Advertisement
Research ArticleCardiovascular

Selectively Engaging β-Arrestins at the Angiotensin II Type 1 Receptor Reduces Blood Pressure and Increases Cardiac Performance

Jonathan D. Violin, Scott M. DeWire, Dennis Yamashita, David H. Rominger, Lisa Nguyen, Kevin Schiller, Erin J. Whalen, Maxine Gowen and Michael W. Lark
Journal of Pharmacology and Experimental Therapeutics December 2010, 335 (3) 572-579; DOI: https://doi.org/10.1124/jpet.110.173005

Abstract

Biased G protein-coupled receptor ligands engage subsets of the receptor signals normally stimulated by unbiased agonists. However, it is unclear whether ligand bias can elicit differentiated pharmacology in vivo. Here, we describe the discovery of a potent, selective β-arrestin biased ligand of the angiotensin II type 1 receptor. TRV120027 (Sar-Arg-Val-Tyr-Ile-His-Pro-d-Ala-OH) competitively antagonizes angiotensin II-stimulated G protein signaling, but stimulates β-arrestin recruitment and activates several kinase pathways, including p42/44 mitogen-activated protein kinase, Src, and endothelial nitric-oxide synthase phosphorylation via β-arrestin coupling. Consistent with β-arrestin efficacy, and unlike unbiased antagonists, TRV120027 increased cardiomyocyte contractility in vitro. In rats, TRV120027 reduced mean arterial pressure, as did the unbiased antagonists losartan and telmisartan. However, unlike the unbiased antagonists, which decreased cardiac performance, TRV120027 increased cardiac performance and preserved cardiac stroke volume. These striking differences in vivo between unbiased and β-arrestin biased ligands validate the use of biased ligands to selectively target specific receptor functions in drug discovery.

Introduction

Drugs targeting G protein-coupled receptors (GPCRs) are used in a wide range of diseases. Indeed, modulation of GPCR function is one of the most common approaches to drug discovery, either with agonists to mimic endogenous receptor activation or antagonists to block endogenous agonist binding. These strategies have successfully targeted dozens of GPCRs in the discovery of new therapeutic agents (Hopkins and Groom, 2002). However, the network of signaling events downstream of a receptor is complex and usually regulates multiple cellular and tissue responses, not all of which are intended targets of classic agonists and antagonists. In addition to off-target side effects, these “on-target” adverse effects have historically presented a major barrier to the development of safe, effective therapeutics.

However, not all GPCR signaling derives from G protein signaling; GPCRs can activate parallel and sometimes distinct signals. The most general of these are mediated by β-arrestins, which bind to activated receptors to desensitize G protein signaling, promote receptor internalization, and activate distinct signal transduction cascades that can be independent of G protein coupling (Reiter and Lefkowitz, 2006; DeWire et al., 2007). Nearly all GPCRs will couple to at least one G protein and at least one β-arrestin, and in many cases the physiological effects of a single agonist activating a GPCR can be separated into β-arrestin- and G protein-mediated components (Schmid and Bohn, 2009).

Other work has shown that the experimental separation of GPCR effects can also be achieved pharmacologically: certain GPCR ligands can selectively activate subsets of receptor signals (Wei et al., 2003; Gesty-Palmer et al., 2006; Mailman, 2007; Tran et al., 2009; Zidar et al., 2009). These “biased ligands” are agonists when assaying some receptor functions, but they are antagonists or even inverse agonists when assaying other receptor functions (Violin and Lefkowitz, 2007; Kenakin, 2009). One of the most studied GPCRs in this regard is the angiotensin II type 1 receptor (AT1R). This receptor engages all three classes of β-arrestin function: desensitization (Violin et al., 2006), internalization (Kule et al., 2004), and signaling (Barnes et al., 2005; DeWire et al., 2008; Ahn et al., 2009). It was also one of the first receptors for which a β-arrestin biased ligand was described: 1Sar, 4Ile, 8Ile-angiotensin II (SII) induces β-arrestin recruitment, receptor internalization, and β-arrestin-mediated signals without activating G protein coupling (Wei et al., 2003; Ahn et al., 2004; Kim et al., 2009). Unfortunately, this peptide displays low receptor binding affinity (Holloway et al., 2002), which has made in vivo characterization of its effects challenging. However, ex vivo studies revealed that SII promotes contractility of isolated cardiac myocytes via the AT1R and β-arrestin2 (Rajagopal et al., 2006) and mitogen-activated protein kinase (MAPK) activation in perfused hearts (Aplin et al., 2007), indicating that β-arrestin biased ligands may have significant impact in a physiological setting. However, it remains unclear whether AT1R signals can be separated by biased ligands in vivo to allow precise targeting of desired AT1R effects. More broadly, despite increasing speculation in the research literature and numerous examples of biased ligands eliciting unique pharmacology in vitro, it remains unclear whether ligand bias can elicit unique, differentiated pharmacology in vivo compared with unbiased ligands by simultaneously activating one signal transduction pathway while antagonizing another at the same receptor.

Thus we set out to discover new β-arrestin biased ligands for the AT1R, seeking potent, selective compounds to enable an in vivo comparison of biased and unbiased ligands to test whether AT1R ligand bias translates from isolated cellular systems to elicit unique integrated responses in vivo. Such a ligand would also allow us to test whether biased ligands truly uncover new opportunities for safer, more efficacious therapeutics for cardiovascular diseases of impaired cardiac performance.

Materials and Methods

Synthesis of Peptides.

Peptides were synthesized by Genscript USA Inc. (Piscataway, NJ) to more than 98% purity; quality control was by high-performance liquid chromatography and mass spectrometry. The peptides described here are as follows: TRV120023, Sar-Arg-Val-Tyr-Lys-His-Pro-Ala-OH; TRV120026, Sar-Arg-Val-Tyr-Tyr-His-Pro-NH2; and TRV120027, Sar-Arg-Val-Tyr-Ile-His-Pro-d-Ala-OH.

Cell Culture and Preparation.

HEK-293 cells were stably transfected to overexpress β-arrestin2 fused to a β-galactosidase fragment and human AT1R or rat AT1aR fused to a complementary β-galactosidase fragment. The growth medium used was MEM with 10% FBS, 1% penicillin/streptomycin, 150 μg/liter of neomycin, and 150 μg/liter of hygromycin. β-Arrestin recruitment and inositol monophosphate (IP1) accumulation were measured in small-volume, 384-well plates; immunoblot experiments were in 12-well or 10-cm plates. U2-osteosarcoma (U2-OS cells) were stably transfected to overexpress rat AT1aR and cultured in MEM with 10% FBS, 1% penicillin/streptomycin, and 150 μg/liter of neomycin. Cells were plated in 6- or 12-well tissue culture plates for immunoblot experiments.

β-Arrestin Recruitment.

The PathHunter protein complementation assay (DiscoveRx Corporation, Fremont, CA) was performed according to the manufacturer's protocol and read for chemiluminescent signaling on a PheraStar reader (BMG Labtech, Durham, NC). In brief, complementary halves of β-galactosidase were genetically fused to the carboxyl termini of the AT1R and β-arrestin2. When cotransfected, the two fusion proteins serve as a proximity sensor; when β-arrestin2 translocates to active receptor, the β-galactosidase fragments interact to form a functional enzyme, which is detected by a chemoluminescent substrate.

Inositol Monophosphate Accumulation.

IP1 was measured with the IP-One Tb HTRF kit (Cisbio, Bedford, MA). Plates were read on a PheraStar reader using a time-resolved fluorescence ratio (665 nm/620 nm).

Immunoblotting.

U2-OS or HEK-293 cells stably expressing the rat AT1aR were grown in MEM supplemented with 10% FBS. After overnight serum starvation, cells were stimulated with 1 μM angiotensin II, valsartan, TRV0100027, TRV0100023, or vehicle for 5 min. Lysates were immunoblotted with phospho-Y416-Src, phospho-T202/Y204-ERK, or phospho-S1177-eNOS antibodies per the manufacturer's instructions (Cell Signaling Technologies, Danvers, MA). For RNA interference experiments, siRNAs for β-arrestin2 were transfected as described previously (DeWire et al., 2008), and cells were plated as above at 48 h after siRNA transfection. The signal transduction protein array assay was performed per the manufacturer's protocol (Proteome Profiler ARY003; R&D Systems, Minneapolis, MN). The activated enzymes detected in this array relied on antibodies to phospho-Y397 on FAK and phospho-S63 on c-Jun. Immunoreactive bands or spots were visualized and quantitated by using GeneSnap and GeneTool software (Syngene, Frederick, MD).

Myocyte Contractility.

Isolated murine cardiomyocytes were isolated and contractility was measured by video edge detection as described previously (Jaleel et al., 2008).

In Vivo Studies.

All in vivo studies were performed at Qtest Labs (Columbus OH) for cardiovascular assays or ChemPartner Co. (Shanghai, China) for pharmacokinetics, under protocols reviewed and approved by the respective Institutional Animal Care and Use Committees.

Rat Hemodynamics during Angiotensin II Dose Escalation.

TRV120027, vehicle, and losartan were dosed to anesthetized (sodium pentobarbital)/ventilated healthy male rats. Under anesthesia, the animals were instrumented with ECG electrodes and micromanometers into a femoral artery and the left ventricle. Mean arterial pressure (MAP) and left ventricular hemodynamic and mechanical indices (end systolic pressure, end diastolic pressure, dP/dtmax, Tau, Vmax) were obtained from the arterial/ventricular pressure signals. In addition, ECG-derived indices of conduction (PQ, QRS) and repolarization duration (QT/QTcF) were measured. Two dose levels of TRV120027, low (1 μg/kg/min) and high (10 μg/kg/min), were assayed. The cardiovascular responses of test article-treated animals to an eight-dose angiotensin II challenge were evaluated and compared against those obtained in animals receiving either losartan (3 mg/kg bolus injection) or vehicle (infusion).

Rat Pressure-Volume Loop Analysis.

Telmisartan and TRV120027 were dosed to anesthetized (sodium pentobarbital)/ventilated healthy male rats. Under anesthesia, the animals were instrumented with ECG electrodes, a micromanometer into a femoral artery, a conductance/micromanometer into the left ventricle, and a hydraulic occluder around the inferior vena cava (via a right thoracotomy). MAP and left ventricular mechanical (end systolic pressure, end diastolic pressure, dP/dtmax, Tau, Vmax) and geometrical (end diastolic volume, stroke volume, dV/dt) indices were obtained from the ventricular pressure/conductance (volume) signals. The myocardial inotropic [end systolic pressure-volume (PV) relationship (ESPVR), preload recruitable stroke work (PRSW)], lusitropic (end diastolic PV relationship), and energetic (PV area, stroke work) state were examined by means of PV curves; PV families/relationships were generated via preload reductions (inferior vena cava occlusion). ECG-derived indices of conduction (PQ, QRS) and repolarization duration (QT/QTcF) were measured.

Statistics.

Statistics were assessed by using Prism version 5 (GraphPad Software Inc., San Diego, CA). Schild analysis fitting the Schild model of competitive antagonism to a series of dose-response curves with varying amounts of antagonist was also performed with Prism software.

Results

The AT1R Is Capable of Potent, Robust Efficacy by β-Arrestin Biased Ligands.

To identify new, higher-potency β-arrestin biased ligands, we initiated an iterative evaluation of custom-synthesized peptides based on SII, using assays of G protein coupling (inositol monophosphate accumulation) and β-arrestin recruitment to the AT1R (enzyme complementation) to characterize the potency, efficacy, and β-arrestin bias of new compounds at overexpressed human AT1R. This search led to the identification of two β-arrestin biased ligands with improved potency compared with SII: TRV120023 and TRV120027 selectively and potently engage β-arrestin2 recruitment (EC50 = 44 and 17 nM, respectively at the human AT1R) without any detectable activation of G protein coupling (data for TRV120027 shown in Fig. 1). This contrasts with the robust G protein and β-arrestin2 signals elicited by angiotensin II (EC50 of 1.1 and 9.7 nM for IP1 accumulation and β-arrestin2 recruitment, respectively) and the complete lack of G protein or β-arrestin2 efficacy of angiotensin receptor blockers (ARBs) such as valsartan. TRV120023 and TRV120027 had very similar potency, efficacy, and bias at the rat AT1aR, with EC50 for β-arrestin2 recruitment of 37 and 23 nM, respectively.

Fig. 1.
Fig. 1.

TRV120027 is a highly potent β-arrestin biased ligand at the AT1R. A–C, β-arrestin2 recruitment (black circles) was measured by chemiluminescent β-galactosidase activity, and G protein coupling (red squares) was measured by IP1 accumulation, in HEK cells expressing human AT1R for angiotensin II (A), the ARB valsartan (B), and TRV120027 (C). D, TRV120027 preincubated at escalating concentrations shifts the angiotensin II dose-response curve for IP1 accumulation, consistent with competitive antagonism. Data are means ± standard error for three independent experiments.

To verify β-arrestin efficacy, we also showed that angiotensin II, TRV120023, and TRV120027 evoked β-arrestin1 and β-arrestin2 recruitment, as measured by β-arrestin–yellow fluorescent protein interaction with human AT1R–cyan fluorescent protein (Rajagopal et al., 2006), whereas ARBs did not, consistent with our enzyme complementation result (data not shown). In these same experiments, we saw clear translocation of β-arrestin2–yellow fluorescent protein from cytosol to membrane compartments and internalization of AT1R–cyan fluorescent protein for angiotensin II-, TRV120023-, and TRV120027-treated cells, but no effect of the ARBs valsartan, losartan, or telmisartan, consistent with β-arrestin2 recruitment and coupling to endocytic machinery (Supplemental Fig. 1A). This was confirmed by whole-cell radioligand binding studies, which showed that TRV120027, but not the ARB losartan, reduced angiotensin II binding sites in HEK cells overexpressing human AT1R (Supplemental Fig. 1B).

We chose TRV120027, the more potent ligand, for more thorough characterization. Escalating concentrations of TRV120027 shifted the EC50 of angiotensin II-evoked G protein coupling without suppressing maximal efficacy, consistent with a purely competitive mechanism of action (Fig. 1D). Fitting the data to a Schild model of competitive antagonism (Lew and Angus, 1995) supported this conclusion (Schild slope = 1.04 ± 0.06), with an estimated Kd of 19 nM. This is consistent with radioligand binding studies with 125I-angiotensin II, which showed a Ki for TRV120027 of 16 nM (Supplemental Table 1). These studies also showed that TRV120027 has apparent first-order binding, with a kon of 3.9 × 106 M−1 · min−1, koff of 4.7 × 10−2 min−1, corresponding to a residence half-time of 16 min, similar to that of losartan. The TRV120027 Kd, as calculated from on and off rates, was 12 nM, consistent with the apparent Ki.

TRV120027 was also remarkably specific for the AT1R: at 58 different receptors and channels 10 μM TRV120027 showed less than 15% inhibition of reference radioligand binding (Supplemental Methods). The only receptor other than the AT1R for which TRV120027 had affinity was the angiotensin II type 2 receptor (AT2R), with apparent Ki as measured by radioligand binding of 7 nM.

The Signaling Profile of β-Arrestin Biased Ligands Is a Subset of Angiotensin II-Mediated Signaling.

Consistent with β-arrestin recruitment, in U2-OS cells overexpressing rat AT1aR, angiotensin II, TRV120023, and TRV120027 activated the MAPK ERK1/2, contrasting with valsartan (Fig. 2A) and a range of other ARBs that were inactive (data not shown). Similar results were found in HEK cells (data not shown), which have been widely used to assess β-arrestin signaling (Ahn et al., 2004); in these cells TRV120027 stimulated ERK1/2 with an EC50 of approximately 1 nM. It is noteworthy that of these compounds only angiotensin II activated p38 MAPK (Fig. 2B), highlighting that β-arrestin biased ligands, devoid of G protein coupling, activate only a subset of the normal complement of AT1R signals. We also found that angiotensin II, TRV120023, and TRV120027, but not valsartan, stimulated activation of both Src (Fig. 2C) and eNOS (Fig. 2D). These data suggested that β-arrestin recruitment to the AT1R could engage a pathway whereby Src phosphorylates and activates Akt, which in turn phosphorylates eNOS (Haynes et al., 2003; Suzuki et al., 2006). eNOS activation by phosphorylation may provide a link between AT1R β-arrestin function and cardiovascular tone via regulation of nitric oxide. Consistent with β-arrestin bias, eNOS activation by both TRV120023 and TRV120027 is eliminated by siRNA silencing of β-arrestin2 (Fig. 2E). In contrast, β-arrestin2 silencing reduced Ang II-stimulated eNOS phosphorylation by approximately 50%, suggesting that AngII activates eNOS by both β-arrestin2-dependent and -independent pathways (data not shown); this may also explain why the β-arrestin biased ligands do not elicit as much eNOS phosphorylation as AngII (Fig. 2D). A second siRNA sequence targeting β-arrestin2 yielded similar results (data not shown). These results demonstrate that this signaling effect of the β-arrestin biased ligands is in fact β-arrestin2-mediated.

Fig. 2.
Fig. 2.

β-Arrestin biased ligands stimulate selective signaling in comparison with an unbiased agonist and an unbiased antagonist in U2-OS cells overexpressing rat AT1aR. A–D, activation by phosphorylation was measured by immunoblot with phospho-specific antibodies for ERK1/2 (A), p38 (B), Src (C), and eNOS (D) activation. E, β-arrestin2 siRNA prevents biased ligand-stimulated eNOS phosphorylation compared with a control siRNA. Data are means ± standard errors of three to seven independent experiments. *, p < 0.05 versus vehicle and valsartan and **, p < 0.05 versus the same compound with control (CTL) siRNA, as measured by ANOVA with the Bonferroni multiple comparison test.

We then searched more broadly for signals engaged by our β-arrestin biased ligands, using an antibody array for 55 different signal transduction proteins, largely selective for phosphorylated “active” or “inactive” proteins. Of these, two were strongly activated by the β-arrestin biased ligands but not valsartan: phospho-c-Jun (Fig. 3A) and phospho-FAK (Fig. 3B). Together, these results establish that β-arrestin biased ligands display a pharmacological profile in vitro that is distinct from classic antagonists and here represents a subset of unbiased agonist signaling.

Fig. 3.
Fig. 3.

FAK and c-Jun phosphorylation stimulated by β-arrestin biased ligands, identified from a signal panning screen for changes in expression or phosphorylation of 55 different signaling proteins. Comparison of TRV120023 and TRV120027 to an unbiased agonist and an unbiased antagonist in U2-OS cells overexpressing rat AT1aR identified c-Jun phosphorylation (A) and FAK phosphorylation (B). Data are means ± standard errors of three to seven independent experiments. *, p < 0.05 versus vehicle and valsartan and **, p < 0.05 versus the same compound with control siRNA, as measured by ANOVA with the Bonferroni multiple comparison test.

β-Arrestin Biased Ligands Stimulate Isolated Cardiomyocyte Contractility.

Previous work showed that SII stimulated contractility of isolated murine cardiomyocytes via the AT1aR and β-arrestin2 (Rajagopal et al., 2006), so we used this experimental system to test the cellular effects of TRV120023 and TRV120027 related to cardiovascular function. In agreement with the effects of SII, we found that TRV120027 stimulated cardiomyocyte contractility, but the unbiased antagonist valsartan did not (Fig. 4A). Valsartan, which is highly selective for the AT1R, including selectivity versus the AT2R (Criscione et al., 1993), blocked TRV120027-stimulated cardiomyocyte contractility, indicating that this effect is AT1R mediated. Separate experiments showed similar results for TRV120023 (Fig. 4B) and a lack of effect for the ARBs losartan and telmisartan at receptor-saturating concentrations (data not shown).

Fig. 4.
Fig. 4.

TRV120027 and TRV120023 increase mouse cardiomyocyte contractility. A, angiotensin II (10 μM) and TRV120027 increased fractional shortening, as measured by video edge detection microscopy. Valsartan (1 μM) had no effect, but completely blocked the effect of TRV120027. B, a separate experiment revealed similar properties for TRV120023 and also compared responses with those elicited by 10 nM isoproterenol. Data are presented as mean ± standard error of the percentage increase in shortening of single myocytes after drug treatment. n = 3 independent experiments, each using a different heart, with two myocytes tested for each drug from each heart. *, p < 0.05 versus vehicle and valsartan and **, p < 0.05 versus TRV120027 alone, as measured by ANOVA with the Bonferroni multiple comparison test.

β-Arrestin Biased Ligands Block the AT1R Pressor Response While Stimulating Cardiac Performance.

In vivo, acute AT1R-mediated blood pressure effects are regulated by G protein-stimulated calcium mobilization (Harris et al., 2007; Wynne et al., 2009). Consistent with this, and with the in vitro findings in Fig. 2, both TRV120027 and the unbiased antagonist losartan competitively antagonized the effect of escalating angiotensin II on MAP in rats, as evidenced by an increase in the angiotensin II EC50 (Fig. 5A). For losartan, this competitive antagonism was the only obvious effect. In contrast, TRV120027 also reduced blood pressure independent of angiotensin II and depressed the maximum angiotensin II-evoked blood pressure increase.

Fig. 5.
Fig. 5.

TRV120027 competitively antagonizes angiotensin II blood pressure increases while stimulating cardiac contractility in normal rats. A, losartan and TRV120027 shift the angiotensin II dose-escalation blood pressure response. B, in the same experiment, before angiotensin II treatment, TRV120027, but not losartan, increased cardiomyocyte contractility, as measured by the maximum rate of myocardial contractility vehicle (Vmax). Data are means ± standard errors of seven animals per group. *, p < 0.05 versus vehicle, as measured by ANOVA with the Bonferroni multiple comparison test.

In the same experiment, when we evaluated effects of losartan and TRV120027 alone, before the addition of angiotensin II, we noted that myocardial shortening velocity (Vmax), a measure of cardiac contractility relatively insensitive to changes in vascular or arterial pressure, was increased by TRV120027 but not by losartan (Fig. 5B). This is consistent with our cardiomyocyte contractility study (Fig. 4), a known β-arrestin2-mediated effect (Rajagopal et al., 2006), indicating that the β-arrestin agonist efficacy of TRV120027 modestly stimulates cardiac contractility, a property not shared by ARBs. TRV120027 did not cause any chronotropic effects and did not affect PQ, QRS, or QT intervals, indicating no effect on conduction and instead suggesting that any effect on contractility is probably a result of favorable effects on the mechanics or energetics of myocyte contractility.

To more directly assess how TRV120027 affects cardiac performance, we compared TRV120027 to an ARB in rats by using left ventricular PV loop analysis (PV loops). This technique measures pressure and conductance (corresponding to blood volume) of the left ventricle during increasing occlusion of the inferior vena cava to reduce cardiac-filling pressure. The family of PV loops, with each loop representing the cycle of pressure and volume changes during a single heart beat, allows direct assessment of cardiac performance independent of preload (filling pressure) and afterload (resistance from aortic pressure) (Kohout et al., 2001). The effects of TRV120027 on systolic, diastolic, and cardiac conductance functions are shown in Table 1. For this study we tested the ARB telmisartan because its high solubility allowed for infusion with a volume matched to TRV120027 to eliminate potential nondrug effects; the effects of telmisartan on cardiac function are shown in Supplemental Table 2.

View this table:
TABLE 1

Effect of TRV120027 on cardiovascular parameters

Data are mean ± SEM of three animals.

TRV120027 and telmisartan both produced dose-dependent decreases in MAP (Fig. 6A). Neither compound increased heart rate (Fig. 6B), but TRV120027, unlike telmisartan, increased cardiac contractility: TRV120027 dose-dependently increased the slope of ESPVR (Fig. 6C), a load-independent measure of left ventricular contractility (Little, 1985; Georgakopoulos et al., 1998). Consistent with these effects, TRV120027 also preserved stroke volume (Fig. 6D) and increased normalized PRSW (Table 1). TRV120027 reduced dP/dtmax, a commonly used measure of contractility, but because this parameter correlates with preload it probably reflects the actions of TRV120027 on preload (Schmidt and Hoppe, 1978). In contrast to these effects, telmisartan reduced cardiac performance, causing reduced ESPVR slope and stroke volume. These responses are consistent with the failure of other ARBs to increase cardiac performance (Chan et al., 1992). Representative PV loop families for vehicle control, telmisartan, and TRV120027 are shown in Supplemental Fig. 2. It is noteworthy that TRV120027-enhanced cardiac performance coincided with reduced stroke work, suggesting improved myocardial efficiency. Furthermore, TRV120027 did not cause any electrocardiographic changes, and its effects were rapidly reversible: 5 to 7 min after ceasing infusion, MAP and ESPVR had returned nearly to baseline values (Supplemental Fig. 3). This rapid reversibility is consistent with the short half-life of TRV120027 in vivo: after infusion into normal rats, TRV120027 had a half-life of 1.5 min (Supplemental Methods).

Fig. 6.
Fig. 6.

TRV120027 reduces blood pressure while increasing cardiac performance in normal rats. A, PV loop analysis in normal rats infused with either TRV120027 or telmisartan revealed similar effects on MAP. B, neither compound substantially changes heart rate. C, TRV120027 increased, whereas telmisartan decreased, load-independent cardiac performance, as measured by the slope of ESPVR. D, telmisartan, but not TRV120027, significantly reduced stroke volume. Data shown are mean ± standard errors from three independent experiments with three to four animals per group. *, p < 0.05 versus baseline, as measured by ANOVA with the Bonferroni multiple comparison test.

Similar to TRV120027, TRV120023 reduced MAP while increasing ESPVR (Supplemental Fig. 4, A and B). TRV120026, a third β-arrestin biased ligand with very similar potency for β-arrestin recruitment (EC50 = 24.5 nM at the human AT1R) but 80-fold higher relative specificity for binding the AT1R over the AT2R, showed similar effects, at the same doses, as TRV120027 and TRV120023 on MAP and ESPVR (Supplemental Fig. 4, C and D). Indeed the biggest difference between TRV120026 and either TRV120027 or TRV120023 is an enhanced effect on ESPVR by TRV120026, which in light of this compound's reduced affinity for the AT2R is consistent with an AT1R-mediated effect. The effects of the affinity of TRV120027 for the AT2R, if any, remain unclear; however, these findings, along with the blockade of cardiomyocyte contractility in vitro by an AT1R-selective antagonist, suggest that the enhancement of cardiac contractility of TRV120027 is AT1R-mediated.

Together, the findings presented here demonstrate that TRV120027 depresses blood pressure while increasing cardiac performance, consistent with its antagonist efficacy against angiotensin II pressor activity (Fig. 5), its β-arrestin-mediated stimulation of cardiac contractility in vitro (Fig. 4), and Vmax and ESPVR in rats (Figs. 5 and 6). The stimulation of contractility by TRV120027 is modest compared with the classic inotrope dobutamine, which raises dP/dtmax and more strongly increases ESPVR and also increases heart rate (Supplemental Table 3). This modest efficacy of TRV120027, in conjunction with its afterload reduction, may explain the apparently beneficial energetics of reduced stroke work. This profile suggests that TRV120027 and related β-arrestin biased AT1R agonists could have a beneficial impact on cardiovascular syndromes characterized by hypertension and insufficient cardiac performance.

Discussion

The results presented here describe the unique in vitro and in vivo pharmacology of TRV120027, a novel, selective, and potent β-arrestin biased ligand of the AT1R. This work demonstrates that biased and unbiased ligands are pharmacologically distinct in vivo. TRV120027 functions as an antagonist, analogous to ARBs, when evaluated for G protein coupling, other cellular signals such as p38 MAPK activation, and effects on blood pressure. However, when assessed for efficacy in recruiting β-arrestins to the AT1R, and for downstream effects including receptor internalization, β-arrestin-mediated signals, and cardiomyocyte contractility, TRV120027 is an agonist. This bias is not merely an artifact of differentially amplified assay readouts: IP1 accumulation in response to angiotensin II is more sensitive than β-arrestin2 recruitment, as measured by potency, but undetectable despite strong β-arrestin responses for TRV120023 and TRV120027. This is a hallmark of intrinsic ligand bias, reflecting a stabilization of different receptor conformations than are stabilized by angiotensin II or the unbiased ARBs (Kenakin, 2007). This assay-dependent efficacy violates classic models of molecular pharmacology that assume that all efficacies are correlated and instead requires newer more complicated models of drug action (Kenakin, 2009; Tran et al., 2009).

Because TRV120027 reduces MAP while increasing cardiac contractility, it is intriguing to speculate what benefits it might have in acute heart failure, a syndrome usually comprising low cardiac output, normal or elevated blood pressure, and an elevated renin-angiotensin system. This syndrome represents a vicious cycle in which insufficient cardiac output causes angiotensin II-mediated vasoconstriction, further impairing cardiac output by raising the work required to maintain end organ perfusion, thus leading to further deterioration of cardiac performance. TRV120027 could block the vasoconstrictive effects of renin-angiotensin system activation while promoting cardiac performance via β-arrestin signaling, potentially breaking this vicious cycle and improving patient outcomes. This is a unique profile, which in light of reduced afterload (Ea), myocardial oxygen demand (PV area), and work (stroke work) during TRV120027 treatment, suggests improved energetic or mechanical function of the heart. This is reflected in improved ventriculo-arterial coupling (Ea/ESPVR slope) for TRV120027 not seen with telmisartan (Table 1). Future studies will test the effects of TRV120027 in disease models and elucidate the molecular mechanism whereby these compounds increase cardiac contractility.

Nevertheless, this work clearly establishes TRV120027 as a biased ligand with unique physiological effects compared with unbiased ligands. Our data strongly support two major themes of contemporary GPCR research: the importance of β-arrestin signaling as a distinct set of pathways operating parallel to classic G protein signaling and the utility of biased ligands to selectively engage desired signal transduction pathways from a single receptor to elicit specific physiological effects. Because biased ligands have been identified for numerous receptors (Violin and Lefkowitz, 2007; Kenakin, 2009), ligand bias may be a general strategy that can dissect the biology of GPCRs as both research tools and investigational new drugs. This has important implications for GPCR-targeted drug discovery by suggesting that in many cases the optimal evaluation of GPCR ligands requires characterization across a range of signal transduction pathways, linking the resulting profiles to the relevant pathophysiological setting.

More broadly, as the interplay of β-arrestin and G protein signaling networks continues to be elucidated, biased ligands offer a general strategy for more precisely targeting GPCR functions. The findings presented here demonstrate successful identification of biased ligands with unique pharmacology that translates from in vitro to in vivo studies. Such biased ligands allow an unprecedented level of control of receptor functions, enabling both basic research into GPCR signal transduction mechanisms and potential development of safer, more efficacious therapeutics.

Acknowledgments

We thank Steven Houser and Xiong Wen Chen for performing the isolated cardiomyocyte contractility studies and Robert Lefkowitz, Howard Rockman, and David Soergel for critically reviewing the manuscript.

Footnotes

  • All work was funded by Trevena Inc., a privately held drug discovery company.

  • Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

    doi:10.1124/jpet.110.173005.

  • ↵Embedded Image The online version of this article (available at http://jpet.aspetjournals.org) contains supplemental material.

  • ABBREVIATIONS:

    GPCR
    G protein-coupled receptor
    AngII
    angiotensin II
    AT1R
    angiotensin II type 1 receptor
    AT2R
    angiotensin II type 2 receptor
    SII
    1Sar, 4Ile, 8Ile-angiotensin II
    ARB
    angiotensin receptor blocker
    eNOS
    endothelial nitric-oxide synthase
    MAPK
    mitogen-activated protein kinase
    FAK
    focal adhesion kinase
    MAP
    mean arterial pressure
    PRSW
    preload recruitable stroke work
    PV
    pressure volume
    ESPVR
    end systolic pressure volume relationship
    ANOVA
    analysis of variance
    HEK
    human embryonic kidney
    MEM
    modified Eagle's medium
    FBS
    fetal bovine serum
    IP1
    inositol monophosphate
    RVU
    relative volume unit
    ERK
    extracellular signal-regulated kinase
    U2-OS
    U2-osteosarcoma
    siRNA
    small interfering RNA
    TRV120023
    Sar-Arg-Val-Tyr-Lys-His-Pro-Ala-OH
    TRV120026
    Sar-Arg-Val-Tyr-Tyr-His-Pro-NH2
    TRV120027
    Sar-Arg-Val-Tyr-Ile-His-Pro-d-Ala-OH.

  • Received July 15, 2010.
  • Accepted August 25, 2010.

References

  1. ↵
    1. Ahn S,
    2. Kim J,
    3. Hara MR,
    4. Ren XR,
    5. Lefkowitz RJ
    (2009) β-Arrestin-2 mediates anti-apoptotic signaling through regulation of BAD phosphorylation. J Biol Chem 284:8855–8865.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    1. Ahn S,
    2. Shenoy SK,
    3. Wei H,
    4. Lefkowitz RJ
    (2004) Differential kinetic and spatial patterns of β-arrestin and G protein-mediated ERK activation by the angiotensin II receptor. J Biol Chem 279:35518–35525.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Aplin M,
    2. Christensen GL,
    3. Schneider M,
    4. Heydorn A,
    5. Gammeltoft S,
    6. Kjølbye AL,
    7. Sheikh SP,
    8. Hansen JL
    (2007) The angiotensin type 1 receptor activates extracellular signal-regulated kinases 1 and 2 by G protein-dependent and -independent pathways in cardiac myocytes and langendorff-perfused hearts. Basic Clin Pharmacol Toxicol 100:289–295.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Barnes WG,
    2. Reiter E,
    3. Violin JD,
    4. Ren XR,
    5. Milligan G,
    6. Lefkowitz RJ
    (2005) β-Arrestin 1 and Gαq/11 coordinately activate RhoA and stress fiber formation following receptor stimulation. J Biol Chem 280:8041–8050.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Chan DP,
    2. Sandok EK,
    3. Aarhus LL,
    4. Heublein DM,
    5. Burnett JC Jr.
    (1992) Renal-specific actions of angiotensin II receptor antagonism in the anesthetized dog. Am J Hypertens 5:354–360.
    OpenUrlPubMed
  6. ↵
    1. Criscione L,
    2. de Gasparo M,
    3. Bühlmayer P,
    4. Whitebread S,
    5. Ramjoué HP,
    6. Wood J
    (1993) Pharmacological profile of valsartan: a potent, orally active, nonpeptide antagonist of the angiotensin II AT1-receptor subtype. Br J Pharmacol 110:761–771.
    OpenUrlCrossRefPubMed
  7. ↵
    1. DeWire SM,
    2. Ahn S,
    3. Lefkowitz RJ,
    4. Shenoy SK
    (2007) β-Arrestins and cell signaling. Annu Rev Physiol 69:483–510.
    OpenUrlCrossRefPubMed
  8. ↵
    1. DeWire SM,
    2. Kim J,
    3. Whalen EJ,
    4. Ahn S,
    5. Chen M,
    6. Lefkowitz RJ
    (2008) β-Arrestin-mediated signaling regulates protein synthesis. J Biol Chem 283:10611–10620.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Georgakopoulos D,
    2. Mitzner WA,
    3. Chen CH,
    4. Byrne BJ,
    5. Millar HD,
    6. Hare JM,
    7. Kass DA
    (1998) In vivo murine left ventricular pressure-volume relations by miniaturized conductance micromanometry. Am J Physiol Heart Circ Physiol 274:H1416–H1422.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Gesty-Palmer D,
    2. Chen M,
    3. Reiter E,
    4. Ahn S,
    5. Nelson CD,
    6. Wang S,
    7. Eckhardt AE,
    8. Cowan CL,
    9. Spurney RF,
    10. Luttrell LM,
    11. et al
    . (2006) Distinct β-arrestin- and G protein-dependent pathways for parathyroid hormone receptor-stimulated ERK1/2 activation. J Biol Chem 281:10856–10864.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Harris DM,
    2. Cohn HI,
    3. Pesant S,
    4. Zhou RH,
    5. Eckhart AD
    (2007) Vascular smooth muscle G(q) signaling is involved in high blood pressure in both induced renal and genetic vascular smooth muscle-derived models of hypertension. Am J Physiol Heart Circ Physiol 293:H3072–H3079.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Haynes MP,
    2. Li L,
    3. Sinha D,
    4. Russell KS,
    5. Hisamoto K,
    6. Baron R,
    7. Collinge M,
    8. Sessa WC,
    9. Bender JR
    (2003) Src kinase mediates phosphatidylinositol 3-kinase/Akt-dependent rapid endothelial nitric-oxide synthase activation by estrogen. J Biol Chem 278:2118–2123.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Holloway AC,
    2. Qian H,
    3. Pipolo L,
    4. Ziogas J,
    5. Miura S,
    6. Karnik S,
    7. Southwell BR,
    8. Lew MJ,
    9. Thomas WG
    (2002) Side-chain substitutions within angiotensin II reveal different requirements for signaling, internalization, and phosphorylation of type 1A angiotensin receptors. Mol Pharmacol 61:768–777.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Hopkins AL,
    2. Groom CR
    (2002) The druggable genome. Nat Rev Drug Discov 1:727–730.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Jaleel N,
    2. Nakayama H,
    3. Chen X,
    4. Kubo H,
    5. MacDonnell S,
    6. Zhang H,
    7. Berretta R,
    8. Robbins J,
    9. Cribbs L,
    10. Molkentin JD,
    11. et al
    . (2008) Ca2+ influx through T- and L-type Ca2+ channels have different effects on myocyte contractility and induce unique cardiac phenotypes. Circ Res 103:1109–1119.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Kenakin T
    (2007) Functional selectivity through protean and biased agonism: who steers the ship? Mol Pharmacol 72:1393–1401.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Kenakin TP
    (2009) 7TM receptor allostery: putting numbers to shapeshifting proteins. Trends Pharmacol Sci 30:460–469.
    OpenUrlPubMed
  18. ↵
    1. Kim J,
    2. Ahn S,
    3. Rajagopal K,
    4. Lefkowitz RJ
    (2009) Independent β-arrestin2 and Gq/protein kinase Cζ pathways for ERK stimulated by angiotensin type 1A receptors in vascular smooth muscle cells converge on transactivation of the epidermal growth factor receptor. J Biol Chem 284:11953–11962.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Kohout TA,
    2. Takaoka H,
    3. McDonald PH,
    4. Perry SJ,
    5. Mao L,
    6. Lefkowitz RJ,
    7. Rockman HA
    (2001) Augmentation of cardiac contractility mediated by the human β3-adrenergic receptor overexpressed in the hearts of transgenic mice. Circulation 104:2485–2491.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Kule CE,
    2. Karoor V,
    3. Day JN,
    4. Thomas WG,
    5. Baker KM,
    6. Dinh D,
    7. Acker KA,
    8. Booz GW
    (2004) Agonist-dependent internalization of the angiotensin II type one receptor (AT1): role of C-terminus phosphorylation in recruitment of β-arrestins. Regul Pept 120:141–148.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Lew MJ,
    2. Angus JA
    (1995) Analysis of competitive agonist–antagonist interactions by nonlinear regression. Trends Pharmacol Sci 16:328–337.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Little WC
    (1985) The left ventricular dP/dtmax-end-diastolic volume relation in closed-chest dogs. Circ Res 56:808–815.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Mailman RB
    (2007) GPCR functional selectivity has therapeutic impact. Trends Pharmacol Sci 28:390–396.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Rajagopal K,
    2. Whalen EJ,
    3. Violin JD,
    4. Stiber JA,
    5. Rosenberg PB,
    6. Premont RT,
    7. Coffman TM,
    8. Rockman HA,
    9. Lefkowitz RJ
    (2006) β-Arrestin2-mediated inotropic effects of the angiotensin II type 1A receptor in isolated cardiac myocytes. Proc Natl Acad Sci USA 103:16284–16289.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Reiter E,
    2. Lefkowitz RJ
    (2006) GRKs and β-arrestins: roles in receptor silencing, trafficking and signaling. Trends Endocrinol Metab 17:159–165.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Schmid CL,
    2. Bohn LM
    (2009) Physiological and pharmacological implications of β-arrestin regulation. Pharmacol Ther 121:285–293.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Schmidt HD,
    2. Hoppe H
    (1978) Preload dependence of dP/dtmax, VCEmax and calculated Vmax compared to the inotropic sensitivity of these indices of cardiac contractility. Basic Res Cardiol 73:380–393.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Suzuki H,
    2. Eguchi K,
    3. Ohtsu H,
    4. Higuchi S,
    5. Dhobale S,
    6. Frank GD,
    7. Motley ED,
    8. Eguchi S
    (2006) Activation of endothelial nitric oxide synthase by the angiotensin II type 1 receptor. Endocrinology 147:5914–5920.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Tran JA,
    2. Chang A,
    3. Matsui M,
    4. Ehlert FJ
    (2009) Estimation of relative microscopic affinity constants of agonists for the active state of the receptor in functional studies on M2 and M3 muscarinic receptors. Mol Pharmacol 75:381–396.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Violin JD,
    2. Lefkowitz RJ
    (2007) β-Arrestin biased ligands at seven-transmembrane receptors. Trends Pharmacol Sci 28:416–422.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Violin JD,
    2. Dewire SM,
    3. Barnes WG,
    4. Lefkowitz RJ
    (2006) G protein-coupled receptor kinase and β-arrestin-mediated desensitization of the angiotensin II type 1A receptor elucidated by diacylglycerol dynamics. J Biol Chem 281:36411–36419.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Wei H,
    2. Ahn S,
    3. Shenoy SK,
    4. Karnik SS,
    5. Hunyady L,
    6. Luttrell LM,
    7. Lefkowitz RJ
    (2003) Independent β-arrestin 2 and G protein-mediated pathways for angiotensin II activation of extracellular signal-regulated kinases 1 and 2. Proc Natl Acad Sci USA 100:10782–10787.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Wynne BM,
    2. Chiao CW,
    3. Webb RC
    (2009) Vascular smooth muscle cell signaling mechanisms for contraction to angiotensin II and endothelin-1. J Am Soc Hypertens 3:84–95.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Zidar DA,
    2. Violin JD,
    3. Whalen EJ,
    4. Lefkowitz RJ
    (2009) Selective engagement of G protein coupled receptor kinases (GRKs) encodes distinct functions of biased ligands. Proc Natl Acad Sci USA 106:9649–9654.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools

Research ArticleCardiovascular

Selectively Engaging β-Arrestins at the Angiotensin II Type 1 Receptor Reduces Blood Pressure and Increases Cardiac Performance

Jonathan D. Violin, Scott M. DeWire, Dennis Yamashita, David H. Rominger, Lisa Nguyen, Kevin Schiller, Erin J. Whalen, Maxine Gowen and Michael W. Lark
Journal of Pharmacology and Experimental Therapeutics December 1, 2010, 335 (3) 572-579; DOI: https://doi.org/10.1124/jpet.110.173005
Reddit logo Twitter logo Facebook logo Mendeley logo

Jump to section

Advertisement