Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Phospholipase D (PLD) is widely distributed in mammalian cells and has been shown to be involved in signal transduction, protein trafficking, and cell proliferation and differentiation (Frohman and Morris, 1999; Liscovitch et al., 2000; Exton, 2002). It is activated by several extracellular signals and catalyzes the hydrolysis of phosphatidylcholine into phosphatidic acid and choline. Activation of PLD by various agents has been shown to involve small G proteins of the Arf and Rho families, protein kinase C (PKC), and phosphatidylinositol 4,5-biphosphate (Frohman and Morris, 1999; Liscovitch et al., 2000; Exton, 2002). Several PLD isoforms have been cloned or purified. PLD1 exhibits a low basal activity; is activated by Arf, Rho, and PKC; and is localized in perinuclear area (Golgi, endoplasmic reticulum, and late endosomes). PLD2 has a high basal activity, is localized in plasma membrane, requires phosphatidylinositol 4,5-biphosphate, and is not activated by Arf, Rho, or PKC (Frohman and Morris, 1999; Liscovitch et al., 2000; Exton, 2002).

Stimulation of ₁-adrenergic receptors (ARs) has been shown to increase PLD activity in rat brain slices (Llahi and Fain, 1992), rat tail arteries (Labelle et al., 1996), and Madin-Darby canine kidney cells (Balboa and Insel, 1998). In rat-1 cells expressing different subtypes of ₁-AR (A, B, and D), it has been shown that, although all these receptor subtypes are coupled to PLD activation, _1A is more effective than _1B- or _1D-AR (Ruan et al., 1998). Although stimulation of ₁-AR promotes arachidonic acid (AA) release via activation of phospholipase A₂ in many tissues (Muthalif et al., 1996), AA may also be released through PLD activation. In rat-1 cells expressing _1A-AR, phenylephrine (PHE) promotes AA release via PLD but not by phospholipase A₂ (Ruan et al., 1998).

The involvement of PKC in PLD regulation has also been documented both in vivo and in vitro (Kiss, 1996; Zhang et al., 1999; Exton, 2002). PKC isoforms have been classified on the basis of their protein sequences and biochemical properties (Mellor and Parker, 1998). The classical PKC isoforms (, 1, 2, and ) are activated by phosphatidylserine and DAG or phorbol esters in a calcium-dependent manner. The novel PKC isoforms (, , , and ) are activated by DAG or phorbol esters in the presence of phosphatidylserine and in the absence of calcium. The atypical PKC isoforms (/ and ) are both calcium- and DAG-independent. Finally, a new family of protein kinases, the PKC-related kinases (PRK or PKN), share a high homology with the catalytic domain of PKCs and, like atypical PKCs, are not regulated by calcium or DAG (Amano et al., 1996) and have been shown to regulate PLD1 activity in vitro (Oishi et al., 2001).

Although only classical PKC subtypes have been implicated in PLD activation in vitro (Ohguchi et al., 1996) or in cells overexpressing classical PKC (Mukherjee et al., 1996), there are reports indicating receptor-mediated PLD activation that is independent of PKC (Yang et al., 1998; Muthalif et al., 2000). PLD activation by classical PKCs does not involve a phosphorylation mechanism in vitro (Singer et al., 1996). It is currently unclear whether the noncatalytic mechanism by which PKC and  activate PLD1 in vitro accounts for PKC-dependent increases in PLD activity in intact cells (Frohman and Morris, 1999; Zhang et al., 1999; Exton, 2002). There is also a discrepancy among results obtained with PKC inhibitors (Frohman and Morris, 1999; Exton, 2002) and the absence of an ATP requirement for PLD activation by PKC in vitro (Singer et al., 1996). Therefore, the involvement of PKC and the underlying mechanism of PLD activation in intact cells are currently unclear. The present study was conducted to investigate in a systematic manner the possible contribution of PKC isoforms and the PKC-related kinase PKN to PLD activation elicited by PHE in rat-1 cells expressing the _1A-AR subtype. Our study demonstrates that PKC, , , and  do not mediate PLD activation caused by _1A-AR stimulation, despite the inhibitory effect of two widely used PKC inhibitors, bisindolylmaleimide BIM (I) and Ro 31-8220 (Toullec et al., 1991; Davies et al., 2000), on PLD activity. Moreover, our study provides evidence for the involvement of the PKC-related kinase family in PLD activation and a nonspecific effect of BIM I and Ro 31-8220 on calcium entry that lead to the inhibition of PLD activation, independent of their inhibition of PKC activity.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. Phenylephrine was obtained from Sigma-Aldrich (St. Louis, MO); BIM I and V, Ro 31-8220, Gö 6976, Gö 6983, calphostin, myristoylated PKC inhibitor 19-27 [mPKC(19-27)], and PMA were from Calbiochem (San Diego, CA); [³H]oleic acid (50 Ci/mmol) was from American Radiolabeled Chemicals (St. Louis, MO); [³H]arachidonic acid (100 Ci/mmol) was from PerkinElmer Life Sciences (Boston, MA); and [-³²P]ATP (3000 Ci/mmol) was from Amersham Biosciences (Piscataway, NJ).

Cell Culture. Rat-1 fibroblasts were stably transfected with _1A-adrenergic receptors (a kind gift of Dr. Allen from Dr. Lefkowitz's laboratory; Duke University, Durham, NC) as described previously (Allen et al., 1991). Cells were maintained under 5% CO₂ at 37°C in Dulbecco's modified Eagle's medium (DMEM) containing 50 units of penicillin, 50 µg/ml streptomycin, and 10% fetal bovine serum. Selection and maintenance of stably transfected cells were carried out with 400 µg/ml G418 (Invitrogen, Carlsbad, CA).

AA Release and PLD Activity. AA release and PLD activity were measured according to methods used in our laboratory and previously described for rat-1 fibroblast (Ruan et al., 1998).

Transfection of Cells with PKC Oligonucleotides and PKN Plasmids. Rat-1 cells were transiently transfected in serum-free DMEM with 10 µM antisense (AS) or scrambled (SC) oligonucleotides for 2, 4, 6, or 9 days with new pulses every 48 h. AS and SC oligonucleotides were synthesized at the Molecular Resource Center Synthesis Facility (The University of Tennessee, Memphis, TN). These oligonucleotides have been used by other investigators to specifically down-regulate PKC (Herbert et al., 1996), PKC (Pessino et al., 1995; modified for rat isoform), PKC (Traub et al., 1997; modified for rat isoform), and PKC (Chen et al., 1998), or were designed against the translation initiation site (nucleotides 136-155) of PKN (Mukai and Ono, 1994). The sequences used were as follows: PKC AS, 5'-CAGCCATGGTTCCCCCCAAC-3'; PKC SC, 5'CCAGTCACTCGCACCATCGC-3'; PKC AS, 5'-ACGGTGCCATGATGGA-3'; PKC SC, CGAGTAGTTACAGCGG-3'; PKC AS, 5'-CATGAGAGCAGATCTGACCT-3'; PKC SC, 5'-AACGCATAACTCG CTTGAGG-3'; PKC AS, 5'-CTGCTGCCGGAGCCCCGA-3'; PKC SC, CCGAGAGCCGCCGTCGTC; PKN AS, 5'-AGGTTCACTCTGCA CGGCGT-3'; and PKN SC, 5'-AGTATTCCGGTCGAGCCCTG-3'. After treatment, cells were lysed and 50 µg of protein from cell lysates was subjected to SDS-PAGE followed by Western blot analysis, or the cells were labeled with [³H]oleic acid for determination of PLD activity. Rat-1 cells were transiently transfected with pMhPKN7 (wild-type PKN) or pMhPKN PK-2 (kinase-deficient PKN) (gift from Dr. Y. Ono; Kobe University, Kobe, Japan) vectors using FuGENE 6 (Roche Applied Science, Mannheim, Germany). pZeoSV2/lacZ vector was cotransfected as a control, and -galactosidase activity was determined on cell lysates (Invitrogen).

Western Blot Analysis. Antibodies to PKC, PKC1, PKC2, PKC, PKC, PKC, PKC, PKC, PKC, and PKN were from Santa Cruz Biotechnology (Santa Cruz, CA). Total cell lysate was prepared and protein concentration was determined. Proteins in 50 µg of lysates in Laemmli buffer were separated by SDS-PAGE and blotted onto nitrocellulose membranes. Blots were incubated with the indicated antibodies at dilutions recommended by the manufacturer. The blots were visualized with the enhanced chemiluminescence detection system (Amersham Biosciences). For experiments involving cell fractionation, cells were scraped in fractionation buffer (10 mM Tris-HCl, pH 7.5, 10 mM NaCl, 0.5 mM EDTA, 0.5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, 10 µg/ml aprotinin, 1 mM sodium orthovanadate, and 10 µg/ml trypsin inhibitor), subjected to 25 passes in a Potter-type Teflon-on-glass homogenizer with a large fitting and centrifuged at 700g for 5 min. The supernatant was centrifuged at 100,000g for 30 min, and the cytosolic fraction was stored at 80°C. The particulate fraction was resuspended in 1% Triton fractionation buffer, sonicated, and stored at 80°C.

PKC Assay. Cells for PKC activity measurement were grown in 150-mm-diameter tissue culture dishes and arrested with DMEM containing 0.1% fetal bovine serum at least for 20 h. Cells were treated with inhibitors and agonists as indicated in the figure legends. For PKC activity measurement in whole cell lysate, cells from each dish were lysed with 0.5 ml of extraction buffer (20 mM Tris-HCl, pH 7.5, 0.5 mM EDTA, 0.5 mM EGTA, 0.5% Triton X-100, 25 µg of aprotinin, and 25 µg/ml leupeptin). Activation of PKC was determined using a classical PKC Assay System (Invitrogen) with [-³²P]ATP (3000 Ci/mmol) according to the manufacturer's instructions. For PKC activity measurement in cell fractions, cells from two dishes were lysed with 1 ml of fractionation buffer with 300 mM sucrose and 10 mM -mercaptoethanol and treated as described under "Western Blot Analysis". The concentration of protein was determined by Bradford assay (Bio-Rad, Hercules, CA). The particulate fraction was resuspended by sonication and centrifuged at 2000g for 5 min. PKC activity was determined in aliquots of cytosolic and membrane fractions.

Calcium Measurement. Cells were plated on coverslips and grown overnight in DMEM containing 10% serum. Serum-deprived cells were exposed to 5 µM Fura2/acetoxymethyl ester (Molecular Probes, Eugene, OR) in Krebs-Ringer buffer containing 0.05% bovine serum albumin for 30 min, washed, and further incubated 30 min without Fura2/acetoxymethyl ester. Intracellular calcium concentrations were determined from the ratio of emissions measured at 510 nm after excitation at 340 and 380 nm using a luminescence spectrometer (PerkinElmer LS50B). All coverslips were treated with 5 µM ionomycin as a positive control after the experimental treatment.

Data Analysis. The results are expressed as mean ± S.E. The data were analyzed by one-way analysis of variance. An unpaired Student's t test was applied to determine the difference between two groups and Newman-Keuls a posteriori test to determine the difference between multiple groups. A value of P  0.05 was considered significant. The increase in fractional [³H]AA release elicited by agonists was expressed as percentage above basal level. PLD activity was expressed as the fold increase from basal. The protein level for the antisense experiments was estimated by densitometric analysis of the Western blots and performed on three different blots using NIH Image software, and expressed as a percentage (mean ± S.D.) of the control, arbitrarily chosen as 100%.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of PKC Inhibitors on PHE- and PMA-Induced Increase in PLD Activity and AA Release. BIM I, a highly selective PKC inhibitor structurally related to staurosporine, acts as a competitive inhibitor for the ATP-binding site (Toullec et al., 1991) and inhibits DAG-dependent PKC isoforms at submicromolar concentrations (Martiny-Baron et al., 1993). In our study, BIM I, at 0.5 µM, inhibited PHE- and attenuated PMA-induced increases in PLD activity (Fig. 1A). BIM V, an inactive analog of BIM I, did not decrease PLD activity (data not shown). PHE-induced AA release was attenuated by 77% in the presence of 0.5 µM BIM I, whereas AA release elicited by PMA was almost abolished (>90% inhibition) (Fig. 1B). Ro 31-8220 is also structurally related to staurosporine but at submicromolar concentrations inhibits atypical PKC isoforms, in addition to DAG-dependent isoforms (Standaert et al., 1997). PLD activity induced by PHE and PMA was attenuated by 5 µM Ro 31-8220 by 83 and 71%, respectively (Fig. 1A), thereby confirming results obtained with BIM I. Ro 31-8220 at 10 µM was able to totally inhibit PHE-induced PLD activity (data not shown). Ro 31-8220 at 5 µM inhibited PHE-induced AA release by 81% and blocked AA release elicited by PMA (Fig. 1B). Gö 6983, another broad-range PKC inhibitor (Gschwendt et al., 1996), also blocked PHE-induced PLD activation [2 µM PHE = 4.11 ± 0.07-fold increase versus 20 µM Gö 6983 + PHE = 1.10 ± 0.08-fold increase, basal = 4.15 ± 0.12 (× 1000 × [³H]PEt/[³H]total lipids].

View larger version (48K): [in this window] [in a new window]   Fig. 1.   Effect of PKC inhibitors BIM I, Ro 31-8220, Gö 6976, and mPKC(19-27) on PLD activity (A) and AA release (B) elicited by PHE or PMA. For PLD activity, cells were labeled with [3H]oleic acid for 18 h in serum-free DMEM and preincubated with 0.5 µM BIM, 5 µM Ro 31-8220, and 10 µM Gö 6976 for 30 min or 10 µM mPKC(19-27) for 1 h. Cells were then incubated with DMEM containing 2 µM PHE or 1 µM PMA in the presence of 200 mM ethanol for measurement of [3H]phosphatidylethanol formation. For AA release, cells were incubated overnight with [3H]AA and preincubated with PKC inhibitors then exposed to PHE or PMA for 15 min and AA release was determined as described under Materials and Methods. Basal values of [3H]AA and [3H]phosphatidylethanol are presented beneath each group. Data are expressed as mean ± S.E. of three experiments performed in sextuplate for AA release and three to four independent experiments performed in duplicate for PLD assay. *, value significantly different from that obtained in the presence of vehicle (PHE or PMA).

Effect of PHE, PMA, and PKC Inhibitors on PKC Activity. PMA increased PKC activity by 70%, whereas PHE did not affect significantly basal PKC activity (Fig. 2); BIM I reduced PKC activity in the presence of PHE and PMA by 80 and 95%, respectively. Gö 6976 also reduced PKC activity in the presence of PHE and PMA by 80 and 85%, respectively. Gö 6976 at 10 µM inhibited PKC to a similar degree as at 1 µM. Basal PKC activity was also reduced by both BIM I and Gö 6976. Ro 31-8220 and mPKC(19-27) inhibited PKC activity to the same extent as BIM I and Gö 6976 (data not shown). PKC activity was also determined in cytosolic and membrane cell fractions. PHE did not alter the distribution of PKC activity between membrane and cytosol, whereas PMA-induced PKC activity was translocated from the cytosol to the membrane fraction (Table 1). BIM I inhibited basal membrane, but not cytosolic, PKC activity. Membrane translocation of PKC activity induced by PMA was blocked by BIM I. These results indicate that PMA, but not PHE, stimulates classical PKCs.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Effect of PHE and PMA on PKC activity. Cells were arrested overnight and preincubated with 0.5 µM BIM I or 1 µM Gö 6976 for 30 min, and then exposed to 2 µM PHE or 1 µM PMA for 15 min. The assay of PKC activity is described under Materials and Methods. Basal PKC activity in resting cells was 13.27 ± 0.44 pmol/min/mg protein. Relative PKC activity is expressed as the ratio of values over basal and taken as zero. Data are expressed as mean ± S.E. of four to six different experiments for PKC assay performed in triplicate. *, value significantly different from the basal. , value significantly different from that obtained in the absence of inhibitor (vehicle).

View this table: [in this window] [in a new window]   TABLE 1 Distribution of PKC activity in cytosol and membrane fractions in rat-1 cells expressing 1A-adrenergic receptors Cell growth was arrested overnight in serum-free DMEM. Cells were treated with PHE or PMA for 15 min. BIM I was preincubated 30 min before adding agonists. Cells were fractionated at 100,000g, and PKC activity was determined in particulate and cytosolic fractions from two experiments performed in triplicate. Data are expressed in picomoles per minute per milligram of protein.

Effect of PKC Isoform-Specific AS Oligonucleotides on PHE-Induced PLD Activation. To rule out nonspecific effect of PKC inhibitors and to determine the involvement of specific PKC isoforms in PHE-induced PLD activation in rat-1 cells expressing _1A-ARs, we investigated the distribution of PKC isoforms in response to PHE and PMA (15-min exposure) and the effect of PKC isoform-specific AS and SC oligonucleotides on PKC protein levels and on PHE-induced PLD activation.

Classical PKC Isoforms. PKC but not PKC 1, 2, or  was detected in rat-1 cells. PKC was translocated from the cytosol to the particulate fraction in response to PMA but not to PHE (Fig. 3A). These results are in agreement with the data shown in Table 1, indicating that PKC activity in rat-1 cells is derived mainly from classical PKCs. To determine the possible contribution of PKC to PHE-induced PLD activation, cells were treated with specific AS oligonucleotides directed against PKC mRNA, thereby inhibiting the neosynthesis of PKC. The specificity of the AS was measured after 2, 4, and 6 days of treatment (Fig. 3B). Densitometric analysis revealed that PKC protein level was reduced by 68% after 4 days and 80% after 6 days of PKC AS treatment. A scrambled oligonucleotide used as control did not affect PKC protein level after 6 days of treatment (Fig. 7C). The PHE-induced increase in PLD activity was not reduced by PKC AS or SC oligonucleotide, both after 2 and 6 days of treatment (Fig. 3C). In addition, PKC depletion did not promote any up-regulation of the expression of PKC1 or 2. Efficiency of PKC depletion by the antisense at 6 days was confirmed using PMA as a PLD activator instead of PHE [1 µM PMA = 2.40 ± 0.10-fold increase, PKC AS + PMA = 1.40 ± 0.05-fold increase, PKC SC + PMA = 2.39 ± 0.13-fold increase, basal = 7.39 ± 0.26 (× 1000 × [³H]PEt/[³H]total lipids]. These data indicate that PKC does not mediate PHE-induced PLD activation. It should be noted that control PLD activity was reduced after 6 days, independently of any treatment, probably due to cell growth arrest.

View larger version (43K): [in this window] [in a new window]   Fig. 3.   Effect of PHE and PMA on PKC translocation and of PKC depletion on PLD activity. A, PKC is translocated to the particulate fraction by PMA but not by PHE. Cells arrested overnight were incubated with 2 µM PHE or 1 µM PMA for 15 min. Cytosolic and particulate fractions were prepared as described under Materials and Methods. Cell fractions (50 µg) in Laemmli buffer were loaded and proteins were separated by 10% SDS-PAGE and blotted on nitrocellulose membranes. Membranes were blocked overnight and then incubated 1 h with a PKC polyclonal antibody (1/200). After extensive washing and incubation with secondary antibody, bands corresponding to PKC (82 kDa) were revealed using enhanced chemiluminescence. B, PKC protein levels are reduced by specific AS oligonucleotide. The protein level estimated by densitometric analysis is shown above the blot and expressed as a percentage of the control without treatment. PKC AS sequence and treatment procedures, are described under Materials and Methods. All blots shown are representative of two to three different experiments. C, PLD activity is not reduced with PKC AS treatment. Cells were incubated in DMEM containing 10 µM PKC AS or SC oligonucleotides (ODN), and PHE-induced PLD activity was measured after 2 or 6 days of treatment as described previously. Basal PLD activity, shown beneath each group and expressed as 103 × [3H]PEt/[3H]total lipids, was not affected by AS treatment. Data are expressed as mean ± S.E. of three independent experiments performed in duplicate for PLD assay. *, value significantly different from the basal.

Novel PKC and Atypical PKC Isoforms. The novel PKC isoforms , , and  but not  were detected in rat-1 cells. PKC  (Fig. 4A) and  (Fig. 5A) but not  (Fig. 6A) translocated to the particulate fraction upon stimulation by PMA. PHE did not stimulate PKC or  translocation, whereas PKC was slightly translocated from the cytosolic to the particulate fraction. PKC and to a lesser extent PKC, but not PKC, were also localized in the particulate fraction in untreated cells. PKC protein level was reduced by 78% after 6 days of treatment with a specific PKC AS oligonucleotide (Fig. 4B) but not with PKC SC (Fig. 7C). The PHE-induced increase in PLD activity was not reduced by PKC AS at 2 or 6 days of treatment (Fig. 4C). These data indicate that PKC is not involved in PLD activation elicited by _1A-AR stimulation. PKC protein level was reduced by 78% after 6 days of treatment with a specific PKC AS oligonucleotide (Fig. 5B). PLD activity was not reduced after PKC AS treatment at 2 or 6 days (Fig. 5C). Because PKC has been proposed as a candidate for participation in PLD activation (Pfeilschifter and Huwiler, 1993), we performed an additional experiment using a PKC translocation inhibitor peptide (Johnson et al., 1996). The peptide was incorporated into subconfluent cells by osmotic loading. PLD activity was unchanged in cells treated with PKC inhibitor (Fig. 5C), confirming that PKC is not involved in PLD activation by PHE in rat-1 cells. PKC AS did not significantly reduce the amount of corresponding protein after 6 days of treatment. Extension of treatment with AS for 9 days caused a reduction of 82% in PKC protein levels (Fig. 6B). However, the PHE-induced increase in PLD activity was not reduced by the specific PKC AS even after 9 days of treatment (Fig. 6C). Levels of PKC and PKC were not affected after similar treatment with PKC or PKC SC oligonucleotides (Fig. 7C). These data suggest that PKC is not involved in PHE-induced PLD activation. The protein levels of PKC were also measured after treatment with PKC or PKC AS, as a control for AS cross-reactivity (Fig. 7). The blockade of PKC or PKC synthesis did not alter PKC protein level.

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Effect of PHE and PMA on PKC translocation and of PKC depletion on PLD activity. A, PKC is translocated to the particulate fraction by PMA but not by PHE. PKC is already localized in the particulate fraction in untreated cells. The protocol is similar to Fig. 3 with a PKC polyclonal antibody (1/200). B, PKC protein levels are reduced by specific AS oligonucleotide. The decrease in protein level is indicated as a percentage of the control without treatment. PKC AS sequence and treatment procedures are described under Materials and Methods. All blots shown are representative of two to three different experiments. C, PLD activity is not reduced with PKC AS treatment. Cells were incubated in DMEM containing 10 µM PKC AS oligonucleotides and PLD activity (vehicle or stimulated with PHE) was measured after 2 or 6 days of treatment as described previously. Basal PLD activity, shown beneath each group and expressed as 103 × [3H]PEt/[3H]total lipids, was not affected by AS treatment. Data are expressed as mean ± S.E. of three independent experiments performed in duplicate for PLD assay. *, value significantly different from the basal.

View larger version (42K): [in this window] [in a new window]   Fig. 5.   Effect of PHE and PMA on PKC translocation and of PKC depletion on PLD activity. A, PKC is translocated to the particulate fraction by PMA but not by PHE. A small part of PKC is already localized in the particulate fraction in untreated cells. The protocol is similar to Fig. 3 with a PKC polyclonal antibody (1/200). B, PKC protein levels are reduced by specific antisense oligonucleotide. The decrease in protein level is indicated as a percentage of the control without treatment. PKC AS sequence and treatment procedures are described under Materials and Methods. All blots shown are representative of three to four different experiments. C, PLD activity is not reduced with PKC AS treatment. Cells were incubated in DMEM containing 10 µM PKC AS oligonucleotides, and PLD activity (vehicle or stimulated with PHE) was measured after 2 or 6 days of treatment as described previously. Basal PLD activity, shown beneath each group and expressed as 103 × [3H]PEt/[3H]total lipids, was not affected by AS treatment. PHE-induced PLD activity is also not altered by a specific PKC translocation inhibitor peptide (500 µg/ml) incorporated by osmotic loading. All procedures are described in the Fig. 3 legend and under Materials and Methods. Data are expressed as mean ± S.E. of three independent experiments performed in duplicate for PLD assay. *, value significantly different from the basal.

View larger version (40K): [in this window] [in a new window]   Fig. 6.   Effect of PHE and PMA on PKC translocation and of PKC depletion on PLD activity. A, cytosolic PKC localization is not altered by PHE or PMA. The protocol is similar to Fig. 3 with a PKC polyclonal antibody (1/200). B, PKC protein levels are reduced by specific antisense oligonucleotide. The decrease in protein level is indicated as a percentage of the control without treatment. PKC AS sequence and treatment procedures are described under Materials and Methods. All blots shown are representative of three different experiments. C, PLD activity is not reduced with PKC AS treatment. Cells were treated for 2, 6, or 9 days with 10 µM PKC AS oligonucleotides. All procedures are described in the Fig. 3 legend and under Materials and Methods. Basal PLD activity, shown beneath each group and expressed as 103 × [3H]PEt/[3H]total lipids, was not affected by AS treatment. Data are expressed as mean ± S.E. of three independent experiments performed in duplicate for PLD assay. *, value significantly different from the basal.

View larger version (64K): [in this window] [in a new window]   Fig. 7.   Effect of PKC and PKC depletion on PKC protein level. Membranes from experiments corresponding to Fig. 4B (PKC AS) (A) and Fig. 5B (PKC AS) (B) were incubated in a stripping buffer (100 mM 2-mercaptoethanol, 2% sodium dodecyl sulfate, and 62.5 mM Tris-HCl, pH 6.7) for 30 min at 50°C, washed twice with Tris-buffered saline-Tween 20, and exposed to a film to ensure removal of antibodies and reprobed with a PKC antibody. Blots shown are representative of two different experiments. The same experiment conducted with a membrane from Fig. 3B (PKC AS) and reprobed with PKC1 or PKC1 did not detect any bands. C, PKC protein levels are not altered by SC oligonucleotides. PKC, PKC, PKC SC (6-day exposure), and PKC SC (9-day exposure) oligonucleotide sequences and treatment procedures are described under Materials and Methods. Blots shown are representative of two to three different experiments.

Role of PKN. A possible candidate for mediating PLD activation is protein kinase N (PKN also called PRK1) because 1) it is downstream of RhoA (Amano et al., 1996) and this small G protein has been shown to mediate agonist-dependent activation of PLD in many cell types (Du et al., 2000; Exton, 2002); 2) its structure is related to PKC in that their catalytic domain are highly homologous (Mellor and Parker, 1998); and 3) PKC inhibitors selective for the ATP-binding domain also reduced PKN activity (Standaert et al., 1998). In rat-1 cells, an antisense oligonucleotide corresponding to nucleotides 136 to 155 of PKN1 cDNA sequence (Mukai and Ono, 1994) slightly but significantly reduced PLD activity (Fig. 8B). However, it did not reduce PKN protein level. This discrepancy may have multiple causes, such as the design of the antisense, its affinity for PKN, and the half-life of PKN protein. The PKN oligonucleotide is not complementary to other known cDNA sequences according to a Blast search. An alternative approach was transfection with a kinase-deficient form of PKN (Amano et al., 1996). The efficiency of transfection was determined by measuring -galactosidase activity after cotransfection with pZeoSV2/lacZ vector (control = 13 ± 2 -gal unit/well, transfected cells = 206 ± 39 -gal unit/well). Two days after transfection, cells were tested for PLD activity. PHE-induced PLD activity was reduced by 50% in cells transfected with kinase-deficient PKN (Fig. 8A). PMA-induced PLD activation was also reduced with kinase-deficient PKN. These data suggest a function for PKN in mediating ₁-AR-induced PLD activation in rat-1 cells.

View larger version (58K): [in this window] [in a new window]   Fig. 8.   PLD activity is decreased by transient transfection with a kinase-deficient PKN. A, cells were arrested overnight and transfected with 2 µg/well of pMhPKN7 expressing wild-type PKN (wt PKN) or with pMhPKN PK-2 plasmid expressing a kinase-deficient PKN (KD PKN). Cells were allowed to express PKN for 48 h before measuring PLD activity, as described under Materials and Methods. Data are expressed as mean ± S.E. of three independent experiments performed in duplicate for PLD assay. *, value significantly different from vehicle (basal, PHE, or PMA). B, cells were incubated in DMEM containing 10 µM PKN AS or SC oligonucleotides for 3, 6, or 9 days of treatment and then challenged with 2 µM PHE for 15 min before measuring PLD activity. Basal PLD activity was not affected by AS treatment. Data are expressed as mean ± S.E. of three independent experiments performed in duplicate for PLD assay. *, value significantly different from PHE without oligonucleotide treatment.

Effect of PKC Inhibitors on Calcium Signal. Agonist-mediated PLD activation is dependent on the presence of extracellular calcium (Frohman and Morris, 1999; Liscovitch et al., 2000; Exton, 2002). Removal or chelation of extracellular calcium abolished PHE-mediated increase in PLD activity in rat-1 cells [2 µM PHE = 3.92 ± 0.23-fold increase versus 0.5 mM EGTA + PHE = 1.26 ± 0.47-fold increase, basal = 8.29 ± 2.29 (× 1000 × [³H]PEt/[³H]total lipids]. Therefore, the potential effect of PKC inhibitors on [Ca²⁺]_i was studied. Stimulation of _1A-AR with PHE induces a rapid rise in [Ca²⁺]_i, mainly caused by release of Ca²⁺ from intracellular stores, followed by a slowly declining phase due to capacitative calcium entry across the plasma membrane (Fig. 9, a and b). Pretreatment with mPKC(19-27) and Gö 6976 slightly decreased the amplitude of the initial calcium peak, but delayed the second phase of Ca²⁺ entry required for refilling intracellular stores (Fig. 9, c and d). However, these inhibitors did not block [Ca²⁺]_i entry. On the other hand, BIM I and Ro 31-8220 blocked the calcium response to PHE (Fig. 9, e and f).

View larger version (24K): [in this window] [in a new window]   Fig. 9.   PKC inhibitors alter calcium signal. Fura2-loaded rat-1 cells were treated with 2 µM PHE (a) or PHE without or with extracellular calcium (b). Cells were also treated with the following PKC inhibitors: 10 µM mPKC(19-27) for 1 h (c), 10 µM Gö 6976 for 30 min (d), 0.5 µM BIM I for 30 min (e), or 5 µM Ro 31-8220 for 30 min (f), and then challenged with 2 µM PHE (horizontal bar) for 80 s. Data are expressed as the ratio of the absorbance at 340 and 380 nm divided by Imax (maximal stimulation found with 5 µM ionomycin). The data are representative of four independent experiments and the results were reproducible.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study demonstrates that ₁-AR-stimulated PLD activation is independent of PKC, , , and  isoforms and is mediated by a PKC-related kinase, PKN. Moreover, BIM I and Ro 31-8220 decreased PLD activity, independent of PKC isoforms, and inhibited ₁-AR stimulation of [Ca²⁺]_i increase. The evidence is based on the following findings: 1) BIM I and Ro 31-8220, but not Gö 6976 or myristoylated PKC(19-27) peptide inhibitor, decreased PLD activity, a strong suggestion of a critical difference among PKC inhibitors; 2) PKC activity and translocation are not increased by 2 µM PHE, the optimal concentration for maximal PLD activation; 3) isoform-specific AS oligonucleotides depleted PKC isoforms but did not decrease PLD activity; 4) kinase-deficient PKN decreased PLD activation elicited by ₁-AR stimulation; and 5) PHE-mediated increase in [Ca²⁺]_i was abolished by BIM I and Ro 31-8220, but not by Gö 6976 or mPKC(19-27).

Ten different PKC isoforms have been cloned in mammalian tissues (Mellor and Parker, 1998). Rat-1 cells express only PKC isoforms , , , , and  but not 1 and 2, , , or . PKC and 1/2 are the only isoforms that have been reported to activate PLD in vitro (Mukherjee et al., 1996; Ohguchi et al., 1996). Therefore, the absence of effect of Gö 6976, mPKC(19-27), and PKC AS on PLD activity, the lack of effect of PKC depletion on PKC1/2 and PKC or PKC depletion on PKC level, and the inability of PHE to activate classical PKCs rules out PKC involvement in PLD activation. Although it is well established that PKC or  is able to bind and activate purified recombinant PLD1 in vitro (Frohman and Morris, 1999; Liscovitch et al., 2000), and that a PKC-nonresponsive PLD1 mutant loses most of its sensitivity to m3 muscarinic acetylcholine receptor stimulation in HEK293 cells (Zhang et al., 1999), it does not exclude the possibility that other proteins, related or not to PKCs, may substitute for PKC or  in other cell types or with different G protein-coupled receptor stimulation. In addition, RhoA is able to stimulate PLD1 independently of PKC in vivo and in vitro (Frohman and Morris, 1999; Du et al., 2000; Exton, 2002). Whether RhoA is involved in PHE-stimulated PLD activation in rat-1 cells is not known.

In the search for a PKC isoform involved in PLD activation, our antisense study ruled out involvement of the isoforms , , or , in addition to PKC. Our results do not exclude a role for PKC in mediating PLD activation. PKD (PKCµ in mice) does not seem to be involved in PHE-induced PLD activation because Gö 6976, which also inhibits PKCµ in addition to classical PKC (Gschwendt et al., 1996), did not alter the PHE-induced increase in PLD activity. From these results, it follows that the PKC isoforms expressed in rat-1 cells are not involved in _1A-AR-stimulated PLD activation and that the catalytic inhibitors BIM I and Ro 31-8220 seem to reduce PHE-induced PLD activity by a mechanism independent of PKC or its activity.

PKN (PRK1) is related to the PKC family, activated by GTP-bound RhoA and proteolysis (Amano et al., 1996) and sensitive to Ro 31-8220 inhibition (Standaert et al., 1998), rendering it attractive as a possible target for BIM I or Ro 31-8220. Transfection of rat-1 cells with a kinase-deficient mutant of PKN decreased PLD activity. An antisense oligonucleotide directed against the PKN translation initiation site caused a small but significant decrease in PLD activity. Therefore, it is possible that PKN is involved in PLD activation in our cell line model, although we did not detect a significant decrease in PKN protein level after AS treatment. It has been recently reported that PKN regulates PLD1 through direct interaction, independent of ATP (Oishi et al., 2001). We do not exclude the possibility that a member of the PRK family directly activates PLD in rat-1 cells. However, because the catalytic activity of PKN is required for PLD activation in our model and PKN stimulates PLD in an ATP-independent manner in vitro (Oishi et al., 2001), a distal function seems more likely. Another target of RhoA, -kinase, has recently been shown to be involved in the stimulation of PLD by the m₃ muscarinic receptor (Schmidt et al., 1999), suggesting that RhoA-activated kinases may mediate PLD activation.

PKC inhibitors, such as BIM and Ro 31-8220 (Toullec et al., 1991), have been widely used to study the involvement of PKC in PLD activation in intact cells. These inhibitors, acting on the catalytic site of all PKC subtypes with different IC₅₀ values, are well characterized in vitro (Davies et al., 2000; for review, see Hofmann, 1997). However, these compounds are not isoenzyme-specific and have also been found to inhibit a growing number of other protein kinases, in addition to PKC (Davies et al., 2000). For example, BIM I and Ro 31-8220 inhibit PKA, MAPKAP kinase 1, and p70 S6 kinase (Alessi, 1997). In rat-1 cells expressing _1A-AR, extracellular signal-regulated kinases and MAPKAP kinase-1 are not activated by PHE or norepinephrine (Lin et al., 1998). Moreover, MAPKAP 1 and p70 S6 kinase do not function downstream of PKC in activating PLD (Morreale et al., 1997).

The role of calcium in G protein-coupled receptor activation of PLD has been studied (Frohman and Morris, 1999; Parmentier et al., 2001; Exton, 2002). In rat-1 cells, the chelation of extracellular calcium by EGTA inhibited _1A-AR-mediated PLD activation. Because, BIM I and Ro 31-8220 also block PHE-mediated [Ca²⁺]_i and PLD activity, our data suggest that calcium is required for PLD activation in rat-1 cells and that BIM I and Ro 31-8220 alter the calcium signal, without PKC involvement. BIM I has been shown to inhibit [Ca²⁺]_i increase by reducing basal filling of the intracellular Ca²⁺ stores, thereby delaying the capacitative calcium entry in MF-2 cells (Sipma et al., 1996). In addition, calphostin C, an inhibitor of the regulatory domain of PKC, directly blocks Ca²⁺ channels (Hartzell and Rinderknecht, 1996). On the other hand, calcium is not directly required for PLD1 activation in vitro (Frohman and Morris, 1999; Exton, 2002). These observations together with our results argue for an early requirement for calcium in the mediation of PHE-induced PLD activation in rat-1 cells. The role of calcium in the _1A-AR signaling pathway in PLD activation is currently under investigation.

In conclusion, this is the first systematic study demonstrating that PKC-related kinase PKN, but not PKC, , , and  isoforms, is involved in G protein-coupled receptor-stimulated PLD activation. Moreover, it demonstrates a novel nonspecific inhibitory effect of BIM I and Ro 31-8220 on calcium signaling. In light of these findings, previous studies using catalytic inhibitors of PKCs as specific tools for determining their contribution to various neurohumoral agents and growth factors may need to be reevaluated.