INTRODUCTION

Dopamine is an endogenous catecholamine that exerts its actions by occupancy of specific receptors. Based on pharmacological and biochemical studies, dopamine receptors were classified initially into two main groups: the dopamine receptor linked to stimulation of adenylyl cyclase (D₁), and the dopamine receptor not linked to adenylyl cyclase or linked to its inhibition (D₂). The cloned dopamine receptors fall into these two groups. Thus, in mammals, the two D₁-like receptors (D₁ and D₅ in humans; D_1A and D_1B, respectively, in rats) are linked to stimulation of adenylyl cyclase, whereas the three D₂-like receptors (D₂ or D_2A, D₃ or D_2B, D₄ or D_2C) are linked to inhibition of adenylyl cyclase (1, 2, 3, 4, 5). Two other D₁-like receptors linked to stimulation of adenylyl cyclase have been cloned in the amphibian and avian classes, D_1C and D_1D (6, 7). Another D₁-like receptor has also been linked to stimulation of PLC¹ activity (8, 9, 10, 11, 12, 13, 14), whereas the D₂ receptor can be linked to inhibition or stimulation of PLC depending on the cell that expresses it (15, 16).

Several PLC isoenzymes have been purified, molecularly cloned, and sequenced (17, 18). These isoforms have been grouped into three families: PLC-, PLC-, and PLC-, with several members in each (e.g. PLC-₁, PLC-₂, PLC-₃, PLC-₄). They are linked to G proteins and tyrosine kinases (17, 18, 19, 20, 21, 22, 23, 24). The G_q family of G proteins, which are pertussis toxin-insensitive, has been linked to activation of PLC-₁, whereas tyrosine kinases activate PLC-₁. PLC- isoforms (PLC-₃ > PLC-₂ > PLC-₁) can also be activated by G protein / subunits independent of G-protein  subunits, PLC-₄ being an exception (18, 25). We reported recently the linkage of dopamine and -adrenergic receptors to PLC isoforms in renal cortical tissue (26). We found that a 3-4-h intravenous or intrarenal arterial infusion of norepinephrine or a D₁ agonist, which produces an anti-natriuresis or natriuresis, respectively, increases PLC-₁ and activity in renal cortical membranes similar to other G_q protein-linked receptors (24, 26). We also found that G protein-linked receptors alter the protein level and activity of PLC-₁ (26), confirming an earlier report (27). Thus, we reported that infusion of two chemically unrelated dopamine receptor agonists, the D₁ agonist fenoldopam and the D₁/D₂/D₃ receptor agonist pramipexole, decreases PLC- protein and activity in the membrane. Norepinephrine infusion, on the other hand, increases PLC-₁ protein and activity but does not affect PLC- activity in the membrane. These actions of dopamine agonists do not occur in medullary membranes. Furthermore, there are no effects of dopamine agonists on PLC- in either cortex or medulla. The putative D₁ receptor linked to PLC stimulation is yet to be cloned. Initial studies with the cloned D₁-like receptors (D_1A, D_1B, D_1C, D_1D) failed to show linkage with PLC (1, 2, 3, 4, 6, 7). However, the human, rat, and goldfish D_1A receptors were found to stimulate phosphoinositol hydrolysis or increase intracellular calcium (28, 29). In the current studies, we determined that the D_1A receptor stimulates a specific isoform of PLC and elucidated the mechanism of this stimulatory effect.

EXPERIMENTAL PROCEDURES

Monoclonal antibodies to the PLC isozymes and PKC were purchased from Upstate Biotechnology Inc., Lake Placid, NY. Forskolin, S_p-8-CTP-cAMP, phorbol 12-myristate 13-acetate (PMA), and calphostin C were from Calbiochem. Chelerythrine was from Biomol Research Laboratories Inc., Plymouth Meeting, PA. Fenoldopam and SKF83742 were from Smith Kline Beecham Pharmaceuticals, Philadelphia. ECL Western blotting detection reagents and RPN2106 were from Amersham Corp. All media, sera, and geneticin (G418) were from Life Technologies, Inc. [³H]Phosphatidylinositol 4,5-bisphosphate was from DuPont NEN. The PKC-ELISA kit was from Kamiya Biomedical Co., Thousand Oaks, CA. All other chemicals were from Sigma.

A rat D_1A receptor cDNA was subcloned in the expression vector pRc/CMV (Invitrogen, San Diego) at the XbaI site (30). The resulting construct was used to transfect stably LTK and HEK-293 cells (cell lines LTKD_1A, and HEKD_1A, respectively) as described (30). The LTKD_1A cells were maintained in -minimal essential medium supplemented with 10% fetal bovine serum and geneticin (0.25 mg/ml) under a humidified atmosphere of 95% air and 5% CO₂ at 37 °C. The cells were subcultured for the experiments by planting them at approximately 10⁵ cells/ml in a 100-mm tissue culture dish. At 70% confluence, the cells were placed in the medium containing 10 mM butyric acid, pH 7.2, and cultured for another 48 h (30). In LTKD_1A cells, the B_max was 0.23 ± 0.13 pmol/mg of protein, and the K_d was 1.35 ± 0.56 nM (n = 2) performed in triplicate. The B_max is close to that found in renal proximal tubules (31). After washing twice with phosphate-buffered saline, they were incubated for the indicated time periods at 37 °C with various drug concentrations in -minimal essential medium containing 1% dialyzed fetal bovine serum. The incubations were then terminated by washing with ice-cold phosphate-buffered saline three times, and the cells were lysed by adding lysis buffer containing (in mM) 20 Tris-HCl, pH 7.4, 2 EDTA, 2 phenylmethylsulfonyl fluoride, 25 sodium pyrophosphate, 20 sodium fluoride, and 10 µg/ml each leupeptin and aprotinin. The cells were sonicated for a few seconds and centrifuged at 14,000 rpm for 20 min at 4 °C; the supernatant represented the cytosol. The pellets were extracted by the addition of lysis buffer containing 1.0% sodium cholate for membrane PLC assays or containing 1.0% Triton X-100 for membrane PKC assays.

The cytosol and membranes were assayed for PLC activity according to the method of Nishibe et al. (32). In brief, the assay was performed in a 50-µl reaction mixture containing 30,000 cpm of [³H]PIP₂, 0.15% n-octyl--D-glucopyranoside, 0.05% Triton X-100, and (in mM) 0.8 EGTA, 0.8 CaCl₂, 35 NaH₂PO₄, pH 6.8, 70 KCl, and 10 µl (5-10 µg) of protein. The reactions proceeded for 15 min at 37 °C and were terminated by adding a stop solution containing 100 µl of 1% bovine serum albumin and 500 µl of 10% trichloroacetic acid. PIP₂-hydrolyzing radioactivity was determined by liquid scintillation counting. PLC activity was expressed as cpm/mg of protein/min.

The cell preparations were assayed for PKC activity by a PKC-ELISA kit. Samples were preincubated at 25 °C in water bath for 5 min with PKC assay buffer containing 0.3 mM ATP and phosphatidylserine. The mixture was transferred to an ELISA plate that was coated with glial fibrillary acidic protein and allowed to incubate for another 5 min. The plate was then incubated with mouse monoclonal antibody YC-10 after three washings. The resulting immunocomplexes were detected by peroxidase conjugated to secondary antibody, and the intensity of the color was measured photometrically at 450 nm.

Since currently available D₁ antagonists do not have receptor subtype specificity, we determined the involvement of the D_1A receptor using antisense phosphorothioate oligodeoxynucleotides purified by high performance liquid chromatography (GENSET SA, Paris, France). Thus, sense (5-ATG GCT CCT AAC ACT TCT ACC-3) (5 µM) and antisense (5-GGT AGA AGT GTT AGG AGC CAT-3) (5 µM) oligodeoxynucleotides were incubated with LTKD_1A cells for 2 days in culture medium at 37 °C. Following two washings, the cells were treated with vehicle or fenoldopam (5 µM). The ability of antisense but not sense oligonucleotides to prevent the expression of the D_1A receptor was verified by immunocytochemistry using anti-D_1A antibodies directed against the third extracytoplasmic loop (33).

B_max and K_d were calculated from Scatchard plots of specific ¹²⁵I-SCH 23982 binding (defined by 1 µM SCH 23390) of LTK membranes (30).

Immunoblotting was performed as described previously (26). Essentially, the proteins were separated by electrophoresis on 7.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred electrophoretically to nitrocellulose membranes. The transblots were probed with indicated antibodies and detected by peroxidase-labeled secondary antibody and chemiluminescence detection reagents. Quantification of the immunoblots was performed as reported with modifications (26); the density of the area of each immunoblot was quantified using Quantiscan (Biosoft, Ferguson, MO).

RESULTS

To determine the effect of a D₁ agonist on PLC activity, LTKD_1A cells were stimulated with the selective D₁ agonist fenoldopam (SKF82526, 5 µM), for indicated time periods, and PLC activity was measured. In the cytosol, PLC activity increased by up to 50 ± 7% (n = 5) in 30 s but slowly fell back to base line after 4 h. Similarly, in the membranes, PLC activity increased rapidly within 30 s (36 ± 13%, n = 5) and returned to baseline at 10 min. After a 24-h incubation, both cytosolic and membrane PLC activity decreased below base line (Fig. 1). These data are similar to our previous in vivo studies (26) where PLC activity increased in cytosol and decreased in membranes following the intravenous infusion of fenoldopam for 3-4 h. Fenoldopam had no effect on PLC activity in nontransfected LTK cells (Fig. 1).

Mouse monoclonal antibodies specific to PLC isoforms which recognize mouse, rat, bovine, and human PLCs were used (34, 35). In preliminary studies, we found that PLC-1 (145 kDa) and PLC-1 (85 kDa) but not PLC-1 were expressed in LTK cells. Fenoldopam moderately increased PLC-1 protein in a time-dependent pattern, which lasted longer in cytosol than in membranes (Fig. 2A). The effect was also concentration-dependent (10⁸-10⁴ M) (Fig. 2B). No such effects were seen in nontransfected LTK cells (data not shown). On the other hand, PLC- protein was not changed in either cell line following fenoldopam treatment (data not shown).

8

4

To determine if the effect of fenoldopam is mediated by the occupation of D_1A receptors, studies were performed in the presence of the D₁/D₂ antagonist SKF83742 (5 µM); the selective D₁ antagonist SCH 23390 was not used since this drug has partial agonist effects² (36), and LTK cells do not express D₂ receptors. As expected, treatment of LTKD_1A cells with fenoldopam (5 µM) for 20 min (Fig. 3A) resulted in increased PLC- protein (p < 0.05) which was blocked by SKF83742; SKF83742 alone had no effect, indicating involvement of a D₁-like receptor. Since currently available D₁ antagonists do not have receptor subtype specificity, we studied involvement of the D_1A receptor using antisense phosphorothioate oligodeoxynucleotides. Under these conditions, antisense but not sense oligonucleotides blocked D_1A receptor determined by immunocytochemistry (data not shown). As shown in Fig. 3B the antisense oligonucleotide, which, by itself, did not decrease PLC-1 isoform, blocked the stimulatory effect of fenoldopam. Sense oligonucleotide not only failed to block the stimulatory effect of fenoldopam but rather seemed to enhance PLC-1 protein level.

Stimulation of PLC- via PKA and PKC

Whether D_1A receptor-mediated stimulation of PLC- protein represents a direct mechanism or is secondary to adenylyl cyclase stimulation was addressed next. The stimulatory effect of fenoldopam on PLC- and activity was blocked by the PKA inhibitor R_p-cAMP-S (50 µM) (Fig. 4, A and B, respectively), indicating involvement of PKA. In addition, this process also involved PKC since the PKC inhibitor calphostin C (10 nM) prevented the stimulatory effect of fenoldopam on PLC- protein and activity (Fig. 4, C and D, respectively). Another PKC antagonist, chelerythrine (10 µM), had a similar effect but only in the membrane and not in the cytosol (data not shown). These results show, for the first time, that PKA and PKC pathways are involved in the D_1A receptor-mediated stimulation of PLC- protein and activity.

Effect of fenoldopam and PKA and PKC inhibitors on PLC- protein and activity in LTK

Effect of Fenoldopam on PLC- Isoform in HEK D_1A

To ensure that the results obtained with LTK cells were not cell-specific, we also studied the effect of fenoldopam on PLC- protein in HEK-293 cells transfected with the D_1A cDNA. Using the same conditions as those used with the LTKD_1A cells, fenoldopam (5 µM) also increased PLC- protein, but only in membranes; this effect was blocked by 10 nM calphostin C (data not shown).

Forskolin Increases PLC- as Well as PKC Activity

If D₁ agonists stimulate PLC- via PKA, then direct stimulation of cAMP production should also stimulate PLC- isoform. This is indeed the case, since treatment of LTKD_1A cells with forskolin (200 µM/20 min/37 °C) increased PLC- protein and activity (Fig. 5, A and B). The effect on PLC- protein and activity in membranes was blocked by a PKA antagonist, R_p-cAMP-S, but produced only a modest and nonsignificant attenuation in forskolin-induced increase in PLC activity in cytosol (Fig. 5B). The PKA agonist S_p-cAMP-S increased PLC activity in both cytosol and membrane, an effect that was blocked by R_p-cAMP-S (Fig. 5C). Thus, PKA is involved in stimulating PLC- protein and activity, especially in membranes. In addition, PKC is also involved since the stimulatory effect of forskolin (200 µM/20 min/37 °C) was blocked by the PKC antagonist calphostin C (10 nM) (Fig. 5D), although this achieved significance only in the membrane fraction. In addition, forskolin increased PKC- in membrane, an effect that preceded the increase in PLC- (Fig. 5E; three other experiments gave similar results). Forskolin (200 µM/5 min/37 °C) increased PKC activity by 32% in membranes (but not in cytosol); the increase in response to forskolin was blocked by calphostin C (10 nM) (Table I). The stimulatory effect of forskolin on PKC activity could also be mimicked by S_p-8-CTP-cAMP (50 µM), a nonhydrolyzable PKA activator (27 ± 2%, n = 6).

Effect of forskolin and PKA and PKC inhibitors on PLC- and PKC- protein in LTKD_1A cells. Cells were treated with 200 µM forskolin and/or 50 µM R_p-cAMP-S (PKA antagonist), or 10 nM calphostin for 20 min at 37 °C; the assay for PLC activity and immunoblotting were performed as described under ``Experimental Procedures.'' Forskolin increased PLC- protein and activity in both cytosol and membrane; the stimulatory effect of forskolin on PLC- protein was prevented by R_p-cAMP-S in membrane and cytosol (panel A, n = 6). R_p-cAMP-S also prevented the stimulatory effect of forskolin on PLC activity in membranes but only insignificantly attenuated it in the cytosol (panel B). However, R_p-cAMP-S did block the stimulatory effect of the PKA agonist S_p-cAMP-S on PLC activity in both cytosol and membrane (panel C, n = 6). The stimulatory effect of forskolin on PLC- protein also involved PKC since this action was prevented by calphostin in membrane (panel D, n = 6). Data are the means ± S.E. #p < 0.05 versus control; *p < 0.05 versus control or combination and p < 0.05 versus control or R_p-cAMP-S, repeated measures ANOVA and Scheffe's test. In addition, forskolin increased PKC-, which preceded the increase in PLC- (panel E; three other experiments gave similar results). Panels A-D: , control. Panel A: , 200 µM FORSKOLIN; , 1 µM R_p-cAMP-S; , forskolin + R_p-cAMP-S. Panel B: , 200 µM forskolin; , 50 µM R_p-cAMP-S; , forskolin + R_p-cAMP-S. Panel C: , 50 µM S_p-cAMP-S; , 50 µM R_p-cAMP-S; , R_p-cAMP-S + S_p-cAMP-S. Panel D: , 200 µM forskolin; , 10 nM calphostin; , forskolin + calphostin.

Table I. Effect of forskolin on PKC activity in LTK  D1A cells PKC activity was measured by a PKC-ELISA kit as described under ``Experimental Procedures.'' The results are means ± S.E. of four experiments performed in duplicate. Experimental condition Optical density % of control Control 0.3409  ± 0.09 Forskolin (200 µM) 0.4239  ± 0.05a 32  ± 6.3 Calphostin C (10 nM) 0.3100  ± 0.09  4.7  ± 9.3 Forskolin + calphostin 0.3640  ± 0.08  7.3  ± 8.8 a  p < 0.05 versus control or calphostin C, repeated measures ANOVA, and Scheffe's test.

PMA Increases PLC- Protein and Activity

Stimulation of LTKD_1A cells with PMA (1 µM/20 min/37 °C) induced the translocation of PKC- from cytosol to membrane (data not shown). PMA also increased the protein level (Fig. 6) and activity (Fig. 4D) of PLC- in membranes; this effect was blocked completely by calphostin C (10 nM) (Fig. 4D).

Effect of PMA or calphostin on PLC- protein in LTK

DISCUSSION

Several studies have shown that D₁-like receptors are linked to stimulation of both adenylyl cyclase and PLC activity (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14). However, initial studies of the cloned D₁-like receptors (D_1A, D_1B, D_1C, and D_1D) could only demonstrate linkage to stimulation of adenylyl cyclase but not to PLC in COS-7 cells (1, 2, 3, 4, 6, 7). Several investigators have now shown that the cloned human and goldfish D_1A receptors expressed in LTK and HEK-293 cells increase intracellular calcium in response to D₁ agonists (28, 29). Furthermore, mobilization of intracellular calcium due to stimulation of D_1A receptors expressed in LTK cells correlates with stimulation of PLC activity (28). However, whether stimulation of PLC activity by D₁ agonists in these transfected cells was primary or secondary to stimulation of adenylyl cyclase activity was not determined. In this regard, the ability of D₁ agonists to stimulate phosphoinositide hydrolysis has been shown to be mediated by a cholera toxin-sensitive process (28). The stimulatory effect of the thyrotropin-releasing hormone receptor expressed in Xenopus oocytes on PLC has also been shown to be coupled to G_s (37). G_q is not involved since LTK cells do not express PLC-,² the isoform that is linked to G_q (19, 24).

The current report confirms earlier findings (28) that short term stimulation of the D_1A receptor in transfected LTK cells is linked to increased PLC- protein and activity. The effect of the D₁ agonist fenoldopam on PLC- protein and activity was time-dependent in both membrane and cytosol; the stimulatory effect on PLC activity was quite transient in membranes, and an inhibitory effect was noted after 24 h of incubation. The fenoldopam-induced changes in PLC activity were paralleled by similar changes in PLC- protein; the effect on activity occurred earlier than the effect on protein. No changes in PLC- protein were noted. These effects are mediated through PKA, since forskolin increased PLC- protein, and a PKA inhibitor blocked the stimulatory effect of either fenoldopam or forskolin. The ability of D_1A receptors to increase PLC activity via cAMP is not unique to these receptors. Glucagon, calcitonin, muscarinic and -adrenergic receptors, forskolin, AlF₄, and cAMP analogs 8-Br-cAMP and S_p-8-CPT-cAMP have also been reported to stimulate phosphatidylinositol 1,4,5-trisphosphate production in hepatocytes, neuroepithelioma, and skeletal muscle cells (38, 39, 40). These studies suggested that the increase in phosphatidylinositol 1,4,5-trisphosphate production was secondary to elevation of cAMP. However, the PKC agonist PMA had an inhibitory effect in hepatic cells (39), and these studies did not determine the mechanism by which cAMP enhances phosphoinositide hydrolysis. Although PKCs can regulate phosphatidylcholine-PLC, this pathway is not directly involved in our studies since the substrate used was PIP₂, the substrate of phosphatidylinositol-specific PLC (41).

In our studies, fenoldopam increased PLC- protein via a PKA/PKC pathway (Fig. 7). PKA stimulation of PKC probably mediates membranous PLC activation, since both forskolin and a PKA agonist increase PKC activity. The involvement of PKC in the activation of PLC- is supported further by the observation that forskolin activates PKC- prior to the stimulation of PLC- protein. Moreover, PKC blockers prevent the stimulatory effect of fenoldopam, forskolin, or PMA on PLC-. This conclusion is particularly relevant to the studies in membranes. The ability of forskolin to increase PLC- protein and activity was blocked by a PKA antagonist only in membrane but not in cytosol. However, the PKA agonist S_p-cAMP-S increased PLC- activity in both membrane and cytosol, effects that were blocked by a PKA antagonist. It is possible that effects of forskolin other than stimulation of cAMP production may stimulate PLC- protein and activity by some other undefined pathway. PMA, which stimulated PLC- protein in both membrane and cytosol, stimulated PLC activity only in the membrane fraction. The failure of PMA to increase PLC activity in cytosol at a time when PLC- protein increased is not understood. However, the PMA-induced increase in PLC- protein in cytosol was not blocked by calphostin, suggesting that this effect may not be PKC-mediated.

Schematic representation of the pathway of PLC- protein in LTK

Although the present finding indicates that PLC stimulation in these transfected cells is secondary to adenylyl cyclase stimulation, several studies in tissues have implicated the existence of a D₁-like receptor that is linked to phospholipase activation independent of adenylyl cyclase (9, 11, 13, 14). Activation of D₁-like receptors decreases sodium transport by cAMP-dependent (42, 43, 44) and cAMP-independent mechanisms (45, 46, 47, 48). For example, dopamine, via D₁-like receptors, can inhibit Na⁺/H⁺ exchange activity in renal brush-border membranes by a cAMP-dependent and -independent/G_s-linked mechanism (48). Another cAMP-independent pathway of sodium transport inhibition is mediated by PLC (45, 46, 47). Thus, D₁ agonists stimulate PLC and PKC activity in renal cortical tubules and membranes (9, 49, 50) independent of adenylyl cyclase. cAMP-independent PLC stimulation by D₁ agonists has also been shown in the retina and striatum (12, 13), although the latter remains controversial (51). The effects of fenoldopam on PLC activity and protein in LTK cells heterologously expressing the D_1A receptor have both similarities and differences compared with in vivo studies in the rat (26). A 2-4-h infusion of dopaminergic drugs with agonist effects on D₁-like receptors increased PLC- activity and protein in cytosol but decreased it in membrane (26). The time frame of these fenoldopam-induced changes in PLC- protein is in accord with the findings in LTKD_1A cells. Changes in PLC- protein and activation are notable differences. In rat studies, D₁ receptor stimulation of PLC activity at 4 h was linked to stimulation of PLC- in membranes (26). LTK cells, however, do not express PLC-. It is, therefore, possible that a D₁-like receptor uniquely linked to PLC- isoform exits. In the kidney, this D₁ receptor is linked to a pertussis toxin-insensitive G protein (9), presumably G_q; PLC- is coupled to G_q (19, 24). Furthermore, in the striatum, the size of the mRNA associated with PLC stimulation and calcium mobilization is different from the transcripts of the cloned D₁ receptors (12). A D₁-like receptor that is not linked to either PLC or adenylyl cyclase activation but stimulates K⁺ efflux has also been reported (52). It is possible that other as yet uncloned D₁-like receptors exist including one that is linked primarily to PLC- and independent of adenylyl cyclase.

In summary, D_1A receptors are linked to stimulation of both adenylyl cyclase and PLC activity. The early increase in PLC activity is associated with an increase in PLC- protein. Based on the results of the present study, we suggest that D_1A-mediated stimulation of PLC occurs as a result of PKA activation. PKA then stimulates PLC- protein in cytosol and membrane. The stimulatory effect on PLC- in membrane is secondary to an increase in PKC activity (Fig. 7). These findings cannot determine if there is a different D₁-like receptor linked to PLC- isoform since this isoform was not expressed in the cell lines in which the D_1A receptor was heterologously expressed.