The apelin receptor inhibits the angiotensin II type 1 receptor via allosteric trans-inhibition

Abstract

Background and purpose: The apelin receptor (APJ) is often co-expressed with the angiotensin II type-1 receptor (AT1) and acts as an endogenous counter-regulator. Apelin antagonizes Ang II signalling, but the precise molecular mechanism has not been elucidated. Understanding this interaction may lead to new therapies for the treatment of cardiovascular disease.

Experimental approach: The physical interaction of APJ and AT1 receptors was detected by co-immunoprecipitation and bioluminescence resonance energy transfer (BRET). Functional and pharmacological interactions were measured by G-protein-dependent signalling and recruitment of β-arrestin. Allosterism and cooperativity between APJ and AT1 were measured by radioligand binding assays.

Key results: Apelin, but not Ang II, induced APJ : AT1 heterodimerization forced AT1 into a low-affinity state, reducing Ang II binding. Likewise, apelin mediated a concentration-dependent depression in the maximal production of inositol phosphate (IP(1) ) and β-arrestin recruitment to AT1 in response to Ang II. The signal depression approached a limit, the magnitude of which was governed by the cooperativity indicative of a negative allosteric interaction. Fitting the data to an operational model of allosterism revealed that apelin-mediated heterodimerization significantly reduces Ang II signalling efficacy. These effects were not observed in the absence of apelin.

Conclusions and implications: Apelin-dependent heterodimerization between APJ and AT1 causes negative allosteric regulation of AT1 function. As AT1 is significant in the pathogenesis of cardiovascular disease, these findings suggest that impaired apelin and APJ function may be a common underlying aetiology.

Linked article: This article is commented on by Goupil et al., pp. 1101-1103 of this issue. To view this commentary visit http://dx.doi.org/10.1111/bph.12040.

Figure 1

Physical interaction between endogenous APJ and AT1 receptors in HEK293 cells. (A) Immunoprecipitation of AT1 using α-AT1 antibody followed by Western blot with α-APJ antibody; lane 1: whole cell lysate immunoprecipitated with α-AT1; lane 2: whole cell lysates subjected to precipitation with AT1 antibody followed by exposure to the chemical cross-linker DTSSP; lane 3: whole cell lysate subjected to precipitation with AT1 antibody prior to cross-linking. Left panel, APJ immunoreactivity, Middle panel shows AT1 immunoreactivity; right panel, merged of left and middle panels. M: Marker. (B) In the reciprocal experiment, APJ receptor immunoprecipitated with α-APJ antibodies, and co-enriched AT1 receptor was detected by Western blot; lane 1: whole cell lysate immunoprecipitated with α-APJ; lane 2: whole cell lysates subjected to precipitation with APJ antibody followed by exposure to the chemical cross-linker DTSSP; lane 3: whole cell lysate subjected to precipitation with APJ antibody prior to cross-linking. Left panel, AT1 immunoreactivity; middle panel, APJ immunoreactivity; right panel, merged of left and middle panels. Triangles indicate the position at which monomers and dimers migrated through the gel.

Figure 2

Specific physical interaction between APJ and AT1. (A) Immunoprecipitation of tagged APJ using α-HA antibody followed by Western blot with α-Flag antibody to detect tagged AT1 revealed a co-enrichment of the two receptors (lanes 1, 2). In the reciprocal experiment, AT1 receptors were precipitated with α-Flag antibodies, and co-enriched APJ receptors were detected by Western blot (lanes 3, 4) with α-HA antibodies. (B) Control precipitation and cross-linking studies reveals specificity of the APJ interaction with AT1. Top panel, AT1 immunoreactivity using antibodies directed at the native receptor. Middle panel, APJ immunoreactivity using antibodies directed at the native receptor. Bottom panel, top and middle panels merged. M: Marker; lane 1: negative control, untransfected HEK293 cell lysate; lane 2: whole cell lysate from cells transfected with pHA-APJ alone; lane 3: whole cell lysate from cells transfected with pFlag-AT1 alone; lane 4: IP control, lysate from cells transfected with pHA-APJ and pFlag-AT1 and precipitated with α-c-Myc; lane 5: untransfected cell lysate precipitated with α-HA; lane 6: whole cell lysate from cells transfected with pHA-APJ and pFlag-AT1 and subjected precipitation with α-HA antibodies followed by cross-linking with DTSSP; lane 7: whole cell lysate from cells transfected with pHA-APJ and pFlag-AT1 and subjected precipitation with α-Flag antibodies followed by cross-linking with DTSSP. The use of antibodies directed against the native APJ and AT1 reveal trace levels of endogenous APJ and AT1 in lanes 1, 2, 3. Triangles indicate the position at which monomers and dimers migrated through the gel. Lanes 1–5 are from the same blot. Lanes 6, 7 are from a separate experiment. All lanes are presented as discrete panels to delineate the different conditions under which each sample was prepared and to facilitate comparison between the cross-linking samples and those that were obtained without cross-linking. Ap13, but not Ang II, enhances the interaction between APJ and AT1. Immunoprecipitation and Western blot of HEK293T cells transfected with pHA-APJ and pFlag-AT1 and incubated with to Ap13 (100 nM) or Ang II (100 nM), alone or in combination. (C) Immunoprecipitation of HA-tagged APJ using α-HA antibody followed by Western blot with α-AT1 antibodies revealed a co-enrichment of the two receptors under vehicle (PBS)-treated conditions (lane 1). Immunoreactivity for AT1 was slightly decreased in the presence of Ang II (lane 2), but significantly increased in the presence of Ap13 alone (lane 3) and in combination with Ang II (lane 4). (D) Immunoprecipitation of Flag-tagged AT1 using α-Flag antibody followed by Western blot with α-APJ antibodies revealed a similar co-enrichment of the two receptors under vehicle (PBS)-treated conditions (lane 1). Immunoreactivity for APJ was slightly decreased in the presence of Ang II (lane 2), but significantly increased in the presence of Ap13 alone (lane 3) and in combination with Ang II (lane 4).Quantification of individual band intensity by optical density analysis is presented below each panel. *P < 0.01 when compared with control, *P < 0.001 when compared with Ang II as determined by anova with Bonferroni's multiple comparison test. IP, immunoprecipitation; WB, Western blot.

Figure 3

Ap13 but not Ang II induces APJ : AT1 heterodimerization in living cells. (A) HEK-293 cells transfected with APJ-Rluc and AT1-eYFP treated with either Ap13 (100 nM), Ang II (100 nM), Ap13 + Ang II (100 nM each). *P < 0.001 when compared with Ang II, ^‡P < 0.001 when compared with Ap13 as determined by anova with Bonferroni's multiple comparison test. (B) Saturation of donor to acceptor BRET ratio in the presence of vehicle, Ap13 (100 nM) and Ang II (100 nM). Data presented are from two or three independent experiments each with an n = 3.The peak light intensities of APJ-RLuc and AT1-eYFP were measured at 480 and 530 nM respectively. The BRET signal is determined by calculating the ratio of the light intensity emitted by AT1-eYFP over the light intensity emitted by the APJ-RLuc, and the netBRET ratio is calculated by subtracting the vehicle induced BRET ratio from the ligand induced BRET ratio.

Figure 4

Effects of Ap13/APJ on Ang II binding. APJ alone has no significant effect on [¹²⁵I]-Tyr⁴-Ang II binding to membranes from cells expressing AT1 receptor alone (A), or from cells expressing AT1 and APJ (B). Six different concentrations of [¹²⁵I]-Tyr⁴-Ang II were incubated with 0.5 μg of membranes for 1 h at room temperature before quenching the reaction by rapid filtration to separate bound Ang II from free Ang II. Nonspecific binding was determined by using excess unlabelled Ang II (10 μM). Specific binding was calculated by subtracting nonspecific binding from total binding. K_D and B_max values are reported in Table 1. Ap13-activated APJ decreases binding of [¹²⁵I]-Tyr⁴-Ang II to AT1. (C) Increasing concentrations of apelin reduced the B_max of Ang II to the AT1 receptor when co-expressed with APJ. An extra sum-of-squares F test determined that estimates of K_D were not significantly different, and therefore were shared, across data sets. In contrast, the B_max was significantly different across data sets (P < 0.05). (D) Increasing concentrations of Ang II had no effect on the B_max of Ap13 to APJ when co-expressed with AT1. An extra sum-of-squares F test determined that estimates of K_D were not significantly different and therefore were shared, across data sets. Similarly, the B_max was not significantly different across data sets (P > 0.05). Each symbol represents the mean specific activity ± SEM (n = 3) from three independent experiments. Curves superimposed on the data represent the best fit of the data from equation 1 obtained by nonlinear regression analysis performed using GraphPad Prism 5. Details of the analysis are provided in the data analysis section of Methods.

Figure 5

Ap13-activated APJ modulates the affinity state of AT1. (A) When AT1 is co-expressed with APJ, Ap13 decreases the affinity of [¹²⁵I]-Tyr⁴-Ang II for AT1. Competition binding assays were performed on membranes from cells expressing both APJ and AT1 by using 0.28 nM of [¹²⁵I]-Tyr⁴–Ang II in the presence of increasing concentrations of unlabelled Ang II or in the presence of increasing concentrations of unlabelled Ang II plus 100 nM of unlabelled Ap13. GraphPad Prism5 was used to calculate IC₅₀ using equation 2 as described in the Methods section. (B) Competitive binding between [¹²⁵I]-Sar², Ile⁸-Ang II and unlabelled Ang II in the absence and presence of Ap13 (1 μM). The curve was preferentially fit using equations 3 and 2, as determined by an extra sum-of-squares F test, to a two- and one-site inhibition mass–action curve respectively. Each symbol represents the mean percentage of specific binding ± SEM (n = 4–6) from two independent experiments.

Figure 6

APJ : AT1 heterodimerization antagonizes AT1 function. (A) Inhibition of Gα_q signalling pathway of AT1 by Ap13-activated APJ. Inositol-1-Phosphate (IP₁) production was measured using CHO-K1-APJ : AT1 cells stably and equivalently expressing AT1 and APJ receptors in the presence of Ang II (0–2.5 μM) alone or in the presence of a fixed concentration of Ap13. (B) Inhibition of β-arrestin recruitment to AT1 by Ap13-activated APJ. PathHunter AT1 β-Arrestin EFC cells were transfected with pHA-APJ and incubated with Ang II (0.0–100 μM) alone or with a fixed concentration of Ap13. Each symbol represents the mean percent activation of AT1 ± SEM (n = 6) by Ang II at the indicated concentration. In both panels, curves represent the best global fit of the operation model of allosterism as determined using Equation 4.

Figure 7

Apelin-dependent formation of an APJ : AT1 heterodimer results in negative allosteric modulation of AT1. (A) In the absence of Ap13, Ang II binds to the AT1 receptor (seven-transmembrane spanning receptor), activating the receptor and the subsequent transduction of signals through G-protein-dependent and -independent pathways. Even in the absence of Ap13, some APJ receptor may exist in a confirmation that favours heterodimerization with AT1 (as demonstrated in Figures 1 and 2). This is indicated by the dashed arrow in panel A. (B) Ap13 binding to APJ (seven-transmembrane spanning receptor) stimulates both APJ signalling and the formation of the APJ : AT1 heterodimer; a mechanism that is consistent with the role of apelin and APJ as counter-regulators of Ang II signalling via AT1. (C) The physical interaction of APJ and AT1 results in the transduction of an antagonist effect from APJ through AT1 to inhibit Ang II binding and AT1 signalling (red X). Thus, Ap13 functions allosterically (via APJ) to alter the binding and function of AT1.