S49G and R389G polymorphisms of the β₁-adrenergic receptor influence signaling via the cAMP-PKA and ERK pathways

Abstract

Two functionally important β₁-adrenergic receptor (β₁AR) polymorphisms have been identified. The R389G polymorphism influences coupling to the Gs-cAMP pathway. R³⁸⁹-β₁ARs display enhanced activation of cAMP/PKA; they provide short-term inotropic support but also cause a predisposition to cardiomyopathic decompensation. A second S49G polymorphism is implicated in the evolution of heart failure, but the mechanism remains uncertain. This study shows that position 49 and 389 polymorphisms function in a coordinate manner to influence agonist-dependent cAMP/PKA and ERK responses. cAMP/PKA and ERK responses are more robust in HEK293 cells that heterologously overexpress G⁴⁹-β₁ARs, compared with S⁴⁹-β₁ARs. However, this phenotype is most obvious on a G³⁸⁹-β₁AR background; the more robust agonist-dependent cAMP/PKA and ERK responses in R³⁸⁹-β₁AR cells effectively obscure the effect of the S49G polymorphism. We also show that isoproterenol (Iso) and carvedilol activate ERK via a similar EGFR-independent mechanism in cells expressing various β₁AR haplotypes. However, Iso activates ERK via an Src-independent pathway, but carvedilol-dependent ERK activation requires Src. Since the S49G polymorphism has been linked to changes in β₁AR trafficking, we examined whether β₁AR polymorphisms influence partitioning to lipid raft membranes. Biochemical fractionation studies show that all four β₁AR variants are recovered in buoyant flotillin-enriched membranes; the distinct signaling phenotypes of the different β₁AR variants could not be attributed to any gross differences in basal compartmentalization to lipid raft membranes. The allele-specific differences in β₁AR signaling phenotypes identified in this study could underlie interindividual differences in responsiveness to β-blocker therapy and clinical outcome in heart failure.

β₁-adrenergic receptors (β₁ARs) are the principle mediators of catecholamine-dependent changes in the force and rate of cardiac contraction. β₁ARs adjust cardiac output by activating a Gs-adenylyl cyclase (AC) pathway that increases cAMP, activates protein kinase A (PKA), and phosphorylates substrates involved in excitation-contraction coupling (16). However, heightened β₁AR drive in the setting of heart failure (HF) activates signaling pathways that promote cardiomyocyte hypertrophy/apoptosis, interstitial fibrosis, disordered energetics, and contractile dysfunction (8, 16). βAR inhibitors that prevent these maladaptive βAR responses have become standard therapy for HF.

While there is ample evidence that βAR blockers decrease morbidity and mortality in HF, there is considerable interindividual variability in drug responsiveness that is not readily attributable to known clinical or demographic factors. This interindividual variability has been attributed at least in part to two common nonsynonymous single nucleotide polymorphisms (SNPs) at positions 49 and 389 of the human of the β₁AR receptor (3). Most studies have focused on the R389G polymorphism at the conformationally sensitive Gs binding domain in the juxtamembrane region of the COOH-terminal cytoplasmic tail. Studies to date indicate that R³⁸⁹-β₁ARs display enhanced coupling to the Gs-cAMP pathway and enhanced agonist-dependent desensitization (relative to G³⁸⁹-β₁ARs) when overexpressed in cultured fibroblasts (10). Transgenic cardiac-specific R³⁸⁹-β₁AR overexpression also leads to enhanced receptor signaling and contractile function in young mice, compared with age-matched G³⁸⁹-β₁AR hearts (11). These findings in cell culture and cardiomyocyte-targeted transgenic mouse models are consistent with clinical studies showing that the R³⁸⁹-β₁AR (in association with a α_2c-AR variant that regulates presynaptic release of norepinephrine) is a risk factor for human HF (23) and that the R³⁸⁹-β₁AR allele predicts responsiveness to βAR inhibitor therapy in patients with HF (7, 11).

A second S49G polymorphism in the relatively short NH₂ terminus of the β₁AR has been implicated as a genetic determinant of βAR inhibitor responses and clinical outcome in HF (1, 9, 17, 25–27). However, the molecular basis for the clinical impact of this polymorphism is less obvious, since the relatively short/featureless βAR NH₂ terminus generally is ignored in structure-function studies. While several studies have provided consistent evidence that G⁴⁹-β₁ARs are more susceptible, and S⁴⁹-β₁ARs are relatively resistant, to agonist-dependent downregulation, the available literature on the effects of the S49G polymorphism on β₁AR affinity for ligands or β₁AR coupling to cAMP is more ambiguous (6, 18, 21).

Despite considerable evidence that R³⁸⁹-β₁ARs act as “gain-of-function” variants for the AC/cAMP/PKA pathway (relative to G³⁸⁹-β₁ARs), there is only scant information on whether position 389 or 49 polymorphisms influence the magnitude or mechanism for β₁AR coupling to effectors that may contribute to cardioprotection, such as ERK (4). Moreover, studies to date that typically focus on the independent effect of a single polymorphic variation in isolation may have only limited relevance to clinical phenotypes that result from the distinct haplotypes that exist in clinical populations. This study identifies allele-specific differences in β₁AR coupling to cAMP/PKA and ERK pathways that have implications for the pathogenesis and treatment of clinical HF phenotypes.

MATERIALS AND METHODS

Materials.

Antibodies were from the following sources: β₁-AR and flotillin-1 were from Santa Cruz Biotechnology. ERK1/2, p-ERK1/2, and p-PKA substrate antibodies were from Cell Signaling Technology (Danvers, MA). Isoproterenol (Iso), forskolin, and EGF were from Sigma (St. Louis, MO). All other chemicals were reagent grade.

Plasmids.

A plasmid that drives expression of the human S⁴⁹R³⁸⁹-β₁-AR, with an NH₂-terminal Flag, was obtained from Addgene. Site-directed mutagenesis of β₁-AR-Flag with the QuikChange mutagenesis system (Agilent Technologies) was performed to generate G⁴⁹R³⁸⁹-β₁ARs, S⁴⁹G³⁸⁹-β₁ARs, or G⁴⁹G³⁸⁹-β₁ARs. All constructs were confirmed by sequencing.

Cell culture and transfection.

HEK293 cells were cultured in DMEM containing 10% FCS, and 100 units/ml penicillin-streptomycin. Cells were transfected with plasmids that drive expression of human β₁-AR polymorphic variants with the Effectene Transfection reagent (Qiagen) according to the instruction manual. Stable cell lines were generated by selection with G418 (400 μg/ml).

Western blotting.

Immunoblotting was performed on cell extracts according to methods described previously or manufacturer's instructions. Each panel in each figure represents the results of a single gel (exposed for a uniform duration); detection was with enhanced chemiluminescence. All results were replicated in at least three experiments on separate culture preparations.

Preparation of caveolar membranes.

Caveolin-enriched fractions were prepared according to a detergent-free purification scheme described previously (20). All steps were carried out at 4°C. In brief, cells from five 100 mm diameter dishes were washed twice with ice-cold phosphate-buffered saline and then scraped into 0.5 m sodium carbonate, pH 11.0 (0.5 ml/dish). Cells from five dishes were combined for each preparation. The extract was sequentially disrupted by homogenization with a loose-fitting Dounce homogenizer (10 strokes), a Polytron tissue grinder (three 10 s bursts), and a tip sonicator (three 20 s bursts). The homogenate was then adjusted to 40% sucrose by adding an equal volume of 80% sucrose prepared in MES-buffered saline (25 mm MES, pH 6.5, and 0.15 m NaCl), placed on the bottom of an ultracentrifuge tube, overlaid with a 5–30% discontinuous sucrose gradient (3 ml of 5% sucrose and 4 ml of 35% sucrose, both in MES-buffered saline containing 0.25 m sodium carbonate), and centrifuged at 260,000 g for 16–18 h in a SW40 rotor (Beckman Coulter, Palo Alto, CA). After centrifugation, 12 1 ml fractions were collected. A pooled caveolar fraction (fractions 4–5), a pooled fraction 8–12 (which contains the bulk of the cellular material including the cytosol and most of the particulate membrane fraction), and the insoluble pellet (which is solubilized in SDS-PAGE sample buffer) were subjected to SDS-PAGE and immunoblotting.

Measurements of βAR density and cAMP accumulation.

Radioligand binding experiments with [¹²⁵I]CYP were performed on membranes prepared from each stable cell line according to published methods (24). cAMP measurements also were performed according to standard methods. In brief, cells were cultured in six-well plates, grown to confluence, and then serum deprived for 12 h. Cells were preincubated with 10 mM theophylline for 60 min and then challenged for 5 min with vehicle, Iso (10⁻⁶ M), or forskolin (10⁻⁵ M) as indicated. Assays were terminated by aspiration of the incubation buffer and addition of 0.5 ml of 100% ice-cold ethanol to each well. Cell lysates were dried in a spin vacuum, and cAMP in the residue was determined with a commercially available enzyme-linked immunosorbent assay kit (R & D Systems, Minneapolis, MN) according to manufacturer's instructions.

Statistics.

Results are shown as means ± SE and were analyzed by Student's t-test followed by Bonferroni adjustment for multiple comparisons with GraphPad Prism software.

RESULTS

The initial studies compared signaling responses induced by activation of individual β₁-AR haplotypes transiently overexpressed at similar levels in HEK293 cells. Figure 1 shows that Iso induces a modest increase in ERK phosphorylation in vector-infected HEK293 that contain low levels of endogenous β₂ (but not β₁) ARs; HEK293 cells also display a modest increase in ERK phosphorylation in response to direct activation of AC by forskolin. β₁AR overexpression leads to an increase in Iso-dependent ERK phosphorylation without any associated changes in ERK phosphorylation in response to forskolin or EGF (included as controls in the experiment). Studies with an anti-phospho-PKA substrate antibody (that recognizes protein phosphorylation at the PKA consensus RXXpS motif, used as a surrogate to track agonist-dependent activation of cAMP) show that heterologously overexpressed β₁ARs also couple to the cAMP/PKA pathway. Iso induces a modest increase in anti-phospho-PKA substrate immunoreactivity in vector-infected HEK293. The Iso-dependent increase in anti-phospho-PKA substrate immunoreactivity is enhanced by heterologous overexpression of β₁-ARs, without any changes in anti-phospho-PKA substrate immunoreactivity in response to treatment with forskolin (included as control in the experiments). Importantly, the magnitude of the Iso-dependent responses differed considerably for the different molecular forms of the β₁AR. R³⁸⁹-β₁ARs (with either S or G at position 49) displayed robust ERK and PKA substrate phosphorylation responses. Both responses are lower in cells that heterologously overexpress G³⁸⁹-β₁ARs. The effect of an S49G polymorphism is most obvious in this context, where a β₁AR harboring a position 49 glycine residue induces more robust ERK and PKA substrate phosphorylation responses than the G³⁸⁹-β₁AR with a position 49 serine residue.

Fig. 1.

Agonist-dependent ERK and PKA responses in HEK293 cells that heterologously overexpress S⁴⁹G³⁸⁹, G⁴⁹G³⁸⁹, S⁴⁹R³⁸⁹, and G⁴⁹R³⁸⁹ haplotypes of the β₁-adrenergic receptor (β₁AR). Cell lysates from cells treated for 5 min with vehicle, isoproterenol (Iso, 1 μM), forskolin (Forsk, 10 μM), or EGF (100 ng/ml) were subjected to immunoblot analysis for pERK and pPKA substrate, as well as ERK protein (to verify equal protein loading) and β₁AR (to track transgene expression) according to materials and methods. Results (means ± SE) from 3 separate experiments on separate culture preparations are quantified. *P < 0.05 vs. uninfected. #P < 0.05 vs. S⁴⁹R³⁸⁹ or G⁴⁹R³⁸⁹. S⁴⁹R³⁸⁹ and G⁴⁹R³⁸⁹ do not differ from each other.

The importance of S49G and R389G polymorphisms was interrogated further in clonal HEK293 cell lines that overexpress similar levels of HA-tagged S⁴⁹G³⁸⁹-, G⁴⁹G³⁸⁹-, or S⁴⁹R³⁸⁹-β₁ARs (B_max values 776 ± 52, 735 ± 33, and 755 ± 41 fmol/mg protein, respectively). We compared signaling in response to activation of S⁴⁹G³⁸⁹-β₁ARs vs. S⁴⁹R³⁸⁹-β₁ARs to resolve the functional importance of the R389G polymorphism. Since the effect of the position 49 polymorphism is most pronounced on a G³⁸⁹-β₁AR background, we resolved the functional importance of the S49G polymorphism by comparing signaling responses that result from activation of S⁴⁹G³⁸⁹-β₁ARs vs. G⁴⁹G³⁸⁹-β₁ARs. Figure 2A shows that all three β₁AR variants activate the cAMP and ERK pathways. In each case, Iso-dependent ERK phosphorylation is rapid (maximal at 2–5 min) and transient (wanes with 30–60 min of continuous agonist stimulation, Fig. 2B). However, the magnitude of the Iso-dependent responses is influenced by both R389G and S49G polymorphisms. Iso-dependent increases in cAMP accumulation and ERK phosphorylation are higher in S⁴⁹R³⁸⁹-β₁AR cells than in S⁴⁹G³⁸⁹-β₁AR cells. However, a position 389 glycine residue does not necessarily impose a low level of signaling, since G³⁸⁹-β₁ARs that harbor a position 49 glycine residue elicit more robust cAMP and ERK phosphorylation responses. Control studies verify that the differences in Iso-dependent activation of cAMP and ERK reflect the distinct signaling phonotypes of the individual heterologously overexpressed β₁ARs, since forskolin-dependent cAMP accumulation and EGF-dependent ERK phosphorylation are similar in all three cell lines.

Fig. 2.

S⁴⁹G³⁸⁹, G⁴⁹G³⁸⁹, and S⁴⁹R³⁸⁹ variants of β₁-AR recruit the cAMP and ERK pathways. A: HEK293 cell lines that stably expressing similar amounts of S⁴⁹G³⁸⁹-, G⁴⁹G³⁸⁹-, or S⁴⁹R³⁸⁹-β₁ARs were treated for the indicated times with Iso (1 μM). Cell lysates were subjected to immunoblot analysis for pERK and ERK protein according to materials and methods. The results are from a single experiment and are representative of data obtained in three separate experiments. B: cAMP and pERK responses were measured in HEK293 cells treated for 5 min with vehicle (Cont), Iso (1 μM), carvedilol (Carv, 1 μM), or forskolin (10 μM) according to materials and methods. The results (means ± SE) represent data from 5–7 (for cAMP) or 3 (for ERK) separate experiments performed on separate culture preparations. *P < 0.05 vs. response to cognate agonist in G⁴⁹G³⁸⁹ or S⁴⁹R³⁸⁹ cells.

β₁ARs adopt distinct “active” conformations that differ in their ability to activate G protein-dependent vs. G protein-independent (β-arrestin-dependent) responses. Ligands that selectively activate only certain signaling pathways are referred to as “biased” ligands. Carvedilol has been characterized as a β-arrestin-biased ligand for the mouse (S⁴⁹G³⁸⁹) β₁AR; carvedilol acts as an antagonist for the classical β₁AR-Gs-cAMP pathway but stabilizes a β₁AR conformation that activates the β-arrestin-ERK pathway (5, 12). Figure 2 shows that carvedilol acts as a partial agonist at S⁴⁹G³⁸⁹-, G⁴⁹G³⁸⁹-, and S⁴⁹R³⁸⁹-β₁ARs. In each case, carvedilol treatment leads to an increase in ERK phosphorylation that is not associated with a detectable increase in cAMP. Carvedilol-dependent ERK phosphorylation responses are ∼⅓ as robust as the cognate Iso-dependent ERK phosphorylation responses in any given cell line. Control studies established that carvedilol-dependent increases in ERK phosphorylation are maximal (do not increase further at higher carvedilol concentrations) and are specifically mediated by β₁ARs (are inhibited by propranolol; data not shown).

Previous studies have attributed β₁AR-dependent activation of ERK to a mechanism involving Src and the activation of a matrix metalloproteinase that cleaves HB-EGF and trans-activates EGFRs (12). However, these studies were performed on HEK293 cells that heterologously overexpress the mouse S⁴⁹G³⁸⁹-β₁ARs variant and EGFRs. Studies that rigorously compare the magnitude and mechanism for ERK activation by individual polymorphic variants of the β₁AR at more physiologically relevant levels of EGFR expression are illustrated in Figure 3. The results show that AG1478 (an EGFR kinase inhibitor) completely abrogates EGF-dependent ERK phosphorylation, but AG1478 does not block either Iso- or carvedilol-dependent increases in ERK phosphorylation in β₁AR cells that do not heterologously overexpress EGFRs. These results indicate that all three β₁AR variants couple to an ERK pathway that does not require EGFR activity (i.e., the EGFR transactivation presumably is recruited only at high level of EGFR overexpression). The PKA inhibitor H89 also does not interfere with β₁AR-dependent ERK activation (under conditions where H89 pretreatment completely abrogated Iso-dependent increases in phospho-PKA substrate immunoreactivity; data not shown). However, an agonist-dependent difference in the mechanism for ERK activation by Iso and carvedilol was exposed in studies with PP1, an inhibitor of Src kinase activity. Figure 3 shows that carvedilol activates ERK via an Src-dependent pathway that is inhibited by PP1, whereas the Iso-dependent pathway for ERK activation persists in PP1-treated cells. These results indicate that Iso and carvedilol stabilize β₁ARs in slightly different active conformations that couple to distinct signal transduction pathways and signaling responses.

Fig. 3.

Iso and Carv recruit the ERK phosphorylation response via distinct pathways. HEK293 cells stably expressing similar amounts of S⁴⁹G³⁸⁹-, G⁴⁹G³⁸⁹-, or S⁴⁹R³⁸⁹-β₁ARs were pretreated for 30 min with vehicle AG1478 (AG, 10 μM), PP1 (10 μM), or H89 (10 μM) prior to a 5 min challenge with 1 μM Iso, 1 μM Carv, or EGF (100 ng/ml). The results of immunoblot analyses from a single representative experiment are depicted on top, with the data from 3 separate experiments (means ± SE) quantified at bottom. *P < 0.05 vs. cognate agonist response in G⁴⁹G³⁸⁹ or S⁴⁹R³⁸⁹ cells. #P < 0.05 vs. agonist alone.

We previously demonstrated that β₁ARs partition to lipid raft membranes (19). Cell lysates were solubilized in detergent-free alkaline sodium carbonate buffers and subjected to isopycnic centrifugation on a discontinuous sucrose gradient to determine whether S49G or G389R polymorphisms alter signaling phenotypes by influencing β₁AR partitioning to buoyant lipid raft membranes. Figure 4 shows that β₁AR variants are recovered almost exclusively in the light sucrose gradient fraction that contains the bulk of the cellular flotillin immunoreactivity (a lipid raft recovery marker) and excludes >98% of total cell protein (including Golgi and other intracellular organelle markers). We did not identify any gross allele-specific differences in β₁AR partitioning to lipid raft microdomains.

Fig. 4.

S⁴⁹G³⁸⁹-, G⁴⁹G³⁸⁹-, S⁴⁹R³⁸⁹-, and G⁴⁹R³⁸⁹-β₁AR haplotypes display similar basal localization to lipid raft membranes. Cav, caveolin; F, fraction.

DISCUSSION

There is recent evidence that polymorphisms at positions 49 and 389 in the coding region of the β₁AR act as genetic modifiers of clinical outcome in various cardiovascular disorders. Most studies have focused on the role of the R389G polymorphism as a predictor of pharmacologic responsiveness to β-AR inhibitor therapy, showing that R³⁸⁹-β₁ARs (the more common allele in individuals of European descent) adopt a more active conformation, display enhanced coupling to the Gs-cAMP pathway, and confer enhanced sensitivity to β-blocker therapy in HF, relative to G³⁸⁹-β₁ARs. The position 389 polymorphism could in theory also impact on clinical outcome by influencing signaling via cardioprotective pathways involving ERK, but a role for the R389G polymorphism as a modifier of signaling via G protein-independent signaling pathways has not previously been considered. The S49G polymorphism has received considerably less attention, with the analysis largely confined to epidemiologic studies that implicate the S49G polymorphism as a genetic determinant of clinical outcome in HF. Two laboratories have identified allele-specific differences in agonist-dependent downregulation kinetics in studies performed in reductionist models (6, 18). A single study from Levin et al. (6) also showed that a position 49 glycine residue confers enhanced basal and agonist-dependent coupling to the AC/cAMP pathway (i.e., the β₁AR harboring a position 49 glycine residue is a constitutively active receptor), but this finding was not replicated by others (18). Studies to date also have not rigorously considered a possible interplay between the polymorphisms at positions 49 and 389. Therefore, we examined the signaling profiles of β₁AR polymorphic variants harboring either serine or glycine at position 49 and either glycine or arginine at position 389.

Studies reported herein show that 1) agonist-dependent cAMP/PKA and ERK responses are more robust in R³⁸⁹-β₁AR cells, than in G³⁸⁹-β₁AR cells and that 2) the effect of the S459G polymorphism is most obvious on a G³⁸⁹-β₁AR background; G³⁸⁹-β₁ARs with a glycine at position 49 display enhanced agonist-dependent cAMP/PKA and ERK responses, compared with S⁴⁹G³⁸⁹-β₁ARs. Of note, the two previous studies that linked the S49G polymorphism to changes in β₁AR trafficking and desensitization kinetics also were performed on a G³⁸⁹ background (6, 18); effects of a position 49 SNP on signaling by R³⁸⁹-β₁ARs have never been considered. This oversight may be important, since clinical studies to date do not provide an entirely coherent picture of the importance of β₁AR polymorphisms as genetic modifiers of cardiac phenotypes. It is tempting to speculate that discrepancies in the literature could be due to the stratification of patients according to a single polymorphic locus (either position 49 or position 389), rather than the haplotypes that exist in the general population. This is important for two reasons. First, glycine is the minor allele at both loci, but there are important racial differences in allele frequencies for both S49G and R389G polymorphisms that could confound an analysis. Second, there is strong linkage disequilibrium between S49G and R389G loci that functions to limit the haplotypes (combinations of specific polymorphisms) in the general population. Most studies identify S⁴⁹/R³⁸⁹-, S⁴⁹/G³⁸⁹-, and G⁴⁹/R³⁸⁹-β₁ARs as the most prevalent haplotypes in Caucasians; the G⁴⁹/G³⁸⁹ haplotype appears to be quite rare (14, 22). Hence, while the results of this and previous studies in reductionist models that characterize the signaling properties of the S49G polymorphism on a G³⁸⁹ background are of interest from a theoretical standpoint, the physiological relevance of these findings will require further study.

Since the S49G polymorphism has been linked to changes in β₁AR trafficking, and the spatial relationship between the βAR and its downstream effectors can function to either facilitate or restrict signaling responses, we examined whether polymorphic variants of the β₁AR compartmentalize differently between lipid raft and nonraft membranes. Lipid rafts are a buoyant membrane fraction that are highly enriched in βARs, Gs, several AC isoforms, EGFRs, and components of the ERK-MAPK pathway; lipid rafts have been implicated as platforms that “launch” signaling to the ERK cascade. Our fractionation studies show that the various polymorphic variants of the β₁AR are recovered exclusively in the buoyant lipid raft membrane fraction; we did not identify an obvious difference in the partitioning of β₁AR variants between lipid raft and nonraft membranes that might explain their distinct signaling phenotypes. However, these studies were performed on resting cells that had not been exposed to agonist ligands; they do not rule out possible allele-specific differences in β₁AR trafficking patterns following agonist exposure that would influence signaling responses.

HF is associated with elevated catecholamine levels and chronic/persistent β₁AR stimulation; this heightened sympathetic tone contributes to the pathogenesis of various HF syndromes. The cardiotoxic actions of β₁AR have been attributed to Gs-dependent activation of AC and enhanced accumulation of cAMP. Drugs that inhibit cardiotoxic β₁AR-mediated responses (such as metoprolol, bisoprolol, and carvedilol) have been approved for the treatment of HF syndromes. However, their clinical efficacy is not equivalent, with evidence that carvedilol affords survival advantage over other approved pharmacologic inhibitors (15). While this has been ascribed to carvedilol's ancillary antioxidant, anti-inflammatory, antiproliferative, or anti-α₁-AR actions (2, 13, 29), a mechanism related to carvedilol's unique pharmacologic profile at β₁ARs also is possible. Carvedilol acts like other βAR inhibitors to block the βAR-Gs-cAMP pathway, but carvedilol also exhibits an additional action as a partial agonist for the G protein-independent cardioprotective pathway that activates ERK (28). Previous literature examined carvedilol's pharmacologic actions in cells that overexpress only certain molecular forms of the β₁AR. This study is the first to extend the analysis and show that carvedilol exerts a similar action as a biased partial agonist for the ERK pathway at all β₁AR variants. However, we identified allele-specific differences in the magnitude of the carvedilol-dependent ERK phosphorylation response that mirror the allele-specific differences in ERK phosphorylation by Iso. We also identified an agonist-dependent difference in the mechanism for β₁AR-dependent activation of ERK. Notably, while previous studies showed that Iso and carvedilol recruit similar EGFR-/Src-dependent pathways to activate ERK in HEK293 cells that heterologously overexpress both β₁ARs and EGFRs (12), we show that β₁ARs activate ERK via an EGFR-independent mechanism at endogenous levels of EGFR expression. However, an agonist-dependent difference in the mechanism for β₁AR activation of ERK activation was identified under these conditions. We show that carvedilol activates ERK via a Src-dependent mechanism, whereas the Iso-dependent ERK pathway is Src independent; the pharmacologic profiles for Iso- and carvedilol-dependent ERK activation are similar across the various β₁AR polymorphic variants. The EGFR-independent mechanisms for ERK activation by β₁ARs, which may or may not involve Src and presumably are obscured at high levels of EGFR overexpression, may contribute to β₁AR-dependent cardioprotection in the heart and are the focus of further study.

In summary, this study identifies allele-specific differences in signaling phenotypes for different β₁AR polymorphic variants. These differences in the cellular actions of individual β₁AR variants could contribute to the pathogenesis of clinical HF phenotypes and may have bearing on the success of β-blocker therapy.

GRANTS

This work was supported by U.S. Public Health Service National Heart, Lung, and Blood Institute Grants HL-93343 and HL-95605.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: F.Z. performed experiments; F.Z. and S.F.S. analyzed data; F.Z. and S.F.S. interpreted results of experiments; F.Z. and S.F.S. prepared figures; F.Z. and S.F.S. drafted manuscript; F.Z. and S.F.S. approved final version of manuscript; S.F.S. conception and design of research; S.F.S. edited and revised manuscript.