GIPC interacts with the β1-adrenergic receptor and regulates β1-adrenergic receptor-mediated ERK activation

β1-adrenergic receptors, expressed at high levels in the human heart, have a carboxyl-terminal ESKV motif that can directly interact with PDZ domain-containing proteins. Using the β1-adrenergic receptor carboxyl terminus as bait, we identified the novel β1-adrenergic receptor-binding partner GIPC in a yeast two-hybrid screen of a human heart cDNA library. Here we demonstrate that the PDZ domain-containing protein, GIPC, co-immunoprecipitates with the β1-adrenergic receptor in COS-7 cells. Essential for this interaction is the Ser residue of the β1-adrenergic receptor carboxyl-terminal ESKV motif. Our data also demonstrate that β1-adrenergic receptor stimulation activates the mitogen-activated protein kinase, ERK1/2. β1-adrenergic receptor-mediated ERK1/2 activation was inhibited by pertussis toxin, implicating Gi, and was substantially decreased by the expression of GIPC. Expression of GIPC had no observable effect on β1-adrenergic receptor sequestration or receptor-mediated cAMP accumulation. This GIPC effect was specific for the β1-adrenergic receptor and was dependent on an intact PDZ binding motif. These data suggest that GIPC can regulate β1-adrenergic receptor-stimulated, Gi-mediated, ERK activation while having no effect on receptor internalization or Gs-mediated cAMP signaling.

␤ 1 -adrenergic receptors, expressed at high levels in the human heart, have a carboxyl-terminal ESKV motif that can directly interact with PDZ domain-containing proteins. Using the ␤ 1 -adrenergic receptor carboxyl terminus as bait, we identified the novel ␤ 1 -adrenergic receptor-binding partner GIPC in a yeast two-hybrid screen of a human heart cDNA library. Here we demonstrate that the PDZ domain-containing protein, GIPC, co-immunoprecipitates with the ␤ 1 -adrenergic receptor in COS-7 cells. Essential for this interaction is the Ser residue of the ␤ 1 -adrenergic receptor carboxyl-terminal ESKV motif. Our data also demonstrate that ␤ 1 -adrenergic receptor stimulation activates the mitogen-activated protein kinase, ERK1/2. ␤ 1 -adrenergic receptor-mediated ERK1/2 activation was inhibited by pertussis toxin, implicating G i , and was substantially decreased by the expression of GIPC. Expression of GIPC had no observable effect on ␤ 1 -adrenergic receptor sequestration or receptor-mediated cAMP accumulation. This GIPC effect was specific for the ␤ 1 -adrenergic receptor and was dependent on an intact PDZ binding motif. These data suggest that GIPC can regulate ␤ 1 -adrenergic receptorstimulated, G i -mediated, ERK activation while having no effect on receptor internalization or G s -mediated cAMP signaling.
␤-Adrenergic receptors (␤ARs) 1 play a critical role in the regulation of cardiovascular function (1). There are three subtypes of ␤AR, ␤ 1 AR, ␤ 2 AR, and ␤ 3 AR. Both the ␤ 1 AR and ␤ 2 AR subtypes are expressed in the heart and appear to regulate cardiac function through similar intracellular signaling pathways (2). These receptors are members of the seven transmembrane spanning receptor superfamily and are thought to signal largely through heterotrimeric G proteins; primarily G s . Acti-vation of G s stimulates adenylate cyclase increasing intracellular cAMP. Increased cAMP activates cAMP-dependent protein kinase (PKA), which is believed to mediate many of the cellular responses associated with ␤AR activation (3,4).
Despite similarities in signaling, there are several striking physiological differences between the ␤ 1 AR and ␤ 2 AR subtypes (5,6). For example, ␤ 1 AR and ␤ 2 AR subtypes play opposing roles in regulating cardiac myocyte apoptosis: stimulation of the ␤ 1 AR increases apoptosis, whereas stimulation of the ␤ 2 AR inhibits apoptosis (7,8). ␤ 1 AR-induced apoptosis is implicated in the transition from cardiac hypertrophy to heart failure (9,10). The differential regulation of cardiac cell survival by these ␤AR subtypes has been partially explained by the ability of the ␤ 2 AR to couple to the pertussis toxin (PTX)-sensitive G protein, G i (11). ␤ 2 ARs in human and mouse cardiomyocytes have been shown to activate both G s and G i , whereas ␤ 1 AR stimulation has, thus far, only been shown to activate G s (12). Activation of the ␤ 2 AR, stimulates a G i -mediated PI3K-Akt-dependent cellsurvival signaling pathway and prevents cardiomyocytes from undergoing G s -mediated apoptosis (11). Recent studies suggest that G s to G i switching by the ␤ 2 AR is the consequence of PKA-mediated receptor phosphorylation (13,14). ␤ 2 AR/G i coupling has also been implicated in receptor-meditated ERK activation (13,15). ␤ 2 AR-stimulated ERK1/2 activation has been reported in cultured HEK-293 and COS-7 cells, and in isolated cardiac myocytes (16,17). This ␤ 2 AR-stimulated ERK1/2 activation is mediated by ␤␥ subunits of PTX-sensitive G proteins (G i ) through a pathway involving the non-receptor tyrosine kinase c-Src, small G protein Ras and Raf-1 kinase (15,18). The mechanism(s) of ␤ 1 AR-stimulated ERK activation is somewhat more controversial. A number of groups have reported that the ␤ 1 AR is unable to stimulate ERK activation, a conclusion based on the perceived inability of the ␤ 1 AR to couple to G i (19,20). However, recent data from cardiac myocytes suggest that the ␤ 1 AR can activate ERK and p38 in a G i -dependent manner; even though the ␤ 1 AR is less potent in stimulating ERK activation than the ␤ 2 AR (11,21). The mechanisms behind the observed differential activation of ERK are still unknown.
Recent data suggest that ␤ARs may transduce signals that are both dependent and independent of heterotrimeric G proteins (22,23). Well known receptor-interacting proteins, such as ␤-arrestins and G protein-coupled receptor kinases (GRKs), which were thought previously only to regulate receptor desensitization, have been implicated in signaling pathways that may be G protein-independent (24 -26). Another G proteinindependent mechanism underlying ␤AR-mediated cell signaling has recently been demonstrated. Both ␤ 1 ARs and ␤ 2 ARs have carboxyl-terminal PDZ binding motifs reported to bind to PDZ domain-containing proteins (27,28). The ␤ 2 AR has been shown to interact with Na ϩ /H ϩ exchanger regulatory factors (NHERF) in an agonist-dependent manner via this PDZ binding domain (27); while the ␤ 1 AR interacts with the postsynaptic density-enriched proteins, PSD-95 and MAGI-2 (28,29). Association with PSD-95 regulates agonist stimulated ␤ 1 AR internalization and may also provide a molecular mechanism by which ␤ 1 ARs are localized to the synapse, regulating synaptic plasticity (28). Interactions between the ␤ 1 AR and PDZ domain-containing proteins can be regulated by GRK5 (30). Furthermore, studies in cardiac myocytes have recently demonstrated that the PDZ binding motif of the ␤ 1 AR modulates receptor trafficking and signaling (31).
In order to identify proteins that interact with the PDZ binding domain of the ␤ 1 AR, we performed a yeast two-hybrid screen on a human heart cDNA library, using the ␤ 1 AR carboxyl-terminal tail as bait. From this screen we identified GIPC (GAIP-interacting protein, carboxyl (C) terminus), a PDZ-containing protein, as a novel binding partner of the ␤ 1 AR. Here we characterize this interaction and its functional consequences.
Polyclonal (A-14) and monoclonal (9E10) anti-Myc antibodies, and polyclonal anti-ERK2 antibody were from Santa Cruz Biotechnologies. Anti-FLAG M2 and Anti-FLAG M2 affinity gel were from Sigma. Polyclonal anti-phospho-ERK1/2 antibody was from Cell Signaling Technology. Horseradish peroxidase (HRP)-conjugated anti-mouse IgG and anti-rabbit IgG secondary antibodies were from Amersham Biosciences.
Yeast Two-hybrid Screening-Plasmid pAS2-1/␤ 1 AR-CT was used as bait to screen a human heart cDNA library (in pACT2) (Clontech). Bait plasmid and library cDNAs were co-transformed into the yeast strain PJ69 -4A using a standard yeast transformation protocol. Yeast were plated on selective medium (S.D.-Leu/Trp/His, ϩ10 mM 3AT) and allowed to grow for 4 -6 days at 30°C. Positive colonies were then restreaked on selective medium (S.D.-Leu/Trp/Ade or S.D.-Leu/Trp/His) plates. Plasmid DNA was rescued from positive colonies that grew on both -His and -Ade plates and subject to further sequence analysis. To further confirm positive interactions, isolated library cDNAs were cotransformed back into yeast together with a bait plasmid, either pAS2-1/␤ 1 AR-CT, empty vector pAS2-1, or other test plasmids as indicated in the text. Yeast was then subjected to growth tests on selective plates.
Cell Culture and Transfection-All tissue culture media and related reagents were purchased from Invitrogen. COS-7 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin in a 37°C incubator under 5% CO 2 . Cells in 100-mm dishes were transfected with Lipo-fectAMINE (at 4:1 ratio with DNA) according to the manufacturer's protocol. HEK-293 cells were maintained under the same conditions as COS-7 cells except minimal essential medium (MEM) was used. HEK-293 cells were transfected with FuGENE 6 (at 2.5:1 ratio with DNA) according to the manufacturer's protocol (Roche Applied Science). After transfection, cells were grown 36 -48 h before agonist stimulation.
Cellular Immunoprecipitations and Immunoblotting-Transfected COS-7 cells in 100-mm plates were incubated in serum-free medium for 60 min before agonist stimulation. The cells were then incubated in the presence or absence of 10 M isoproterenol (ISO) for 10 min at 37°C. The cells were rinsed with cold Dulbecco's phosphate-buffered saline (DPBS) and replaced with 1 ml of cross-linking buffer (DPBS containing 10 mM HEPES, pH 7.4, and 2.5 mM of the cell permeable cross-linking reagent dithiobis(succinimidyl propionate) (DSP, Pierce)). For the stimulated cells, 10 M ISO was present in the cross-linking buffer. The cells were incubated for 30 min at room temperature with continuous slow rocking. The cross-linking reaction was terminated by quickly removing the cross-linking buffer and replacing it with 1 ml of ice-cold lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 2 mM EDTA, 10% glycerol, 0.5% Nonidet P-40, and protease inhibitors). The cells were lysed by incubation on ice for 30 min and then clarified by centrifugation at 21,000 ϫ g for 12 min at 4°C. The clarified supernatants were then used in cellular coimmunoprecipitation experiments. 50 l of each supernatant was diluted into an equal amount of 3ϫ SDS-PAGE sample buffer to serve as whole cell extract controls.
1 ml of clarified cell extract was then incubated with 25 l of anti-FLAG M2 affinity gel slurry at 4°C with gentle rotation for 4 h to overnight. Beads were washed four to five times with ice-cold lysis buffer, and the bound proteins were eluted with 50 l of 3ϫ SDS-PAGE sample buffer. Equivalent amounts of sample in 3ϫ sample buffer were resolved on precast 4 -20% polyacrylamide gels (Invitrogen) and then transferred to nitrocellulose membranes by semi-dry blotting. Nitrocellulose membranes were blocked with 5% fat-free milk in TBST (20 mM Tris, pH 7.4, 500 mM NaCl, and 0.1% Tween 20) and incubated with the appropriate primary antibody at room temperature for 1 h. After washing with TBST, membranes were incubated for 1 h with horseradish peroxidase-conjugated anti-mouse or rabbit IgG secondary antibody (1:2000). Protein bands were visualized via SuperSignal Chemiluminescence Substrate (Pierce). Quantitation of band density was performed with Bio-Rad Fluor-S MultiImager.
ERK Phosphorylation-Twenty-four hours after transfection, cells were split into 6-well dishes and then incubated in serum-free media (DMEM for COS-7 and MEM for HEK-293, 0.1% bovine serum albumin, 10 mM HEPES, pH 7.4) overnight before agonist stimulation. Agonist stimulation was performed at 37°C in serum-free media for the times indicated in the figure legends. The medium was removed, and cells were then solubilized in 3ϫ SDS-PAGE sample buffer. The whole cell lysates were sonicated and resolved by SDS-PAGE. ERK phosphorylation was detected by Western blot using anti-phospho-ERK1/2 antibody as described above. The total ERK2 in the cell lysates was detected using anti-ERK2 polyclonal antibody. Protein bands were visualized via SuperSignal Chemiluminescence substrate and quantitated with Bio-Rad Fluor-S MultiImager.
Receptor Internalization and Cyclic AMP Accumulation Assays-For receptor internalization assays, HEK-293 cells in 100-mm dishes were transiently transfected with pcDNA3/FLAG-␤ 1 AR in the presence and absence of Myc-GIPC. One day after transfection, cells were split into poly-D-lysine-coated 6-well plates (Biocoat) and grown overnight at 37°C. Cells were serum-starved for 1 h before stimulation with 10 M ISO for 30 min at 37°C. Cells were placed on ice and cell surface FLAG-tagged receptors were detected with anti-FLAG M2 antibody followed by FITC-conjugated anti-mouse IgG as described previously (35). Receptor internalization was defined as the percentage of cell surface receptors lost after agonist stimulation, measured by cell flow cytometry.
For cyclic AMP accumulation assays, HEK-293 cells in 100-mm dishes were transiently transfected with pcDNA3/FLAG-␤ 1 AR in the presence and absence of transfected Myc-GIPC. Twenty-four hours after transfection, cells were split into 12-well collagen-coated plates and then labeled with modified essential medium supplemented with 5% fetal bovine serum and 2 Ci/ml [ 3 H]adenine for 4 h to overnight. Cells were serum-starved for 30 min and then stimulated with ␤ 1 AR selective agonist dobutamine or 10 M forskolin for 10 min. Cyclic AMP accumulation was quantitated by chromatography and expressed as a percentage of 3 H incorporated into cyclic AMP as described previously (35).

GIPC Binds Specifically to the ␤ 1 AR Carboxyl-terminal
Tail-In our search for potential ␤ 1 AR binding partners, we used the ␤ 1 AR carboxyl-terminal tail (␤ 1 AR-CT) as bait in a yeast two-hybrid screen of a human heart cDNA library. From a total of 4 million independent colonies screened, 24 positive clones were obtained. Four of the positive clones yielded identical sequence encoding a portion of the gene for human GIPC (hGIPC), with an N-terminal truncation at residue 74. We also isolated rat GIPC from a yeast two-hybrid screen of rat brain. The isolated rat sequence (rGIPC) encoded full-length GIPC plus a portion of the 5Ј-untranslated region (Fig. 1).
The specificity of the interaction between GIPC and the ␤ 1 AR-CT was confirmed by further yeast two-hybrid analysis. GIPC clones were transformed back into yeast strain PJ69-4A together with baits (either ␤ 1 AR-CT or ␤ 2 AR-CT). Transformed yeast containing both bait and GIPC were then subjected to growth tests on selective medium. As shown in Fig. 1, only yeast hosting both GIPC and the ␤ 1 AR-CT were able to grow on both -His and -Ade plates. The interaction between the ␤ 1 AR and GIPC is specific, since the ␤ 2 AR-CT did not support yeast growth under the same conditions when co-transformed with GIPC.
The ␤ 1 -Adrenergic Receptor Interacts with GIPC Through Its Carboxyl-terminal PDZ Binding Motif-In order to determine whether the full-length ␤ 1 AR could associate with GIPC in the cell, COS-7 cells were co-transfected with FLAG-␤ 1 AR and Myc-GIPC. Immunoprecipitation of the ␤ 1 AR followed by Western blotting for Myc-GIPC revealed robust co-immunoprecipitation of a GIPC/␤ 1 AR complex (Fig. 2). Treatment of the cells with the ␤-adrenergic receptor agonist ISO had no significant effect on the GIPC/␤ 1 AR association (data not shown). Consistent with the yeast two-hybrid results, GIPC did not co-immunoprecipitate with the full-length FLAG-␤ 2 AR when the two proteins were co-expressed in COS-7 cells.
To examine the structural determinants of the GIPC/␤ 1 AR interaction, we used several ␤ 1 AR mutants, in which three out of the last five residues were replaced with Ala individually (28). Similar to that observed with the ␤ 1 AR/PSD-95 interaction, when the Ser residue at the Ϫ2 position of the ␤ 1 AR was mutated to either Ala (S475A) or Asp (S475D), the association of GIPC with the ␤ 1 AR could no longer be detected. When GIPC was co-expressed with the ␤ 1 AR mutant S473A, co-immunoprecipitation of GIPC was reduced only slightly relative to the wild-type ␤ 1 AR. However, the ␤ 1 AR mutant V477A, which is unable to bind with PSD-95, showed a very strong interaction with GIPC. These studies demonstrate by both yeast two-hybrid screening and cellular co-immunoprecipitation experiments that the GIPC specifically interacts with the ␤ 1 AR through the PDZ domain. Unlike the association of the ␤ 1 AR with PSD-95, the Val in the last position (Val-477) of the receptor is not required for the interaction with GIPC.
GIPC Has No Effect on Receptor-stimulated Cyclic AMP Accumulation-Next we examined the effect of GIPC on ␤ 1 ARmediated whole cell cAMP accumulation. HEK-293 cells were transiently transfected with wild-type ␤ 1 AR, and either Myctagged GIPC or vector alone. Dose-response curves of cAMP accumulation induced by the selective ␤ 1 AR agonist dobu-

FIG. 1. Identification of GIPC as a novel binding partner of the ␤ 1 AR carboxyl terminus in yeast two-hybrid screens.
A, domain structure of GIPC and two isolated GIPC clones. The ␤ 1 AR carboxyl terminus was used as bait to screen a human heart cDNA library in a yeast two-hybrid system. The GIPC clones isolated from a human library (hGIPC) encoded a partial sequence of GIPC; while the clone isolated from the rat library (rGIPC) encoded a full-length GIPC plus some 5Ј-untranslated region. The PDZ domain of GIPC has been highlighted for comparison. B, GIPC interacts specifically with the ␤ 1 AR-CT in a yeast two-hybrid system. Yeast strain PJ69 -4A was co-transformed with bait DNAs (pAS2-1, pAS2-1/␤ 1 ARCT, or pAS2-1/␤ 2 AR-CT) together with isolated preys. Yeast colonies were subjected to growth tests on selective medium -His (S.D.-Leu/Trp/His) or -Ade (S.D. -Leu/Trp/Ade). Yeast growth rates were rated from fast (ϩϩϩ) to slow (ϩ) to no growth (Ϫ).  Fig. 3. Similar to previous results with PSD-95, GIPC expression had no effect on ␤ 1 AR-mediated whole cell cAMP accumulation, changing neither the maximal level of cAMP accumulation nor the EC 50 . Because PDZ domain-containing proteins have previously been shown to regulate receptor internalization (28,29), we also examined the effect of GIPC on ␤ 1 AR internalization in HEK-293 cells. GIPC expression had no effect on agonist stimulated ␤ 1 AR internalization (data not shown).
␤ 1 AR-stimulated ERK Activation-In order to examine ␤ 1 AR-mediated ERK activation, we used a plasmid encoding an ERK2-GFP fusion protein. This allowed for GFP-ERK2 to be easily separated from endogenous ERK1/2 by SDS-PAGE. In COS-7 cells transiently transfected with pcDNA3/FLAG-␤ 1 AR and GFP-ERK2, ISO stimulation induced robust ERK phosphorylation/activation compared with non-stimulated cells (Fig.  4A). In control cells transfected with GFP-ERK2 alone, ISOstimulated ERK phosphorylation was not detected using phospho-ERK antibody.
Similar to previous studies of the ␤ 2 AR, ␤ 1 AR-stimulated ERK activation was a relatively fast process, peaking at around 5 min. At maximum, ISO-stimulated ERK activation was 6 -8 fold greater than that of non-stimulated control cells (Fig. 4B). After peaking at 5 min, ␤ 1 AR-stimulated ERK activation steadily decreased to ϳ50% of maximum at 15 min. ␤ 1 ARstimulated ERK activation was observed over a range of agonist concentrations. The dose response curve for ISO-stimulated ERK activation is shown in Fig. 4C.
Next we examined the effect of different agonists on ␤ 1 ARmediated ERK activation in COS-7 cells. The ␤ 1 AR has a high affinity for both norepinephrine and epinephrine (2). As predicted, norepinephrine (NE) stimulation dramatically increased ERK activation (Fig. 4D). The ␤ 1 AR specific agonist dobutamine (Dobu) also stimulated ERK activation, but was much less active than either NE or ISO. As a control, PMA, the direct activator of PKC, demonstrated robust ERK activation, whereas the direct activator of adenylate cyclase, forskolin (Forsk), showed no demonstrable ERK activation. As expected, ISO-stimulated ERK activation was completely blocked by the ␤-adrenergic receptor antagonists propranolol (PRO) and alprenolol (ALP) (Fig. 4E).
Treatment of cells with 10 M PTX overnight decreased ␤ 1 AR-mediated ERK activation considerably. To further demonstrate that PTX inhibited the activation of ERK by inhibiting the ␣ i subunit of the G protein, we overexpressed a G protein ␣ i 1 subunit (C351S), which cannot be ADP-ribosylated and is therefore unaffected by PTX treatment. As shown in Fig. 4F, overexpression of G i (C351S), while having little effect on the overall ISO-stimulated ERK activity, leads to a marked increase in ISO-stimulated ERK activation observed in the presence of PTX. The residual effect of PTX is presumably due to inhibition of endogenous G i . These data further support the idea that the ␤ 1 AR activates ERK through a G i -mediated pathway, similar to that reported for the ␤ 2 AR (13). In contrast to findings with the ␤ 2 AR (13), the PKA inhibitor H89 had no significant effect on ␤ 1 AR-mediated ERK activation.
GIPC Regulates ␤ 1 AR-mediated ERK Activation-GIPC is a PDZ domain-containing protein that interacts with a G i -specific RGS protein called GAIP (G␣-interacting protein) (34); therefore, GIPC could be involved in the regulation of G i -mediated signaling. Here we tested the effect of GIPC on ␤ 1 ARstimulated ERK activation, which may be mediated by the PTX-sensitive G protein, G i . As shown in Fig. 5A, expression of GIPC substantially decreased ISO-stimulated ␤ 1 AR-mediated ERK activation. ISO-stimulated ERK activation in the presence of GIPC was decreased by 45% compared with control cells lacking GIPC (Fig. 5, A and D). Under similar conditions, expression of GIPC had little effect on ␤ 2 AR-mediated ERK activation (Fig. 5B). The inhibitory effect of GIPC appeared to be specific for the ␤ 1 AR. GIPC had no significant effect on the ability of the ␤ 1 AR mutant (S475A), which is unable to interact with GIPC, to activate ERK (Fig. 5C). These data suggest that the ability of GIPC to regulate ␤ 1 AR-mediated ERK activation is dependent on its ability to interact directly with the ␤ 1 AR via its PDZ domain. DISCUSSION ␤ 1 ARs are of primary importance in regulating heart rate and contractility. Through a yeast two-hybrid screen of a human heart cDNA library, we identified GIPC as a novel binding partner for the ␤ 1 AR. GIPC is a PDZ domain-containing protein that was originally identified and named as a protein that interacts with GAIP, an RGS (regulator of G protein signaling) protein (34). In addition to GAIP, GIPC has also been reported to interact with several membrane proteins through its PDZ domain; these include the Glut-1 transporter, transmembrane Semaphorin-F (M-SemF) (36), neurophilin-1 (37), insulin-like growth factor 1 receptor (38), gp75 (tyrosinase-related protein-1), transforming growth factor ␤ (TGF-␤) receptor (39), 5T4 transmembrane glycoprotein (40), integrins ␣ 5 , ␣ 6A ␣ 6B (41,42), and the viral protein TAX (43).
Here we report a specific interaction between the ␤ 1 AR and GIPC. Despite the fact that both the ␤ 1 AR and ␤ 2 AR contain similar PDZ binding motifs, GIPC interacts only with the ␤ 1 AR, as demonstrated by both yeast two-hybrid screening and co-immunoprecipitation experiments. The GIPC PDZ domain facilitates the interaction with the ␤ 1 AR. Mutation of Ser-475 to either Ala (S475A) or Asp (S475D) completely eliminated the interaction. However, unlike that reported for PSD-95, changing the last residue Val-477 of the ␤ 1 AR to Ala had no effect on the ␤ 1 AR/GIPC association. TAX, a protein containing a similar carboxyl-terminal ETEA-COOH motif has also been shown to interact with GIPC (43). GIPC expression has no significant effect on ␤ 1 AR-induced cAMP accumulation. Therefore, similar solubilized in 3ϫ SDS-PAGE sample buffer. Phosphorylation of the ERK in the whole cell lysates was detected by Western blot analysis using an anti-phospho-ERK1/2 antibody. The data presented in A-D are representative of a minimum of three independent experiments. B, dose-dependent ISO-stimulated ␤ 1 AR-stimulated ERK activation. Serum-starved COS-7 cells were incubated with different concentration of ISO for 5 min and then phospho-ERK was determined by anti-phospho-ERK1/2 antibody. C, time course of the ␤ 1 AR-stimulated ERK activation. Serum-starved COS-7 cells were incubated with 10 M ISO for the indicated times, and then ERK phosphorylation in the whole cell lysates was detected using an anti-phospho-ERK1/2 antibody. D, effect of different agonist stimulation on the ␤ 1 AR-mediated ERK activation. Serum-starved COS-7 cells were incubated with 10 M ISO, norepinephrine (NE), dobutamine (Dobu), 1 M PMA or 10 M forskolin (Forsk) for 5 min. The ERK phosphorylation in the whole cell lysates was detected using anti-phospho-ERK1/2 antibody. The amount of phosphorylated-ERK from each sample was normalized to that from unstimulated cells. E, ␤ 1 AR-stimulated ERK activation was inhibited by ␤ antagonist or PTX pretreatment, but not by the PKA inhibitor H89. Serum-starved COS-7 cells were treated with 100 ng/ml PTX overnight, or 10 M H89, 10 M propranolol (Pro) or 10 M alprenolol (ALP) for 30 mins prior to ISO stimulation. The amount of ERK phosphorylation from ISO-stimulated cells was set as control. The amount of phosphorylated-ERK from each sample was expressed as a percentage of control. The data are presented as means Ϯ S.E. from four independent experiments. F, ␤ 1 AR-stimulated ERK activation was inhibited by PTX. The effects of PTX treatment were reversed by overexpressing a mutant G i (C351), which could not be ADP-ribosylated. Serum-starved COS-7 cells were treated with 100 ng/ml of PTX overnight, stimulated with 10 M ISO, and run on Western blots as described previously. The amount of phosphorylated-ERK from each sample was expressed as a percentage of control in cells expressing ␤ 1 AR. *, p Ͻ 0.05.
to PSD-95, the association of GIPC with the ␤ 1 AR had no effect on ␤ 1 AR coupling to G s . In contrast to PSD-95, which inhibits receptor internalization in HEK-293 cells (31), GIPC had little to no effect on agonist-stimulated receptor internalization.
Recently, many GPCRs have been reported to activate MAP kinase cascades regulating cell growth and/or proliferation (18). In COS-7 cells, it has been clearly demonstrated that stimulation of the ␤ 2 AR activates the MAP kinase ERK through a G i -mediated pathway (13,44). In cardiac myocytes, ␤ 2 AR stimulation has been shown to activate the MAP kinases ERK, JNK, and p38 (21,45). However, ␤ 1 AR-mediated ERK activation or coupling to G i is still quite controversial. Here, we demonstrate ␤ 1 AR stimulated, PTX-sensitive, ERK activation; strongly suggesting that the ␤ 1 AR, like the ␤ 2 AR, can signal through both G s and G i .
GIPC, similar to many PDZ domain-containing proteins, has been shown to play an important role in organizing signaling cascades (46), anchoring proteins in specific subcellular compartments (40), and regulating cell signaling (39). GIPC regulates the expression of the TGF-␤ receptor at the cell surface and enhances the cellular response to TGF-␤ (39). GIPC has also been linked to G protein signaling by its ability to interact with the RGS protein GAIP (46). Overexpression of GIPC in PC12 cells decreases NGF induced MAP kinase (ERK1/2) activation (38). Interestingly, a GIPC-like protein Kermit, which interacts with the Frizzed receptor, has been reported to regulate frizzed 3 signaling in neural crest development (47). Here we provided further evidence that GIPC regulates G protein signaling by directly interacting with the ␤ 1 AR in a PDZ-dependent manner. Expression of GIPC decreased the ability of the ␤ 1 AR to stimulate ERK activation, but had no effect on ␤ 1 AR-mediated cAMP accumulation. This regulatory effect of GIPC on ERK activation is specific for the ␤ 1 AR and depends on a PDZ-mediated interaction, since expression of GIPC has no effect on the ERK activation stimulated by either the ␤ 2 AR or the ␤ 1 AR mutant (S475A). These data suggest that the GIPC/␤ 1 AR association affects receptor G i -but not G s -mediated signaling.
In mouse cardiac myocytes, stimulation of the ␤ 1 AR leads to a PKA-dependent increase in the rate of contraction (48). ␤ 2 AR stimulation shows a biphasic effect on the rate of cardiac myocyte contraction, with an initial PKA-dependent increase followed by a PTX-sensitive decrease, mediated by G i (48). Disruption of the PDZ binding motif in the ␤ 1 AR changed the signaling profile of the ␤ 1 AR in cardiac myocytes to more resemble that of the ␤ 2 AR (31). These data suggest that the ␤ 1 AR is capable of coupling to G i , but that association with PDZcontaining proteins such as PSD-95 could prevent this interaction from occurring. GIPC interacts with the ␤ 1 AR through its PDZ domain and thereby specifically inhibits ␤ 1 AR-stimulated G i -mediated signaling. Thus, the GIPC effect on ␤ 1 AR-mediated ERK activation might explain the differential effects of ␤ 1 AR and ␤ 2 AR stimulation on the rate of cardiac myocyte contraction.
The effect of GIPC on ␤ 1 AR/G i -mediated signaling could also explain the differential effects of ␤ 1 AR and ␤ 2 AR stimulation on cardiac myocyte apoptosis. ␤ 1 AR stimulation leads to cardiac myocyte apoptosis, whereas stimulation of the ␤ 2 AR concurrently activates pro-apoptotic and anti-apoptotic signals, the net effect being increased cell survival (8). Differential coupling to G s and G i has been used to explain this differential effect of ␤ 1 AR versus ␤ 2 AR stimulation (10). Here we demonstrate that the ␤ 1 AR can couple to both G s and G i , similar to a number of other GPCRs including the histamine, serotonin, and glucagon receptors (20). The continued study of receptorinteracting proteins, such as PDZ-containing proteins, will further our understanding of differential signaling by receptors.
FIG. 5. GIPC specifically inhibits ␤ 1 AR-stimulated ERK activation. COS-7 cells were transiently transfected with ␤ 1 AR, ␤ 2 AR, or ␤ 1 AR S475A mutant and GIPC plus GFP-ERK2. Forty-eight hours after transfection, cells were treated with serum-free medium overnight. Serum-starved cells were stimulated with 10 M ISO for 5 min at 37°C. ERK activation was assayed as described under "Experimental Procedures." Expression of GIPC dramatically inhibits ␤ 1 AR-mediated ERK activation (A), but has little effect on ␤ 2 AR-(B) or ␤ 1 AR S475A (␤ 1 AR(M)) mutant (C)-mediated ERK activation. D, summary of the effect of GIPC on ␤ 1 AR-, ␤ 2 AR-, and ␤ 1 AR S457A mutant (␤ 1 AR(M))mediated ERK activation. ISO-stimulated ERK activation in the absence of GIPC was set as control. The amount of phospho-ERK in the presence of GIPC was expressed as a percentage of the control. The data are presented as means Ϯ S.E. from four independent experiments.