The ghrelin receptor agonist AC01 improves cardiomyocyte contractility through Gαi signalling and without Ca2+ mobilization

Time of primary review: 56 days Reduced cardiac contractility is a central feature of heart failure (HF) with reduced ejection fraction (HFrEF). Increased cardiomyocyte contractility can be achieved by increasing intracellular [Ca2+] and/or increasing the sensitivity of myofilaments to Ca2+. The latter occurs by altering the function of regulatory proteins associated with thin and thick filaments,1 which enhance the cross-bridge function in response to Ca2+. While this mechanism can improve systolic performance, it may also impair myocytes relaxation due to prolonged calcium offloading from the myofilaments, however in vivo myocardial function is governed by complex myofilament regulatory mechanisms involving dynamic conformational changes in key proteins, thus Ca2+ sensitivity alone does not always reliably predict overall cardiac performance.2 Ghrelin is the endogenous ligand of the growth hormone secretagogue receptor 1a (GHSR1a), a G-protein coupled receptor expressed in different tissues including heart3,4 and cardiomyocytes.5 Acyl (‘activated’) ghrelin improved cardiac output in human HFrEF6 and increased contractility of mouse cardiomyocytes without increasing Ca2+ mobilization suggesting a lower risk of calcium-related adverse events.6 AC01 (previously denoted HM01)7 is an orally available, small molecule with high affinity for the human GHSR1a receptor (Ki = 1.42 ± 0.36 nM)7 and potent agonistic activity (EC50 = 1.3 nM). We aimed to investigate the effects of AC01 in primary cardiomyocytes isolated from healthy control and HFrEF mice to explore its potential role in increasing contractility. This study was approved by the Stockholm North Ethical Committee on Animal Experiments (2155-2020) according to Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes or the current NIH guidelines. C57BL/6J male mice (12–16 weeks old) underwent myocardial infarct (MI) or sham procedures.6 Mice were anesthetized with a gas mixture of oxygen and isoflurane (1.5–2.0% for maintenance; up to 3% for induction) and received analgesic treatment with buprenorphine (0.05 mg/kg, s.c.) 30 min before and for 2 days after surgery every 8–12 h. MI mice, 4–6 weeks after surgery, showed reduced left ventricle ejection fraction (EF) on echo-cardiography (Figure 1A) and higher weights of heart (131.8 ± 7.6 vs. 226.7 ± 47.6 mg, P < 0.05) and lung (120.8 ± 6.0 mg vs. 221.0 ± 52.5 mg, P < 0.05), consistent with post-infarction cardiac remodelling and pulmonary congestion of HFrEF. Mice were euthanized by isoflurane (5%) and underwent exsanguination and organ collection. Hearts were mounted in a Langendorff system, and retrogradely perfused with Ca2+ free buffer containing collagenases (Worthington #LS004177) for 15 min at 37°C followed by mechanical dissociation with micro forceps to facilitate ventricular cardiomyocytes isolation.8,9 Cardiomyocytes were treated with AC01, vehicle, or AC01 plus the specific ghrelin receptor antagonist D-Lys3-GHRP-6 (D-Lys, #G4535, Sigma Aldrich) (Figure 1A). Fluorescent probe Fluo-3 AM was used to trace cytoplasmic Ca2+. Cardiomyocytes were plated on a perfusion and electrical stimulation chamber mounted on a confocal microscope. Cardiomyocytes were continuously perfused at room temperature with O2/CO2 (95/5%) bubbled Tyrode solution (NaCl 121, KCl 5.0, CaCl2 1.8, MgCl2 0.5, NaH2PO4 0.4, NaHCO3 24, EDTA 0.1, and glucose 5.5 mM). Cardiomyocytes were stimulated to contract at 1 Hz using voltage pulses delivered via two platinum electrodes attached to the chamber. Measurements were only performed on cardiomyocytes that responded to electrical stimulation without spontaneous contraction, and displayed normal morphology (e.g. striated, ‘‘brick shaped’’). Fluorescence was measured using a confocal microscope (Bio-Rad MRC1024; 40× oil immersion lenses) in line scan mode with 491 nm/ > 515 nm wavelength used for excitation, and emitted light respectively. The line scan was oriented along the long axis of the cardiomyocyte. Bio-Rad LaserSharp 2000 software was used for image acquisition, and Fiji/ImageJ software for analyses. Contractility was assessed by measuring cell length in the relaxed, and the maximally contracted state in order to calculate the length fractional shortening. The comparisons between groups are based on measurements performed at the single-cell level, where each variable (e.g. FS% and Ca²⁺ transients) reflects the functional response of an individual cardiomyocyte. Given that these parameters are intrinsic to each cell, they were treated as independent observations in the analysis. Inotropic effect of AC01 in mouse cardiomyocytes through ghrelin receptor signalling. Bar graph data are shown as mean ± SD of the number of mice (A), cardiomyocytes (B–C) or aliquots of cardiomyocytes (D–E) treated as stated. N denotes technical replicates and n denotes biological replicates (e.g. number of animals used for independent cardiomyocyte isolations in each experiment). Mann–Whitney test was used to compare differences between the SHAM and MI groups (A). One-way ANOVA analysis was used for multiple comparisons between more than two groups (B–E). A P < 0.05 was considered statistically significant where * indicates P < 0.05; **P < 0.01; *** P < 0.001. (A) Protocol. Mice underwent myocardial infarction (MI) through left anterior descendant (LAD) coronary artery ligation or sham procedures and after 4–6 weeks, development of congestive HFrEF was assessed through echocardiography. Cardiomyocytes were isolated from hearts of SHAM or MI mice through retrograde perfusion with Ca2+ free buffer containing collagenases (Worthington #LS004177) for 15 min at 37°C followed by mechanical dissociation to disrupt the cardiac structure. Cardiomyocytes were treated as indicated to perform line scan confocal experiments or biochemistry experiments as indicated. Figure created using icons obtained from Biorender.com B. AC01 in sham cardiomyocytes (including AC01 titration). Representative confocal line scan images (top), AC01 increased the depth of edges invaginations in the intense green areas, representing shortening of the cardiomyocyte during systole. Scale bars: vertical = 50 μm, horizontal = 1 second. AC01 increased contractility of cardiomyocytes dose dependently; the ghrelin receptor antagonist D-Lys blocked the effect of AC01 (n = 5, N = 19–44); the mean ± SD of Ca2+ transient amplitude (n = 5, N = 19–44) and representative Ca²⁺ transient curves were unchanged. Data obtained from 5 independent experiments using 5 sham mice. (C) AC01 in HFrEF post MI cardiomyocytes. Representative confocal line scan images (top). Scale bars: vertical = 50 μm, horizontal = 1 second. AC01 increased contractility, which was blocked by D-Lys (n = 3, N = 23–59). AC01 alone and with the ghrelin receptor antagonist decreased the mean of Ca2+ transient amplitude (n = 3, N = 23–59) and in representative Ca²⁺ transient curves. Data obtained from 3 independent experiments using 3 HFrEF mice. (D) Protein kinase A activity AC01 treatment reduced PKA activity. This effect was blocked by D-Lys and by PTX (n = 5, N = 6–8, 100000 cells or more per condition). (E) cTnI phosphorylation. Representative Western blot (left) and mean ± SD (right) for phospho-(Ser 23–24) (Cell Signalling, #4004S) and total cTnI (Cell Signalling #4002S); 10 μg of protein were loaded per line in Western blot gel. AC01 reduced PKA-mediated-phosphorylation of cTnI. This effect was blocked by D-Lys and by PTX (n = 5, N = 6–7, 100000 cells or more per condition). The total incubation time of PTX was 1 h. In D and E, Data obtained from 5 independent experiments using 5 sham mice. Figure created with Biorender.com. AC01 increased fractional shortening of sham cardiomyocytes in a dose-dependent manner with a statistically significant response at 0.5 and 1 μM of AC01. Pre-treatment with D-Lys (3 μM), blocked the inotropic effect of AC01 (Figure 1B). The fluorescent signal (F/Fo) was calculated as the ratio between resting fluorescence (Fo) and the peak fluorescence (F) of the [Ca2+]i transient, which enables comparison between cells. The amplitudes [Ca2+]i transients were not affected by treatment with AC01 (Figure 1B). Thus, AC01 increased contractility without enhancing Ca2+ transients amplitudes. AC01 increased contractility also in ventricular cardiomyocytes isolated from HFrEF post MI and this was again blocked by D-Lys. The amplitudes of Ca2+ transients were not increased; instead, they appeared to decrease with AC01 treatment alone or in combination with D-Lys (Figure 1C). Cardiomyocytes isolated from sham mice were aliquoted and randomized to treatment with Tyrode buffer alone (vehicle), AC01 alone, D-Lys plus AC01, or pre-treatment (1 h) with the specific Gαi blocker pertussis toxin (PTX, #P2980-50UG, Sigma Aldrich) plus AC01 (Figure 1D). An ELISA-based PKA kinase activity assay (ab139435, Abcam) assessed PKA activity in cardiomyocytes. Western blots were performed on protein lysates to measure phospho-(Ser 23-24) cTnI (Cell Signalling, #4004S) and total cTnI (Cell Signalling #4002S). Cardiomyocytes treated with AC01 displayed decreased PKA activity and reduced phosphorylation of cTnI (Serine 23-24) compared to vehicle treated cardiomyocytes; this effect was blocked by pre-treatment with D-Lys (3 μM) or with PTX (12.5 µM) (Figure 1E and F). Ghrelin has previously been shown to increase cardiac output and cardiomyocyte contractility without increasing Ca2+ transients and to reduce cTnI phosphorylation.6 Here we show that the ghrelin receptor agonist AC01, similarly to ghrelin, increases contractility in cardiomyocytes without increasing Ca2+ transients, and reduces cTnI phosphorylation. The AC01 effects were blocked by the specific ghrelin receptor antagonist D-Lys, confirming that the effect on contractility is controlled by ghrelin receptor signalling. AC01 signalling was also blocked by the Gαi inhibitor, PTX, suggesting that AC01 signals through GHSR-Gαi. Unrelated to the primary findings, in HFrEF cardiomyocytes, AC01 treatment appeared to cause a reduction of Ca2+ transient amplitudes (Figure 1C). This may result from GHSR-Gαi signalling counterbalancing a presumedly elevated Gαs and Gαq/11 signalling axis in the early stage of post-ischemic HF, or it may stem from potential interaction with other cellular mediators involved in GHSR signalling. Activation of Gαi inhibits adenylyl cyclase, leading to reduced cAMP production and PKA activity, which in turn could dampen SR Ca2+ overload and decrease potassium currents.10 In a broader context, targeting Gαi activation might reduce the predisposition to ventricular arrhythmia ad atrial fibrillation.11,12 Ghrelin receptor has been showed to be constitutively coupled to multiple G-proteins, in particular Gαq/11, initiating distinct cascades such as PI3K/AKT, MAPK, and PKC pathways.13 In our study, AC01 appears to preferentially decrease cTnI phosphorylation, suggesting a biased agonist profile through Gαi. Moreover, in HF cardiomyocytes, D-Lys in combination with AC01 further lowered Ca2+ transient amplitudes, suggesting complex ghrelin receptor signalling and potential crosstalk with other HF-related signalling. Ca2+ transients appeared to be slightly higher in the HF condition than in sham cardiomyocytes, which may be consistent with cardiomyocyte adaptations to stressors from HF.6,14 This is not consistent with long-standing HFrEF and reduced cardiomyocyte contractility, and this is a limitation of our model. Yet, this did not affect its contractile effect since AC01 increased fractional shortening in both sham and HFrEF. Importantly, reducing Ca²⁺ overload in HFrEF cardiomyocytes may have significant therapeutic implications. Elevated intracellular Ca²⁺ promotes spontaneous sarcoplasmic reticulum Ca²⁺ release events that can trigger delayed afterdepolarizations and ventricular arrhythmias.15 Moreover, sustained calcium overload exacerbates oxidative stress and mitochondrial dysfunction, ultimately leading to apoptosis and progressive myocardial injury, while ghrelin treatment has an anti-apoptotic effect in cardiomyocytes.16 Thus, AC01 may offer a safer inotropic approach in HFrEF by normalizing Ca²⁺ handling and reducing the risk of calcium-mediated adverse outcomes, including arrhythmia and cytotoxicity. Future investigations employing different HF models associated with increased expression of Gαi17 could provide valuable insights into the potential benefits or limitations of AC01 in enhancing Gαi signalling. Changes in phosphorylation of cTnI have previously been shown to alter myofilament Ca2+ sensitivity in skinned myocytes through a cAMP-PKA-dependent mechanism.18 AC01 reduced cTnI phosphorylation and PKA activity, an effect blocked by the ghrelin receptor antagonist and by PTX, which catalyses the ADP-ribosylation of the α subunits of the heterotrimeric guanine nucleotide regulatory proteins Gαi, Gαo, and Gαt, thus blocking their coupling and activation. Nevertheless, AC01 may activate additional compensatory mechanisms, beyond those explored in our study, that contribute to maintain Ca2+ transients within physiological levels. The enhanced sensitivity of myofibrils to Ca2+, however, likely represents the primary driver of the inotropic response. The ghrelin receptor agonist AC01 exerts inotropic effects in cardiomyocytes without increasing Ca2+ transients through functional selectivity to Gαi pathway following activation of ghrelin receptor(s). Our findings reinforce the potential of AC01 in the ongoing Phase 1b/2a clinical trial in patients with HFrEF (GOAL-HF1; NCT05642507). G.P. and L.H.L. conceived the research study. G.P., L.H.L, C.H., M.C, P.C, M.S., P.S., G.W., E.L., A.G., J.S., D.C.A. contributed to the acquisition, providing reagents, analysis, or interpretation of data. G.P., and L.H.L. drafted the manuscript and all the authors revised it and contributed to its intellectual content. All the authors approved the version to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This work was supported by grants to LHL from The Swedish Research Council (grant 523-2014-2336), the Swedish Heart-Lung foundation (grants 20150557 and 20190310), Karolinska Institutet (grant 2-70/2014), Stockholm County Council (grant 20140220). GP and D.C.A. were supported by the Swedish Heart-Lung foundation (grants 20210607 and 20210498). MC was supported by Swedish Research Council (2020-01645 and 2024-02930) and the Swedish Heart-Lung foundation (20210431 and 20240906). AnaCardio and Helsinn provided the AC01 molecule. The data underlying this article will be shared on reasonable request to the corresponding author.