelial cells to the Ca2+ ionophore ionomycin did not promote the tyrosine phosphorylation of PKCδ (46), there was a possibility that elevations in [Ca2+]i supported the stimulatory effects of carbachol and substance P on PKCδ tyrosine phosphorylation. To examine this possibility, we loaded parotid acinar cells with the Ca2+ chelator BAPTA, which buffers the carbachol-promoted rise in [Ca2+]i in these cells (45). BAPTA-loaded cells responded to carbachol in a similar manner as those cells that were not loaded with BAPTA (Fig.2). The carbachol-promoted increase in PKCδ tyrosine phosphorylation was also observed in BAPTA-loaded cells that were exposed to EGTA to deplete extracellular Ca2+(not shown). These results indicate that a rapid rise in [Ca2+]i was not required for the stimulatory effect of carbachol on PKCδ tyrosine phosphorylation and demonstrate that another aspect of muscarinic receptor activation was responsible for the rapid increase in PKCδ tyrosine phosphorylation. In conjunction with the inhibitory effects of U-73122 on the substance P- and carbachol-promoted PKCδ tyrosine phosphorylations (Fig. 1) and the positive effect of PMA on PKCδ tyrosine phosphorylation, these studies are consistent with DAG production playing the main role in the receptor-mediated increases in the tyrosine phosphorylation of PKCδ.

Fig. 2.

Carbachol-promoted tyrosine phosphorylation of PKCδ is not reduced in 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA)-loaded parotid acinar cells. Parotid acinar cells were exposed to 0.1% DMSO (−) or BAPTA-AM (25 μM) for 30 min and then to vehicle (−) or carbachol (20 μM, 1 min). Proteins were immunoprecipitated with polyclonal anti-PKCδ antibody and immunoblotted with anti-P-Tyr antibody (top). Blots were stripped and reprobed with anti-PKCδ antibody (bottom).

Tyrosine phosphorylation of PKCδ increases the activity of PKCδ in parotid acinar cells and PC-12 cells.

We performed in vitro substrate phosphorylation assays using PKCδ immunoprecipitated from cells exposed to various stimuli. The exposure of parotid acinar cells to carbachol (10−4 M, 1 min) increased the PKCδ activity to 1.8 ± 0.1 (n = 4) times the basal level found in unstimulated cells. PMA (100 nM, 5 min) and pervanadate (100 μM, 1 min) each increased PKCδ activity by a much greater amount, to 4.1 ± 1.0 (n = 6) and 4.6 ± 1.0 (n = 4), respectively, times the basal level. These results suggested that the tyrosine phosphorylation of PKCδ promoted an increase in PKCδ activity in salivary acinar cells.

To determine whether these results were specific to salivary gland acinar cells, we performed similar experiments using PC-12 cells. PMA (100 nM, 5 min) increased PKCδ activity to 2.5 ± 0.1 (n = 6) times the basal level in PC-12 cells. PMA (100 nM) added with pervanadate (100 μM; both for 5 min) increased the PKCδ activity to 3.5 ± 0.5 (n = 5) times the basal level. Under both conditions (± pervanadate), PMA produced an increase in tyrosine phosphorylation of PKCδ (below).

The stimuli-dependent changes in PKCδ activity were consistent with the conclusion that increases in the tyrosine phosphorylation of PKCδ produce increases in PKCδ activity in parotid and PC-12 cells. To test this hypothesis directly, we examined the effects of dephosphorylating the tyrosine-phosphorylated PKCδ by using PTP1B, a protein tyrosine phosphatase, to see if the increases in PKCδ activity were reversible. In these experiments, we collected anti-PKCδ immunoprecipitates from cells exposed to agents that increased PKCδ activity and tyrosine phosphorylation and exposed the immunoprecipitates to PTP1B to dephosphorylate PKCδ in vitro before measuring PKCδ activity (see materials and methods). Immunoprecipitates that were not exposed to PTP1B were treated in an identical manner, except that PTP1B was not present in the dephosphorylation step. The effects of PTP1B on tyrosine phosphorylation were also analyzed. PTP1B dephosphorylated PKCδ to near basal levels in immunoprecipitates from PMA- and pervanadate-treated parotid acinar cells (Fig.3 A). PTP1B also dephosphorylated PKCδ immunoprecipitated from carbachol-treated parotid acinar cells (not shown). Moreover, the dephosphorylations were accompanied by large reductions in the increased PKCδ activity (Fig.3 B). PTP1B treatment reduced the increases in response to carbachol, PMA, and pervanadate by 98.0 ± 19.7% (n = 4), 86.2 ± 10.7% (n = 6), and 79.6 ± 21.8% (n = 4), respectively. PTP1B did not significantly alter the basal PKCδ activity in untreated parotid acinar cells (Fig. 3 B), consistent with the low or absent tyrosine phosphorylation of PKCδ in untreated cells. These results demonstrate conclusively that increases in tyrosine phosphorylation produce increases in PKCδ activity that can be reversed when tyrosine residues are dephosphorylated.

Fig. 3.

PTP1B, a protein tyrosine phosphatase, reverses stimulus-dependent increases in PKCδ tyrosine phosphorylation and activity in parotid acinar cells. Parotid cells were exposed to carbachol (10−4 M, 1 min), PMA (100 nM, 5 min), pervanadate (PV; 100 μM, 1 min), or left untreated (B, basal). PKCδ was immunoprecipitated with an anti-PKCδ MAb. Immunoprecipitates were exposed to a dephosphorylation buffer with PTP1B or without (mock) PTP1B for 30 min, and the immunoprecipitates were washed and subjected to a substrate phosphorylation assay to quantify activity (B). After the assay, immunoprecipitates were subjected to SDS-PAGE on a 7% separating gel, transferred to nitrocellulose, and immunoblotted. A: immunoprecipitated PKCδ was immunoblotted with an anti-P-Tyr antibody (top). Blots were stripped and reprobed with polyclonal anti-PKCδ antibody (bottom). The open arrowhead indicates the tyrosine-phosphorylated form of PKCδ, and the closed arrowhead indicates the nontyrosine-phosphorylated form. B: alterations in PKCδ activity in anti-PKCδ immunoprecipitates exposed (PTP1B) or not exposed (mock) to PTP1B. All activity values are normalized to the activity found under basal conditions without PTP1B treatment. For carbachol, PMA, and pervanadate, n = 4, 6, and 4, respectively. ## P < 0.001 and# P < 0.005 vs. basal (mock). **P < 0.001 and *P < 0.005 vs. stimulus + mock treatment.

We also examined the effects of PTP1B-mediated dephosphorylation on PKCδ in PC-12 cells to determine whether the effects of tyrosine phosphorylation on PKCδ activity were reversible. Exposure of PC-12 cells to PMA or PMA plus pervanadate produced large increases in PKCδ tyrosine phosphorylation, and exposure of the immunoprecipitates to PTP1B dephosphorylated PKCδ (Fig.4 A). PTP1B also reduced the increase in PKCδ activity. The increases in PKCδ activity in cells treated with PMA or PMA plus pervanadate were reduced by 66.9 ± 0.2% (n = 6) and 60.4 ± 11.5% (n = 5), respectively (Fig.4 B).2 PTP1B did not significantly affect the basal PKCδ activity in PC-12 cells (Fig.4 B). These results are similar to those obtained using salivary gland acinar cells and conclusively demonstrate that tyrosine phosphorylation of PKCδ can increase the activity of PKCδ in a reversible manner in multiple types of cells.

Fig. 4.

PTP1B reverses increases in PKCδ tyrosine phosphorylation and activity in PC-12 cells exposed to PMA and pervanadate. PC-12 cells were exposed to vehicle or PMA (100 nM) ± pervanadate (100 μM) for 5 min. PKCδ was immunoprecipitated with anti-PKCδ MAb. Immunoprecipitates were exposed to a dephosphorylation buffer with or without PTP1B for 30 min, and they were washed and subjected to a substrate phosphorylation assay to quantify activity (B). After assay, immunoprecipitates were subjected to SDS-PAGE on a 7% separating gel, transferred to nitrocellulose, and immunoblotted. A: immunoprecipitated PKCδ was immunoblotted with an anti-P-Tyr antibody (top). Blots were stripped and reprobed with polyclonal anti-PKCδ antibody (bottom). The open arrowhead indicates the tyrosine-phosphorylated form of PKCδ, and the closed arrowhead indicates the nontyrosine-phosphorylated form. B: alterations in PKCδ activity in anti-PKCδ immunoprecipitates exposed (PTP1B) or not exposed (mock) to PTP1B. All activity values are normalized to the activity found under basal conditions without PTP1B treatment. For PMA and PMA + pervanadate, n = 6 and 5, respectively. ## P < 0.001 vs. basal (mock). **P < 0.001 vs. stimulus + mock treatment.

Increases in PKCδ tyrosine phosphorylation retard the migration of PKCδ through SDS-polyacrylamide gels, as observed previously in parotid acinar cells (46). This produced a doublet in PKCδ when anti-PKCδ immunoprecipitates were immunoblotted using anti-PKCδ antibody, especially when a significant fraction of PKCδ was phosphorylated on tyrosine (Fig. 4 A). The tyrosine-phosphorylated PKCδ comigrated with the top band. The in vitro dephosphorylation of PKCδ by PTP1B decreased the appearance of the top band in the anti-PKCδ immunoblots (Fig. 4 A) and decreased the amount of tyrosine-phosphorylated PKCδ when the anti-PKCδ immunoprecipitates were immunoblotted using anti-P-Tyr antibody (Fig. 4 A). These results suggest that the top band of PKCδ consists nearly exclusively of the tyrosine-phosphorylated protein.

Stimuli-dependent increases in PKCδ activity did not occur in presence of lipid cofactors.

The alterations in PKCδ activity in parotid and PC-12 cells presented above were measured in the absence of lipid cofactors and reflect an increase in PKCδ activity that is intrinsic to the immunoprecipitated protein and not due to increases in protein levels. When PS was present in the assay mixture (see materials and methods), the PKCδ activity in immunoprecipitates from PMA (100 nM, 5 min)-stimulated PC-12 cells was significantly greater than the activity in unstimulated cells, but the degree of stimulation was less than that measured in assays conducted in the absence of added lipids (Fig.5). Note, however, that the basal activity in the presence of PS in vitro was approximately three times that measured in the absence of PS (see Fig. 5 legend). When DAG and PS were both present in the assay mixture, the basal PKCδ activity in anti-PKCδ immunoprecipitates from unstimulated PC-12 was increased to 28.9 ± 4.2 (n = 3) times the activity measured in the absence of lipid cofactors. However, in the presence of DAG plus PS, the PKCδ activity found in immunoprecipitates from stimulated PC-12 cells was not enhanced compared with the activity in immunoprecipitates from unstimulated cells. For example, under these conditions (PS+DAG) the activity in PMA (100 nM, 5 min)-treated cells was 0.9 (n = 2) times that in untreated cells (Fig. 5), and the activity in PMA (100 nM, 5min) plus pervanadate (100 μM, 5 min)-treated cells was 1.0 ± 0.1 (n = 4) times the activity in untreated cells. Thus the contribution of tyrosine phosphorylation to PKCδ activity was not manifested when DAG and PS were both present in the assay, as was also reported by others (21, 36).

Fig. 5.

Increases in PKCδ activity in PMA-treated cells are reduced in the presence of phosphatidylserine (PS). PKCδ activity was measured in PKCδ immunoprecipitated from unstimulated (basal) PC-12 cells or PC-12 cells exposed to PMA (100 nM) for 5 min. In paired studies, in vitro activities were measured in the absence or presence of PS (20 μg/ml) or PS (20 μg/ml) combined with diacylglycerol (DAG; 10 μM). To compare the relative effects of PMA under these 3 conditions, the basal activities (± PS/DAG) were each normalized to 100. Basal PKCδ activity assayed in the presence of PS was 2.9 ± 1.2 (n = 4) times that measured in the absence of lipids, and basal activity in the presence of DAG+PS was 28.9 ± 4.2 (n = 3) times the activity measured in the absence of lipids. **P < 0.005 vs. basal, PS absent (n = 4). *P < 0.05 vs. basal, PS present (n = 4).

Similar findings were made using parotid acinar cells. When DAG plus PS were present in the assay mixture, the basal PKCδ activity in anti-PKCδ immunoprecipitates from unstimulated parotid cells was increased to 24.4 ± 4.3 (n = 3) times the activity in the absence of lipids. Also, in assays conducted with DAG plus PS in the assay mixture, the PKCδ activity found in immunoprecipitates from unstimulated cells was the same as the activity in cells treated with carbachol, PMA, or pervanadate (not shown). These results bear on the biological significance of the activation of PKCδ (see discussion).

Src phosphorylates PKCδ and increases its enzymatic activity.

Previously, we found that a kinase inhibitor, staurosporine, and genistein, a nonselective tyrosine kinase inhibitor, blocked the tyrosine phosphorylation of PKCδ (5). Because several investigators have suggested a role for Src, Src-related kinases, and other tyrosine kinases in the tyrosine phosphorylation of PKCδ, we conducted several experiments to determine whether Src-related proteins are involved in the activation of PKCδ by receptor-mediated and nonreceptor-mediated agents in parotid cells and PC-12 cells. In one series of in vivo experiments, parotid acinar cells were treated with PP1, an inhibitor of Src and Src family members (16), and then the cells were exposed to agents that promoted increases in PKCδ tyrosine phosphorylation and activity. Treatment of parotid cells with PP1 blocked carbachol- and pervanadate-promoted increases in tyrosine phosphorylation (Fig. 6 A). PP1 also blocked the effects of PMA on PKCδ tyrosine phosphorylation (not shown). PP1 treatment of parotid cells also blocked the increases in PKCδ activity promoted by carbachol, PMA, and pervanadate by 78.2% (n = 2), 83.1 ± 6.6% (n = 3), and 67.3 ± 21.3% (n = 3), respectively (Fig.6 B). PP1 treatment did not significantly reduce the basal activity in parotid cells (Fig. 6 B), consistent with a role for Src in the increase of PKCδ activity but not in basal activity. PP1 blocked the tyrosine phosphorylation of PKCδ in PMA (± pervanadate)-treated PC-12 cells (Fig. 6 C), and it also blocked the PMA-promoted increase in PKCδ activity by 74.8% (n = 2). These results suggest that the in vivo activity of Src or an Src-like protein is responsible for the increases in PKCδ tyrosine phosphorylation and the resulting increases in PKCδ activity in both parotid acinar cells and PC-12 cells. These results also suggest that a Src-related kinase is involved in tyrosine phosphorylation that is promoted by diverse agents: PLC-linked receptor agonists, phorbol ester, and a phosphatase inhibitor.

Fig. 6.

Increases in PKCδ tyrosine phosphorylation and activity in parotid cells and PC-12 cells are blocked by 4-amino-5-(4-methylphenyl)-7-(t-butyl) pyrazolo[3,4-d]pyrimidine (PP1). Parotid acinar cells or PC-12 cells were treated with 10 μM PP1 or vehicle (DMSO) for 15 min before exposure to various agents. PKCδ was immunoprecipitated with an anti-PKCδ MAb. Immunoprecipitates were subjected to SDS-PAGE, transferred to nitrocellulose, and immunoblotted (A andC). In some experiments (B), immunoprecipitates were first subjected to a PKCδ activity assay (see materials and methods). A: parotid acinar cells were not stimulated (−) or exposed to carbachol (10−4 M, 1 min) or pervanadate (100 μM, 1 min). PKCδ immunoprecipitates were sequentially immunoblotted with anti-phosphotyrosine antibody (top) and a polyclonal anti-PKCδ antibody (bottom). Arrows indicate location of tyrosine-phosphorylated form of PKCδ. B: alterations in PKCδ activity in anti-PKCδ immunoprecipitates from parotid cells exposed to PP1 or vehicle (mock) before carbachol (10−4 M, 1 min), PMA (100 nM, 5 min), or pervanadate (100 μM, 1 min). All activity values are normalized to the activity found under basal conditions without PP1 treatment. For basal with PP1, n= 5. For carbachol, PMA, and pervanadate, n = 2, 3, and 4, respectively. ## P < 0.001 and# P < 0.05 vs. basal (control). *P < 0.01 vs. stimulus without PP1. C: PC-12 cells were not stimulated (basal) or were exposed to 100 nM PMA ± 100 μM pervanadate (5 min) after treatment with PP1 or vehicle (−). PKCδ immunoprecipitates were sequentially immunoblotted with anti-phosphotyrosine antibody (top) and a polyclonal anti-PKCδ antibody (bottom). See text for PKCδ activity measurements in PC-12 cells measured under some of these conditions. The open arrowhead indicates the tyrosine-phosphorylated form of PKCδ, and the closed arrowhead indicates the nontyrosine-phosphorylated form.

In some systems, Src-related proteins may coimmunoprecipitate with PKCδ. Although we immunoblotted anti-PKCδ immunoprecipitates for Src and several Src-related proteins and vice versa, we were unable to observe any association between these proteins in either PC-12 cells or parotid acinar cells in response to various stimuli or under basal conditions (Fig. 7 A and unpublished observations; see materials and methods). However, the in vivo experiments shown in Fig. 6 suggested that a Src-related tyrosine kinase was involved in the activation of PKCδ. Therefore, we examined whether Src was able to increase the tyrosine phosphorylation and activity of PKCδ in vitro. In these experiments, we immunoprecipitated PKCδ, Src, or PKCδ plus Src from unstimulated parotid acinar cell lysates using antibodies to both PKCδ and Src. Some of the immunoprecipitated proteins were subjected to in vitro conditions (incubation with ATP and divalent cations) that promote substrate phosphorylation by tyrosine kinases, and then the activity of PKCδ was assayed. There was a substantial tyrosine phosphorylation of PKCδ (Fig. 7 A) only when immunoprecipitated Src was also present. There was not a significant degree of tyrosine phosphorylation of PKCδ when it was immunoprecipitated by itself (no Src antibody), even if the immunoprecipitate was subjected to conditions that promote tyrosine phosphorylation (Fig. 7 A). The tyrosine phosphorylation of the PKCδ that was immunoprecipitated jointly with Src was blocked if PP1 was present.

Fig. 7.

PKCδ tyrosine phosphorylation and activity is increased by Src in parotid acinar cells. PKCδ was immunoprecipitated from unstimulated parotid acinar cell lysates with either an anti-PKCδ MAb alone or simultaneously with an anti-Src MAb. Some immunoprecipitates were subjected to conditions (ATP) to promote tyrosine phosphorylation (see materials and methods). Immunoprecipitates were washed and then subjected to a substrate phosphorylation assay to quantify PKCδ activity. When present, PP1 was added during both the in vitro tyrosine phosphorylation and the activity assay. A: immunoprecipitated proteins were subjected to SDS-PAGE, transferred to nitrocellulose, and immunoblotted with anti-P-Tyr antibody (top). Blots were stripped and reprobed with anti-PKCδ antibody (middle) and anti-Src (MAb 327) antibody (bottom). The open arrowhead indicates the tyrosine-phosphorylated form of PKCδ, and the closed arrowhead indicates the nontyrosine-phosphorylated form. B: PKCδ activity in anti-PKCδ immunoprecipitates with and without immunoprecipitated Src (see materials and methods). All activity values are normalized to the activity found in anti-PKCδ immunoprecipitates that were not subjected to tyrosine phosphorylation reaction. *P < 0.01 vs. basal (PKC, no ATP);## P < 0.005 vs. PKC + ATP;+ P < 0.05 vs. no PP1. For each condition,n = 3 experiments.

In these in vitro experiments, there was a significant increase in PKCδ activity under conditions (joint immunoprecipitation of Src and PKCδ) that promoted the increase in tyrosine phosphorylation of PKCδ (Fig. 7 B). This increase in activity was blocked by PP1 in vitro. No increases in PKCδ activity were observed if immunoprecipitated Src was not present, even if PKCδ was subjected to in vitro phosphorylation conditions before the activity assay. Immunoprecipitates of Src without PKCδ did not display measurable PKC activity (not shown). Thus when these changes in activity are compared with the alterations in tyrosine phosphorylation, they indicate that the tyrosine phosphorylation of PKCδ has a positive effect on PKCδ activity in a Src-dependent manner. The in vitro PP1-sensitive Src-promoted increases in PKCδ activity and tyrosine phosphorylation (Fig. 7) are consistent with the inhibitory effects of in vivo PP1 on PKCδ activity (Fig. 6) being due to the inhibition of Src or a Src-like protein. The results of these in vitro and in vivo experiments indicate that Src or a Src-related protein increases the tyrosine phosphorylation of PKCδ in cells in which GPCRs are activated, protein tyrosine phosphatases are inhibited, or PKC is activated by phorbol ester. These increases in tyrosine phosphorylation increase the enzymatic activity of PKCδ.

Tyrosine phosphorylation of PKCδ does not alter constitutive serine and threonine phosphorylations.

PKCδ is also phosphorylated on particular serine and threonine residues that can affect PKCδ activity. To put PKCδ tyrosine phosphorylation in the context of Thr-505 and Ser-662 phosphorylation, we examined the phosphorylation of PKCδ on Thr-505 and Ser-662 in rat parotid acinar cells exposed to carbachol and PMA. Because PtdIns 3-kinase and mTOR activities are upstream of the phosphorylation of these sites (39, 40), we treated cells with the PtdIns 3-kinase inhibitor LY-294002 and the mTOR inhibitor rapamycin. In parotid cells, Thr-505 and Ser-662 sites were constitutively phosphorylated under basal conditions. These phosphorylations were not reduced by exposure of cells to LY-294002 (100 μM) or rapamycin (20 nM; Fig. 8) for times up to 30 min. The PtdIns 3-kinase inhibitor wortmannin (100 nM) gave similar results to LY-294002 (not shown). No alterations in the constitutive Thr-505 or Ser-662 phosphorylations were observed in cells exposed to 100 μM carbachol for 1–10 min (not shown) or PMA for 5 min (Fig.8) or 10 min (not shown) in control cells or in cells exposed to PtdIns 3-kinase and mTOR inhibitors. The increases in PKCδ tyrosine phosphorylation promoted by PMA (Fig. 8) and carbachol (not shown) were not affected by the presence of these inhibitors. In stimuli-treated cells, the more slowly migrating top band of PKCδ was phosphorylated on Thr-505 and Ser-662 in addition to tyrosine residues, but the bottom band was only phosphorylated on Thr-505 and Ser-662. These results indicate that tyrosine phosphorylation of PKCδ in freshly isolated parotid cells occurs in the absence of any change in Thr-505 and Ser-662 phosphorylation.

Fig. 8.

Thr-505 and Ser-662 sites on PKCδ are constitutively phosphorylated and not affected by LY-294002, rapamycin (Rap), or increases in PKCδ tyrosine phosphorylation. Parotid acinar cells were treated with 100 μM LY-294002, 20 nM rapamycin, or vehicle (0.1% DMSO) for 30 min and then exposed to 100 nM PMA or vehicle (0.1% DMSO) for 5 min. PKCδ was immunoprecipitated with an anti-PKCδ MAb. Immunoprecipitates were subjected to SDS-PAGE, transferred to nitrocellulose, and sequentially immunoblotted with a phosphospecific antibody for Thr-505 on PKCδ, an antibody that recognizes Ser-662 on PKCδ, anti-P-Tyr antibody, and polyclonal anti-PKCδ antibody. Arrows indicate location of the tyrosine-phosphorylated form of PKCδ. Results are representative of 2 other experiments.

DISCUSSION

A major reason for undertaking these studies was to examine the contribution of tyrosine phosphorylation to PKCδ enzyme activity in rat parotid acinar cells, which respond to secretory stimuli with an increase in PKCδ tyrosine phosphorylation. We also examined PC-12 cells as a second cell model. From our results, we conclude the following: 1) the activation of PLC and the production of DAG account for muscarinic and substance P-mediated increases in PKCδ tyrosine phosphorylation in parotid acinar cells; 2) increases in the tyrosine phosphorylation of PKCδ by receptor-mediated and nonreceptor-mediated agents have a positive effect on PKCδ enzyme activity in both parotid and PC-12 cells;3) the activation of PKCδ is reversible upon the dephosphorylation of the tyrosine residues; 4) Src or Src-related proteins phosphorylate PKCδ on tyrosine in vivo and in vitro, and this increases its activity; and 5) the stimuli-dependent tyrosine phosphorylation of PKCδ in freshly isolated salivary cells occurs subsequent to its phosphorylation on Thr-505 and Ser-662.

PKCδ was activated by receptor-mediated stimuli (carbachol), by PMA, which bypasses PLC (Fig. 1), and by pervanadate. Pervanadate does not appear to act by stimulating PLC (Fig. 1); presumably, it promotes the tyrosine phosphorylation of PKCδ by inhibiting tyrosine phosphatase activity. Increases in carbachol-mediated PKCδ tyrosine phosphorylation in parotid cells were blocked by inhibiting PLC (Fig.1) but were not blocked by preventing the elevation of [Ca2+]i (Fig. 2), suggesting that DAG production initiated the activation of PKCδ in parotid acinar cells in a Ca2+-independent manner. Previously, we reported that the tyrosine kinase inhibitor genistein blocked increases in the tyrosine phosphorylation of PKCδ (5). The data presented in this paper suggests that Src or a Src-related kinase is the tyrosine kinase that phosphorylates PKCδ in these cells. In addition, these studies suggest that both protein tyrosine kinases and phosphatases affect the tyrosine phosphorylation and activity of PKCδ.

A model for PKCδ tyrosine phosphorylation is as follows: the production of DAG in response to PLC-linked receptor activation, or, bypassing this step, the exposure of cells to PMA promotes the recruitment of PKCδ to the plasma membrane, where PKCδ is a substrate for Src or other Src-like proteins and is phosphorylated on tyrosine residues that contribute to increases in PKCδ activity. Previously, we demonstrated that PMA promoted the translocation of PKCδ from the cytosol to a membrane compartment of salivary gland acinar cells, and the tyrosine phosphorylated form of PKCδ was exclusively found in the membrane compartment (46). Other investigators have made similar observations concerning the localization of tyrosine-phosphorylated PKCδ (13, 19, 24,58), although an alternative finding of PMA-promoted tyrosine-phosphorylated PKCδ in the cytosol has also been reported (42).

Presumably, membrane-localized PKCδ is a substrate for Src or a Src-related protein that is constitutively active and able to phosphorylate PKCδ in cells exposed to GPCRs, PMA, and pervanadate. Other proteins known to be Src substrates are constitutively phosphorylated on tyrosine residues in unstimulated parotid cells (unpublished observations). Src can also be directly activated by α-subunits of G proteins (31), but this cannot account for the stimulation promoted by PMA or pervanadate. Under the conditions of our experiments, we did not find a physical association between Src or Src-related proteins and PKCδ. If this association occurred, it did not survive our immunoprecipitation protocol or was too subtle to detect by immunoblot techniques. In other cellular systems, PKCδ was coimmunoprecipitated with Src and members of the Src family (42, 47, 58). PKCδ and c-Abl associate in response to oxidative stress (H2O2), and each of these two proteins affects the activity of each other (49).

Studies of PKCδ using serum-starved cells indicate that serum can promote increases in the phosphorylation of Thr-505 and Ser-662. Under these conditions, these sites were crucial for PKCδ activity, and the phosphorylation of these sites were dependent on pathways involving PtdIns 3-kinase and mTOR (39, 40). In our experiments using freshly isolated parotid cells, these sites were constitutively phosphorylated and were not changed by exposure to PtdIns 3-kinase and mTOR inhibitors (Fig. 8). Presumably, cellular serine/threonine phosphatases were ineffective in dephosphorylating Thr-505 and Ser-662 during the relatively brief (30-min) exposure of cells to these inhibitors. This may not be surprising, because cultured cells must be serum starved (23) and maintained in suspension (38,40) to reduce the phosphorylation of these sites before stimulation by serum. Phosphorylations of the basal serine and threonine phosphorylation sites were also not affected by secretory stimuli (carbachol) and other PKC-activating agents (PMA). These PKCδ sites had already been phosphorylated in the salivary cells, and tyrosine phosphorylation occurred subsequent to threonine and serine phosphorylation. The basal phosphorylation status of the cells did not allow us examine alterations in tyrosine phosphorylation under conditions in which the enzyme was not phosphorylated on Thr-505 and Ser-662, and, therefore, we cannot conclude that tyrosine phosphorylation and its effects on PKCδ activity absolutely require the prior phosphorylation of these threonine and serine sites. It also should be noted that a contrasting argument has been made that Glu-500, an acidic residue in the activation loop, rather than phosphorylated Thr-505, is required for the catalytic activity of PKCδ (48).

There have been contrasting studies concerning the contribution of tyrosine phosphorylation to the activity of PKCδ. Because there are multiple tyrosine residues on PKCδ, phosphorylation of the different residues could have quite different effects on the absolute activity of PKCδ and/or the biological activity of PKCδ. Evidence of the latter was recently suggested (22). Thus our findings may be restricted to activation of PKCδ by PLC-linked receptor activation, phorbol ester, and pervanadate treatment of these cells. In CHO-K1 cells, PKCδ activity was increased by activation of a G protein-coupled ATP-binding receptor by PMA and by H2O2 (37). The degree of maximal activation by these stimuli was H2O2 ≥ PMA > ATP, and the relative amount of tyrosine phosphorylation also followed this relative order, although no significant degree of phosphorylation was detected for ATP. These results are similar to the relative order of stimuli acting on salivary gland cells: pervanadate ≥ PMA > carbachol.

The exposure of parotid cells to 100 μM pervanadate produced a much larger increase in PKCδ tyrosine phosphorylation than did exposure of the cells to 100 nM PMA (Figs. 1 C and 4 A), yet both agents produced about the same degree of increase in PKCδ activity (Fig. 4 B). This suggests that multiple tyrosine residues on PKCδ can be phosphorylated but that not all of the phosphorylated tyrosine residues contribute to increases in PKCδ activity. This may be true for increases in tyrosine phosphorylation promoted by PMA and GPCR ligands as well as by pervanadate. On the basis of mutational studies and an analysis of tryptic digests of PKCδ, Blake et al. (2) concluded that Src phosphorylated Tyr-311 but that additional tyrosine sites were subsequently phosphorylated. Thus the relative degree of stimuli-dependent increases in PKCδ activity is not necessarily quantitative with the relative increases in tyrosine phosphorylation, especially when different stimuli are compared.

The tyrosine phosphorylation of PKCδ produced at most an increase in activity to levels four to five times the basal level (Fig.6 B). However, the tyrosine-phosphorylated form of PKCδ, which migrated more slowly through an SDS-PAGE gel (Figs.3 A, 4, A and C, and 6 A), constituted at most one-half of the total PKCδ protein in cells exposed to PMA and pervanadate in both cell types. Thus the tyrosine-dependent increases in PKCδ activity would be at least twice the values as those reported here if all of the PKCδ in the samples were composed of the tyrosine-phosphorylated form. In parotid cells exposed to carbachol, the tyrosine-phosphorylated form of PKCδ made up a much lower fraction of the total PKCδ (46), and this would appear to account for the relatively smaller increase in PKCδ activity promoted by carbachol compared with PMA or pervanadate. Stimuli-dependent increases in activity due to tyrosine phosphorylation were not observed in the presence of DAG and PS, which increased the assayed activity to a much greater level than that produced by tyrosine phosphorylation, but stimuli-dependent increases in activity were observed (albeit to a lesser relative extent) in the presence of PS alone (Fig. 3). Presumably, tyrosine phosphorylation and lipid production both affect PKCδ activity in vivo, and the membrane lipid composition in vivo may not be that which is idealized under in vitro conditions. In vivo, PKCδ may localize at a site that has a lipid composition much different from that which permits the maximal activation of the enzyme in vitro. PKCδ also may have a biological role in a nonmembrane compartment or in a membrane compartment distinct from the plasma membrane. In fact, PKCδ was found to sequentially localize to nuclear and Golgi membranes after it translocated to the plasma membrane in response to PMA and other phorbol ester derivatives (51). In some cells, phorbol ester initiated the translocation of PKCδ to the mitochondria where it was involved in promoting apoptosis (32). The translocation of PKCδ to a new location will increase PKCδ activity at that location. In addition to affecting PKCδ activity, tyrosine phosphorylation can alter the substrate specificity of PKCδ (15) as well as the lipid dependence (19). Thus changes in location and changes in various enzymatic parameters may affect the biological activity of PKCδ.

PKCδ is an important signaling protein that produces various effects on cells (for review, see Ref. 13), including alterations of cell growth (29, 52). In PC-12 cells, PKCδ modulated the upregulation of L-type Ca2+ channels in ethanol-treated cells (11) and also appeared to play a role in nerve growth factor-mediated neurite outgrowth (36). PKCδ was involved in receptor-mediated activation of the Na+-K+-Cl cotransport system in human tracheal epithelial cells (25) and affected the permeability of tight junctions in LLC-PK1epithelial cells (34). We did not directly address the physiological role of PKCδ in rat parotid acinar epithelial cells, but the activation of parotid muscarinic receptors, which increases PKCδ activity, stimulates both fluid secretion as well as a Ca2+-sensitive component of exocytosis of protein secretory granules (56). PKCδ may play a role in these events. PKCδ plays a critical role in etoposide-induced apoptosis in a cultured rat parotid acinar cell line (41). However, this was mediated by the caspase-dependent cleavage of PKCδ to its constitutively activated 40-kDa form, which does not play a role in the increases in PKCδ activity presented in our studies.

In conclusion, these results demonstrate that Src-related protein kinases and tyrosine phosphorylation participate in cellular events initiated by secretagogues acting on freshly isolated salivary acinar cells. DAG production, Src tyrosine kinases, and tyrosine phosphatases play integrated roles in the regulation of PKCδ activity. In both salivary gland epithelial cells and PC-12 cells, two cell types that contain neither overexpressed PKCδ nor overexpressed Src family members, the tyrosine phosphorylation of PKCδ produced a positive effect on PKCδ activity. This property of PKCδ is important in our understanding of the biological function of PKCδ in these and other cells.

Acknowledgments

We thank Yue Zheng for excellent technical assistance.

Footnotes

  • This work was supported in part by National Institute of Dental Research Grant DE-10877 (to S. P. Soltoff).

  • Address for reprint requests and other correspondence: S. P. Soltoff, Division of Signal Transduction, Dept. of Medicine, Beth Israel Deaconess Medical Center, Harvard Institutes of Medicine, Rm. 1025, 330 Brookline Ave., Boston, MA 02215 (E-mail:ssoltoff{at}caregroup.harvard.edu).

  • 1  The exposure of parotid cells to 100 μM pervanadate for 1 min produced a larger increase in PKCδ tyrosine phosphorylation than the effects of the exposure of these cells to PMA (100 nM) or carbachol (10−4 M) for any time between 0 and 15 min.

  • 2  The relatively lower effectiveness of PTP1B treatment in reducing the enhanced PKCδ activity in PC-12 cells compared with salivary cells was due to the lower effectiveness of one particular lot of PTP1B that was used to dephosphorylate PKCδ immunoprecipitated from PC-12 cells.

  • 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