Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Beta-adrenergic receptors are well known regulators of vascular tone and myocardial contractility. Activation of these receptors by agonists stimulates adenylate cyclase, generating an increase of intracellular concentrations of adenosine cAMP in various cell lines (Karl and Divald, 1996; Kasis et al., 1985; Ho and Chik, 1995; Nakada et al., 1990). It has been recently described that alterations in the regulation of this pathway can lead to vascular complications such as hypertensive heart disease in which adenylyl cyclase desensitization occurs (Castellano and Bohm, 1997) or hypertension by exerting trophic effects on the vasculature and the heart (Buchholz et al., 1991; Aviv, 1994). PKC is known to play a central role in regulating diverse cellular functions and can modify the inherent properties of beta-adrenergic receptors. For instance, it was reported that activation of PKC by angiotensin II decreases the responsiveness of the rat heart to subsequent activation by beta-adrenergic agonists, leading to hypertension-induced cardiac hypertrophy and ultimately, heart failure (Schwartz and Naff, 1997). However, activation of PKC can exhibit beneficial effects in ischemic preconditioning that protects patients against acute myocardial infarction (Cohen and Downey, 1996).

The mechanisms by which PKC exerts its effects have been studied in detail, and these studies have shown that activation of PKC attenuates beta-adrenergic agonist-mediated cAMP accumulation (Keller et al., 1984; Kassis et al., 1985; Aiyar et al., 1987; Nakada et al., 1990). The mechanism by which phorbol esters regulate receptor function is thought to involve translocation of PKC from the cytosol to the membrane (Chen et al., 1995), followed by either direct phosphorylation on the beta-adrenergic receptor itself (Keller et al., 1984; Sibley et al., 1984), the adenylate cyclase (Yoshimasa et al., 1987; Simmoteit et al., 1991), on the G proteins (Heyworth et al., 1984; Iyengar, 1993) or ARK (Winstel et al., 1996). However, little is known about the roles of individual PKC isozymes in this process and their role in disease states.

PKC is, in fact, a family of serine/threonine kinases comprised of at least 12 isozymes that are divided into four subfamilies based upon domain structure. The classical or conventional PKCs (PKC-, -I, -II and -) contain the putative Ca⁺⁺ binding region C-2 and are Ca⁺⁺ responsive. The novel or new PKCs (PKC-, - and -) lack the Ca⁺⁺ binding region. The atypical PKCs (PKC-µ) and PKD lack the Ca⁺⁺ binding region and have only one cysteine-rich zinc-finger-like motif in the C-1 region. Recently described PKC-µ and PKD appear to share a high degree of identity and may constitute a separate family (Valverde et al., 1994). Although many PKC inhibitors are available to probe the role of PKC in several pathologies and experimental settings, progress in determining the isotypic pharmacology of this family of enzymes in various biological processes has been hindered by the lack of isozyme-specific PKC inhibitors.

To bridge this technological gap, first generation antisense phosphorothioate oligodeoxynucleotides have been identified to selectively inhibit PKC- mRNA and protein expression (Dean and McKay, 1994; Dean et al., 1994, 1996; McKay et al., 1996; Levesque et al., 1997a). These new, potent and selective tools demonstrated that the PKC- isozyme plays a major role in the induction of ICAM-1 expression, tumor growth and regulation of calcium mobilization. These oligonucleotides were chemically modified to improve stability and potency by introducing 2'-O-propyl and 2'-O-methoxyethyl chimeric oligonucleotides (McKay et al., 1996; Levesque et al., 1997a).

In an effort to deplete the expression of PKC- protein expression in the murine Swiss 3T3 cell line, we have chemically modified ISIS 4189, a murine PKC- phosphorothioate oligonucleotide. A chimeric phosphorothioate oligonucleotide consisting of 2'-O-methoxyethyl modifications (fig. 1) in the wings (3' and 5' portions) and an oligodeoxynucleotide gap in the center was created. This molecule retained the ability to create a substrate for RNase H when binding to the target mRNA. These second generation antisense oligonucleotides contain phosphorothioate linkages throughout the molecule, rendering the oligonucleotides extremely stable. They are more potent and more stable than phosphorothioate oligodeoxynucleotides because the 2'-O-methoxyethyl wings enhance affinity for the target RNA and resistance to nuclease digestion (Altmann et al., 1996).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Position and structure of chimeric 2'-O-methoxyethyl modified phosphorothioate oligonucleotides. A, Sequence of the phosphorothioate oligonucleotides; boxed sequences represent 2'-O-methoxyethyl nucleotides. B, Structure of the 2'-O-methoxyethyl modification.

Our purpose was to assess the role of PKC- in the regulation of cAMP production induced by beta-adrenergic agonists, using an antisense oligonucleotide that specifically depletes PKC-. A previous study based on the translocation patterns of individual PKC isozymes after TPA exposure showed that activation of PKC- and PKC-, but not PKC- was implicated in a enhanced isoproterenol- and adrenaline-stimulated cAMP production in human placental trophoblasts (Karl and Divald, 1996). In contrast, activation of PKC by phorbol esters in several cell lines such as the 3T3-L1 (Nakada et al., 1990) has been reported to reduce the subsequent production of cAMP induced by isoproterenol. In this study, we demonstrate that cells treated with phorbol esters reduced cAMP production induced by isoproterenol and this inhibition was completely prevented by depletion of PKC- by antisense oligonucleotides. We have found that selective depletion of PKC- in murine Swiss 3T3 fibroblasts completely restored cAMP production induced by isoproterenol, a nonselective beta-adrenergic agonist, when cells are pretreated with phorbol esters. Moreover, selective depletion of PKC- shifts the concentration response curve for isoproterenol to the left. This suggests that PKC- plays a role in the regulation of beta-adrenergic receptors in this cell type and that appropriately designed antisense oligonucleotides may be useful tools in isotypic pharmacology.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Cell culture. Murine Swiss 3T3 cells were obtained from the American Type Culture Collection (Rockville, MD) and cultured in Dulbecco's modified Eagle's medium (Gibco, Grand Island, NY). The medium was supplemented with 10% fetal bovine serum and antibiotics (0.1 mg/ml penicillin and 0.1 mg/ml streptomycin; Gibco). Cells were routinely passaged at 85 to 95% confluency in T-175 flasks. The cells were plated in 100-mm² culture dishes or T-75 flasks for Western and Northern analysis, respectively. For the cAMP assay, cells were plated in 24-well plates at a density of 50,000 cells/well.

Oligonucleotide synthesis. 2'-O-methoxyethyl-substituted oligonucleotides were prepared as described previously (Levesque et al., 1997a) Oligonucleotide sequences and their chemical modifications are listed in figure 1. The melting temperature of each oligonucleotide was determined in triplicate as previously described (Lesnik et al., 1993).

Treatment of cells with oligonucleotides. Swiss 3T3 cells at 85 to 90% confluency were washed once with prewarmed Dulbecco's modified Eagle's medium. A Dulbecco's modified Eagle's medium solution containing oligonucleotides and DOTMA-DOPE (Gibco; 2.5 µg/ml/100 nM oligonucleotide) was then added to the cells and incubated at 37°C for 4 hr. The DOTMA-DOPE/oligonucleotide mixture was aspirated off the cells and replaced with media containing 0.4% fetal bovine serum. Control cells (no oligonucleotide) were treated with the lipofectin concentration corresponding to that used with the highest concentration of oligonucleotide in the experiment.

Measurement of PKC mRNA levels. PKC- mRNA expression in Swiss 3T3 cells was evaluated as previously described (Dean and McKay, 1994). Briefly, cells were lysed in a 4 M guanidinium isothiocyanate solution then layered over a cesium chloride gradient and centrifuged overnight at 18°C at 150 000 × g. The resulting total RNA (20-25 ng) was electrophoresed on a 1.2% agarose gel containing 1.1% formaldehyde and transferred to nylon membranes (Hybond). The membranes were then probed in Quikhyb solution (Stratagene, La Jolla, CA) using [-³²P] dCTP radiolabeled PKC-, PKC-, PKC- cDNA (ATCC). To confirm equal loading, the membranes were stripped in boiling 0.1% SSC/0.1% SDS solution for 2 min and then reprobed with a radiolabeled human G3PDH. Hybridizing bands were visualized and quantified using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).

Immunoblotting of PKC isozymes. Immunoblotting was performed as previously described (Levesque et al., 1997a). Briefly, cells were washed once with ice-cold PBS and lysed in 250 µl of lysis buffer (20 mM Tris, pH 7.4; 1% (v/v) Triton X-100; 5 mM EGTA; 2 mM EDTA; 2 mM dithiothreitol; 50 mM NaF; 10 mM Na₂HPO₄) supplemented with leupeptin (2 µg/ml) and aprotinin (1 µg/ml) at 4°C. Samples were loaded equally on gel, as determined by Bradford protein assay (Bio-Rad, Hercules, CA), and electrophoresed through a 12% acrylamide gel and then electroblotted. The levels of PKC- (79 kDa) and G3PDH (33 kDa) were simultaneously determined by use of anti-PKC- (1:2000; Upstate Biotechnology, Lake Placid, NY) and anti-G3PDH (1:50000; Advanced ImmunoChemical Inc., Long Beach, CA) monoclonal antibodies. After a minimum of 2 hr incubation with the primary antibody, the membranes were incubated with either 5 µCi of ¹²⁵I-labeled goat anti-mouse or ¹²⁵I-labeled goat anti-rabbit antibodies (ICN Radiochemicals, Costa Mesa, CA) for 1 hr. Hybridizing bands were visualized and quantified using a PhosphorImager.

cAMP assay. The cAMP content of the cells was assessed using the cAMP RIA kit from Du Pont (Boston, MA). Cells were pretreated with PDBu (1 µM) and/or GF-109203X (5 µM; Alexis Biochemicals, San Diego, CA) by adding 10 µl of either drug to the wells for the indicated times. Media were aspirated and 250 µl of PBS supplemented with dextrose (4.5 g/l), bovine serum albumin (0.2%) and IBMX (1 mM) containing either isoproterenol, forskolin or cholera toxin (preactivated with 20 mM dithiothreitol for 10 min at 37°C) was added. The reaction was stopped by adding 250 µl of 10% TCA. The extracts were then centrifuged at 4°C for 5 min. The supernatant was extracted three times with 2 ml of water-saturated ether and lyophilized. The pellet was suspended in 200 µl of cAMP buffer provided with the kit. The determination of the cAMP content of the samples was then performed as instructed by the manufacturer's protocol.

Membrane preparation. Cells were transfected in T-175 flasks with 200 nM of oligonucleotide as described above and incubated in a low serum medium for 72 hr. Cells were washed twice in the lysis buffer (5 mM Tris-HCl, pH 7.4; 2 mM MgCl₂; 1 mM EDTA) then 3 ml were added and the flasks were kept 10 min on ice. The cells were then scraped using a policeman then homogenized using a tissue triator (Biospec Products Inc, Racine, WI). An aliquot was collected and lyophilized for determination of PKC- depletion by immunoblotting. The homogenate was centrifuged 10 min at 300 × g at 4°C. The supernatant was then centrifuged at 40,000 × g for 20 min at 4°C. The pellet was resuspended in a Tris-sucrose buffer (75 mM Tris-HCl, pH 7.4; 12.5 mM MgCl₂; 1.5 mM EDTA; 0.5 mM dithiothreitol; 250 mM sucrose) and passed through a 22-gauge syringe, four to five times, then stored at 80°C. The protein content of samples was determined with the Bradford protein assay (BioRad).

Beta-Adrenergic receptor binding. The binding experiments were performed as previously described (Nakada et al., 1990). Briefly, 65 µg of membrane protein were incubated in the binding buffer (75 mM Tris-HCl, pH 7.4; 12.5 mM MgCl₂; 1.5 mM EDTA) with 5 to 1000 pM of ¹²⁵I-CYP (NEN, Boston, MA) and competing ligand, when indicated. The total volume for assays was 500 µl. The nonspecific binding was determined in a parallel set of matching tubes containing an excess of pindolol (5 µM), a nonselective beta-1 and beta-2 adrenergic antagonist. After 45 min of incubation at 37°C, membrane-bound radioligand was separated from unbound fluid phase by rapid vacuum filtration through glass fiber filters (Whatmann GF/C, Brandell Corp., Gaithersburg, MD) with a 24-channel cell harvester (model M-24, Brandell Crop.). Filters were washed four times with ice-cold binding buffer, removed from the harvester template and placed into 12 x 75 mm tubes. The radioactivity was quantified in a gamma counter (Beckman Gamma 5500B, Fullerton, CA).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of PdBu and GF-109203x on isoproterenol-induced cAMP production. Isoproterenol elicited a concentration-dependent increase of cAMP production in murine Swiss 3T3 fibroblasts (fig. 2A). PdBu treatment reduced the maximal cAMP production induced by isoproterenol and shifted the dose response curve only slightly to the right (control EC₅₀: 732 ± 183 nM; Pdbu EC₅₀: 1327 ± 418 nM).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Effect of PdBu on the kinetics and the concentration-dependent increases of cAMP levels induced by isoproterenol in Swiss 3T3 cells. Cells were serum deprived for 24 hr. A, When indicated PdBu (1 µM) was then added to the cells 45 min prior stimulation with various concentrations of isoproterenol. Control (); PdBu (). Values are mean ± S.E.M. of three determinations from a representative experiment selected from two individual experiments. B, When indicated, GF-109203X (5 µM) was added 30 min before stimulation with PdBu. PdBu (1 µM) was then added to the wells for either 0, 5, 10, 15, 30, 45 or 60 min before addition of 1 µM of isoproterenol. Control (); GF-109203X (). A and B, Accumulation of cAMP induced by isoproterenol for 10 min at 37°C. cAMP was measured by radioimmunoassay as described in methods. Values are mean ± S.E.M. of three determinations from a representative experiment selected from three individual experiments.

Reduction of PKC- mRNA and protein levels by antisense oligonucleotides. To assess the possible role of the PKC- isozyme in the regulation of cAMP production by phorbol esters, we specifically reduced PKC- mRNA and protein expression using antisense oligonucleotides. Treatment of Swiss 3T3 cells with ISIS 14012 reduced the expression of both the 4- and the 8.5-kb species of PKC- mRNA in a concentration-dependent fashion after 72 hr, exhibiting a IC₅₀ of ~50 nM (fig. 3). This time was chosen to show the persistence of the mRNA reduction by the oligonucleotide. A previous report demonstrated that oligonucleotides containing 2'-O-methoxyethyl modifications reduced the mRNA levels up to 72 hr after treatment (Levesque et al., 1997a). Moreover, a prolonged treatment is necessary to reduce PKC- protein levels in cells (Levesque et al., 1997a). At concentrations as high as 200 nM, ISIS 13818, a 13-base mismatch control of ISIS 14012, had no effect on the levels of PKC- transcripts. The oligonucleotides (100 nM) had no effect on the mRNA expression on PKC- and PKC- transcripts, demonstrating selectivity of ISIS 14012 for PKC- (fig. 3D). None of the oligonucleotides tested affected G3PDH mRNA levels, demonstrating selectivity for the targeted mRNA (fig. 3).

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Inhibition of PKC- mRNA expression in Swiss 3T3 cells. A, B and C, Concentration-effect relationship for the reduction of PKC- mRNA expression 72 hr after oligonucleotide treatment. A, Representative blots of cells treated ISIS 14012 or ISIS 13818. Top bands, PKC- mRNA transcripts. Bottom band, G3PDH mRNA transcripts, demonstrating equal loading in each lane. B and C, Levels of PKC- 4 Kb (B) and 8.5 Kb (C) transcripts from the gels that were quantified with a PhosphoroImager, normalized for G3PDH loading, and expressed as percentage of control. Values are the average of two determinations. () ISIS 14012. () ISIS 13818. D, Effect of ISIS 14012 and ISIS 13818 on the expression of PKC- and PKC- in cells that were treated with 100 nM of oligonucleotides, and the mRNA was extracted after 72 hr.

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Inhibition of immunoreactive PKC- protein expression in Swiss 3T3 cells. A and B, Kinetic analysis of the reduction in PKC- mRNA expression by oligonucleotides. A, Representative blots of cells treated with 200 nM of ISIS 14012 or ISIS 13818 for 24, 48 and 72 hr. Top bands, immunoreactive PKC- protein. Bottom band, immunoreactive G3PDH, demonstrating equal loading in each lane. B, Levels of PKC- from the above gels that were quantified with a PhosphorImager, normalized for G3PDH loading, and expressed as percentage of control. Values are the mean ± S.E.M. of three determinations. Open bars, no oligonucleotide; solid bars, ISIS 14012; hatched bars, ISIS 13818. C and D, Concentration-effect relationship for the reduction of PKC- protein expression by oligonucleotides. C, Representative blots of cells treated with ISIS 14012 or ISIS 13818. Top bands, immunoreactive PKC- protein. Bottom band, immunoreactive G3PDH, demonstrating equal loading in each lane. D, Levels of PKC- from the above gels that were quantified with a PhosphoImager, normalized for G3PDH loading, and expressed as percentage of control. Values are the average of two determinations. () ISIS 14012. () ISIS 13818.

Effect of PKC- depletion on PdBu regulation of cAMP accumulation induced by isoproterenol. Cells were exposed to oligonucleotides as described above and the cAMP accumulation assay was performed at 72 hr after treatment, when PKC- protein reduction by ISIS 14012 was maximal (fig. 5). Treatment of cells with PdBu was shown to depress the maximal induction of cAMP accumulation (fig. 2). However, when cells were pretreated with ISIS 14012 (200 nM), the effects of PdBu were reversed because ISIS 14012 restored the maximal induction of cAMP by isoproterenol when treated with PdBu (1 µM) 45 min before agonist challenge (fig. 5A). As described above, oligonucleotide treatment exhibited only a marginal effect on the EC₅₀ (no oligonucleotide: 440 ± 66 nM; ISIS 14012: 606 ± 257 nM; ISIS 13818: 456 ± 149 nM).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Effect of oligonucleotides on the kinetics and the concentration-dependent increases of cAMP levels induced by isoproterenol in Swiss 3T3 cells. Approximately 72 hr after treatment with no oligonucleotide () or 200 nM of ISIS 14012 () or ISIS 13818 (), cells were treated with PdBu (1 µM). A, PdBu was added 45 min before challenge with various concentrations of isoproterenol. B, PdBu was added to the wells for the time indicated, except for control (0 min). Accumulation of cAMP was then stimulated for 10 min at 37°C with isoproterenol. cAMP was measured by radioimmunoassay as described in methods. Values are mean ± S.E.M. of six determinations from two separate experiments.

Effect of PKC inhibition by GF-109203X and by antisense oligonucleotide on isoproterenol-induced cAMP accumulation. The effect of PKC- depletion on isoproterenol-induced cAMP accumulation was assessed in cells untreated with PdBu (fig. 6A). The oligonucleotides (200 nM) did not alter the maximal cAMP production induced by isoproterenol. However, ISIS 14012, but not the 13-base mismatch ISIS 13818, shifted the concentration-response curve of isoproterenol to the left, sensitizing the cells and producing a EC₅₀ of 116 ± 90 nM for the ISIS 14012, compared to an EC₅₀ of 458 ± 153 nM and 949 ± 306 nM for control and ISIS 13818. GF-109203X had no effect on the concentration-response curve of isoproterenol (fig. 6B).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Effect of oligonucleotides (A) or GF-109203X (B) on the concentration-dependent increases of cAMP levels induced by isoproterenol in Swiss 3T3 cells. A, Approximately 72 hr after treatment with no oligonucleotide () or 200 nM of ISIS 14012 () and ISIS 13818 (), accumulation of cAMP was stimulated for 10 min at 37°C with varying concentrations of isoproterenol in the presence of IBMX (1 mM). B, GF-109203X was added 60 min before stimulation with isoproterenol. Control (), GF-109203X (). cAMP was measured by radioimmunoassay as described in methods. Values are mean ± S.E.M. of nine determinations from three experiments (A) or six determinations from two experiments (B).

Effect of PKC- depletion and PdBu treatment on beta-adrenergic receptor affinity and density. To better understand the mechanisms of regulation of beta-adrenergic receptors after PKC- depletion or PKC activation, we have assessed the affinity and density of the beta-adrenergic receptor population following PKC- depletion and PdBu treatment. Saturation curves were performed in the Swiss 3T3 cells by varying the labeled ligand concentration from 2 to 1000 pM and the data results are shown in tables 1 and 2. Table 1 represents cells that were treated with ISIS 14012 or ISIS 13818 and compared to nontreated cells; the experiments were designed to obtain an identical statistical weight from each cell line represented in the three saturation curves. PKC- protein depletion did not significantly alter the K_d and B_max of the receptors. Table 2 represents cells that were treated with PdBu (1 µM) for 30 min. No effect was seen on the receptor affinity and density, which is consistent with a previous report (Nakada et al., 1990).

View this table: [in this window] [in a new window]   TABLE 1 Effect of PKC- depletion on -adrenergic receptor affinity and density

View this table: [in this window] [in a new window]   TABLE 2 Effect of phorbol esters on -adrenergic receptor affinity and density

Effect of PKC inhibition by GF-109203X and by antisense oligonucleotide on cholera toxin-induced cAMP accumulation. Cholera toxin catalyses an NAD⁺ -dependent ribosylation of G_s (Gilman, 1987). This covalent modification of G_s reduces its intrinsic GTPase activity, promoting a persistent activation of G_s and thereby of adenylate cyclase. Cholera toxin stimulated cAMP accumulation in the Swiss 3T3 cell line after a incubation period of 60 min (fig. 7). Levels of cAMP accumulation were further increased when cells were treated simultaneously with cholera toxin and PdBu. The nonspecific PKC inhibitor, GF-109203X (fig. 7A), reversed the effects of PdBu although depletion of PKC- by ISIS 14012 (fig. 7B) had no effect on the activation by PdBu. Neither GF-109203X nor ISIS 14012 had any effect on cells treated only with cholera toxin.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Effect of GF-109203X (A) or oligonucleotides (B) on PdBu-induced potentiation of cholera toxin (CTX) induction of cAMP levels in Swiss 3T3 cells. When indicated, PdBu (1 µM) was added 30 min prior addition of cholera toxin (50 µg/ml). Accumulation of cAMP was then stimulated for 60 min at 37°C with cholera toxin in the absence (open bars) or in the presence (solid bars) of PdBu. cAMP was measured by radioimmunoassay as described in "Materials and Methods." Values are mean ± S.E.M. of nine (A) or six (B) determinations from three and two separate experiments, respectively.

Effects of PdBu and GF-109203X on forskolin-induced cAMP production. We have studied the effects of PKC activation and inhibition on the beta-adrenergic receptor and the G_s protein. To study the effect of PKC on direct activation of adenylate cyclase, forskolin was used in the presence or absence of PdBu and GF-109203X. Forskolin elicited a concentration-dependent increase of cAMP production in murine Swiss 3T3 fibroblasts that was not altered by GF-109203X (fig. 8A). Concentrations of 300 nM forskolin consistently resulted in maximal elevations of cAMP production. PdBu and GF-109203X had no significant effect on forskolin-induced cAMP levels (fig. 8B). The effect of the oligonucleotides was not tested for forskolin because the PKC inhibitor, GF-109203X and PdBu had no effect on the cAMP accumulation induced by this agonist.

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Effect of GF-109203X increases of cAMP levels induced by forskolin in Swiss 3T3 cells. GF-109203X was added 60 min before stimulation with forskolin. A, Accumulation of cAMP was stimulated for 10 min at 37°C with varying concentrations of forskolin in presence () or absence () of 5 µM GF-109203X. B, PdBu (1 µM) was added to the wells for the time indicated, except for control (0 min). Accumulation of cAMP was then stimulated with 300 nM forskolin in presence () or absence () of 5 µM GF-109203X. cAMP was measured by radioimmunoassay as described in methods. Values are mean ± S.E.M. of three determinations from a representative experiment (A and B).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The effects of phorbol esters on cAMP accumulation in control Swiss 3T3 cells were consistent with previous studies which found that beta-adrenergic receptor-mediated responses were attenuated after phorbol ester treatment (Nakada et al., 1990). We report that short term (5-60 min) PdBu treatment of the Swiss 3T3 cell line attenuated the maximal increase of cAMP levels, but does not affect the EC₅₀ of isoproterenol, a nonselective beta-1 and beta-2 receptor agonist. This reduction of isoproterenol-induced cAMP production by phorbol esters was completely reversed with GF-109203X, a bisindolylmaleimide inhibitor of PKC (Toullec et al., 1991), which is consistent with previous reports using other chemical classes of PKC inhibitors.

To evaluate the putative role of PKC- in beta-adrenergic receptor regulation by phorbol esters, we used an antisense oligonucleotide to selectively deplete PKC- protein expression in the Swiss 3T3 cells. A previously reported 20-mer phosphorothioate oligodeoxynucleotide (ISIS 4189) was shown to specifically reduce murine PKC- expression in tissue culture by hybridizing to the target mRNA, creating a DNA-RNA duplex that becomes a substrate for RNase H cleavage (Dean et al., 1996). In a subsequent study, this sequence was chemically modified by incorporating 2'-O-propyl modification in nucleotides situated at the 5' and 3' extremities of the oligonucleotide, leaving a center gap of deoxynucleotides. This center gap assures that the formed duplex is a substrate for RNase H although the chemically modified nucleotides at the 5' and 3' extremities enhance affinity for the sense strand and increase nuclease resistance. Incorporation of 2'-O-propyl modifications enhanced the thermodynamic hybridization properties of the parent oligonucleotide from 66.4 to 73.9°C. These findings correlated well with the increase of potency of the compound, for which the IC₅₀ was increased 2- to 3-fold to ~75 nM, compared to ~200 for the original phosphorothioate oligonucleotide. In this study, the 2'-O-methoxyethyl modifications increased melting temperatures to 80.2°C and yielded an IC₅₀ for PKC- mRNA reduction of ~50 nM. Thus, one can conclude that higher melting temperatures enhanced affinity for target RNA and hence, correlated with the increase of potency for mRNA reduction. Additionally, because such oligonucleotides are more stable, the reductions in PKC- mRNA and protein levels were more prolonged than with a phosphorothioate oligodeoxynucleotide. Using this pharmacological tool, ISIS 14012 reduced the immunoreactive PKC- protein more than 85%, 72 hr after oligonucleotide treatment.

The depletion of PKC- protein expression by ISIS 14012 completely reversed the effect of phorbol ester on cAMP production induced by isoproterenol, to the same extent as the non-selective PKC inhibitor GF109203X. We can conclude that PKC- plays a major role in this process because the reversal was complete. However, further confirmation should be pursued by depleting other PKC isozymes with antisense oligonucleotides and assessing their roles.

PKC- protein depletion was shown to alter isoproterenol-induced cAMP production in cells untreated with PdBu. The oligonucleotide did not affect the maximal response but did shift the potency for isoproterenol by 3- to 4-fold. However, a nonisozyme-specific PKC inhibitor, GF-109203X, did not alter the cAMP levels induced by beta-adrenergic agonists in the absence of phorbol ester activation. One possible explanation for these observations is that in the basal state, coupling of beta-adrenergic receptors to adenylate cyclase is regulated by several PKC isozymes, some of which are stimulatory although others are inhibitory. Non-specific inhibition of PKC then would have no effect on overall coupling. Alternatively, GF-109203X may have effects on Swiss 3T3 cells in addition to its effects on PKC that mitigate its activity in cells untreated with PdBu. When cells are treated with PdBu, however, PKC- is clearly the key isozyme stimulated and both ISIS 14012 and GF-109203X inhibit the effects of PdBu.

PKC may either phosphorylate the -adrenergic receptor itself (Keller et al., 1984; Sibley et al., 1984), the G_s protein (Heyworth et al., 1984; Iyengar, 1993) or the adenylate cyclase (Yoshimasa et al., 1987; Simmoteit et al., 1991). Recently, it has been proposed that PKC phosphorylates the beta-adrenergic receptor kinase, which not only activates cytosolic beta-adrenergic receptor kinse-1, but also translocates beta-adrenergic receptor kinase immunoreactivity from the cytosol to the membrane fraction (Winstel et al., 1996). In any case, PKC activation is known to alter subsequent induction of cAMP by beta-adrenergic receptor agonists. Binding studies indicate that neither receptor affinity nor density was altered by the oligonucleotide treatment or by phorbol ester stimulation of cells. Because forskolin, a direct activator of adenylate cyclase, was unaffected by PKC activation, we can assume that the site of action of PKC- in this cell model would be at G_s protein level. To verify this possibility, we studied cholera toxin-induced cAMP accumulation in the Swiss 3T3 cell line. Surprisingly, PdBu enhanced the cAMP accumulation induced by cholera toxin. This potentiation was reversed by the nonspecific PKC inhibitor, GF-190203X, but depletion of PKC- by ISIS 14012 failed to alter cAMP levels induced by cotreatment with cholera toxin and PdBu. These data suggest that several PKC isozymes regulate coupling of G_s and G_i proteins to adenylate cyclase that could explain the lack of effect of PKC- depletion by ISIS 14012. Cholera toxin activates one component of this system, the G_s protein. In the case of beta-adrenergic receptor activation, several regulatory proteins are involved, which could explain the opposite effects of phorbol esters on isoproterenol and cholera toxin-induced cAMP levels. Taken together, the lack of effect of PKC activation by phorbol esters on the beta-adrenergic receptor population and adenylate cyclase, and the effects of PKC on cholera toxin-induced cAMP levels would indicate that the site of action of PKC- is located at the G protein level.

In conclusion, these results suggest that PKC- plays a important role in regulation of the beta-adrenergic receptors. We have previously shown that PKC- can also regulate the bradykinin B2 receptor when activated by phorbol esters (Levesque et al., 1997a). In another study, PKC- depletion did not alter MAP kinase activation by bombesin or phorbol esters (Levesque et al., 1997b) in A549 human lung carcinoma cells. MAP kinase is down stream from the receptor-effector system that is known to be activated by PKC. Bombesin is a cell surface receptor agonist that is known to activate phospholipase C through G_q protein. This pathway is known to activate MAP kinase through PKC and in this system, PKC- was not involved. These limited data suggest that PKC- is a regulator of early steps in receptor-mediated events while other isoforms may be involved in later events, such as activation of MAP kinase or be involved in other signaling pathways. It has also been demonstrated that PKC- is involved in the regulation of intercellular adhesion molecule 1 mRNA expression (Dean et al., 1994). Obviously, the effects of PKC- inhibition in other pathways and cellular processes must be determined before a definitive understanding of the various roles of this isotype in cellular physiology and pathophysiology can be achieved, but antisense inhibitors may provide excellent tools to facilitate this process.