Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Natriuretic peptides were discovered as a natriuretic entity present in atrial granules (de Bold et al., 1981). The initially identified natriuretic peptide, termed atrial natriuretic peptide, consisted of 28 amino acids with a 17-amino- acid cyclic ring (Flynn et al., 1983). Subsequently, brain natriuretic peptide (Sudoh et al., 1988) and C-type natriuretic peptide (Sudoh et al., 1990) were identified as additional natriuretic peptides with distinct structures. Both atrial natriuretic peptide and brain natriuretic peptide activate a guanylyl cyclase termed natriuretic peptide receptor (NPR)-A (Chinkers et al., 1989; Koller et al., 1991), whereas C-type natriuretic peptide selectively activates a distinct guanylyl cyclase termed NPR-B (Chang et al., 1989; Suga et al., 1992). All three peptides bind to the third natriuretic peptide receptor designated NPR-C (Koller et al., 1991; Suga et al., 1992). The NPR-C is a truncated version of the guanylyl cyclase receptors and it lacks the consensus guanylyl cyclase catalytic or protein kinase-like domains present in NPR-A and NPR-B (Schenk et al., 1987; Porter et al., 1990). The NPR-C contains a short cytoplasmic extension of only 37 amino acids (Porter et al., 1990); thus, the NPR-C typically is discounted as a signal-transducing receptor. Rather, it has been proposed to perform a buffering or clearance function to regulate natriuretic peptide levels in blood (Maack et al., 1987). This study challenges this contention and defines a biological effect of the NPR-C protein to suppress evoked catecholamine efflux.

Natriuretic peptides suppress sodium reabsorption (de Bold et al., 1981), vasoconstriction, and secretion of renin, aldosterone, and catecholamines (Anand-Srivastava and Trachte, 1993). Characteristically, natriuretic peptide actions are attributed to activation of particulate guanylyl cyclases to elevate cGMP concentrations to effect intracellular responses (Lewicki and Protter, 1995). However, the inhibitory neuromodulatory effect of natriuretic peptides has been attributed to NPR-C activation (Drewett et al., 1990; Babinski et al., 1995). This putative activity of the NPR-C was based on both the ability of selective NPR-C binding agents to reproduce the inhibitory effects of natriuretic peptides (Drewett et al., 1990) and a dissociation of the inhibitory effects from guanylyl cyclase activation (Trachte and Drewett, 1994). The major problem with this reasoning involves the potential existence of other unidentified NPRs mediating the natriuretic peptide neuromodulatory effects.

This study defines the role of the NPR-C in neuromodulatory actions of natriuretic peptides by the technique of suppressing NPR-C levels with antisense oligodeoxynucleotides. The results are consistent with the NPR-C responding to natriuretic peptides to attenuate evoked dopamine efflux.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture. Pheochromocytoma (PC12) cells were grown in Dulbecco's modified Eagle's medium as previously described (Trachte et al., 1995). The cells were differentiated by the addition of 200 ng/ml 7S nerve growth factor and 1% fetal bovine serum. The differentiated cells were grown in flasks coated with collagen. All experiments reported in this study were performed on differentiated cells.

Antisense Knockdown. The NPR-C was depleted by treating cells with an 18-mer oligodeoxynucleotide of the following composition: 5'-CAGCAGCAGGGACCGCAT-3'. This antisense treatment was designed to hybridize with the initiation region of the NPR-C mRNA, as reported by Engel et al. (1994). The missense control oligodeoxynucleotide consisted of the following bases: 5'-CAGCAGCAGGCAGCGTAC-3'. These oligodeoxynucleotides were purchased from Oligos Etc (Wilsonville, OR) and were phosphorothioate derivatives to increase stability. The oligodeoxynucleotides were administered with 2.5 mg/l GS cytofectin (Glen Research, Sterling, VA) to facilitate oligodeoxynucleotide delivery to the intracellular compartment of the cell. The oligodeoxynucleotide incubation lasted for 12 h and cells were used for experiments 48 h later. All NPR-C knockdown experiments involved paired cultures being treated with either the antisense or missense oligodeoxynucleotides in the presence of the GS cytofectin.

Western Blotting. Levels of the NPR-C were assessed by Western blotting with an antibody supplied by Dr. David Lowe (Genentech, South San Francisco, CA). The antibody was generated against a peptide representing 16 amino acids in the extracellular region of the NPR-C (i.e., NH₂ Ala-Ala-Ala-Arg-Gly-Ala-Pro-Asp-Leu-Ile-Leu-Gly-Pro-Val-Cys COOH; Bennett et al., 1991). Cells were prepared for Western blotting by sonication in 100 mM sucrose, followed by centrifugation at 15,000g for 30 min. The pellets were digested in a solubilization buffer consisting of the following: Triton X-100 (1%), deoxycholic acid (0.5%), SDS (0.1%), sodium chloride (150 mM), Tris (50 mM), ethylenediamine-tetraacetic acid (1 mM), leupeptin (2 mg/l), antipain (4 mg/l), benzamidine (20 mg/l), and aprotinin (18 trypsin inhibitory units). After protein analysis with the Bio-Rad D_C protein assay kit, samples were placed in SDS (4%), Tris (0.125 M), glycerol (22%), bromphenol blue (0.04%), water (10%), and -mercaptoethanol (1.42 M); boiled; and resolved by electrophoresis in 10% polyacrylamide gels. The proteins were transferred to nitrocellulose membranes and blocked for 2 h with 3% milk, followed by incubation with the primary antibody (1:200 dilution) overnight. Bound antibodies were detected with enhanced chemiluminescence (Amersham, Arlington Heights, IL) with horseradish peroxidase-conjugated secondary antibodies generated against rabbit IgG (Jackson Research Laboratories, West Grove, PA). The light generation was recorded on photographic film (Hyperfilm; Amersham) in a film cassette and developed by a Konica QX-70 medical film processor. The density was quantified with the NIH Image program.

Catecholamine Analysis. Catecholamine efflux from the PC12 cells was induced by depolarization with Krebs' buffer containing 40 mM potassium chloride, as previously described (Trachte et al., 1995) or with nicotine (10 µM to 60 mM) dissolved in Krebs' buffer containing 4.5 mM potassium chloride. The effect of C-type natriuretic peptide on this evoked release of dopamine was investigated by including the C-type natriuretic peptide with the depolarizing Krebs' buffer at varying concentrations. The incubations lasted for 5 min. Results are expressed as percentage release of total cellular dopamine contents or as percentage of the control release in the absence of C-type natriuretic peptide. These catecholamine efflux experiments were performed after either missense or NPR-C antisense treatment.

Guanylyl Cyclase Activity. Guanylyl cyclase activity was measured indirectly by assessing cGMP concentrations within the cells. These experiments were performed identically to the catecholamine efflux experiments except that isobutylmethylxanthine (2.5 mM; Sigma Chemical Co., St. Louis, MO) was used in the depolarizing Krebs' buffer to inhibit phosphodiesterase activity. Cyclic nucleotides were extracted with 1 ml of ethanol. The ethanol was evaporated and the cyclic nucleotides were dissolved in assay buffer and assayed with the TRK.500 assay kit (Amersham).

Statistical Analyses. All experiments were performed in a paired fashion such that curves could be compared by ANOVA for repeated measures. Comparisons of individual responses to control values were compared by the Student's paired t test with correction for multiple comparisons.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Treatment of PC12 cells with the antisense oligodeoxynucleotide reduced NPR-C expression, as shown in Fig. 1. The NPR-C migrated with a molecular mass of 67 kDa. The antisense treatment suppressed NPR-C immunoreactivity by an average of 52 ± 4% (n = 3), as measured by densitometry with the NIH Image program, relative to missense treatment. Equal amounts of protein were loaded in each lane, as indicated by a nonspecific band migrating at 77 kDa in Fig. 1. The secondary antibody used failed to react detectably with any proteins in the absence of the primary antibody (data not shown).

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Western blot demonstrating NPR-C levels after antisense or missense oligodeoxynucleotide treatment. The NPR-C migrated at an approximate molecular mass of 67 kDa in the reducing gels used. The NPR-C protein was less abundant in cells treated with antisense oligodeoxynucleotides (100 nM). The reduction averaged 52 ± 4% after analysis by NIH Image. The presence of a nonspecific band at 77 kDa confirms that equal amounts of protein were loaded in corresponding lanes for the different treatment groups. Lanes are labeled as antisense or missense treated. Molecular mass (Mr) markers are indicated on the right. Duplicate samples from cells treated with either antisense or missense oligodeoxynucleotides are shown.

The effect of the antisense oligodeoxynucleotide on neurotransmission is shown in Fig. 2. Reduction of NPR-C levels was associated with a statistically significant attenuation of potassium-induced catecholamine efflux of 33 ± 6%, relative to missense treatment. Potassium-evoked dopamine efflux averaged 16.0 ± 1.4% of total dopamine content in cells treated with the missense oligodeoxynucleotide and 10.2 ± 0.5% (P < .001) in cells receiving the antisense oligodeoxynucleotide (Fig. 2). A similar reduction in nicotine-induced dopamine efflux also was observed in cells treated with the antisense oligodeoxynucleotide, relative to missense treatment, at nicotine concentrations of 10, 100, and 1000 µM (data not shown; n = 2). This approximate 37% reduction in catecholamine efflux was not observed at higher nicotine concentrations (i.e., 10 and 60 mM; n = 2; data not shown).

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Dopamine efflux evoked by a depolarizing stimulus (40 mM KCl) in cells treated with either the missense or antisense oligodeoxynucleotide (100 nM) for 2 days. Evoked dopamine efflux was significantly reduced by antisense treatment (***P < .001). Values are mean ± S.E. The number of preparations in each group is indicated by the N.

The C-type natriuretic peptide reduced evoked dopamine efflux 39 ± 3 and 32 ± 5% at concentrations of 10¹⁰ and 10⁹ M, respectively, in cells treated with the missense oligodeoxynucleotide as shown in Fig. 3. These reductions were statistically significant (P < .05) and concentration dependent, with an EC₅₀ of 18 ± 5 pM. In contrast, C-type natriuretic peptide failed to alter evoked dopamine efflux in cells depleted of their normal complement of NPR-C. The two curves differed statistically (P < .01). Dopamine content of the cells was unchanged by the antisense treatment, averaging 221 ± 34 and 232 ± 42 ng of dopamine per culture in cells treated with either the missense or antisense oligodeoxynucleotide, respectively (P = .59).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of C-type natriuretic peptide (CNP) on evoked dopamine efflux after treatment with either the missense or antisense oligodeoxynucleotides (100 nM) for 2 days. The two curves differed significantly by ANOVA (**P < .01). All values are mean ± S.E. The number of preparations in each group is indicated by the N.

The function of the neuromodulatory pathway emanating from the NPR-C was assessed with an intracellular NPR-C fragment previously shown to produce neuromodulatory actions (Kanwal et al., 1999). The peptide fragment (Amino[1-15]) consisted of the following amino acids: RKKYRITIERRNHQE. These experiments required permeabilization of the cells with digitonin and catecholamine efflux was induced with 10 µM calcium. Basal dopamine release averaged 12.0 ± 2.9 and 11.8 ± 3.3% of total dopamine contents in permeabilized cells exposed to either the missense or antisense oligodeoxynucleotides, respectively. The addition of calcium increased dopamine efflux to 29.5 ± 1.8 and 28.6 ± 3.5% in the cells treated with either the missense or antisense oligodeoxynucleotides (data not shown). The elevation in dopamine efflux initiated by the calcium was statistically significant in both groups (P < .01). Figure 4 indicates that the active NPR-C fragment significantly suppressed evoked dopamine efflux in both groups of cells, indicating that the neuromodulatory pathway distal to the NPR-C remained intact. The effect of the NPR-C fragment was concentration dependent (P = .02), albeit extremely steep. The precipitous and potent concentration-response curve has been noted previously, although these effects are slightly more potent than our previous findings, indicating an EC₅₀ of 497 ± 97 fM (Kanwal et al., 1999).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effect of an active fragment of the NPR-C in permeabilized cells. The pentadecapeptide (Amino 1-15) corresponding to the first 15 amino acids of the cytosolic region of the NPR-C suppressed evoked neurotransmitter efflux equally in cells treated with either the missense or antisense oligodeoxynucleotides (100 nM) for 2 days. All values are means with the S.E. indicated for the antisense-treated cells. The S.E. was omitted for the cells treated with the missense oligodeoxynucleotide because of excessive overlap. The number of preparations in each group is indicated in parentheses.

An additional test of the specificity of the antisense knockdown investigated the neuromodulatory effect of L-NAME. The results are shown in Fig. 5. The L-NAME suppressed evoked dopamine efflux at concentrations of 20 and 200 µM regardless of the oligodeoxynucleotide treatment. The effects were concentration dependent (P = .001) but there was no statistical difference between the curves either in terms of absolute effects (P = .89) or slope (P = .78).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Effect of L-NAME on evoked dopamine efflux. The L-NAME attenuated evoked dopamine efflux equally in cells treated with either the missense or antisense oligodeoxynucleotides (100 nM) for 2 days. Values are mean ± S.E. The number of preparations per group is indicated by the N.

The antisense treatment also could influence guanylyl cyclase stimulation by eliminating natriuretic peptide uptake into the PC12 cells. Thus, we evaluated guanylyl cyclase activation after treatment with either the missense or antisense oligodeoxynucleotides. The results are shown in Fig. 6. Basal concentrations of cGMP averaged 6.6 ± 2.2 and 5.4 ± 1.6 pmol/culture flask treated with either the missense or antisense oligodeoxynucleotide, respectively. The stimulation of cGMP accumulation by C-type natriuretic peptide was essentially identical in the presence of the two oligodeoxynucleotides (Fig. 6). The two curves were not statistically distinguishable by ANOVA (P = .89). Furthermore, the curves possessed nearly identical slopes (P = .99). The C-type natriuretic peptide effects were concentration dependent (P = .008). The maximal stimulation of cGMP accumulation ranged from 3- to 20-fold in different cell passages, accounting for the large variability at the highest concentration of C-type natriuretic peptide.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Guanylyl cyclase activation by C-type natriuretic peptide (CNP). Elevations in cGMP concentrations were equivalent in cells treated with either the missense or antisense oligodeoxynucleotides (100 nM) for 2 days. All values are mean ± S.E. The number of preparations per group is indicated by the N.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results presented in this report provide strong support for the hypothesis that the NPR-C mediates biological responses to natriuretic peptides and is not restricted to the role of a transport protein. Prior work implicating the NPR-C as a transducing entity relied on the pharmacological selectivity of truncated natriuretic peptide analogs for the NPR-C. More recently, Anand-Srivastava et al. (1996) demonstrated that the cytoplasmic region of the NPR-C suppresses adenylyl cyclase activity in membrane preparations, providing strong support for the hypothesis that this receptor regulates adenylyl cyclase. These receptor fragments also attenuated evoked dopamine efflux from permeabilized cells (Kanwal et al., 1999) and activated GTP-binding proteins in guinea pig taenia coli (Murthy and Maklouf 1999), further strengthening the support for the NPR-C as a signaling entity. We reasoned that the strongest evidence favoring a signal-transducing role for this receptor would involve an inhibitor of the receptor eliminating the suspected biological activity. No NPR-C ligand consistently inhibits the receptor, and most NPR-C ligands have been reported to stimulate activity (Anand-Srivastava et al., 1990; Drewett et al., 1990; Isales et al., 1992). Therefore, we attempted to ablate the receptor with specific antisense oligodeoxynucleotides. The antisense treatments both suppressed receptor levels and eliminated neuromodulatory effects of C-type natriuretic peptide (Figs. 1 and 3). The NPR-C ablation failed to alter the neuromodulatory effects of either an active fragment of the NPR-C in permeabilized cells or L-NAME. Furthermore, guanylyl cyclase activation by C-type natriuretic peptide persisted after antisense oligodeoxynucleotide treatment. Thus, NPR-C ablation selectively ablates neuromodulatory effects of natriuretic peptides but does not produce nonspecific effects to alter either effects of other neuromodulators or guanylyl cyclase activation. These data provide the strongest support for a neuromodulatory role of the NPR-C.

Biological actions of natriuretic peptides characteristically are attributed to activation of guanylyl cyclase, leading to elevated cGMP concentrations, protein kinase G activation, and cellular responses (Levin et al., 1998). In contrast, we consistently have observed natriuretic peptides to act independently of alterations of cGMP concentrations in PC12 cells to suppress evoked neurotransmitter efflux (Drewett et al., 1990; Trachte and Drewett, 1994). Specifically, an NPR-C selective ligand, cyclic atrial natriuretic peptide (cANP), mimicked the effects of ANP in PC12 cells without altering cGMP concentrations (Drewett et al., 1990). Furthermore, C-type natriuretic peptide suppressed evoked dopamine efflux in the absence of changes in cGMP concentrations (Trachte et al., 1995), as observed in the present study (compare effective concentrations in Figs. 3 and 6). Natriuretic peptides also use the NPR-C to suppress catecholamine efflux in response to nicotinic receptor stimulation of bovine adrenal chromaffin cells (Babinski et al., 1995), although guanylyl cyclase stimulation also has been invoked as the dominant mechanism (Rodriguez-Pascual et al., 1996). These discrepancies necessitated a more definitive test of the mechanism responsible for the neuromodulatory effect of natriuretic peptides. The dearth of potent selective antagonists of the three recognized NPRs required alternate approaches to this problem. The knockdown of the NPR-C with oligodeoxynucleotides presented the best opportunity to accomplish a selective ablation.

Recent in vivo studies using mice deficient in either ANP or GC-A also prompted this study. Mice lacking ANP exhibit a sodium-dependent hypertension, characterized by an inability of sodium to suppress renin release (Melo et al., 1998). In contrast, mice lacking NPR-A exhibit a sodium-independent form of hypertension (Oliver et al., 1997). The difference in sodium sensitivity of these two strains has been suggested to involve a natriuretic peptide receptor other than NPR-A mediating the suppression of renin release (Melo et al., 1998). Indeed, Devlin and Leckie (1994) have demonstrated an inhibitory effect of natriuretic peptides, including the NPR-C selective ligand cANP on renin efflux from a nephroblastoma cell line. The implication from their work is that the NPR-C is the NPR-regulating renin release. Collectively, these reports provide support for biologically active NPRs in addition to the particulate guanylyl cyclase receptors. The NPR-C represents an attractive candidate in mediating inhibitory effects on renin release and neurotransmission.

An unexpected finding of our study involved the effect of NPR-C depletion to suppress evoked dopamine efflux. As indicated in Fig. 2, NPR-C depletion reduced evoked dopamine efflux by one-third. This result might indicate that the NPR-C is involved in exocytosis of neurotransmitter. The NPR-C is widely recognized as a transport protein facilitating the entry of natriuretic peptides into the cell interior (Maack et al., 1987). The results from this study could be consistent with the NPR-C operating in conjunction with the exocytotic apparatus to facilitate the egress of neurotransmitter as well. This mechanism was not explored but the observation indicates the potential of this receptor to influence neurotransmission.

The possibility that NPR-C agonists, such as C-type natriuretic peptide, suppress evoked neurotransmitter efflux by down-regulating surface NPR-C receptors in a similar manner to the antisense treatment is unlikely because active fragments of the receptor also suppress evoked dopamine efflux. Furthermore, these fragments were active even after NPR-C depletion with antisense oligodeoxynucleotides (Fig. 4). Another neuromodulator, L-NAME, also suppressed evoked dopamine efflux after antisense depletion of NPR-C. Therefore, the reduction in evoked dopamine efflux caused by the NPR-C antisense treatment did not preclude the effects of inhibitory neuromodulators but appeared to ablate natriuretic peptide effects selectively. The actual mechanism accounting for the neuromodulatory effect of the NPR-C was not discerned in this study but we have found the GTP-binding protein G_o to be essential for neuromodulatory activity (Takida et al., 1999).

Antisense treatments to eliminate the expression of a given protein have been observed to suppress functional responses more than protein expression (Mohuczy et al., 1999). The rationale for this divergence is usually attributed to the presence of spare receptors not coupled to the response. In this study, we observed a complete ablation of the neuromodulatory response to C-type natriuretic peptide, whereas protein concentrations were suppressed only 50% (compare Figs. 1 and 3). Presumably, the NPR-C remaining after antisense treatment lacked a functional connection to the neuromodulatory pathway. Cohen et al. (1996) reported that one-third of the NPR-C present in renal or transfected Chinese hamster ovary cells is intracellular. If a similar situation exists in PC12 cells, one would predict that at least one-third of the receptors present are not coupled to the neuromodulatory pathway involving C-type natriuretic peptide because they would be inaccessible to the extracellular peptide. Such a scenario might explain the divergence between receptor expression and functional responses observed in this study.

One expected and important observation of this study involves the sustained guanylyl cyclase activation by C-type natriuretic peptide after NPR-C depletion (Fig. 6). Particulate guanylyl cyclases have been perceived to respond to natriuretic peptides independently of the NPR-C; however, other investigators have suggested the possibility of heteromers involving both guanylyl cyclases and the NPR-C. Our observation of intact guanylyl cyclase activity after NPR-C reduction is a novel observation to the best of our knowledge. Previous studies using transfected guanylyl cyclases typically used tissues possessing the NPR-C receptor; therefore, they did not address the specific issue of NPR-C interactions with guanylyl cyclases.

This study has defined the NPR-C as essential for C-type natriuretic peptide effects on evoked neurotransmitter efflux. Novel aspects of the study include the following: 1) an effective oligodeoxynucleotide sequence to suppress NPR-C expression, 2) an inhibitory effect of NPR-C depletion on evoked neurotransmitter efflux, 3) an elimination of neuromodulatory effects of natriuretic peptides on evoked neurotransmitter efflux after NPR-C depletion, and 4) the persistence of guanylyl cyclase activation by C-type natriuretic peptide after NPR-C depletion. These data indicate that the NPR-C appears to participate in the evoked efflux of neurotransmitters and mediates natriuretic peptide effects to attenuate evoked neurotransmitter efflux. Thus, the NPR-C is not solely a transport protein but appears to participate in cellular signaling mechanisms as well.