Identification of the G Protein-activating Domain of the Natriuretic Peptide Clearance Receptor (NPR-C)*

We have shown recently that the 37-amino acid intracellular domain of the single-transmembrane, natriuretic peptide clearance receptor, NPR-C, which is devoid of kinase and guanylyl cyclase activities, activates selectively Gi1 and Gi2 in gastric and tenia coli smooth muscle. In this study, we have used synthetic peptide fragments of the N-terminal, C-terminal, and middle regions of the cytoplasmic domain of NPR-C to identify the G protein-activating sequence. A 17-amino acid peptide of the middle region (Arg469–Arg485), denoted Peptide 4, which possesses two N-terminal arginine residues and a C-terminalB-B-X-X-Bmotif (where B and X are basic and non-basic residues, respectively) bound selectively to Gi1 and Gi2, activated phospholipase C-β3 via the βγ subunits, inhibited adenylyl cyclase, and induced smooth muscle contraction, in similar fashion to the selective NPR-C ligand, cANP4–23. A similar sequence (Peptide 3), but with a partial C-terminal motif, had minimal activity. Sequences which possessed either the N-terminal basic residues (Peptide 1) or the C-terminalB-B-X-X-Bmotif (Peptide 2) were inactive. Peptide 2, however, inhibited G protein activation and cellular responses mediated by the stimulatory Peptide 4 and by cANP4–23, suggesting that theB-B-X-X-Bmotif mediated binding but not activation of G protein, thus causing Peptide 2 to act as a competitive inhibitor of G protein activation.

(eNOS) (9), NPR-C selectively bound G i1 and G i2 (G i1 Ͼ G i2 ), activated eNOS, and inhibited adenylyl cyclase via the ␣ subunits of both G proteins and activated PLC-␤ presumably via the ␤␥ subunits (7,8). A synthetic peptide corresponding to the 37-amino acid intracellular domain of NPR-C inhibited adenylyl cyclase activity in cardiac membranes in a pertussis toxinsensitive fashion implying that this domain was the locus of G protein binding and activation (10). The specific motifs within this domain responsible for G protein activation have not been identified.
Both single-and multitransmembrane receptors possess intracellular sequences capable of activating G proteins. Okamoto et al. (11,12) have identified a 14-amino acid intracellular sequence (Arg 2410 -Lys 2423 ) of the human insulin-like growth factor (IGF) II/mannose 6-phosphate receptor that activates G proteins with an order of potency of G i2 Ͼ G i1 ϭ G i3 Ͼ G o . The sequence is characterized by the presence of two N-terminal basic residues and a C-terminal B-B-X-X-B motif, where B and X represent basic and non-basic residues, respectively. A 9amino acid peptide sequence lacking the N-terminal basic residues inhibited activation of G proteins by both IGF-II and the 14-amino acid stimulatory peptide (13). Recently, a similar G i2 -activating sequence was identified in the C-terminal region of the 7-to 11-transmembrane polycystin-1 receptor (14). A consensus sequence (Arg 259 -Lys 273 ) present in the terminal region of third cytoplasmic loop of the seven-transmembrane ␤ 2 -adrenergic receptor couples preferentially to G s ; phosphorylation of Ser 262 by cAMP-dependent protein kinase decreases coupling to G s and enhances coupling to G i1 (15)(16)(17).
In the present study, we have used peptide fragments corresponding to the N-terminal, C-terminal, and middle regions of the cytoplasmic domain of NPR-C to determine the locus of G protein binding and activation. A 17-amino acid peptide of the middle region (Arg 469 -Arg 485 ), which possesses the consensus sequence B-B. . . . . . . B-B-X-X-B, was shown to bind selectively to G i1 and G i2 , activate PLC-␤3 via the ␤␥ subunits, and inhibit adenylyl cyclase in similar fashion to the selective NPR-C ligand, cANP4 -23. A C-terminal peptide (Gly 479 -Ala 496 ), which included the B-B-X-X-B motif at its N-terminal, inhibited activation by the stimulatory peptide and cANP4 -23.

EXPERIMENTAL PROCEDURES
Synthesis of Partial Sequences Corresponding to the Cytoplasmic Domain of NPR-C-Four peptide fragments corresponding to the Nterminal region (Peptide 1, Arg 460 -Arg 470 ), C-terminal region (Peptide 2, Gly 479 -Ala 496 ), and middle region (Peptide 3, Ile 467 -Arg 482 ; and Peptide 4, Arg 469 -Arg 485 ) of the 37-amino acid cytoplasmic domain of NPR-C were synthesized by the solid phase method and highly (95-99%) purified by high performance liquid chromatography (Chiron Technologies) (Fig. 1). The lyophilized synthetic peptides were dissolved in distilled water.
Preparation of Freshly Dispersed and Cultured Smooth Muscle Cells-Muscle cells were isolated from guinea pig tenia coli by sequential enzymatic digestion, filtration, and centrifugation as described previously (7,9). After washing, the cells were allowed to disperse spontaneously for 30 min and then harvested by filtration through 500-m Nitex and centrifuged twice at 350 ϫ g for 10 min. In some experiments, the cells were permeabilized by incubation for 5 min with saponin (35 g/ml) in a low Ca 2ϩ (100 nM) medium as described previously (7) and resuspended in saponin-free medium with 1.5 mM ATP and ATP-regenerating system (5 mM creatine phosphate and 10 units/ml creatine phosphokinase).
Dispersed muscle cells were cultured as described previously (7,9) in Dulbecco's modified Eagle's medium containing 10% fetal calf serum. The muscle cells in confluent primary cultures were trypsinized, replated at a concentration of 2.5 ϫ 10 5 cells/ml, and cultured under the same conditions. All experiments were done on cells in first passage.
Identification of Receptor-activated G Proteins in Solubilized Membranes-G proteins selectively activated by the synthetic peptides were identified by an adaptation of the method of Okamoto et al. (18), as described previously (19,20). Cultured muscle cells (2 ϫ 10 6 cell/ml) were homogenized in 20 mM HEPES medium (pH 7.4). After centrifugation at 25,000 ϫ g for 15 min, the membranes were solubilized at 4°C in 20 mM HEPES medium (pH 7.4) and 1% CHAPS. The membranes were incubated with 60 nM [ 35 S]GTP␥S in a medium containing 10 mM HEPES (pH 7.4), 100 M EDTA, and 10 mM MgCl 2 for 20 min at 37°C in the presence or absence of peptides (100 M). The reaction was stopped with 10 volumes of 100 mM Tris-HCl medium (pH 8.0) containing 10 mM MgCl 2 , 100 mM NaCl, and 20 M GTP, and the solubilized membranes were incubated for 2 h on ice in wells precoated with specific antibodies to G i1 ␣, G i2 ␣, G i3 ␣, G s ␣, and G q/11 ␣. The wells were washed three times with phosphate buffer containing 0.05% Tween 20, and the radioactivity from each well was counted.
Assay of PLC-␤ Activity in Muscle Membranes-PLC activity was determined as described previously (19) by a modification of the method Uhing et al. (21) in membranes from cultured tenia coli muscle cells prelabeled with myo[ 3 H]inositol. The assay was initiated by addition of 0.4 mg of membrane protein to 25 mM Tris-HCl (pH 7.5), 0.5 mM EGTA, 10 mM MgCl 2 , 300 nM free Ca 2ϩ , 100 M GTP, 5 mM phosphocreatine, 50 units/ml creatine phosphokinase, in a total volume 0.4 ml. After incubation at 31°C for 60 s, the reaction was terminated with 0.6 ml 25% trichloroacetic acid. The supernatant was extracted four times with 2 ml of diethyl ether, and the amount of labeled inositol phosphates in the aqueous phase counted. The trichloroacetic acid-soluble radioactivity at time 0 (100 -150 cpm) was subtracted from all values. PLC activity was expressed as counts/min/mg of protein.
cAMP Assay-Cyclic AMP was measured by radioimmunoassay as described previously (7). Forskolin (10 M) was added to 0.5 ml of cell suspension (10 6 cells/ml) in the presence of 10 M isobutylmethylxanthine, either alone or in combination with various peptides (100 M). The reaction was terminated after 60 s. The results were expressed as picomoles/10 6 cells.
Measurement of Contraction in Permeabilized Muscle Cells-Contraction was measured in permeabilized muscle cells by scanning micrometry, as described previously (7). A 0.25-ml aliquot of cells (10 4 cells/ml) was added to 0.1 ml of medium containing cANP4 -23 (1 M) or various concentrations of partial peptide sequences, and the reaction was terminated after 30 s with 1% acrolein. The effect of the partial peptide sequences on maximal contraction induced by cANP4 -23 was also determined. Under each condition, the lengths of treated muscle cells were compared with the lengths of untreated control cells. Contraction was expressed in micrometers as the mean decrease in cell length from control.
Materials-125 I-cAMP, [ 35 S]GTP␥S, and myo[ 3 H]inositol were obtained from NEN Life Science Products; polyclonal antibodies to G proteins and PLC-␤ isoforms from Santa Cruz Biotechnology; and all other chemicals from Sigma.

Selective Activation of G i1 and G i2 by Peptide Sequences of the Cytoplasmic Domain of NPR-C-
The ability of peptide sequences to activate specific G proteins in solubilized tenia coli smooth muscle membranes was determined from the increase in the binding of [ 35 S]GTP␥S.G␣ complexes to the corresponding G␣ antibodies. At a concentration of 100 M, Peptide 4 (Arg 469 -Arg 485 : RRTQQEESNLGKHRELR; basic residues in bold) significantly increased the binding of [ 35 S]GTP␥S to G i1 ␣ (216 Ϯ 22%) and G i2 ␣ (347 Ϯ 53%), but not to G i3 ␣, G s ␣, or G q/11 ␣ (Table I). Peptide 4 possessed two N-terminal Arg residues and a C-terminal B-B-X-X-B motif, where B and X are basic and non-basic residues, respectively. Peptide 3 (Ile 467 -Arg 482 : IERRTQQEESNLGKHR), which lacked the full Cterminal motif of Peptide 4, was less effective increasing the binding of [ 35 S]GTP␥S to G i1 ␣ and G i2 ␣ by 94 Ϯ 15% and 57 Ϯ 14%, respectively. Peptide 1 (Arg 460 -Arg 470 ) and Peptide 2 (Gly 479 -Ala 496 ) had no effect.
Effect of Peptides on Basal and cANP4 -23-stimulated PLC-␤ Activity-The ability of the peptides to stimulate basal PLC-␤ activity ([ 3 H]inositol phosphate formation) in membranes derived from cultured tenia coli smooth muscle cells paralleled their ability to activate G i1 and G i2 . Peptide 4 increased basal PLC-␤ activity in a concentration-dependent fashion (0.1-100 M) with an EC 50 of 1.3 Ϯ 0.4 M, whereas Peptide 3 was only effective at the highest concentration (100 M) (Fig. 2). The N-terminal Peptide 1 and the C-terminal Peptide 2 had no effect on basal PLC-␤ activity (Fig. 2).
cANP4 -23 increased PLC-␤ activity in a concentration-dependent fashion with an EC 50 of 0.8 Ϯ 0.2 nM and a threshold concentration of Ͻ1 pM. Peptide 4, at an EC 50 concentration of 1 M, augmented the PLC-␤ response to cANP4 -23, shifting the concentration-response curve to the left (Fig. 3). In contrast, Peptide 2 (10 M), which had no effect on basal PLC-␤ activity, inhibited the PLC-␤ response to cANP4 -23, shifting the concentration-response curve to the right (Fig. 3).
Effect of Peptides on Muscle Cell Contraction-The ability of Peptides 1-4 to induce contraction paralleled their ability to stimulate phosphoinositide hydrolysis. Contraction was measured in saponin-permeabilized tenia coli smooth muscle cells by scanning micrometry and expressed as mean decrease in muscle cell length from control. Peptide 4 induced contraction in a concentration-dependent fashion with an EC 50 of 0.4 Ϯ 0.1 M and a maximal contraction of 22.6 Ϯ 0.4 m, whereas Peptide 3 was effective only at high concentrations (maximal contraction 7.2 Ϯ 0.5 m) (Fig. 6). Peptides 1 and 2 had no significant effect on muscle cell length. However, when added in combination with Peptide 4, Peptide 2 (100 M) inhibited Peptide 4-induced maximal contraction by 60 Ϯ 8% (p Ͻ 0.01).
cANP4 -23 induced contraction of muscle cells in a concentration-dependent fashion with an EC 50 of 0.6 Ϯ 0.3 nM and a threshold concentration of 1 pM. Peptide 4, at a near-EC 50 concentration of 0.1 M, augmented contraction induced by cANP4 -23, shifting the concentration-response curve to the left (Fig. 7). In contrast, Peptide 2 (1 M), which had no effect on contraction, inhibited contraction induced by cANP4 -23, shifting the concentration-response curve to the right (Fig. 7).
The effects of various concentrations of Peptide 4 and Pep-tide 2 on maximal contraction induced by cANP4 -23 were also tested. As shown in Fig. 8, Peptide 4 augmented contraction induced by cANP4 -23 in a concentration-dependent fashion. In contrast, Peptide 2 inhibited contraction induced by cANP4 -23 in a concentration-dependent fashion with a maximal inhibition of 52 Ϯ 6% (Fig. 8). Peptides 1 and 3 had no effect on cANP4 -23-induced contraction. Effect of Peptides on Adenylyl Cyclase Activity-Freshly dispersed smooth muscle cells were used to examine the ability of Peptides 1-4 to inhibit forskolin-stimulated cAMP formation. At a concentration of 100 M, Peptide 4 and Peptide 3 inhibited forskolin-stimulated cAMP (19.1 Ϯ 1.1 pmol/10 6 cells) by 64 Ϯ 4% and 23 Ϯ 4%, respectively, whereas Peptides 1 and 2 had no effect (Fig. 9A). cANP4 -23 (0.1 M) also inhibited forskolinstimulated cAMP by 59 Ϯ 4%; the inhibition by cANP4 -23 was accentuated to 84 Ϯ 3% (p Ͻ 0.01) in the presence of Peptide 4 and attenuated to 35 Ϯ 4% (p Ͻ 0.01) in the presence of Peptide 2 (Fig. 9B). DISCUSSION This study shows that a 17-amino acid sequence in the middle region (Arg 469 -Arg 485 ) of the 37-amino acid intracellular  domain of the NPR-C accounts for the ability of this singletransmembrane receptor to activate pertussis toxin-sensitive G proteins in various tissues (2,7,10). The sequence possesses two N-terminal basic residues (Arg 469 , Arg 470 ), and the Cterminal motif, B-B-X-X-B (where B ϭ basic and X ϭ nonbasic residue) (Fig. 1). A synthetic peptide with this sequence (denoted Peptide 4 in this study) activated selectively G i1 and G i2 in tenia coli smooth muscle, stimulated phosphoinositide hydrolysis by activating PLC-␤3 via the ␤␥ subunits of both G proteins, inhibited adenylyl cyclase activity via the ␣ subunits, and induced muscle contraction, mimicking in all instances the properties of the selective NPR-C ligand, cANP4 -23. The peptide also enhanced the ability of cANP4 -23 to activate G i1 and G i2 , stimulate phosphoinositide hydrolysis, induce contraction, and inhibit forskolin-stimulated cAMP. The effects of Peptide 4 alone and in combination with cANP4 -23 were concentration-dependent.
A C-terminal peptide (denoted Peptide 2), which included the B-B-X-X-B motif at its N-terminal (Fig. 1), had no effect by itself but it blocked activation of G i1 and G i2 and all cellular responses induced by Peptide 4 and by cANP4 -23, suggesting that Peptide 2 bound to, but did not activate G i1 and G i2 , thus acting as a competitive inhibitor of G protein activation. The ability of Peptide 2 to inhibit responses to Peptide 4 and cANP4 -23 was concentration-dependent.
The muscle cells were highly sensitive to cANP4 -23 (EC 50 0.8 Ϯ 0.2 and 0.6 Ϯ 0.3 nM for activation of PLC-␤ and stimulation of muscle contraction, respectively). At an EC 50 concentration, Peptide 4 augmented the PLC-␤ and contractile responses to all concentrations of cANP4 -23 (Figs. 3 and 7). The augmentation was additive suggesting additional recruitment of G i1 and G i2 by Peptide 4. Peptide 2 inhibited PLC-␤ and contractile responses to all concentrations of cANP4 -23 (Figs.  3 and 7).
The synthetic peptides were designed to include or exclude specific residues in the stimulatory consensus sequence. Thus, Peptide 1, which included at its C terminus the two arginine residues present in the N terminus of Peptide 4, had no effect. Peptide 3, which closely resembled Peptide 4 and included the two N-terminal arginine residues but only a part (i.e. B-B) of the C-terminal B-B-X-X-B motif, was only partially active at the highest concentrations (100 M), emphasizing the requirement for a complete N-terminal motif. Peptide 2, which retained the complete B-B-X-X-B motif at its C terminus, maintained the ability to bind but not activate G proteins: the pattern emphasized the significance of the location of the B-B-X-X-B motif at the C terminus, as well as the requirement for N-terminal arginine residues.
As noted above (1, 2), NPR-C, unlike NPR-A or NPR-B, is devoid of an intracellular guanylyl cyclase domain. Its truncated intracellular sequence possesses a G i1 /G i2 binding domain that induces activation or inhibition of other effector enzymes. Activation of NPR-C in tenia coli smooth muscle causes inhibition of adenylyl cyclase and activation of PLC-␤3, resulting in stimulation of inositol 1,4,5-trisphosphate-dependent Ca 2ϩ release and muscle contraction (8). The activation of PLC-␤3 is mediated by the ␤␥ subunits of G i1 and G i2 ; this conforms to a pattern of preferential activation of this PLC-␤ isozyme by the ␤␥ subunits of inhibitory G proteins, as shown for other smooth muscle receptors coupled to G i1 (somatosta-  23), and G i3 (adenosine A1 (24), muscarinic m2 (20), and purinergic P 2Y2 receptors (19)). Activation or inhibition of other regulatory enzymes involved in cell signaling by the ␤␥ subunits of G proteins has been well documented (25)(26)(27)(28)(29).
Unlike tenia coli smooth muscle cells, gastric and intestinal smooth muscle cells express eNOS (9). Activation of NPR-C in these cells results in preferential activation of eNOS by G i1 /G i2 ; the formation of nitric oxide causes sequential activation of soluble guanylyl cyclase and cGMP-dependent protein kinase and results in muscle relaxation (7). Thus, although NPR-C is devoid of a membrane-bound guanylyl cyclase domain, its activation by natriuretic peptides can result in relaxation of gastric and intestinal muscle that expresses a G protein-dependent, constitutive NOS (7,9). It is probable that the same 17-amino acid consensus sequence represented by Peptide 4 is responsible for the ability of NPR-C to bind G i1 and G i2 in gastric smooth muscle and preferentially activate eNOS and inhibit adenylyl cyclase (7). The potency of the G i2 /G i1 -activating sequence of NPR-C (EC 50 ϳ1 M for activation of PLC-␤3 and ϳ0.5 M for stimulation of muscle contraction) was similar to that of the G i2activating sequence located in the C-terminal region of the multitransmembrane polycystin-1 receptor (14). The potency of both sequences may be related to the presence of dual Nterminal arginine residues and arginine or lysine residues in the C-terminal motif. As noted by Okamoto et al. (11), substitution of one basic residue for another altered the potency of the G i2 -activating sequence of IGF II/mannose 6-phosphate receptor in the order of arginine Ͼ lysine Ͼ histidine.
Preferential activation of G i2 appears to be a common feature of the intracellular consensus sequences not only of NPR-C, but also of the polycystin-1 and the IGF II receptors (11,14). A similar consensus in the terminal region of the third cytoplasmic loop of the ␤ 2 -adrenergic receptor couples preferentially to G s ; phosphorylation of Ser 262 by cAMP-dependent protein kinase decreased affinity for G s and enhanced coupling to G i1 (15)(16)(17). It is possible that phosphorylation of serine residues in the middle or C-terminal regions of the intracellular domain of NPR-C could alter its coupling to G proteins.
NPR-C is the predominant natriuretic peptide receptor in visceral and vascular smooth muscle and possesses high affinity for all natriuretic peptides (7,30,31). Binding of these peptides to NPR-C, as a prelude to their recycling and degradation, activates G protein-dependent pathways linked to several effector enzymes. In smooth muscle cells (e.g. gastric and intestinal smooth muscle) that express eNOS, G protein activation leads to NO formation and muscle relaxation. In smooth muscle cells devoid of eNOS (e.g. tenia coli), G protein activation leads to phosphoinositide hydrolysis and muscle contraction.