INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

PROSTAGLANDINS AND OTHER eicosanoid metabolites play an important role in many physiological processes including the modulation of salt and water homeostasis by the collecting duct system of the kidney (3, 12). Availability of the precursor fatty acid, arachidonic acid (AA), as determined by its cleavage from the sn-2 position of phospholipids, represents the rate-limiting step in the synthesis of these mediators. Recent evidence suggests that depending on the agonist or cell type examined, various phospholipase A₂ (PLA₂) forms may be involved in AA hydrolysis (23). Of particular interest to this process is cytosolic PLA₂ (cPLA₂), an acyl hydrolase that responds to extracellular ligand-receptor interaction, requires submicromolar Ca²⁺ concentrations for its translocation to membrane substrate, and displays enhanced catalytic activity following its phosphorylation on serine residues (18, 21, 22). The signaling events responsible for the phosphorylation and resultant activation of this enzyme have not been definitively established, yet numerous studies support a role for protein kinase C (PKC) in this process (7, 11, 22, 30). Although PKC may directly phosphorylate cPLA₂ (21, 24), it often appears that PKC may be acting secondarily through the activation of another kinase, namely mitogen-activated protein kinase (MAPK) (15, 21, 29, 35). In a series of experiments with CHO cells overexpressing cPLA₂, Lin et al. (21) showed that MAPK recognizes a specific consensus sequence within cPLA₂ and that phosphorylation of serine-505 within this domain converts cPLA₂ to its active form. In contrast with those studies implicating PKC and MAPK in the activation of cPLA₂, there are additional models that deny the involvement of either one or both of these enzymes (1, 2, 13, 27). It therefore appears that the regulation of cPLA₂ activity may be both agonist specific and cell specific.

PKC is a ubiquitously expressed serine/threonine kinase known to include a family of isozymes that differ in structure, cofactor requirement, and substrate specificity (14). The PKC family can be divided into the following three groups: the Ca²⁺/diacylglycerol (DAG)-sensitive conventional PKCs (cPKC; PKC, PKC₁, PKC₂, and PKC); the Ca²⁺-insensitive/DAG-sensitive novel PKCs (nPKC; PKC, PKC, PKC, PKC, and PKCµ); and the atypical PKCs (aPKC; PKC and PKC), which are Ca²⁺/DAG insensitive (25). Current interest in the PKC isozymes has been highlighted by recent evidence suggesting that the heterogeneity within this family of enzymes is maintained because of their separate and unique functions in the cell (14, 25). With respect to cPLA₂ regulation, Godson et al. (11) have shown by antisense techniques, that PKC is the isozyme specifically implicated in the release of AA stimulated by the PKC activator, phorbol 12-myristate 13-acetate (PMA).

In the present study, we have used the nonapeptide, bradykinin (BK), a known stimulator of prostaglandin production in the kidney (3), to study the mechanism of cPLA₂ activation in a rabbit cortical collecting duct (RCCD) cell line. Previous results obtained from the study of this cell line within this laboratory have revealed that BK is a potent stimulator of AA release and that this process is mediated through the PKC-dependent serine phosphorylation and activation of cPLA₂ (19). Interestingly, in that study, we found that the BK-induced activation of cPLA₂ was only partially dependent upon PKC, thereby suggesting the involvement of additional activating mechanisms. To establish whether the MAPK cascade was part of the stimulatory pathway, we employed an inhibitor of this cascade, PD-98059, and measured the effect on BK-induced AA release and MAPK activity. In addition, we extended our investigation into the role of PKC by assessing the involvement of individual isozymes in the response to BK. The results of these studies are reported presently and they strongly indicate that PKC and MAPK are both involved in the activation of cPLA₂ downstream of BK-receptor occupation.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Materials. Biodegradable counting scintillant, [5,6,8,9,11, 12,14,15-³H]AA, p42/p44 MAPK enzyme activity assay system, horseradish peroxidase-conjugated donkey anti-rabbit Ig and sheep anti-mouse Ig, enhanced chemiluminescence (ECL) Hyperfilm, Hybond nitrocellulose, and ECL reagents were from Amersham Canada (Oakville, ON, Canada). PMA was obtained from Biomol Research Laboratories (Plymouth, PA). Protein A-agarose was provided by Boehringer-Mannheim (Laval, Quebec, Canada). The agents Ro-31-8220, PD-98059, and 4-phorbol-12,13-didecanoate were supplied by Calbiochem-Novabiochem (La Jolla, CA). Mouse anti-p42-MAPK antibody was bought from Dimension Laboratories (Mississauga, ON, Canada). Phosphatidylcholine (L--1-stearoyl-2-[5,6,8,9,11,12,14,15-³H]arachidonyl) was obtained from DuPont NEN (Mississauga, ON, Canada). Anti-PKC isozyme (, , , , , ) sampler set complete with competing peptides and anti-PKC antibody (used for immunoprecipitations) were bought from GIBCO-BRL (Burlington, ON, Canada). PepTag nonradioactive PKC assay kit was purchased from Promega (Madison, WI). BK and 1-stearoyl-2-arachidonyl-sn-glycerol were purchased from Sigma Chemical (Mississauga, ON, Canada).

Cell culture. Cultures of RCCD cells (4) from passages 3-25 were harvested from DMEM-F12 media supplemented with 10% FCS, 0.4% Pen/Strep solution (GIBCO), 15 mM HEPES, 50 nM hydrocortisone, 2.5 nM 3,5,3'-triiodothyronine, 44 mM sodium bicarbonate, and insulin-transferrin-sodium selenite medium supplement (Sigma) and maintained in an atmosphere of 5% CO₂ at 37°C. Passages were made after 2-3 days. Experiments were performed on cells cultured in either 12-well dishes (for measurement of AA release) or 100 × 20-mm dishes (for PLA₂, PKC, and MAPK assays) prior to forming a complete monolayer of cells.

Measurement of AA release. Cells were serum starved overnight in DMEM-F12 media containing 0.05% (wt/vol) BSA and 0.3 µCi [³H]AA. After an 18- to 24-h incubation period, labeled medium was aspirated, and cells were rinsed of unincorporated [³H]AA by two successive washes with Hanks' balanced saline solution (HBSS) containing 0.05% BSA. This was followed by a 30-min preincubation with or without PKC and MAPK inhibitors. Cells were subsequently stimulated for 15 min at 37°C with BK (in the continued presence of the inhibitors), after which the medium was immediately removed and centrifuged at 5,000 g for 5 min to pellet any cellular debris. An aliquot of the supernatant was measured for [³H]AA release by scintillation counting. Results were normalized for total label incorporated into the cells by dividing the dpm [³H]AA released by the total dpm [³H]AA incorporated into the cells (determined by solubilizing the cells in 5% SDS).

Determination of PLA₂ activity. Cells were serum starved over night in DMEM-F12 media containing 0.05% (wt/vol) BSA. The next day, cultures were washed twice with HBSS + 0.05% BSA and preincubated for 30 min at 37°C with or without PKC and MAPK inhibitors. Following a 2-min treatment with HBSS or BK (with or without inhibitors), stimulation was stopped by placing the dishes on ice, aspirating the medium, and washing the cells twice with ice-cold wash solution containing 50 mM Tris · HCl (pH 7.5), 250 mM sucrose, and 1 mM EGTA. Cells were then scraped off the plates into wash solution, centrifuged at 1,000 g for 5 min, and subsequently resuspended in cell sonication buffer composed of 50 mM Tris · HCl (pH 7.5), 250 mM sucrose, 1 mM EGTA, protease inhibitors [in µg/ml: 100 benzamidine, 20 leupeptin, 2 phenylmethylsulfonyl fluoride (PMSF), and 100 aprotinin], phosphatase inhibitors (in mM: 10 sodium vanadate, 10 sodium pyrophosphate, and 1 levamisole), and 5 mM dithiothreitol. The resuspended cells were then sonicated on ice by two successive pulses of 10 s each using the small probe of an Ultrasonics cell disrupter set at 5, after which protein concentrations were determined by the Bio-Rad protein assay method with BSA as a standard.

Immunoprecipitation and determination of PKC activity. Following overnight incubation in serum-free media, cells were washed twice with HBSS + 0.05% BSA, preincubated for 30 min at 37°C, and stimulated with BK for 2 min. Medium was subsequently aspirated, and cells were rinsed twice with ice-cold wash buffer. Cytosolic and particulate fractions were separated as described above (in SDS-PAGE and immunoblotting), and protein concentrations were determined and diluted to 400 µg/500 µl using immunoprecipitation buffer [homogenization buffer supplemented with 1% Nonidet P-40 (NP40) and 0.5% sodium deoxycholate] in a fresh centrifuge tube. To each aliquot, 50 µl protein A-agarose suspension was added, and the solution was subsequently rocked on a platform for 3 h at 4°C to reduce background that may be caused by nonspecific adsorption of cellular debris. Following a low-speed centrifugation, the resultant supernatant was transferred to a new centrifuge tube and incubated with 5 µl of PKC antibody. After 1 h, 50 µl protein A-agarose was added to the tube, and the mixture was rocked gently overnight. The next day, beads were pelleted and washed twice with 500 µl immunoprecipitation buffer, twice with 500 µl buffer 1 [50 mM Tris · HCl (pH 7.5), 500 mM NaCl, 0.1% NP40, 0.05% sodium desoxycholate], and once with 500 µl buffer 2 [50 mM Tris · HCl (pH 7.5), 0.1% NP40, 0.05% sodium desoxycholate]. Following the last wash, immunoprecipitates were dried with strips of filter paper and resuspended in immunoprecipitation buffer.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

A role for MAPK in BK-mediated cPLA₂ regulation. To address the question of whether MAPK was involved in the events leading to BK-stimulated AA release, we measured the ability of PD-98059, an inhibitor of MAPK activation secondary to its inhibition of upstream MAPK kinase (MEK) (10), to reduce this response. Treatment of cells with 30 µM PD-98059 reduced BK-stimulated AA release by 54% (Fig. 1) while exhibiting an insignificant effect upon basal release. To establish whether PKC and MAPK act independently of one another to regulate AA release, we presented cells with a combination of Ro-31-8220 and PD-98059. The agent, Ro-31-8220, is a specific but not species-selective PKC inhibitor (33). Results shown in Figs. 2 and 3 indicate that the inhibition of BK-stimulated AA release and cPLA₂ activity resulting from preincubation with both inhibitors was similar to that following pretreatment with either one of the inhibitors alone. Neither inhibitor significantly reduced the response when tested under control conditions (result not shown). Since PKC can activate the MAPK cascade (28), it appears possible that the role of MAPK in the BK-mediated release of AA is determined initially by its stimulation via a PKC-dependent pathway. To confirm this conclusion, we looked at the ability of Ro-31-8220 to prevent MAPK activation by BK.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effect of PD-98059 on bradykinin (BK)-stimulated arachidonic acid (AA) release. Rabbit cortical collecting duct (RCCD) cells were labeled overnight with 0.3 µCi [3H]AA in DMEM-F12 defined media containing 0.05% (wt/vol) BSA, then washed twice with HBSS + 0.05% BSA followed by a 30-min pretreatment with 30 µM PD-98059. Cells were subsequently stimulated with 100 nM BK in the presence of inhibitor for 15 min. Media was then removed and counted for [3H]AA release by the cells. The amount of label released was normalized for the total label incorporated into the cells and is presented as the multifold increase compared with control. The data are expressed as means ± SE of 4 independent experiments performed in duplicate. * P < 0.05 vs. BK alone.

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Effect of combined protein kinase C (PKC) and mitogen-activated protein kinase (MAPK) inhibition on BK-induced AA release. RCCD cells were labeled overnight with 0.3 µCi [3H]AA in DMEM-F12 defined media containing 0.05% (wt/vol) BSA, washed twice with HBSS + 0.05% BSA, and preincubated for 30 min with 5 µM Ro-31-8220, 30 µM PD-98059, or 5 µM Ro-31-8220 + 30 µM PD-98059. Cells were subsequently stimulated for 15 min with 100 nM BK in the continued presence of the inhibitors. [3H]AA released into the media was counted and divided by the total label incorporated into the cells. The data are expressed as means ± SD of 3 independent experiments performed in duplicate. * P < 0.05 vs. BK-stimulated condition.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   BK-mediated activation of cytosolic phospholipase A2 (cPLA2) is dependent upon PKC and MAPK. RCCD cells were incubated overnight in DMEM-F12 defined media containing 0.05% (wt/vol) BSA. They were subsequently preincubated with or without 5 µM Ro-31-8220 (Ro), 30 µM PD-98059 (PD), or 5 µM Ro + 30 µM PD for 30 min prior to stimulation with 100 nM BK for 2 min. Total cell lysate PLA2 activity was determined as described under EXPERIMENTAL PROCEDURES. Results are expressed as means ± SE of 4 separate experiments assayed in duplicate. * P < 0.05 vs. any other treatment group.

1

1

1

1

View larger version (43K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   Fig. 4.   Effect of Ro-31-8220 and PD-98059 pretreatment on BK-induced MAPK mobility and activation. RCCD cells were incubated overnight in DMEM-F12 defined media containing 0.05% (wt/vol) BSA and preincubated with or without 5 µM Ro-31-8220 (Ro), 30 µM PD-98059 (PD), or 5 µM Ro + 30 µM PD for 30 min prior to stimulation with 100 nM BK for 2 min. A: immunoblot for MAPK revealing inhibition of BK-induced mobility shift. Position of molecular weight markers is indicated to left of blot. Shown is a representative blot with 10 µg protein loaded per lane. B: total MAPK activity assessed by the phosphorylation of a MAPK-specific synthetic peptide. Data are reported as means ± SE from 4 experiments performed in duplicate. * P < 0.05 vs. any other treatment group.

View larger version (36K): [in this window] [in a new window]   Fig. 5.   Immunodetection of PKC isozymes in RCCD cells. RCCD cells cultured overnight in DMEM-F12 defined media supplemented with 0.05% (wt/vol) BSA were analyzed for the expression of PKC isozymes as described under EXPERIMENTAL PROCEDURES. Blots for PKC, , and  were loaded with 15 µg protein per lane, whereas blots for PKC were loaded with 3 µg protein per lane. The precise location of each of the various PKC isozymes was confirmed by the disappearance of the putative signal by preincubation of each antibody with its competing peptide.

PKC and BK signaling. In many cell types, complete depletion of cellular PKC (downregulation) can be achieved by long-term incubation of cells with PMA, a phorbol ester that mimics the action of diacylglycerol to activate PKC (25). This strategy was currently employed in an attempt to reveal whether a reduction in BK-stimulated AA release could be correlated with the downregulation of individual PKC species. The data presented in Table 1, however, show that this technique was an ineffective tool for RCCD cell assessment, since 24 h PMA pretreatment was unable to diminish BK-stimulated AA release. Shorter and longer preincubation times were also tested but were equally without significant effect. This was surprising, since based on results obtained previously, we were able to implicate PKC in the events leading to cPLA₂ phosphorylation and activation by BK in RCCD cells (present results and Ref. 19). Remarkably, Western blotting of PKC isozymes and measurement of PKC activity following long-term PMA incubation did not reveal any significant difference in either the pattern of PKC expression or the magnitude of PKC activity compared with control cells treated without PMA. Thus it appears that PKC is resistant to long-term phorbol ester treatment or, perhaps more specifically, that a member of the DAG-insensitive aPKC group (possibly PKC) may be responsible for mediating the observed BK responses.

View larger version (57K): [in this window] [in a new window]   Fig. 6.   Redistribution of PKC isozymes to the cytosolic and particulate fractions following exposure of RCCD cells to BK or phorbol 12-myristate 13-acetate (PMA). RCCD cells were serum starved overnight in DMEM-F12 defined media containing 0.05% (wt/vol) BSA. They were then exposed to 100 nM BK or 100 nM PMA for 2 min after which denatured proteins from cytosolic (c) and particulate (p) fractions were separated by 7.5% SDS-PAGE. Protein loading was 15 µg/lane for PKC, , and  and 3 µg for PKC. Following SDS-PAGE and transfer of proteins to nitrocellulose membranes, immunodetection was performed. The signal corresponding to PKC is represented by the top band. Each blot is representative of at least 3 experiments.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   PKC activity associated with PKC immunoprecipitate complexes. RCCD cells were serum starved overnight in DMEM-F12 defined media containing 0.05% (wt/vol) BSA. They were then rinsed and treated with BK, and separated into cytosolic and particulate fractions. Immunoprecipitates of PKC were obtained from each fraction and processed for PKC activity using a nonradioactive PKC assay. PKC activity is expressed as the absorbance associated with each phosphorylated peptide. Results are expressed as the means ± SD from 3 independent experiments.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

Prostaglandins produced from the precursor molecule, AA, are of critical importance to the maintenance of body fluid homeostasis by the collecting duct segment of the nephron (3, 12). We did previously determine that cPLA₂ is the enzyme responsible for the BK-mediated stimulation of AA release in RCCD cells (19), and we have continued this line of research in efforts to better characterize the signaling cascade responsible for activation of this enzyme.

The MAPK signaling cascade is associated with the activation of cPLA₂ in many cell systems (18, 28), but this is not always the case, and evidence indicates that different G protein-coupled receptors can also activate cPLA₂ through MAPK-independent pathways (17, 31, 34). However, as appears most commonly, when PKC is implicated in the events activating cPLA₂, MAPK is also involved. Under the present conditions, the observed results suggest that the BK-induced activation of cPLA₂ is in fact MAPK dependent (Fig. 1), in addition to PKC dependent. Although BK-receptor stimulation can activate MAPK (5, 6), recent studies of BK-stimulated AA release in MDCK cells have either confirmed this conclusion (16) or have indicated that the process occurs independent of MAPK activation (37). Concerning the study of Xing and colleagues (37), BK-induced AA release occurred through the phosphorylation-dependent activation of cPLA₂, and where neither PKC nor MAPK cascade inhibitors were able to prevent BK-stimulated AA release, the authors provide evidence that a tyrosine kinase is involved in the activation of cPLA₂ by this agonist. Interestingly, these results with BK are in contrast to their previously published results regarding cPLA₂ regulation following ₁-adrenergic receptor stimulation of this same cell type (35). When epinephrine was used as the agonist, these investigators (35) found that AA release and cPLA₂ activation was PKC dependent and MAPK dependent. Thus, within a single cell type, agonist-unique regulation of cPLA₂ can occur, and with respect to the present study, results show that a single agonist (BK) can invoke cPLA₂ activation through a unique cell type-specific pathway. Current evidence presented herein therefore provides an example of a cell-specific, BK-stimulated intracellular signaling cascade that results in cPLA₂ activation partially dependent upon PKC and MAPK.

Given the involvement of both of these kinases in the BK-mediated release of AA in RCCD cells and the partial inhibition imparted by both inhibitors (Ro-31-8220 and PD-98059), experiments were designed to determine whether these two kinases were involved independently or sequentially in the activation of cPLA₂. As discussed earlier, PKC-mediated activation of cPLA₂ is often dependent upon subsequent activation of MAPK. However, this proposed direct relationship reveals a variability that appears dependent on the specific coupling between receptor and G protein in native cells. In fact, there are examples of MAPK-dependent cPLA₂ activation occurring independent of PKC activation (13, 15) as well as PKC-dependent/MAPK-independent mechanisms of cPLA₂ activation (36). It was therefore possible that combined PKC and MAPK inhibition could fully blunt RCCD BK-stimulated AA release. The present results, however, support a sequential PKC and MAPK involvement after BK-receptor occupation, since following PKC and MEK inhibition, there resulted no additive reduction to either AA release or cPLA₂ activity (Figs. 2 and 3). Moreover, BK-induced MAPK phosphorylation and activation (Fig. 4) was significantly reduced by PKC inhibition with Ro-31-8220. It appears likely that PKC activation takes place prior to, and is responsible for, MAPK activation following BK treatment. The mechanism through which this occurs was not established in the current study, but it may involve sequential activation of upstream kinases such as Raf-1, MEK, and MEK kinase (28). Interestingly, the inability of combined inhibitor treatment to completely inhibit BK-stimulated AA release suggests that other routes leading to cPLA₂ activation are operational. Although the additional mechanisms remain uncharacterized, they do not appear to be a result of changes in the level of cPLA₂ serine phosphorylation, since our previous results (19) show that Ro-31-8220 is able to completely reverse BK-induced cPLA₂ serine phosphorylation.

Since PKC signaling represents a major step in BK-mediated cPLA₂ activation and AA release in RCCD cells, we sought to reveal whether this agonist exerted its effect through the activation of specific PKC isozymes. Östlund and colleagues (26) have demonstrated by Northern blot techniques that the rat kidney expresses mRNA transcripts coding for four PKC subspecies, PKC, , , and , and they suggest that these isozymes maintain a unique distribution pattern that may be representative of their specific roles in the functional regulation of the mature kidney. Indeed, hormonal activation of PKC is believed to be a major signaling mechanism regulating salt and water transport in the distal nephron (8, 12, 32). Accordingly, DeCoy et al. (8) reported that freshly immunodissected rabbit CCDs possess PKC and PKC and that PKC is implicated in the late inhibition of arginine vasopressin-induced sodium transport. Surprisingly, these investigators found that when cultured, RCCD cells express PKC in addition to the two species seen in the freshly isolated preparation. Subsequent to this observation, another study of PKC expression in rabbit CCD by Wilborn and Schafer (32) revealed the presence PKC, , , -like, and -like isozymes in freshly microdissected CCD segments and in fresh and cultured immunodissected CCD cells. In contrast to the maintained expression pattern seen for PKC, , and  when RCCD cells were grown in culture, these investigators found that the -like isoform was more heavily expressed in cultured CCD cells than in fresh CCD cells and that the -like isoform was strongly expressed in fresh CCDs but only weakly expressed in cultured CCDs. With respect to the present study, Western blots prepared with rabbit polyclonal antibodies show that our immortalized RCCD cell line expresses four PKC isozymes, namely, PKC, , , and  (Fig. 5). These results are comparable to the expression pattern discussed above and support the absence of PKC and PKC from rabbit CCD segments. The occurrence of PKC was not surprising, since this is the one isozyme regarded to be virtually ubiquitous in all tissues. Concerning PKC, it can be mentioned that, although this isozyme often appears restricted to neural tissue, its observed presence is similar to that described for cultured RCCD cells (8). The aforementioned results make it apparent that culturing techniques can markedly reduce or enhance the expression of individual PKC isozymes.

Results shown in Fig. 6 indicate that following BK stimulation, PKC is the only species translocated to the particulate fraction. Results illustrated in Fig. 7 further demonstrate that the PKC translocated represents an increased pool of activity as measured in vitro. Whether this redistribution event results in increased activity in situ, triggering a cascade leading to cPLA₂ activation, remains to be further proved. Interestingly, there is evidence for PKC isozyme-specific participation in the activation of cPLA₂ in CHO cells (7), MDCK cells (11, 36), and FRTL-5 cells (30). Interestingly, in each of the above-mentioned cases, whether the cells were stimulated with an extracellular ligand or artificially by phorbol ester, PKC was the isozyme predicted to play an essential role in the release of AA. A unique role for individual PKC isozymes in the phosphorylation and activation of MAPK has also been previously reported (6, 37). Our results therefore represent, to our knowledge, the first observation possibly linking PKC to cPLA₂ regulation. Since only PKC appears to be translocated following BK stimulation, we intend to further examine whether this PKC species may be responsible for the pending activation of MAPK, which results in the serine phosphorylation and activation of cPLA₂. The use of antisense techniques or of dominant negative PKC mutants should prove very valuable in this respect.

In summary, we have demonstrated that RCCD cells express at least four members of the PKC family of isozymes (PKC, , , and ) and that with respect to BK, PKC is selectively translocated to the membrane fraction with an associated increase in PKC activity. Furthermore, we have also shown that PKC is involved in the sequential activation of MAPK and cPLA₂. These results provide an example of a PKC- and MAPK-dependent process of cPLA₂ activation and further illustrate the inherent variability associated with the intracellular signaling cascade responsible for the regulation of this enzyme in different cell types.