Small molecule-induced ERBB4 activation to treat heart failure

Small molecule-induced ERBB4 activation to treat heart failure | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Small molecule-induced ERBB4 activation to treat heart failure Vincent Segers, Julie Cools, Eline Feyen, Siel Van den Bogaert, and 16 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4175488/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 10 Jan, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Heart failure is a common and deadly disease, requiring new therapeutic approaches. The neuregulin-1 (NRG1)/erythroblastic leukemia viral oncogene homolog 4 (ERBB4) pathway is an interesting target because of its cardioprotective effects. The therapeutic use of recombinant NRG1 has been difficult, because it requires intravenous administration and is non-selective for the ERBB4 receptor. Moreover, development of small-molecule agonists of receptor dimers is generally considered to be challenging. Here, we hypothesized that small-molecule-induced activation of ERBB4 is feasible and can protect against myocardial cell death and fibrosis. To this end, we screened 10,240 compounds for their ability to induce homodimerization of ERBB4. We identified a series of 8 structurally similar compounds (named EF-1 – EF-8) that concentration-dependently induced ERBB4 dimerization, with EF-1 being the most potent. EF-1 decreased in an ERBB4-dependent manner cell death and hypertrophy in cultured atrial cardiomyocytes and collagen production in cultured human cardiac fibroblasts. EF-1 also inhibited angiotensin-II (AngII)-induced myocardial fibrosis in wild-type mice, but not in Erbb4-null mice. Additionally, EF-1 decreased troponin release in wild-type mice treated with doxorubicin (DOX), but not in Erbb4-null mice. Finally, EF-1 improved cardiac function in a mouse model of myocardial infarction (MI). In conclusion, we show that small-molecule-induced ERBB4 activation is possible, displaying anti-fibrotic and cardiomyocyte protective effects in the heart. This study can be the start for the development of small-molecule ERBB4 agonists as a novel class of drugs to treat heart failure. Biological sciences/Drug discovery/Drug screening/High-throughput screening Biological sciences/Physiology/Cardiovascular biology/Cardiovascular diseases/Heart failure Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Over 20 years ago, it was reported that humans treated with anti-ERBB2 antibodies for ERBB2-positive breast cancer, and mice with conditional post-natal deletion of Nrg1 , Erbb2 , or Erbb4 tyrosine kinase receptors, develop cardiomyopathy. 1 – 6 Since then, the ERBB system has been recognized as a cardioprotective system. The ERBB system regulates cardiac embryonic development, preserves normal cardiac function, and is activated during cardiac overload or injury as part of a compensatory mechanism. 7 – 9 The protective effects of the ERBB system are attributed to various mechanisms, including regenerative, pro-survival, anti-fibrotic, and anti-inflammatory mechanisms. These mechanisms act on various cardiac cell types, such as cardiomyocytes, endothelial cells, fibroblasts, and inflammatory cells, all of which express one or more ERBB subtypes. 10 – 14 The ERBB receptor system, with its attractive biological profile, is a promising albeit complex therapeutic target. 15 Multiple endogenous peptide growth factors have been identified as ligands. Each of these ligands binds to one or more ERBB receptors, except for ERBB2, which only functions as a dimerization partner, and operates as either a full or partial agonist. 16 Upon activation, ERBB4 receptors can form homodimers or heterodimers with ERBB2 or ERBB3, which leads to different post-receptor signaling events and biological effects. Many researchers studying the activation of this complex system have used variants of the NRG1 protein, one of the system’s most active ligands. 17 In animal experiments, NRG1 application improves various aspects of cardiac disease, including heart failure mortality, systolic and diastolic dysfunction, myocardial infarct size, and myocardial fibrosis. 9 , 18 – 21 Although progress has been made, these effects of NRG1 remain to be sufficiently reproduced in clinical trials. 22 – 24 Use of recombinant NRG1 (rNRG1) as a therapeutic has three major disadvantages. First, it has a short half-life of 15 minutes. Second, it requires intravenous administration, which limits its applicability in chronic diseases such as heart failure. Third, NRG1 is a non-selective agonist of the ERBB system, binding to both ERBB3 and ERBB4 receptors, and thus it may activate ERBB3 receptors in ERBB2-overexpressing tumor cells, thereby enhancing the formation of ERBB3/ERBB2 complexes, which could induce or accelerate cancer growth. 25 , 26 Although the identification of small molecules that induce cell surface receptor dimerization has been recognized to be challenging, 8 the hypothesis of this study is that small molecule-induced activation of the ERBB system is feasible. Here, we screened for compounds that induce ERBB4 dimerization, for several reasons. First, most cardioprotective effects of the ERBB system are dependent on the activation of ERBB4. 9 , 15 , 27 Second, selective activation of ERBB4 preferentially over ERBB3 has a much safer oncological profile. 28 Specifically, previous studies with engineered bivalent NRG1 that activates the ERBB system by forced ERBB4 homodimerization, reproduced the anti-apoptotic effects of NRG1 in cardiomyocytes, without inducing growth of cancer cells. 29 For this reason, we performed a high-throughput screen (HTS) for ERBB4 ligands using an assay that detects ERBB4 homodimerization upon ligand binding. Here, we report the identification of small molecules that induce ERBB4 homodimerization and show their cardiac activity in vitro and in vivo . Results Identification of small molecule ERBB4 agonists To identify small molecule agonists of the ERBB4 receptor, an HTS was performed with a Pharmacological Diversity Set of 10,240 synthetic molecules using an ERBB4/ERBB4 dimerization assay. A primary screen followed by a confirmation screen resulted in 62 hit compounds (Fig. 1 a). To reduce, confirm and cluster the compounds, maximum common substructure (MCS) analysis was performed resulting in the identification of 368 unique MCSs within the 62 confirmed hits. The outcome of a k-means clustering on these 368 MCSs resulted in 10 unique cluster centers, of which cluster center 4 showed a 458-fold enrichment compared to the Enamine reference library (Suppl. Figure 1a). Cluster center 4 consisted of 61 individual cluster members, 3 of which provided a significant enrichment compared to the reference library. The common MCS derived from these 3 MCSs resulted in a pattern that gives 109-fold enrichment (Suppl. Figure 1b). Four of the compounds of the confirmed hit list contained this pharmacophore, and 16 compounds contained analogous MCS patterns. We generated a dose-response curve (DRC) of these 20 compounds. Only the 4 hit compounds containing the common pharmacophore (named EF-1, EF-2, EF-3, and EF-4) showed a reliable and reproducible DRC with half maximal effective concentration (EC 50 ) values in the micromolar range and a maximum response (E max ) that varied from 10.9 ± 2.2% to 27.9 ± 4.8%, relative to NRG1 (Fig. 1 b, f-h). We screened an additional 111 compounds containing the pharmacophore (selected from the Enamine library), which resulted in 2 additional hits (EF-5 and EF-6) that induced ERBB4 dimerization in a concentration-dependent manner (Fig. 1 i-j). Next, a random-forest machine-learning model was developed using Morgan fingerprints of active and inactive molecules based on the results of the 20 selected hit compounds and the results of the additional screening of 111 analogues. Applying this machine-learning model on the Enamine library led to 34 additional compounds that were predicted to be active. Experimental validation of these 34 predictions resulted in 2 additional compounds (EF-7 and EF-8) that induced ERBB4 dimerization in a concentration-dependent manner (Fig. 1 k-l). Apart from the 8 active pharmacophore-containing ERBB4 agonists, we selected 1 of the compounds that did not activate ERBB4 but contained the same pharmacophore (hereafter named NA-1, Fig. 1 m) to be included in several experiments as a control. The chemical structures and molecular weights (MW) of the 8 selected hit compounds and NA-1 are shown in Suppl. Figure 2. We selected EF-1 for further studies because it showed the highest potency (EC 50 = 10.5 ± 4.5 × 10 − 6 M) and efficacy (E max = 27.9 ± 4.8% of the effect of NRG1; Fig. 1 b) in the ERBB4/ERBB4 dimerization assay. Since ERBB2 is a preferred heterodimerization partner of ERBB4, and because NRG1 can also bind to ERBB3 to induce ERBB2/ERBB3 dimerization, we evaluated the effects of EF-1 on ERBB2/ERBB4 and ERBB2/ERBB3 dimerization. Although the potency of EF-1 in the ERBB2/ERBB4 dimerization assay was low (EC 50 > 32 µM), its efficacy at the highest concentration was like the natural ligand NRG1 (Fig. 1 c). The stimulatory effect of EF-1 on ERBB2/ERBB3 heterodimerization was lower compared to ERBB4 homodimerization (Fig. 1 d), indicating a preferential but not exclusive activation of ERBB4. To evaluate whether EF-1 binds to a similar binding pocket than NRG1, we performed a cell-based competition assay. A flow cytometry–based fluorescence competition assay showed that NRG1 as a positive control (ctrl) reduced binding of fluorescent NRG1 (F-NRG1) to the ERBB4 receptor with a half maximal inhibitory concentration (IC 50 ) of 32 nM (Suppl. Figure 3). EF-1, however, did not reduce binding of F-NRG1 (Suppl. Figure 3), indicating that EF-1 and NRG1 bind to different sites. To evaluate whether EF-1 could influence the effect of NRG1 on ERBB4 dimerization, the ERBB4/ERBB4 dimerization cell line was used to generate DRCs of NRG1 in the presence of 0 to 32 µM of EF-1. EF-1 dose-dependently and significantly potentiated NRG1-induced ERBB4 homodimerization, by 299.5% at a concentration of 32 µM (Fig. 1 e). EF-1 and NRG1 activate similar downstream signaling pathways Canonical pathways activated by NRG1 in cardiomyocytes are the phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) and mitogen-activated protein kinase (MAPK)/ extracellular signal-regulated kinase (ERK) pathways. 30 – 32 To examine the effect of EF-1 on both pathways, phosphorylation of AKT and ERK1/2 was assessed by western blot analysis in condtionally immortalized rat atrial myocytes (iAMs). 33 Compared to cardiomyocytes derived from induced pluripotent stem cells, iAMs are easier to culture and show a more complete and reliable cardiomyogenic differentiation. 33 EF-1 induced a time-dependent phosphorylation of AKT (Fig. 2 a) in cardiomyogenically differentiated iAMs. EF-1 also induced a transient phosphorylation of ERK1/2 (Fig. 2 b). To gain insight into downstream signaling pathways and cell responses activated by small-molecule ERBB4 agonists, we performed bulk RNA sequencing on cardiomyogenically differentiated iAMs and human cardiac fibroblasts (HCF), stimulated with either EF-1, NRG1, or vehicle (Veh). In iAMs, EF-1 significantly upregulated 354 genes and significantly downregulated 1120 genes compared to Veh (log 2 -fold change ≥ 0.58, P adj < 0.05) (Fig. 2 c). The top 50 significant differentially expressed genes (DEGs) between EF-1 and the Veh group are shown in Fig. 2 d. Although there is a substantial overlap in DEGs between NRG1 and EF-1 (Fig. 2 e), many DEGs induced by NRG1 and EF-1 differ, which is also reflected in a cluster analysis (Fig. 2 f). In HCFs, EF-1 exhibited a significant upregulation of 336 genes and a significant downregulation of 507 genes compared to Veh (log2-fold change > 0, adjusted P-value < 0.05) (Fig. 3a). Figure 3b displays the top 50 significantly DEGs between EF-1 and the Veh group. Additionally, nearly two-thirds of the DEGs influenced by NRG1 are shared with DEGs influenced by EF-1 (Fig. 3c and 3d). In HCFs, the transforming growth factor-β (TGF-β) and MAPK/ERK pathways were both downregulated by EF-1 and by NRG1, but only reaching statistical significance with EF-1 (Fig. 3e). The PI3K/AKT pathway was downregulated in HCFs when stimulated with EF-1 or NRG1, but not statistically significant (Fig. 3e). EF-1 decreases TGF-β1-induced collagen expression through ERBB4 in vitro Because NRG1 has been shown to decrease collagen mRNA expression induced by TGF-β1 in fibroblasts, we determined the effects of EF-1 on collagen expression induced by TGF-β1 (10 ng/mL) in HCFs and included an ERBB4 knockdown experiment to test the role of ERBB4. 11 , 34 We first observed that TGF-β1 significantly increased collagen type 3 alpha 1 ( COL3A1) mRNA levels in HCFs by 36%. (Fig. 4 a) and that the expression of ERBB4 in HCFs was downregulated by 60% by ERBB4 silencing RNAs (siRNAs) (Fig. 4 b). Next, EF-1 dose-dependently decreased COL3A1 mRNA levels, induced by TGF-β1, up to 50%, while knockdown of ERBB4 expression significantly attenuated this effect. (Fig. 4 c). In contrast, NA-1, a compound that contains the pharmacophore but does not induce ERBB4 dimerization, did not affect COL3A1 expression in HCFs induced by TGF-β1 (Fig. 4 c). EF-1 reduces cardiomyocyte cell death and hypertrophy in vitro Because NRG1 has been shown to decrease cardiomyocyte cell death and to attenuate cardiomyocyte hypertrophy, we next evaluated the effects of EF-1 on cardiomyocytes and again included an ERBB4 knockdown experiment. 10 , 35 , 36 Cardiotoxicity was induced in iAMs with hydrogen peroxide (H 2 O 2 ). We first observed that exposure of iAMs to 100 µM H 2 O 2 increased total cell death to 62% (Fig. 4 d) and that Erbb4 siRNAs reduced ERBB4 mRNA in iAMs levels by more than 50% (Fig. 4 e). Next, EF-1 dose-dependently decreased H 2 O 2 -induced cardiomyocyte cell death with more than 75% at its highest dose, an effect that was blunted by Erbb4 knockdown (Fig. 4 f). In contrast, NA-1 did not have a significant effect on cardiomyocyte survival (Fig. 4 f). Finally, Ang II induced cardiomyocyte hypertrophy, as indicated by a significant increase in cross sectional area (CSA), and this effect was dose-dependently attenuated by EF-1 (Fig. 4 g). EF-1 prevents myocardial fibrosis Before performing in vivo studies, we evaluated the pharmacological stability of EF-1. EF-1 remained stable in plasma of both human and mouse for at least 6 h (Suppl. Figure 5a). Moreover, the half-life of EF-1 was > 6 h when incubated with human liver microsomes and 15 min when incubated with mouse microsomes (Suppl. Figure 5b). Next, we evaluated the effects of EF-1 in a mouse model of AngII-induced myocardial fibrosis. 11 , 37 The experimental design is shown in Fig. 5 a. To evaluate changes in mRNA expression of markers of fibrosis, EF-1 was administered simultaneously with AngII for 1 week in wild type mice. EF-1 not only significantly reduced Col1a1 and Col3a1 mRNA expression induced by AngII (Fig. 5 b; 5 c), indicating anti-fibrotic and cardioprotective properties, but also significantly decreased atrial natriuretic peptide ( Nppa ) expression (Fig. 5 d), a cardiac stretch marker. To study effects on tissue fibrosis, EF-1 was administered simultaneously with AngII for 4 weeks in wild type mice. Masson's trichrome staining of myocardial tissues showed that EF-1 significantly prevented both interstitial and perivascular fibrosis, induced by AngII (Fig. 5 e). EF-1 did not significantly alter cardiac dimensions on ultrasound (Suppl. Table 1a). To evaluate whether the observed anti-fibrotic effects could be mediated by targets unrelated to ERBB4, we performed a similar experiment with NA-1 (containing the common pharmacophore but without induction of ERBB4 dimerization). NA-1 did not significantly prevent interstitial or perivascular AngII-induced fibrosis (Fig. 5 f). Additionally, we evaluated the effects of EF-1 in Ang II-treated transgenic mice with tamoxifen-induced deletion of Erbb4 ( Erbb4 -null mice). Erbb4 deletion was confirmed in the heart by western blot analysis (Suppl. Figure 5). As expected, Erbb4 deletion resulted in a cardiomyopathy phenotype, with a significant increase in left ventricular internal diameter in diastole (LVIDd) and systole (LVIDs), and a fall in fractional shortening (FS) (Suppl. Table 1c). Since the commonly used AngII dose of 1,000 ng/kg/day induced 80% mortality in the Erbb4-null mice (data not shown), the dose of AngII was lowered to 400 ng/kg/day, 38 EF-1 did not significantly prevent interstitial or perivascular AngII-induced fibrosis in Erbb4 -null mice indicating that ERBB4 is necessary for the effects of EF-1 on cardiac fibrosis (Fig. 5 g). Of note, when iCAGCre-Erbb4 f/f mice were injected with corn oil without tamoxifen, the inhibitory effects of EF-1 on AngII-induced interstitial and perivascular fibrosis were preserved (data not shown). EF-1 prevented cardiomyocyte injury in DOX-treated mice Because ERBB4 activation is known to prevent cell death, we investigated the effects of EF-1 in a mouse model of DOX-induced acute cardiac toxicity. 29 , 39 , 40 The experimental design of this experiment is shown in Fig. 6 a. Osmotic minipumps with either EF-1 or Veh were implanted in wild type and Erbb4-null mice. Four days after the start of the EF-1 or Veh treatment, the animals received an intraperitoneal injection of 20 mg/kg DOX followed 3 days later by the measurement of cardiac troponin I (cTnI) plasma levels. EF-1 significantly prevented the DOX-induced increase in circulating cTnI plasma levels (Fig. 6 b). To verify whether the cardioprotective effects of EF-1 were mediated by ERBB4, we repeated the experiment using Erbb4 -null mice and observed that EF-1 did not prevent the increase in cTnI plasma levels in these mice (Fig. 6 c). EF-1 reduces cardiac remodeling and interstitial fibrosis after MI In view of the above effects induced by EF-1, and also because NRG1 has been shown to attenuate adverse cardiac remodelling in MI models, we assessed the effects of EF-1 in a murine MI model (Fig. 6 d). 27 , 41 , 42 Mice were randomized 1 week after ligating the left anterior descending artery (LAD) to implantation of osmotic minipumps with either EF-1 or Veh. Treatment lasted for 28 days, during which cardiac size and function was evaluated using echocardiography (Suppl. Table 1e). Representative echocardiographic images of all groups are shown in Fig. 6 e. In Veh-treated mice, MI significantly increased left ventricular end-diastolic volume (LVEDV; Fig. 6 f) and left ventricular end-systolic volume (LVESV; Fig. 6 g) and decreased left ventricular ejection fraction (EF, Fig. 6 h). 4-week treatment with EF-1 significantly attenuated the increase in LVEDV and LVESV and resulted in a trend towards an increased EF (p = 0.0858). Additionally, interstitial fibrosis in the remote myocardium of the infarcted hearts remained significantly lower in mice treated with EF-1 (Fig. 6 i; 6 j). Discussion By combining HTS and chemoinformatics, we identified a series of compounds that consist of a common pharmacophore and that induce dimerization of the ERBB4 receptor. The most potent compound, EF-1, partially activated downstream signaling pathways that are activated by NRG1, the natural ligand of ERBB4. EF-1 decreased fibroblast collagen production, decreased cardiomyocyte hypertrophy, and improved cardiomyocyte survival in vitro . In vivo , EF-1 mitigated myocardial fibrosis in an AngII model of cardiac fibrosis, prevented acute cardiomyocyte injury in DOX-treated mice, and reduced cardiac dilation and cardiac fibrosis in mice that underwent MI. We showed that the in vitro effects of EF-1 were ERBB4-dependent since they could be abrogated by siRNAs against ERBB4 . Moreover, the in vivo effects of EF-1 could be abrogated by transgenic deletion of Erbb4 in mice. Finally, a compound containing the same pharmacophore as EF-1 but without inducing ERBB4 dimerization (NA-1), did not have the same in vitro and in vivo effects as EF-1, further supporting that the effects of EF-1 depend on its ability to induce ERBB4 dimerization. Previous studies have shown the importance of the NRG1/ERRB4 signaling pathway in cardiac development, physiology, and adaptation during disease. 8 , 43 , 44 Based on these studies, rNRG1 has been developed as a potential therapy for chronic heart failure (CHF). Despite encouraging results in phase I and II clinical trials over a decade ago, 22 , 23 no results of phase III trials have been published yet. Due to its short plasma half-life, rNRG1 must be administered by continuous intravenous infusion. Moreover, in the published phase I and II trials, 22 , 23 rNRG1 was administered for 10 consecutive days without follow-up treatment. Although, short-term effects of a brief rNRG1 administration were significant, it seems unlikely that this treatment regimen will provide long-lasting benefits for CHF patients. Therefore, development of small-molecule ERBB4 agonists could result in a more efficacious therapy for CHF patients. Receptor dimerization is an established mechanism for the initiation of signal transduction, seen in many cell surface receptors. 45 Examples are protein–tyrosine kinase receptors (including ERBB4), the tumor necrosis factor receptor family, protein-serine/threonine kinase receptors, antigen receptors, and members of the cytokine receptor superfamily. 45 Discovery of small molecules with agonist activity on receptor dimerization has been recognized as a challenging endeavor. 45 There are few publications on small molecules capable of disrupting specific protein–protein interactions (antagonists), and the additional requirement for agonists to bind to but also induce dimerization of two receptor molecules makes this aim even more challenging. 45 For instance, considerable effort has been put into the identification of small non-peptide agonists of the erythropoietin receptor, resulting in the identification of weak activators. 46 Screening for compounds activating the related thrombopoietin receptor (TPOR) resulted in identification of a potent activator, called SB394725. 47 Structural studies have shown that SB394725 interacts with the juxtamembrane residues of the transmembrane region of TPOR, 48 but whether SB394725 directly induces dimerization of TPOR is unclear. The location of the exact binding site of the hit compounds on ERBB4 is currently unknown, but most likely differs from NRG1, as our data indicate that EF-1 does not compete with NRG1 for binding to U2OS ERBB4/ERBB4 dimerization cells and because EF-1 potentiates the effect of NRG1 on homodimerization of ERBB4. This is consistent with a mechanism of action based on ago-allosteric modulation of ERBB4 receptor dimerization. 49 Allosteric activation of ERBB4 and potentiation of the effects of NRG1 could partially explain the remarkable biological effects of EF-1 in vitro and in vivo , despite its potency being lower than that of NRG1. Nevertheless, identification of the binding site of the small molecules could facilitate computational screening of novel small molecules and in silico optimization. Binding pockets are potentially located at the domains involved in receptor dimerization, for instance domain II. Both NRG1 and EF-1 induce ERBB4 receptor homodimerization, although the potency and efficacy of EF-1 is lower than those of NRG1. Remarkably, EF-1 also induced ERBB2-ERBB4 heterodimerization, with the same efficacy as NRG1. EF-1 induces phosphorylation of key proteins in 2 canonical pathways activated by NRG1: ERK1/2 and AKT. ERK1/2 phosphorylation induced by EF-1 is transient with peak levels at 15 min, which is in line with published data on NRG1-induced ERK1/2 phosphorylation in neonatal rat ventricular myocytes. 50 AKT phosphorylation induced by EF-1, however, is much slower compared to NRG1-induced AKT phosphorylation, peaking at 2 h instead of 15 min. 10 Differences in potency, efficacy, signaling kinetics and the ability to induce ERBB dimerization pairs between EF-1 and NRG1 could at least partially explain the differences observed in AKT and ERK1/2 phosphorylation and in gene expression after the treatment of iAMs and HCFs with these compounds. The number of DEGs after EF-1 treatment of iAMs was significantly lower than after stimulation of the cells with the more potent NRG1 unlike the number of DEGs in HCFs, which were significantly higher after EF-1 treatment than after NRG1 stimulation. Finally, although we tested the effect of EF-1 in a number of contexts that required ERBB4 receptors including in cells transfected with ERBB4 targeting siRNA, and in Erbb4 -null mice, and examined effects of non-active compounds that shared the same pharmacophore, off-target effects cannot completely be excluded both in vitro and in vivo . In summary, we showed that small molecules can act as ERBB4 agonists inducing ERBB4 dimerization and triggering ERBB4-mediated biological effects in fibroblasts and cardiomyocytes. We also showed in vivo evidence that these small molecules could be a novel therapeutic strategy for treatment of CHF. As ERBB4 is also important in other diseases like fibrotic, inflammatory, and neurological disorders, 11 , 51 – 55 small-molecule ERBB4 agonists could also be of therapeutic relevance in other diseases. Methods Study approval All animal experiments were approved by the Ethical Committee of the University of Antwerp and conformed to the Guide for the Care and Use of Laboratory Animals, 8th edition published by the US National Institutes of Health in 2011, and to the European Communities Council Directive 2010/63/EU for the protection of animals used for experimental purposes. All animals were fed on a standard chow, were provided with water at libitum , and were housed at a constant temperature of 22°C and humidity of 50% in a 12 h controlled light/dark cycle. Throughout the experimental period, mice were closely monitored for any signs of distress or adverse effects. Cells PathHunter U2OS ERBB4/ERBB4 (Eurofins, 93-0961C3), U2OS ERBB2/ERBB4 (Eurofins, 493-0960C3) and ERBB2/ERBB3 (Eurofins, 93-1042C3) dimerization cell lines were cultured according to the manufacturer’s instructions. Briefly, cells were cultured in Cell Culture Reagent 103 (Eurofins, 92-3103G) supplemented with 250 µg/mL Hygromycin B (Eurofins, 92 − 0029) and 500 µg/mL G418 (Eurofins, 92 − 0030), and were maintained at 37°C in humidified atmosphere of 5% CO 2 . HCFs (Innoprot, P10454) were cultured in fibroblast medium (Innoprot, P60108) supplemented with 10% (v/v) fetal bovine serum (FBS, Innoprot), 10% (v/v) fibroblast growth supplement (Innoprot), and 1% (v/v) penicillin/streptomycin solution (Innoprot). iAMs 33 were cultured in Advanced DMEM F-12 (Thermo Fisher Scientific, 12634028) supplemented with 2% (v/v) heat-inactivated FBS (Thermo Fisher Scientific, 10270106), 1% (v/v) penicillin/streptomycin (10,000 units/mL and 10 mg/mL respectively, Thermo Fisher Scientific, 15140122), 1× GlutaMAX (Thermo Fisher Scientific, 35050061), and 100 ng/mL doxycycline (Tocris, 4090) for proliferation. To induce differentiation, iAMs were transferred to medium without doxycycline. HTS and chemoinformatics An HTS of 10,240 compounds (Pharmacological Diversity Set, Enamine) was performed using the PathHunter U2OS ERBB4/ERBB4 dimerization cell line. U2OS ERBB4/ERBB4 dimerization cells were seeded at a density of 5×10 3 cells/well in 50 µL Cell Plating 0 Reagent (Eurofins, 93-0563R0A) in white 384-well plates (Greiner Bio-One, 781080). Cells were treated with compound (10 µM), NRG1 (positive ctrl, 1 µM; Eurofins, 92-1031), or phosphate-buffered saline (negative ctrl, PBS; Thermo Fisher Scientific, 14040133). All wells contained dimethyl sulfoxide (DMSO; Merck Life Science, D2438) at a final concentration of 1%. The cells were subsequently incubated for 6 h at 37°C in a humidified atmosphere of 5% CO 2 . Then, 25 µL of PathHunter Flash Detection Reagent (Eurofins, 93–0247) was added to each well, and cells were incubated at room temperature (RT) in the dark for 1 h. Subsequently, luminescence was measured using the EnVision plate reader (Revvity). Spotfire (TIBCO) was used for data analysis and visualization and Collaborative Drug Discovery Vault for compound registry and data management. Data from the primary screening was analyzed via the HTS-Corrector software. 56 In HTS-Corrector, intraplate normalization (via median polish) was performed to correct for row, column, or edge effects. The analysis output was the normalized values for each compound. Subsequently, normalized values were used to perform inter-plate normalization, which generated B-scores as final analysis output. The top 80 compounds having the highest B-score were selected and re-tested in a confirmation screen under the same assay conditions as the primary screen. In the confirmation screen, the threshold for hit selection was defined as the average signal of the negative control plus three times the standard deviation (SD) of the negative control. To validate and cluster hit compounds, chemoinformatic methods were used to identify the MCSs between the biological active compounds. To estimate the relative significance of each MCS pattern for biological activity, the enrichment of each MCS pattern was calculated by comparing the occurrence of the respective MCS pattern in both the hitlist and reference set (which was the entire Enamine HTS collection consisting of 1,773,567 compounds). MCS patterns with the highest enrichment were then used to select compounds from the Enamine library for additional screening that contain the respective MCS from the Enamine library, resulting in an additional 111 compounds for follow-up screening. In parallel, we developed a random-forest machine-learning model using Morgan fingerprints of active and inactive molecules of previous screenings. PathHunter dimerization assay in 96-well format U2OS ERBB4/ERBB4 dimerization cells were seeded at a density of 5×10 4 cells/well in 100 µL Cell Plating 0 Reagent, in white 96-well plates (PerkinElmer, 6005680) and incubated for 24 h at 37°C, 5% CO 2 . Cells were treated for 6 h with different concentrations of the 8 hit compounds (Enamine, EF-1 – EF-8) or compound NA-1, NRG1 (Peprotech, 100-03), or PBS. Next, 110 µL PathHunter Flash Detection Reagent was added to each well and incubated at RT for 1 h after which the luminescence signal was measured using the Luminoskan Ascent (Thermo Fisher Scientific). Dose-response curves of the compounds were performed in 2-fold (0.0625–32 µM). The same experimental set-up was used for the U2OS ERBB2/ERBB4 and U2OS ERBB2/ERBB3 dimerization cell lines. For co-administration of NRG1 and EF-1, cells were pretreated for 10 min with 5 µL of EF-1 (1, 10, or 32 µM) before adding 5 µL of different NRG1 concentrations and incubated for 6 h at 37°C, 5% CO 2 . All wells contained DMSO at a final concentration of 0.9%. Fluorescence-based competition binding assay NRG1 was dissolved in PBS at a concentration of 1 mg/mL and labeled with Alexa Fluor 488 using the Alexa Fluor 488 microscale protein labeling kit (Thermo Fisher Scientific, A30006). A Bio-gel P-4 (Bio-Rad, 1504124) fine resin suspended in PBS and a dye:protein molar ratio of 5 was used to purify the labelled NRG1 according to the manufacturer’s instructions. Fluorescent labeling was evaluated by performing a dose-response experiment using F-NRG1 (1–1000 nM) on the U2OS ERBB4/ERBB4 dimerization cell line. Next, the same cell line was used to perform a competition binding assay with F-NRG1 using flow cytometry. Cells were plated in transparent U-bottom 96-well plates (Greiner Bio-One, M9436) at 5×10 5 cells/mL in 50 µL ice-cold buffer (PBS with 0.1% bovine serum albumin (BSA; Sigma-Aldrich, A7906) and 0.05% sodium azide (Sigma-Aldrich, S2002)). Next, cells were centrifuged at 17,968 x g for 4 min at RT and the supernatants were discarded. The cell pellets were washed with 50 µL ice-cold buffer by gently pipetting up and down. Centrifugation and washing were repeated, and the supernatants were discarded. For the competition assay between NRG1 and F-NRG1, cells were treated with NRG1 (1–100 nM) and F-NRG1 (30 nM). For the competition assay between EF-1 and F-NRG, cells were treated with EF-1 (0.1–100 µM) and F-NRG1 (30 nM). All wells contained DMSO at a final concentration of 0.9%. After gentle mixing of each sample, the plate was incubated for 1 h at 4°C on a microplate shaker. Next, the plate was centrifuged at 17,968 x g for 4 min at RT and the supernatants were discarded. The cells were then washed with 100 µL ice-cold buffer, pelleted by centrifugation and washed again. The supernatants were discarded and 100 µL ice-cold buffer was added to the cell pellets and pipetted up and down to generate single-cell suspensions. These suspensions were transferred to 5-mL polystyrene round-bottom tubes, kept on ice, and exposed to minimum light until flow cytometric analysis (BD Accuri C6, BD Biosciences). Unstained cells were used to set the parameters of the flow cytometer. RNA sequencing iAMs were seeded at a density of 6×10 6 cells/mL in differentiation medium in 75-cm 2 cell culture flasks (Greiner Bio-One, 658175) and incubated for 9 days (d0 = day of seeding). HCF were seeded at a density of 3x10 6 cells/mL in fibroblast medium in 75-cm 2 cell culture flasks. Then, cells were incubated for 16 h at 37°C in the presence of EF-1 (32 µM), PBS, or NRG1 (0.1 µM). All wells contained DMSO at a final concentration of 0.9%. Next, cells were lysed using 350 µL buffer RLT (Qiagen) supplemented with β-mercaptoethanol (100:1; Merck, 444203). Total RNA was isolated using the RNeasy Micro Kit (Qiagen, 74104) according to the manufacturer’s protocol, with an extra step of DNase digestion. The concentration and quality of RNA were determined using a Qubit fluorometer (Thermo Fisher Scientific) and 2100 Agilent BioAnalyzer (Agilent Technologies), respectively. Samples with an RNA integrity number > 7 were used for library preparation. Sequencing libraries were prepared by Genewiz/Azenta (Leipzig) on cDNA prepared from polyadenylated mRNA. Libraries were sequenced using a NovaSeq 6000 (Illumina) with a read length of 2×150 bp. Gene expression was quantified at the transcript level using Salmon (v1.10.0) 57 , with the validatMappings and -gcBias parameters switched on, to the Rnor_6.0 or GRCh38 transcriptome. Transcript level counts were aggregated to gene level using the import in the tximport package (v1.26.1) 58 , setting countsFromAbundance to ‘lengthScaledTPM’ in R (v4.1.1). DESeq2 R package (v1.38.3) 59 was used for differential gene expression analysis between different conditions. The batch variability of different sequencing runs was accounted for by defining “batch” as a covariate in the linear model to analyse differential gene expression. Differential gene expression heat-maps were generated by using the pheatmap R package (v1.0.12), and volcano plots by EnhancedVolcano (v1.10.0). The overlapping genes between different conditions were obtained by VennDiagram (v1.7.3) 60 . Functional enrichment of DEGs was determined using a hypergeometric test against the Gene Ontology database by using the ClueGO (v2.5.7) 61 module of Cytoscape (v3.9.1) 62 with Benjamini–Hochberg adjusted (FDR) P < 0.05. GSEA Preranked method was performed to identify the Hallmark pathways. 63 The RNA sequencing data generated has been deposited in NCBI’s Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo) and is accessible through GEO Series accession numbers GSE256024 for iAM data and GSE261219 for HCF data. SiRNA transfection experiment SiRNA transfection was carried out according to the manufacturer’s instructions, using 1.25 µL DharmaFECT 1 transfection reagent (Horizon Discovery, T-2001) and 2.5 µL of 10 µM ERBB4 siRNAs or non-targeting siRNAs (ON-TARGETplus Human ERBB4 or Non-Targeting Pool, Horizon Discovery, D-00180/T-2001) per well. After the assay, cells transfected with and without ERBB4 siRNAs, to determine knockdown efficiency, were lysed using RA1 lysis buffer (Macherey-Nagel, 740955.250) supplemented with β-mercaptoethanol (100:1), scraped with a cell scraper and collected in an Eppendorf tube (Greiner Bio-One, 616201) before RNA isolation for reverse transcription-quantitative polymerase chain reaction (RT-qPCR). In vitro collagen expression HCFs were seeded at a density of 1.5×10 5 cells/well in doxycycline-free medium in 12-well plates (Greiner Bio-One, 665180) and incubated overnight. Next, siRNA transfection was carried out as described above. Cells were incubated for 22 h at 37°C and 5% CO 2 . Next, medium was changed to fresh differentiation medium (without doxycycline), and cells were stimulated for 24 h with either PBS, EF-1 (4–32 µM) or NA-1 (4–32 µM) together with TGF-β1 (10 ng/mL; Peprotech, 100 − 21). All wells had a final concentration of 0.9% DMSO. Compounds were pre-incubated for 10 min before addition of TGF-β1. Next, cells were lysed using 100 µL RA1 lysis buffer supplemented with β-mercaptoethanol (100:1), scraped with a cell scraper, and collected in an Eppendorf tube before RNA isolation for RT-qPCR. H 2 O 2 –induced cell death assay iAMs were seeded at a density of 2.7x10 5 cells/well in 48-well plates (Greiner Bio-One, 677180) and differentiated over 9 days in medium without doxycycline (d0 = day of seeding). Next, siRNA transfection was carried out as described above. After incubation for 24 h at 37°C and 5% CO 2 , the culture medium was refreshed. Next, the cells were pretreated for 10 min with either EF-1 (4–32 µM), PBS, or NA-1 (4–32 µM). Subsequently, H 2 O 2 (100 µM; Merck, H1009) was added and the cells were incubated for 4 h at 37°C. All wells contained DMSO at a final concentration of 0.9%. Next, 150 µL of the culture medium in each well was transferred to a 96-well plate and mixed with 100 µL Toxilight AK detection reagent (Lonza, LT07-217). After incubation at RT for 5 min, luminescence was measured using the Luminoskan Ascent. A 100% lysis control (Lonza, LT27-239) was used to determine the percentage of total dead cells. Cardiomyocyte hypertrophy assay iAMs were seeded at a density of 10 4 cells/well in differentiation medium in 24-well plates (Greiner Bio-One, 662160) and differentiated over 9 days (d0 = day of seeding). Next, the cells were pretreated for 1 h with either EF-1 (4–32 µM) or PBS after which AngII (100 nM; Merck, A9525) was added and the cells were incubated for 24 h at 37°C. All wells contained DMSO at a final concentration of 0.9%. To determine CSA, cells were fixed with 4% paraformaldehyde (Thermo Fisher Scientific, 043368.9M) for 30 min at 4°C. Cells were washed thrice with PBS and permeabilized by incubation with 0.1% Triton X-100 (Merck, 10789704001) for 10 min at RT. iAMs were stained with Alexa Fluor 568 phalloidin (Thermo Fisher Scientific, A12380) in 1% BSA-PBS solution for 1 h at RT. After washing 3 times with PBS, 4’,6-diamidino-2-phenylindole dihydrochloride (DAPI, Merck, D9542) was added. Images were obtained by fluorescence microscopy (Celena S). CSA was quantified using an automated algorithm, custom-made in Python. The mode m D (maximum of the histogram) of DAPI intensities (excluding 0) was determined and binary segmentation of nuclei was created with threshold m D + SD D (with SD D the SD of DAPI intensities). Distinct objects with an area less than 500 pixels and nuclei where the average overlapping iAM signal was less than md A + 0.5 SD A (with md A and SD A the median and SD of the Alexa Fluor 568 intensity, respectively) were removed. Binary segmentation of Alexa Fluor 568 signal was created using a threshold of m A + 0.5 SD A (with m A the mode of Alexa Fluor 568 intensities excluding 0), followed by one pass of binary erosion and two passes of binary dilation, each with a 1-connected neighborhood. Any pixels with intensities less than m A ­ 0.5 SD A were removed from the segmentation. Next, a gradient image was created from the Alexa Fluor 568 phalloidin channel. A grey erosion on the original Alexa Fluor 568 intensities (structuring element 3x3 pixels) was followed by a Gaussian blur ( σ = 3 pixels) and a 2D Scharr filter as implemented by ‘scikit-image’. Intensities of the resulting gradient image were scaled to the full 8-bit range, before another grey dilation was performed with a 5×5 pixel structuring element to smooth the gradient. The gradient was used for watershed segmentation of the individual cells with segmented nuclei as seeds. The implementation provided by ‘scikit-image’ was used with a compactness parameter of 0.1. All objects extending to the image borders were removed. Holes within each remaining individual object were filled, and 3 iterations of binary erosion followed by 3 iterations of binary dilation removed small protrusions. For each cell, the CSA was reported. In vitro assays to assess compound stability in plasma and in liver microsomes A 5-µL aliquot of compound solution (10 mM in DMSO) was added to 995 µL of Non-Swiss Albino Mouse Plasma (Innovative Research, IMSNSAPLAK2E10mL) or Pooled Normal Human plasma (Innovative Research, IPLAK2E10ML) in sodium citrate to obtain a final concentration of 50 µM compound in plasma. The mixture was gently shaken for 6 h at 37°C. Aliquots of 100 µL were taken at various time points (0, 0.5, 1, 2, and 6 h), and diluted with 400 µL of ice-cold acetonitrile (Sigma-Aldrich, AX0156). The resulting suspensions were centrifuged at 17,968 x g for 5 min. Subsequently, 50 µL of the supernatant was diluted with 950 µL of ice-cold acetonitrile and analyzed by liquid chromatography with tandem mass spectrometry (LC-MS/MS; Waters Acquity H-class UPLC system with a Bruker Daltonics Esquire 3000 plus ion trap mass spectrometer and an Agilent 1100 Series LC system). Samples were analyzed in triplicate and plotted against a standard curve (compound at 31–1000 nM in plasma and diluted in ice-cold acetonitrile as described above). A mixture of 713 µL milliQ water (Merck, C85358), 200 µL 0.5 M phosphate buffer (pH 7.4, Becton Dickinson, TBS5034), 50 µL NADPH regenerating system solution A (Becton Dickinson), 10 µL NADPH regenerating system solution B (Becton Dickinson) and 2 µL compound (5 mM in DMSO) was prepared and heated for 5 min at 37°C. A volume of 25 µL human and mouse liver microsomes (0.5 mg protein/mL, Corning Life Sciences, 452117 and 452220, respectively) was added to the mixture and 20 µL samples were withdrawn at 0, 0.25, 0.5, 1, 2, 4, 6 and 24 h. Next, 80 µL of ice-cold acetonitrile was added to the samples. After a 10-min incubation period on ice, the mixtures were centrifuged at 15,493 x g for 5 min at 4°C. Finally, 75 µL of an acetonitrile/water (10/90) mixture was added to 25 µL of supernatant and the resulting samples were analysed in triplicate by LC-MS/MS. Mouse model of AngII-induced myocardial fibrosis Thirteen-week-old C57BL/6N (Charles River, 027) male mice were randomized to the ctrl group (n = 5 mice) or to the groups treated with EF-1 (EF-1 group, n = 4 mice), AngII plus vehicle (AngII/Veh group, n = 5 mice), or AngII plus EF-1 (AngII/EF-1 group, n = 5 mice). AngII (1,000 ng/kg/min in PBS), Veh (DMSO/propylene glycol/50:50), and EF-1 (2 mg/kg/day in Veh) were administered for 4 weeks using subcutaneously implanted micro-osmotic pumps (Alzet, model 1004). Four weeks after implantation, cardiac ultrasound was performed, mice were euthanized, and hearts were collected. A similar set-up was used for the 1-week study (Alzet, model 1007D). In some experiments, EF-1 was replaced by NA-1 (2 mg/kg/day in Veh). Mouse model of acute high-dose DOX-induced cardiotoxicity Twelve-week-old C57BL/6N female mice were randomized to the ctrl group (n = 9 mice), or to the groups treated with DOX plus Veh (DOX/Veh group, n = 8 mice), or DOX plus EF-1 (DOX/EF-1 group, n = 8 mice). EF-1 (2 mg/kg in Veh) or Veh were administered for 1 week using subcutaneous micro-osmotic pumps and started on day 1. DOX (20 mg/kg; Pfizer, 4222) was administered intraperitoneally once on day 4 to induce cardiotoxicity. On day 7, mice were euthanized, serum samples were collected via the retrobulbar sinus and hearts were excised. Mouse model of MI Twelve-week-old female Balb/cJ mice were randomized into either the ctrl or MI group. Mice in the MI group underwent surgical ligation of the LAD, while mice in the ctrl group underwent a sham procedure (n = 8 mice). Briefly, all mice received an injection of 0.1 mg/kg buprenorphine (Produlab Pharma) before induction of anesthesia with 8% sevoflurane (Zoetis). Anesthesia was maintained with 4.5% sevoflurane and mice were intubated and ventilated. An incision of 15 mm was made on the left side of the thorax and the thoracic cavity was opened at the third intercostal space using blunt forceps. The LAD was permanently ligated with a 8/0 polypropylene monofilament sutures (Ethicon, F1894) after which the thoracic cavity and skin were closed with 6/0 polypropylene monofilament sutures (Ethicon, F1841). and 5/0 polyamide 6 sutures (Ethilon, F2412H), respectively. Mice received another dose of 0.1 mg/kg buprenorphine 6–8 h after the initial dose. Two days after surgery, cardiac ultrasound was performed to exclude mice without successful MI (i.e. displaying hypokinesia in ≥ 2 out of 5 segments). Mice with successful MI were randomized into 2 groups: an MI/EF-1 group (n = 9 mice) and an MI/Veh group (n = 8 mice). Seven days after MI surgery, the mice were equipped with osmotic mini-pumps containing either EF-1 (2 mg/kg/day) or Veh, as described above. Cardiac ultrasound was performed weekly. After 4 weeks of treatment, mice were euthanized, and hearts were collected. Transgenic mouse models Floxed Erbb4 mice ( Erbb4 f/f ; B6; 129-Erbb4tm1Fej/Mmucd, MMRRC, #010439-UCD) were crossed with CAGGCre-ER™ mice (Jackson Laboratory, 004682) containing the Tg(CAG-cre/Esr1*)5Amc transgene that expresses Cre recombinase under the control of a chicken beta actin promoter/enhancer coupled to the human cytomegalovirus immediate-early gene enhancer, resulting in expression of tamoxifen-inducible Cre-ERT in most cell types. 64 To induce Cre recombinase-mediated deletion of Erbb4 , 11-week-old female and male i CAGGCre-ER TM /Erbb4 f/f mice were intraperitoneally injected with tamoxifen (Merck, T5648; 10 mg/kg; in corn oil) daily for 5 consecutive days. Mice were used in aforementioned studies two weeks after the start of the tamoxifen injections. As a control experiment, 11-week-old female and male i CAGGCre-ER TM /Erbb4 f/f mice were injected with corn oil without tamoxifen. Cardiac ultrasound Echocardiography was performed using the Vevo F2-LAZRX (FUJIFILM VisualSonics) and UHF57x probe. Mice were anesthetized with 1.5% isoflurane (Alvira, BE-V512222). Parasternal long-axis B-mode images and M-mode images were obtained and analyzed using VevoLAB software (FUJIFILM VisualSonics, Version 5.7.1). For the AngII experiments, measurements were obtained through analysis of the short-axis M-mode. For the MI experiments, measurements were obtained through analysis of the parasternal long-axis B-mode, of which the area and end-volume of the left ventricular cavity in diastole and systole were determined by tracing the endocardial border. Acquisitions, measurements and analyses were performed blinded. Serum cTnI enzyme-linked immunoassay Serum samples were centrifuged for 15 min (1,000 × g ) at 4°C and analyzed using a mouse cardiac Troponin I Type 3 ELISA kit (Novus Biologicals, NBP3–00456), according to the manufacturer's instructions. Briefly, a 96-well microplate was coated with an anti-mouse TNNI3/cTnI primary antibody (Capture Antibody Solution). Standard solution or samples (100 µL) were added to the plate and incubated for 90 min at 37°C. A biotinylated anti-mouse TNNI3/cTnI detection antibody was added, and the plate was incubated for 1 h at 37°C followed by 3 washes with Wash Buffer. An avidin–horseradish peroxidase conjugate (Detection Antibody Solution) was then added, and the plate was incubated once more for 30 min at 37°C. After 3 washes with Wash Buffer, Substrate Reagent was added. Following incubation for 15 min at 37°C, the enzymatic reaction was terminated by the addition of Stop Solution. Sample optical densities at 450 nm were converted to tissue concentrations of mouse TNNI3/cTnI using a calibration curve. Western blot analysis iAMs were seeded at a density of 10 6 cells/well in a transparent 6-well plate (Greiner Bio-One, 657160) and treated with either EF-1 (32 µM) or PBS for 0, 15, 60, 120, or 240 min; all wells contained DMSO at a final concentration of 0.9%. Hearts of the aforementioned transgenic mice were removed immediately after the animals had been killed. Cells and heart tissue were collected in RIPA lysis buffer (Thermo Fisher Scientific, 89900) supplemented with protease inhibitors (Merck, 11836153001) and phosphatase inhibitors (Merck, 4906845001) and the lysates were centrifuged at 14,000 × g for 10 min. The supernatants were collected, supplemented with sample buffer, and incubated at 95°C for 5 min. Protein quantification was done with Pierce™ BCA Protein Assay Kits (Thermo Fisher, 23227) according to the manufacturer’s instructions. Equal protein amounts of clarified lysates (20 µg) were subsequently separated by sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (165 V, 400 mA, 60 min) and transferred onto polyvinylidene fluoride membranes by electroblotting (100 V, 400 mA, 90 min). Next, the membranes were incubated for 1 h with Li-Cor blocking buffer (Li-Cor, 927-60001) at RT in Tris-buffered saline (TBS), supplemented with 0.1% Tween (BIO-RAD, 1706531). Next, primary antibodies were added (Suppl. Table 2a) in Li-Cor blocking buffer supplemented with 0.1% Tween, and the membranes were incubated overnight at 4°C on a shaker. Following washing with TBS-0.1% Tween, the membranes were incubated for 1 h at RT with corresponding IRDye-conjugated secondary antibodies (Suppl. Table 2b) in Li-Cor blocking buffer supplemented with 1% SDS and 0.1% Tween. The membrane was visualized with the Odessey imaging system. RT-qPCR analysis of gene expression in fibroblasts, cardiomyocytes, and myocardial tissue RNA was extracted using Nucleospin RNA XS (Macherey-Nagel, 740902.50) according to the manufacturer's instructions. For cDNA synthesis, total RNA was added to a mixture containing buffer with random hexamers and reverse transcriptase enzyme (TaqMan reverse transcription reagents, Applied biosystems, N8080234) after which samples were incubated for 10 min at 25°C, 30 min at 48°C, and 5 min at 95°C. RT-qPCR was performed on QuantStudio 3 Real-time PCR system (Applied Biosystems) using Taqman Universal PCR Master Mix (Applied Biosystems, 4304437) and Taqman primers according to the manufacturer’s instructions. Settings were as follows: 2 min at 95°C followed by 10 min at 95°C, 40 cycles (45 cycles for ERBB4) of denaturation at 95°C for 15 s, and 1 min at 60°C. All reactions were run in duplicate, and all data were normalized against housekeeping genes glyceraldehyde-3-phosphate dehydrogenase ( GAPDH) , beta-actin ( ACTB ), or phosphoglycerate kinase 1 ( PGK1 ). Expression levels were calculated using the comparative cycle method and expressed as fold change (FC) to appropriate controls. The following TaqMan primers were used (Thermo Fisher Scientific): GAPDH (Hs02758991_g1, Mm99999915_g1 and Rn01775763_g1), PGK1 (Hs99999906_m1), COL1A1 (Mm00801666_g1), COL3A1 (Hs00943809_m1 and Mm00802305_g1), ERBB4 (Hs00955525_m1 and Rn00572447_m1), ACTB (Mm02619580_g1) and ANP (Mm01255747_g1). Histology and immunostaining Hearts were fixed in 4% paraformaldehyde, paraffin-embedded, and cut into 5-µm sections. Collagen distribution was visualized by Masson’s trichrome staining. Images were acquired with an Olympus BX43 microscope (Olympus Stream Motion Software) and analyzed with ImageJ 2.14.0 software. Cardiac total and perivascular fibrosis were expressed as the ratio of positively stained fibrotic area (blue) to the total area or vascular lumen area, respectively. Quantification was performed by a person blinded to the treatment protocol. Data analysis and statistics Statistical analysis was performed using GraphPad Prism version 10. Data was checked for normality using the Shapiro-Wilk normality test. Data were expressed as the mean of independent repeats ± SD. Comparison between groups was performed using unpaired t-test or one-way analysis of variance (ANOVA) with Dunnett or Tukey corrections for multiple comparisons. Abbreviations ACTB Beta-actin AKT Protein kinase B AngII Angiotensin-II ANOVA Analysis of variance BSA Bovine serum albumin cDNA copy DNA CHF Chronic heart failure Col1a1 Collagen type 1 alpha 1 Col3a1 Collagen type 3 alpha 1 CSA Cell surface area CST Cell-Signaling Technologies cTnI Cardiac troponin I Ctrl Control DAPI 4',6-Diamidino-2-phenylindole dihydrochloride DEG Differentially expressed gene DMSO Dimethyl sulfoxide DNA Deoxynucleic acid DOX Doxorubicin DRC Dose-response curve EC 50 Half maximal effective concentration EF Ejection fraction Emax Maximum response ERBB Erythroblastic leukemia viral oncogene homolog Erbb4 f/f Floxed Erbb4 ERK Extracellular signal-regulated kinase FBS Fetal bovine serum FC Fold change F-NRG1 Fluorescent NRG1 FS Fractional shortening GAPDH Glyceraldehyde-3-phosphate dehydrogenase H 2 O 2 Hydrogen peroxide HCF Human cardiac fibroblast HTS High throughput screening iAM Conditionally immortalized rat atrial cardiomyocyte IC 50 Half maximal inhibitory concentration IVSd Interventricular septum thickness in diastole IVSs Interventricular septum thickness in systole LAD Left anterior descending coronary artery LC-MS/MS Liquid chromatography with tandem mass spectrometry LVEDV Left ventricular end-diastolic volume LVESV Left ventricular end-systolic volume LVIDd Left ventricular internal diameter in diastole LVIDs Left ventricular internal diameter in systole LVPWd Left ventricular posterior wall thickness in diastole LVPWs Left ventricular posterior wall thickness in systole MAPK Mitogen-activated protein kinase MCS Maximum common substructure MI Myocardial infarction mRNA Messenger RNA MW Molecular weight NA Non-active pharmacophore-containing compound Nppa Atrial natriuretic peptide NRG1 Neuregulin-1 PBS Phosphate-buffered saline PGK1 Phosphoglycerate kinase 1 PI3K phosphatidylinositol 3-kinase RNA Ribonucleic acid RNA-seq RNA sequencing rNRG1 Recombinant neuregulin-1 RT Room temperature RT-qPCR Reverse transcription-quantitative polymerase chain reaction SD Standard deviation SDS Sodium dodecyl sulfate siERBB silencing RNA against ERBB4 siRNA Silencing RNA TBS Tris-buffered saline TGF-β1 Transforming growth factor-β1 TPOR Thrombopoietin receptor Veh Vehicle Declarations Acknowledgments Juan Zhang and Minka Bax (Laboratory of Experimental Cardiology, Leiden University Medical Center, Leiden, the Netherlands) are gratefully acknowledged for arranging the transfer of iAMs and providing protocols for their handling. We also thank Tine Bruyns and Mandy Vermont (University of Antwerp, Antwerp, Belgium) for technical support. Funding This work was supported by a Geconcerteerde onderzoeksactie grant (GOA, PID36444) of the University of Antwerp; by a Senior Clinical Investigator fellowship (to VFS), a PhD fellowship (to JMTC and CC), and research grants of the Fund for Scientific Research Flanders (Application numbers 1842219N, G021019N, G021420N, 1S49323N, and 11PBU24N); VLIR/iBOF Grant 20-VLIR-iBOF-027 (to NV, VFS, HLR, and GWDK). Disclosures Patent "MODULATORS OF ERBB4 IN THE TREATMENT OF DISEASES"; EP20210160742; Inventors: Vincent FM Segers, Gilles W De Keulenaer, Eline Feyen, Hans De Winter. Author contributions V.F.M.S. and G.W.D.K. conceived and designed research; J.MT.C, E.F., S.V.d.B, B.G, C.C, J.v.F, M.T., L.N., performed experiments; J.MT.C., E.F., B.G., Y.F., J.V.H. and E.M.W. analyzed data; J.MT.C., E.F., G.W.D.K. and V.F.M.S. interpreted results of experiments; J.MT.C. and E.F. prepared figures; J.MT.C. and E.F. prepared manuscript; J.MT.C, E.F., S.V.d.B, B.G, C.C, J.v.F, E.M.W., B.V.B, A.A.F.D.V, N.V., D.A.P., D.A., H.L.R., H.D.W., G.W.D.K. and V.F.M.S. edited and revised manuscript; J.MT.C, E.F., S.V.d.B, B.G, Y.F., C.C, J.v.F, M.T., J.V.H., L.N., E.M.W., B.V.B, A.A.F.D.V, N.V., D.A.P., D.A., H.L.R., H.D.W., G.W.D.K. and V.F.M.S. approved final version of manuscript. References Nemeth BT, Varga ZV, Wu WJ, Pacher P (2017) Trastuzumab cardiotoxicity: from clinical trials to experimental studies. Br J Pharmacol 174:3727 Zeglinski M, Ludke A, Jassal DS, Singal PK (2011) Trastuzumab-induced cardiac dysfunction: A ‘dual-hit’. Exp Clin Cardiol 16:70–74 Gassmann M et al (1995) Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor. Nature 1995 378:6555 378, 390–394 Meyer D, Birchmeier C (1995) Multiple essential functions of neuregulin in development. Nature 1995 378:6555 378, 386–390 Lee KF et al (1995) Requirement for neuregulin receptor erbB2 in neural and cardiac development. Nature 378:394–398 Odiete O, Hill MF, Sawyer DB (2012) Neuregulin in cardiovascular development and disease. Circ Res 111:1376–1385 Lemmens K, Doggen K, De Keulenaer GW (2007) Role of neuregulin-1/ErbB signaling in cardiovascular physiology and disease: Implications for therapy of heart failure. Circulation 116:954–960 De Keulenaer GW et al (2019) Mechanisms of the Multitasking Endothelial Protein NRG-1 as a Compensatory Factor During Chronic Heart Failure. Circ Heart Fail 12:6288 Wang Y et al (2022) Neuregulin-1, a potential therapeutic target for cardiac repair. Front Pharmacol 13:945206 Fukazawa R et al (2003) Neuregulin-1 protects ventricular myocytes from anthracycline-induced apoptosis via erbB4-dependent activation of PI3-kinase/Akt. J Mol Cell Cardiol 35:1473–1479 Vermeulen Z et al (2017) Inhibitory actions of the NRG-1/ErbB4 pathway in macrophages during tissue fibrosis in the heart, skin, and lung. Am J Physiol Heart Circ Physiol 313:H934–H945 Ryzhov S et al (2017) ERBB signaling attenuates proinflammatory activation of nonclassical monocytes. Am J Physiol Heart Circ Physiol 312:H907–H918 Bersell K, Arab S, Haring B, Kühn B (2009) Neuregulin1/ErbB4 Signaling Induces Cardiomyocyte Proliferation and Repair of Heart Injury. Cell 138:257–270 D’Uva G et al (2015) ERBB2 triggers mammalian heart regeneration by promoting cardiomyocyte dedifferentiation and proliferation. Nat Cell Biol 17:627–638 Galindo CL, Ryzhov S, Sawyer DB (2014) Neuregulin as a heart failure therapy and mediator of reverse remodeling. Curr Heart Fail Rep 11:40–49 Fuller SJ, Sivarajah K, Sugden PH (2008) ErbB receptors, their ligands, and the consequences of their activation and inhibition in the myocardium. J Mol Cell Cardiol 44:831–854 Falls DL (2003) Neuregulins: functions, forms, and signaling strategies. Exp Cell Res 284:14–30 Zhao Y et al (1998) Neuregulins Promote Survival and Growth of Cardiac Myocytes. J Biol Chem 273:10261–10269 Liu X et al (2006) Neuregulin-1/erbB-Activation Improves Cardiac Function and Survival in Models of Ischemic, Dilated, and Viral Cardiomyopathy. J Am Coll Cardiol 48:1438–1447 Li B et al (2011) Therapeutic effects of neuregulin-1 in diabetic cardiomyopathy rats. Cardiovasc Diabetol 10:69 Dugaucquier L et al (2020) The role of endothelial autocrine NRG1/ERBB4 signaling in cardiac remodeling. Am J Physiol Heart Circ Physiol 319:H443–H455 Jabbour A et al (2011) Parenteral administration of recombinant human neuregulin-1 to patients with stable chronic heart failure produces favourable acute and chronic haemodynamic responses. Eur J Heart Fail 13:83–92 Gao R et al (2010) A phase II, randomized, double-blind, multicenter, based on standard therapy, placebo-controlled study of the efficacy and safety of recombinant human neuregulin-1 in patients with chronic heart failure. J Am Coll Cardiol 55:1907–1914 Lenihan DJ et al (2016) A Phase I, Single Ascending Dose Study of Cimaglermin Alfa (Neuregulin 1β3) in Patients With Systolic Dysfunction and Heart Failure. JACC Basic Transl Sci 1:576 Holbro T et al (2003) The ErbB2/ErbB3 heterodimer functions as an oncogenic unit: ErbB2 requires ErbB3 to drive breast tumor cell proliferation. Proceedings of the National Academy of Sciences 100, 8933–8938 Arteaga CL, Engelman JA (2014) ERBB Receptors: From Oncogene Discovery to Basic Science to Mechanism-Based Cancer Therapeutics. Cancer Cell 25:282–303 Lin Y, Liu H, Wang X (2020) Neuregulin-1, a microvascular endothelialderived protein, protects against myocardial ischemiareperfusion injury (Review). Int J Mol Med 46:925–935 Segers VFM, Dugaucquier L, Feyen E, Shakeri H, De Keulenaer G (2020) W. The role of ErbB4 in cancer. Cell Oncol 43:335–352 Jay SM et al (2013) An engineered bivalent neuregulin protects against doxorubicin-induced cardiotoxicity with reduced proneoplastic potential. Circulation 128:152–161 Fang S-J et al (2010) Neuregulin-1 preconditioning protects the heart against ischemia/reperfusion injury through a PI3K/Akt-dependent mechanism. Chin Med J (Engl) 123:3597–3604 Jie B et al (2012) Neuregulin-1 suppresses cardiomyocyte apoptosis by activating PI3K/Akt and inhibiting mitochondrial permeability transition pore. Mol Cell Biochem 370:35–43 Gallo S, Vitacolonna A, Bonzano A, Comoglio P, Crepaldi T (2019) ERK: A key player in the pathophysiology of cardiac hypertrophy. Int J Mol Sci 20 Liu J et al (2018) Generation and primary characterization of iAM-1, a versatile new line of conditionally immortalized atrial myocytes with preserved cardiomyogenic differentiation capacity. Cardiovasc Res 114:1848 Kirabo A et al (2017) Neuregulin-1β induces proliferation, survival and paracrine signaling in normal human cardiac ventricular fibroblasts. J Mol Cell Cardiol 105:59–69 Lemmens K, Segers VFM, Demolder M, De Keulenaer GW (2006) Role of neuregulin-1/ErbB2 signaling in endothelium-cardiomyocyte cross-talk. J Biol Chem 281:19469–19477 Kuramochi Y et al (2004) Cardiac endothelial cells regulate reactive oxygen species-induced cardiomyocyte apoptosis through neuregulin-1beta/erbB4 signaling. J Biol Chem 279:51141–51147 Hervent A-S, Vandekerckhove L, De Keulenaer GW (2013) Neuregulin-1 antagonizes myocardial fibrosis and diastolic dysfunction in angiotensin-II treated mice. Eur Heart J 34:P2434–P2434 Shakeri H et al (2021) Neuregulin-1 compensates for endothelial nitric oxide synthase deficiency. Am J Physiol Heart Circ Physiol 320:H2416–H2428 Bian Y et al (2009) Neuregulin-1 attenuated doxorubicin-induced decrease in cardiac troponins. Am J Physiol Heart Circ Physiol 297:H1974–H1983 Liu F-F et al (2005) Heterozygous knockout of neuregulin-1 gene in mice exacerbates doxorubicin-induced heart failure. Am J Physiol Heart Circ Physiol 289:H660–H666 Shiraishi M, Yamaguchi A, Suzuki K (2022) Nrg1/ErbB signaling-mediated regulation of fibrosis after myocardial infarction. FASEB J 36:e22150 Hedhli N et al (2011) Endothelium-Derived Neuregulin Protects the Heart Against Ischemic Injury. Circulation 123:2254–2262 Mendes-Ferreira P, De Keulenaer GW, Leite-Moreira AF, Brás-Silva C (2013) Therapeutic potential of neuregulin-1 in cardiovascular disease. Drug Discov Today 18:836–842 Vermeulen Z, Segers VFM, De Keulenaer GW (2016) ErbB2 signaling at the crossing between heart failure and cancer. Basic Res Cardiol 111:60 Boger DL, Goldberg J (2001) Cytokine receptor dimerization and activation: prospects for small molecule agonists. Bioorg Med Chem 9:557–562 Dömling A, Beck B, Baumbach W, Larbig G (2007) Towards erythropoietin mimicking small molecules. Bioorg Med Chem Lett 17:379–384 Erickson-Miller CL et al (2005) Discovery and characterization of a selective, nonpeptidyl thrombopoietin receptor agonist. Exp Hematol 33:85–93 Kim M-J et al (2007) NMR structural studies of interactions of a small, nonpeptidyl Tpo mimic with the thrombopoietin receptor extracellular juxtamembrane and transmembrane domains. J Biol Chem 282:14253–14261 Schwartz TW, Holst B (2007) Allosteric enhancers, allosteric agonists and ago-allosteric modulators: where do they bind and how do they act? Trends Pharmacol Sci 28:366–373 Pentassuglia L et al (2016) Neuregulin-1β promotes glucose uptake via PI3K/Akt in neonatal rat cardiomyocytes. Am J Physiol Endocrinol Metab 310:E782–E794 Zeng F, Miyazawa T, Kloepfer LA, Harris RC (2018) ErbB4 deletion accelerates renal fibrosis following renal injury. Am J Physiol Ren Physiol 314:F773–F787 Jiang Y, Shi J, Zhou J, He C, Gu R (2023) ErbB4 promotes M2 activation of macrophages in idiopathic pulmonary fibrosis. Open Life Sci 18 Walsh T et al (2008) Rare Structural Variants Disrupt Multiple Genes in Neurodevelopmental Pathways in Schizophrenia. Sci (1979) 320:539–543 Iwakura Y, Nawa H (2013) ErbB1-4-dependent EGF/neuregulin signals and their cross talk in the central nervous system: pathological implications in schizophrenia and Parkinson’s disease. Front Cell Neurosci 7 Banerjee A, MacDonald ML, Borgmann-Winter KE, Hahn C-G (2010) Neuregulin 1–erbB4 pathway in schizophrenia: From genes to an interactome. Brain Res Bull 83:132–139 Makarenkov V et al (2006) HTS-Corrector: software for the statistical analysis and correction of experimental high-throughput screening data. Bioinformatics 22:1408–1409 Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C (2017) Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods 14:417–419 Soneson C, Love MI, Robinson MD (2016) Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Res 4, 1521 Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15:550 Chen H, Boutros PC (2011) VennDiagram: a package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinformatics 12:35 Bindea G et al (2009) ClueGO: a Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks. Bioinformatics 25:1091–1093 Shannon P et al (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13:2498–2504 Subramanian A et al (2005) From the Cover: Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102:15545 Hayashi S, McMahon AP (2002) Efficient Recombination in Diverse Tissues by a Tamoxifen-Inducible Form of Cre: A Tool for Temporally Regulated Gene Activation/Inactivation in the Mouse. Dev Biol 244:305–318 Additional Declarations Yes there is potential Competing Interest. Patent "MODULATORS OF ERBB4 IN THE TREATMENT OF DISEASES" EP20210160742 Inventors: Vincent FM Segers, Gilles W De Keulenaer, Eline Feyen, Hans De Winter Supplementary Files Supplementalfiguresandtables.docx Cite Share Download PDF Status: Published Journal Publication published 10 Jan, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4175488","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":285122041,"identity":"1225a63f-067f-4568-ba89-3ddb51016c95","order_by":0,"name":"Vincent Segers","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0001-7796-2222","institution":"University of Antwerp","correspondingAuthor":true,"prefix":"","firstName":"Vincent","middleName":"","lastName":"Segers","suffix":""},{"id":285122042,"identity":"e0f04189-f7d2-4d01-aeb5-ff009644046f","order_by":1,"name":"Julie Cools","email":"","orcid":"https://orcid.org/0000-0001-7661-367X","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Julie","middleName":"","lastName":"Cools","suffix":""},{"id":285122043,"identity":"9db63b1f-2fb9-4d74-b2bd-fe62e5cb5182","order_by":2,"name":"Eline Feyen","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Eline","middleName":"","lastName":"Feyen","suffix":""},{"id":285122044,"identity":"4ce1fd7a-662d-45aa-b829-3a5b016e8cae","order_by":3,"name":"Siel Van den Bogaert","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Siel","middleName":"Van den","lastName":"Bogaert","suffix":""},{"id":285122045,"identity":"719857a5-7401-47e8-95d8-944982946aea","order_by":4,"name":"Bo Goovaerts","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Goovaerts","suffix":""},{"id":285122046,"identity":"62720267-566c-4bb9-84d1-14901f6caa6c","order_by":5,"name":"Yile Fu","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Yile","middleName":"","lastName":"Fu","suffix":""},{"id":285122047,"identity":"c97a44fa-d649-4676-bd19-68d8c2c55abb","order_by":6,"name":"Celine Civati","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Celine","middleName":"","lastName":"Civati","suffix":""},{"id":285122049,"identity":"af01bfb3-2d71-4fe4-a5e1-3c212a92087a","order_by":7,"name":"Jens Van fraeyenhove","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Jens","middleName":"Van","lastName":"fraeyenhove","suffix":""},{"id":285122050,"identity":"86e6583f-58ff-4cd2-9338-c36e6278792e","order_by":8,"name":"Michiel Tubeeckx","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Michiel","middleName":"","lastName":"Tubeeckx","suffix":""},{"id":285122051,"identity":"04796ea4-21ce-4e65-a94b-3fd221323f45","order_by":9,"name":"Jasper Van Haver","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Jasper","middleName":"Van","lastName":"Haver","suffix":""},{"id":285122052,"identity":"4576b100-ff71-4d60-be58-06c608772821","order_by":10,"name":"Long Nguyen","email":"","orcid":"","institution":"University of Ghent","correspondingAuthor":false,"prefix":"","firstName":"Long","middleName":"","lastName":"Nguyen","suffix":""},{"id":285122053,"identity":"6a52083f-fa7f-4ac7-8b3b-df2b3b54ab34","order_by":11,"name":"Eike Wülfers","email":"","orcid":"https://orcid.org/0000-0002-8292-6514","institution":"University of Ghent","correspondingAuthor":false,"prefix":"","firstName":"Eike","middleName":"","lastName":"Wülfers","suffix":""},{"id":285122054,"identity":"b67a4967-a5bb-4a19-a1f2-b59b2bfb055a","order_by":12,"name":"Benji Van Berlo","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Benji","middleName":"Van","lastName":"Berlo","suffix":""},{"id":285122055,"identity":"214a6ac0-82ab-439e-8086-78dc2c952827","order_by":13,"name":"Antoine De Vries","email":"","orcid":"","institution":"Leids Universitair Medisch Centrum","correspondingAuthor":false,"prefix":"","firstName":"Antoine","middleName":"","lastName":"De Vries","suffix":""},{"id":285122056,"identity":"1615a55f-02bf-4a1f-9448-14235e8248ed","order_by":14,"name":"Nele Vandersickel","email":"","orcid":"","institution":"University of Ghent","correspondingAuthor":false,"prefix":"","firstName":"Nele","middleName":"","lastName":"Vandersickel","suffix":""},{"id":285122057,"identity":"a8007589-54e1-4939-a628-aef1d635c16b","order_by":15,"name":"Daniël Pijnappels","email":"","orcid":"https://orcid.org/0000-0001-6731-4125","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Daniël","middleName":"","lastName":"Pijnappels","suffix":""},{"id":285122058,"identity":"99ac4cea-0cd8-4d71-b9fe-76119438c022","order_by":16,"name":"Dominique Audenaert","email":"","orcid":"","institution":"VIB - UGent","correspondingAuthor":false,"prefix":"","firstName":"Dominique","middleName":"","lastName":"Audenaert","suffix":""},{"id":285122059,"identity":"efdcd79b-e73b-4a6f-957a-2a2769714fc3","order_by":17,"name":"Llew Roderick","email":"","orcid":"https://orcid.org/0000-0001-7065-3523","institution":"University of Leuven","correspondingAuthor":false,"prefix":"","firstName":"Llew","middleName":"","lastName":"Roderick","suffix":""},{"id":285122060,"identity":"a7a47e2c-3314-49dd-928f-266ee0788b34","order_by":18,"name":"Hans De Winter","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Hans","middleName":"","lastName":"De Winter","suffix":""},{"id":285122061,"identity":"be2992b5-646f-4ce4-b9c7-8d63c7eae4ba","order_by":19,"name":"Gilles W. De Keulenaer","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Gilles","middleName":"W.","lastName":"De Keulenaer","suffix":""}],"badges":[],"createdAt":"2024-03-27 10:55:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4175488/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4175488/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-024-54908-5","type":"published","date":"2025-01-10T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":54316640,"identity":"8f670959-1c84-445a-a0d2-479dd9988015","added_by":"auto","created_at":"2024-04-08 17:56:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":337185,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHigh-throughput screening to identify small-molecule ERBB4 agonists.\u003c/strong\u003e \u003cstrong\u003e(a) \u003c/strong\u003eAn ERBB4/ERBB4 dimerization assay was used to screen 10,240 pharmacologically diverse compounds for their ability to induce ERBB4 homodimerization. The confirmation screen resulted in 62 hits (green dots) with a luminescence signal above threshold (light blue line, 3.5×10\u003csup\u003e6\u003c/sup\u003e LU). MCS analysis and enrichment analysis of the confirmed hits resulted in the identification of a pharmacophore. After the screening and evaluation of additional compounds, 8 candidate ERBB4 agonists were selected. \u003cstrong\u003e(b-d)\u003c/strong\u003e DRC of the most potent compound EF-1 in \u003cstrong\u003e(b)\u003c/strong\u003e the ERBB4/ERBB4 dimerization assay, \u003cstrong\u003e(c)\u003c/strong\u003e the ERBB2/ERBB4 dimerization assay, and \u003cstrong\u003e(d) \u003c/strong\u003ethe ERBB2/ERBB3 dimerization assay. \u003cstrong\u003e(e)\u003c/strong\u003e DRCs showing the effect of NRG1 on the ERBB4/ERBB4 dimerization assay in the presence of 3 different concentrations of EF-1. \u003cstrong\u003e(f-l)\u003c/strong\u003e DRCs of the 7 other selected hit compounds and \u003cstrong\u003e(m)\u003c/strong\u003e a non-active compound NA-1 on the ERBB4/ERBB4 dimerization assay. The signal in the dimerization assays is plotted relative to the signal obtained with 0.1 µM NRG1. Data are represented as mean ± SD, n = 4–8 per compound. Yellow dots, positive control (NRG1); red dots, negative control (Veh); green dots, hit compounds; DRC, dose response curve; LU, light units; MSC, maximum common substructure; NA, non-active pharmacophore-containing compound; NRG1, neuregulin-1; Veh, vehicle.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/3cfdb38161543e0c41ef3fe7.png"},{"id":54316254,"identity":"73ad1c2d-5b15-41be-a5ff-235338412b53","added_by":"auto","created_at":"2024-04-08 17:48:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":431647,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEF-1 and NRG1 activate similar downstream signaling pathways in cultured cardiomyocytes. (a-b) \u003c/strong\u003eRepresentative western blot images showing the effect of EF-1 (32 µM) on \u003cstrong\u003e(a)\u003c/strong\u003e pAKT/AKT and \u003cstrong\u003e(b) \u003c/strong\u003epERK1/2/ERK1/2 pathways in iAMs stimulated for different times. Bar graphs show the average effects of 4 independent experiments, one-way ANOVA with Dunnett’s multiple comparisons test. The samples derive from parallel experiments and the blots were processed in parallel. \u003cstrong\u003e(c)\u003c/strong\u003e Volcano plot showing differential gene expression (DESeq2, P adj \u0026lt; 0.05) of iAMs in response to EF-1 (n=4). Vertical dashed lines indicated log2 fold change ≥ 0.58. The horizontal dashed line indicates -Log10 (P adj \u0026lt; 0.05). \u003cstrong\u003e(d) \u003c/strong\u003eHeatmap showing log2 fold change estimates for top DEGs in EF-1-treated iAMs relative to the control group. \u003cstrong\u003e(e) \u003c/strong\u003eVenn diagram of overlapping DEGs between the NRG1 and EF-1 group. \u003cstrong\u003e(f)\u003c/strong\u003e Heatmap showing log2 fold change estimates for top DEGs in EF-1- or NRG1-treated iAMs relative to the control group. AKT, protein kinase B; ANOVA, analysis of variance; DEG, differentially expressed gene; ERK, extracellular signal-regulated kinase; FC, fold change; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; iAM, conditionally immortalized rat atrial myocyte; NRG1, neuregulin-1; Veh, vehicle.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/4be235eb6f322a14c34ab943.png"},{"id":54316256,"identity":"94a02161-0730-4af7-b04a-2c6a14f83118","added_by":"auto","created_at":"2024-04-08 17:48:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":259395,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEF-1 and NRG1 activate similar downstream signaling pathways in human cultured cardiac fibroblasts. (a)\u003c/strong\u003eVolcano plot showing differential gene expression (DESeq2, P adj \u0026lt;0.05) of HCFs in response to EF-1 (n=3). The horizontal dashed line indicates -Log10 (P adj \u0026lt; 0.05). \u003cstrong\u003e(b) \u003c/strong\u003eHeatmap showing log2 fold change estimates for top DEGs in EF-1-treated HCFs relative to the control group. \u003cstrong\u003e(c) \u003c/strong\u003eVenn diagram of overlapping DEGs between the NRG1 and EF-1 group. \u003cstrong\u003e(d)\u003c/strong\u003e Heatmap showing log2 fold change estimates for top DEGs in EF-1- or NRG1-treated HCFs relative to the control group. \u003cstrong\u003e(e) \u003c/strong\u003eGSEA of HCFs with EF-1 or NRG1 for gene signatures of TGF-β, PI3K/AKT and MAPK pathway-regulated genes. P-values were adjusted for multiple comparions. AKT, protein kinase B; DEG, differentially expressed gene; FC, fold change; GSEA, gene set enrichment analysis; HCF, human cardiac fibroblast; MAPK, mitogen-activated protein kinase; NES, normalized enrichment score; NRG1, neuregulin-1; PI3K, phosphatidylinositol 3-kinase; TGF-β, transforming growth factor-β.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/f2ad64a89b36420dcb05c48e.png"},{"id":54316641,"identity":"d77adbac-c218-4755-b59f-7350474c2084","added_by":"auto","created_at":"2024-04-08 17:56:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":197330,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEF-1 decreases collagen production in fibroblasts, and decreases cardiomyocyte cell death and hypertrophy. (a)\u003c/strong\u003e Effect of TGF-β1 on \u003cem\u003eCOL3A1 \u003c/em\u003emRNA expression in HCFs. \u003cstrong\u003e(b)\u003c/strong\u003e ERBB4 knockdown efficiency of siRNAs against \u003cem\u003eERBB4\u003c/em\u003e compared to control (i.e. scrambled) siRNAs in HCFs. \u003cstrong\u003e(c)\u003c/strong\u003e \u003cem\u003eCOL3A1\u003c/em\u003e mRNA expression after stimulation with EF-1 (4–32 µM) and TGF-β1 in the presence of scrambled or \u003cem\u003eERBB4\u003c/em\u003e-specific siRNAs, or after stimulation with NA-1. \u003cstrong\u003e(d)\u003c/strong\u003e Effect of H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e on cardiomyocyte cell death. \u003cstrong\u003e(e)\u003c/strong\u003e \u003cem\u003eErbb4\u003c/em\u003e knockdown efficiency of siRNA against \u003cem\u003eErbb4\u003c/em\u003e compared to control (i.e. scrambled) siRNAs in iAMs. \u003cstrong\u003e(f)\u003c/strong\u003e Effect of H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2 \u003c/sub\u003eon total cell death of iAMs in the presence of NA-1, or in the presence of EF-1 (4–32 µM) after transfection with scrambled or \u003cem\u003eErbb4\u003c/em\u003e-specific siRNAs. n = 3–6 in each group, unpaired t-test or one-way ANOVA with Dunnett’s multiple comparisons test against \u003cem\u003esiScr\u003c/em\u003e. P-values shown are between EF-1 + \u003cem\u003esiScr \u003c/em\u003eand EF-1 + \u003cem\u003esiERRB4\u003c/em\u003e. \u003cstrong\u003e(g)\u003c/strong\u003e CSA of iAMs after AngII exposure in the presence or absence of EF-1 (4–32 µM). Scale bar = 100 µm, n = 20 CSA in each group, one-way ANOVA with Tukey’s multiple comparisons test. All data are represented as mean ± SD. AngII, angiotensin II; ANOVA, analysis of variance; COL3A1, collagen type 3 alfa 1; CSA, cross sectional area; FC, fold change; H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e, hydrogen peroxide; HCF, human cardiac fibroblast; iAM, conditionally immortalized rat atrial myocyte; M, molar concentration; NA, non-active compound containing the pharmacophore; SD, standard deviation; si\u003cem\u003eERBB4\u003c/em\u003e, silencing RNA against \u003cem\u003eERBB4\u003c/em\u003e; \u003cem\u003esiScr\u003c/em\u003e, silencing RNA against scrambled control; TGF-β1, transforming growth factor β1; Veh, vehicle.\u003cbr\u003e\n\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/8d45ee5c6165a5df16048e09.png"},{"id":54316258,"identity":"8c0b9b3a-c36b-430c-acdd-0256b9dc8e3c","added_by":"auto","created_at":"2024-04-08 17:48:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":539311,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEF-1 prevents cardiac fibrosis \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ein vivo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e. (a)\u003c/strong\u003e Design of the \u003cem\u003ein vivo\u003c/em\u003e experiments. Hearts were collected after 1 week for mRNA analysis and after 4 weeks for histological analysis.\u003cstrong\u003e (b-d) \u003c/strong\u003eRT-qPCR was performed on fibrosis and cardiac stress markers; normalized FC compared to Ctrl: \u003cstrong\u003e(b)\u003c/strong\u003e \u003cem\u003eCol1a1, \u003c/em\u003e\u003cstrong\u003e(c)\u003c/strong\u003e\u003cem\u003e Col3a1 \u003c/em\u003eand \u003cstrong\u003e(d)\u003c/strong\u003e\u003cem\u003e Nppa\u003c/em\u003e.\u003cstrong\u003e (e) \u003c/strong\u003eRepresentative images of Masson’s trichrome staining of AngII-induced myocardial fibrosis following treatment with EF-1 or Veh and corresponding bar graphs showing the quantitation for total and perivascular fibrosis \u003cstrong\u003e(f)\u003c/strong\u003e Representative images of Masson’s trichrome staining of AngII-induced myocardial fibrosis following treatment with NA-1 and corresponding graphs showing the quantitation for total and perivascular fibrosis. \u003cstrong\u003e(g)\u003c/strong\u003e Representative images of Masson’s trichrome staining of AngII-induced myocardial fibrosis in \u003cem\u003eErbb4-null\u003c/em\u003e\u003csup\u003e \u003c/sup\u003emice, treated with EF-1 or Veh and corresponding graphs showing the quantitation for total and perivascular fibrosis. All data are represented as mean ± SD. n=4–5 in each group, one-way ANOVA with Tukey’s multiple comparisons test. AngII, angiotensin II; ANOVA, analysis of variance; Col1a1, collagen type 1 alpha 1; Col3a1, collagen type 3 alpha 1; Ctrl, control; echo, echocardiography; FC, fold change; NA, non-activating pharmacophore-containing compound; Nppa, atrial natriuretic peptide; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; SD, standard deviation; Veh, vehicle.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/3271db7f35cf86140b76ad54.png"},{"id":54316259,"identity":"8c6f8d62-db11-4c7a-9ec2-08553028940c","added_by":"auto","created_at":"2024-04-08 17:48:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":385433,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEF-1 prevented cardiomyocyte cell death \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ein vivo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e and decreases ventricular dilation after MI. (a) \u003c/strong\u003eDesign of the \u003cem\u003ein vivo \u003c/em\u003eDOX experiment.\u003cstrong\u003e (b-c) \u003c/strong\u003eGraphs showing the cTnI levels measured in plasma samples of untreated control mice and of mice treated for 7 days with EF-1 or Veh and given a single injection of 20 mg/kg DOX on day 4 of treatment for \u003cstrong\u003e(b)\u003c/strong\u003e wild type mice and \u003cstrong\u003e(c)\u003c/strong\u003e \u003cem\u003eErbb4\u003c/em\u003e-null mice. \u003cstrong\u003e(d)\u003c/strong\u003e Design of the \u003cem\u003ein vivo\u003c/em\u003e MI experiment. \u003cstrong\u003e(e)\u003c/strong\u003e Representative echocardiographic images in end-diastole and end-systole of sham-operated control mice and of mice that underwent MI for 4 weeks with EF-1 or Veh. Graphs showing \u003cstrong\u003e(f)\u003c/strong\u003e LVEDV, \u003cstrong\u003e(g)\u003c/strong\u003e LVESV and \u003cstrong\u003e(h)\u003c/strong\u003eEF in the different experimental groups. \u003cstrong\u003e(i)\u003c/strong\u003e Representative images of Masson’s trichrome staining of interstitial fibrosis in the remote zone, and \u003cstrong\u003e(j)\u003c/strong\u003ecorresponding graph quantifying the fibrotic area in the different experimental groups.\u003cstrong\u003e \u003c/strong\u003eData are represented as mean ± SD, n = 8–10 in each group, one-way ANOVA test with Tukey’s correction for multiple testing. ANOVA, analysis of variance; cTnI, cardiac troponin I; Ctrl, control; DOX, doxorubicin; EF, ejection fraction; LVEDV, left-ventricular end-diastolic volume; LVESD, left-ventricular end-systolic volume; MI, myocardial infarction; SD, standard deviation; Veh, vehicle.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/be33a1132247e93baa00af0a.png"},{"id":73547480,"identity":"c72fdee8-b52e-4da3-a99e-bc4a20f41855","added_by":"auto","created_at":"2025-01-11 08:06:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3602590,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/a677f870-c2a8-4aff-9b7f-3e356794fe7e.pdf"},{"id":54316257,"identity":"bde3764f-6198-4ff4-a93f-4d9069de6a64","added_by":"auto","created_at":"2024-04-08 17:48:51","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1035542,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementalfiguresandtables.docx","url":"https://assets-eu.researchsquare.com/files/rs-4175488/v1/2b026a67df7c7dbc254978af.docx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nPatent \"MODULATORS OF ERBB4 IN THE TREATMENT OF DISEASES\"\r\nEP20210160742\r\nInventors: Vincent FM Segers, Gilles W De Keulenaer, Eline Feyen, Hans De Winter","formattedTitle":"Small molecule-induced ERBB4 activation to treat heart failure","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOver 20 years ago, it was reported that humans treated with anti-ERBB2 antibodies for ERBB2-positive breast cancer, and mice with conditional post-natal deletion of \u003cem\u003eNrg1\u003c/em\u003e, \u003cem\u003eErbb2\u003c/em\u003e, or \u003cem\u003eErbb4\u003c/em\u003e tyrosine kinase receptors, develop cardiomyopathy.\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e Since then, the ERBB system has been recognized as a cardioprotective system. The ERBB system regulates cardiac embryonic development, preserves normal cardiac function, and is activated during cardiac overload or injury as part of a compensatory mechanism.\u003csup\u003e\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e The protective effects of the ERBB system are attributed to various mechanisms, including regenerative, pro-survival, anti-fibrotic, and anti-inflammatory mechanisms. These mechanisms act on various cardiac cell types, such as cardiomyocytes, endothelial cells, fibroblasts, and inflammatory cells, all of which express one or more ERBB subtypes.\u003csup\u003e\u003cspan additionalcitationids=\"CR11 CR12 CR13\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThe ERBB receptor system, with its attractive biological profile, is a promising albeit complex therapeutic target.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e Multiple endogenous peptide growth factors have been identified as ligands. Each of these ligands binds to one or more ERBB receptors, except for ERBB2, which only functions as a dimerization partner, and operates as either a full or partial agonist.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e Upon activation, ERBB4 receptors can form homodimers or heterodimers with ERBB2 or ERBB3, which leads to different post-receptor signaling events and biological effects. Many researchers studying the activation of this complex system have used variants of the NRG1 protein, one of the system\u0026rsquo;s most active ligands.\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e In animal experiments, NRG1 application improves various aspects of cardiac disease, including heart failure mortality, systolic and diastolic dysfunction, myocardial infarct size, and myocardial fibrosis.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan additionalcitationids=\"CR19 CR20\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e Although progress has been made, these effects of NRG1 remain to be sufficiently reproduced in clinical trials.\u003csup\u003e\u003cspan additionalcitationids=\"CR23\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eUse of recombinant NRG1 (rNRG1) as a therapeutic has three major disadvantages. First, it has a short half-life of 15 minutes. Second, it requires intravenous administration, which limits its applicability in chronic diseases such as heart failure. Third, NRG1 is a non-selective agonist of the ERBB system, binding to both ERBB3 and ERBB4 receptors, and thus it may activate ERBB3 receptors in ERBB2-overexpressing tumor cells, thereby enhancing the formation of ERBB3/ERBB2 complexes, which could induce or accelerate cancer growth.\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eAlthough the identification of small molecules that induce cell surface receptor dimerization has been recognized to be challenging,\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e the hypothesis of this study is that small molecule-induced activation of the ERBB system is feasible. Here, we screened for compounds that induce ERBB4 dimerization, for several reasons. First, most cardioprotective effects of the ERBB system are dependent on the activation of ERBB4.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e Second, selective activation of ERBB4 preferentially over ERBB3 has a much safer oncological profile.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e Specifically, previous studies with engineered bivalent NRG1 that activates the ERBB system by forced ERBB4 homodimerization, reproduced the anti-apoptotic effects of NRG1 in cardiomyocytes, without inducing growth of cancer cells.\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e For this reason, we performed a high-throughput screen (HTS) for ERBB4 ligands using an assay that detects ERBB4 homodimerization upon ligand binding. Here, we report the identification of small molecules that induce ERBB4 homodimerization and show their cardiac activity \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of small molecule ERBB4 agonists\u003c/h2\u003e \u003cp\u003eTo identify small molecule agonists of the ERBB4 receptor, an HTS was performed with a Pharmacological Diversity Set of 10,240 synthetic molecules using an ERBB4/ERBB4 dimerization assay. A primary screen followed by a confirmation screen resulted in 62 hit compounds (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). To reduce, confirm and cluster the compounds, maximum common substructure (MCS) analysis was performed resulting in the identification of 368 unique MCSs within the 62 confirmed hits. The outcome of a k-means clustering on these 368 MCSs resulted in 10 unique cluster centers, of which cluster center 4 showed a 458-fold enrichment compared to the Enamine reference library (Suppl. Figure\u0026nbsp;1a). Cluster center 4 consisted of 61 individual cluster members, 3 of which provided a significant enrichment compared to the reference library. The common MCS derived from these 3 MCSs resulted in a pattern that gives 109-fold enrichment (Suppl. Figure\u0026nbsp;1b). Four of the compounds of the confirmed hit list contained this pharmacophore, and 16 compounds contained analogous MCS patterns. We generated a dose-response curve (DRC) of these 20 compounds. Only the 4 hit compounds containing the common pharmacophore (named EF-1, EF-2, EF-3, and EF-4) showed a reliable and reproducible DRC with half maximal effective concentration (EC\u003csub\u003e50\u003c/sub\u003e) values in the micromolar range and a maximum response (E\u003csub\u003emax\u003c/sub\u003e) that varied from 10.9\u0026thinsp;\u0026plusmn;\u0026thinsp;2.2% to 27.9\u0026thinsp;\u0026plusmn;\u0026thinsp;4.8%, relative to NRG1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb, f-h). We screened an additional 111 compounds containing the pharmacophore (selected from the Enamine library), which resulted in 2 additional hits (EF-5 and EF-6) that induced ERBB4 dimerization in a concentration-dependent manner (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ei-j). Next, a random-forest machine-learning model was developed using Morgan fingerprints of active and inactive molecules based on the results of the 20 selected hit compounds and the results of the additional screening of 111 analogues. Applying this machine-learning model on the Enamine library led to 34 additional compounds that were predicted to be active. Experimental validation of these 34 predictions resulted in 2 additional compounds (EF-7 and EF-8) that induced ERBB4 dimerization in a concentration-dependent manner (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ek-l). Apart from the 8 active pharmacophore-containing ERBB4 agonists, we selected 1 of the compounds that did not activate ERBB4 but contained the same pharmacophore (hereafter named NA-1, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003em) to be included in several experiments as a control. The chemical structures and molecular weights (MW) of the 8 selected hit compounds and NA-1 are shown in Suppl. Figure\u0026nbsp;2.\u003c/p\u003e \u003cp\u003eWe selected EF-1 for further studies because it showed the highest potency (EC\u003csub\u003e50\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;10.5\u0026thinsp;\u0026plusmn;\u0026thinsp;4.5 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;6\u003c/sup\u003e M) and efficacy (E\u003csub\u003emax\u003c/sub\u003e = 27.9\u0026thinsp;\u0026plusmn;\u0026thinsp;4.8% of the effect of NRG1; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb) in the ERBB4/ERBB4 dimerization assay. Since ERBB2 is a preferred heterodimerization partner of ERBB4, and because NRG1 can also bind to ERBB3 to induce ERBB2/ERBB3 dimerization, we evaluated the effects of EF-1 on ERBB2/ERBB4 and ERBB2/ERBB3 dimerization. Although the potency of EF-1 in the ERBB2/ERBB4 dimerization assay was low (EC\u003csub\u003e50\u003c/sub\u003e\u0026thinsp;\u0026gt;\u0026thinsp;32 \u0026micro;M), its efficacy at the highest concentration was like the natural ligand NRG1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec). The stimulatory effect of EF-1 on ERBB2/ERBB3 heterodimerization was lower compared to ERBB4 homodimerization (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed), indicating a preferential but not exclusive activation of ERBB4.\u003c/p\u003e \u003cp\u003eTo evaluate whether EF-1 binds to a similar binding pocket than NRG1, we performed a cell-based competition assay. A flow cytometry\u0026ndash;based fluorescence competition assay showed that NRG1 as a positive control (ctrl) reduced binding of fluorescent NRG1 (F-NRG1) to the ERBB4 receptor with a half maximal inhibitory concentration (IC\u003csub\u003e50\u003c/sub\u003e) of 32 nM (Suppl. Figure\u0026nbsp;3). EF-1, however, did not reduce binding of F-NRG1 (Suppl. Figure\u0026nbsp;3), indicating that EF-1 and NRG1 bind to different sites. To evaluate whether EF-1 could influence the effect of NRG1 on ERBB4 dimerization, the ERBB4/ERBB4 dimerization cell line was used to generate DRCs of NRG1 in the presence of 0 to 32 \u0026micro;M of EF-1. EF-1 dose-dependently and significantly potentiated NRG1-induced ERBB4 homodimerization, by 299.5% at a concentration of 32 \u0026micro;M (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eEF-1 and NRG1 activate similar downstream signaling pathways\u003c/h2\u003e \u003cp\u003eCanonical pathways activated by NRG1 in cardiomyocytes are the phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) and mitogen-activated protein kinase (MAPK)/ extracellular signal-regulated kinase (ERK) pathways.\u003csup\u003e\u003cspan additionalcitationids=\"CR31\" citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e To examine the effect of EF-1 on both pathways, phosphorylation of AKT and ERK1/2 was assessed by western blot analysis in condtionally immortalized rat atrial myocytes (iAMs).\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e Compared to cardiomyocytes derived from induced pluripotent stem cells, iAMs are easier to culture and show a more complete and reliable cardiomyogenic differentiation.\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e EF-1 induced a time-dependent phosphorylation of AKT (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea) in cardiomyogenically differentiated iAMs. EF-1 also induced a transient phosphorylation of ERK1/2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb).\u003c/p\u003e \u003cp\u003eTo gain insight into downstream signaling pathways and cell responses activated by small-molecule ERBB4 agonists, we performed bulk RNA sequencing on cardiomyogenically differentiated iAMs and human cardiac fibroblasts (HCF), stimulated with either EF-1, NRG1, or vehicle (Veh). In iAMs, EF-1 significantly upregulated 354 genes and significantly downregulated 1120 genes compared to Veh (log\u003csub\u003e2\u003c/sub\u003e-fold change\u0026thinsp;\u0026ge;\u0026thinsp;0.58, P adj\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec). The top 50 significant differentially expressed genes (DEGs) between EF-1 and the Veh group are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed. Although there is a substantial overlap in DEGs between NRG1 and EF-1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee), many DEGs induced by NRG1 and EF-1 differ, which is also reflected in a cluster analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef).\u003c/p\u003e \u003cp\u003eIn HCFs, EF-1 exhibited a significant upregulation of 336 genes and a significant downregulation of 507 genes compared to Veh (log2-fold change\u0026thinsp;\u0026gt;\u0026thinsp;0, adjusted P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Fig.\u0026nbsp;3a). Figure\u0026nbsp;3b displays the top 50 significantly DEGs between EF-1 and the Veh group. Additionally, nearly two-thirds of the DEGs influenced by NRG1 are shared with DEGs influenced by EF-1 (Fig.\u0026nbsp;3c and 3d). In HCFs, the transforming growth factor-β (TGF-β) and MAPK/ERK pathways were both downregulated by EF-1 and by NRG1, but only reaching statistical significance with EF-1 (Fig.\u0026nbsp;3e). The PI3K/AKT pathway was downregulated in HCFs when stimulated with EF-1 or NRG1, but not statistically significant (Fig.\u0026nbsp;3e).\u003c/p\u003e \u003cp\u003e \u003cb\u003eEF-1 decreases TGF-β1-induced collagen expression through ERBB4\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e\u003c/p\u003e \u003cp\u003eBecause NRG1 has been shown to decrease collagen mRNA expression induced by TGF-β1 in fibroblasts, we determined the effects of EF-1 on collagen expression induced by TGF-β1 (10 ng/mL) in HCFs and included an \u003cem\u003eERBB4\u003c/em\u003e knockdown experiment to test the role of ERBB4.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e We first observed that TGF-β1 significantly increased collagen type 3 alpha 1 (\u003cem\u003eCOL3A1)\u003c/em\u003e mRNA levels in HCFs by 36%. (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ea) and that the expression of \u003cem\u003eERBB4\u003c/em\u003e in HCFs was downregulated by 60% by \u003cem\u003eERBB4\u003c/em\u003e silencing RNAs (siRNAs) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Next, EF-1 dose-dependently decreased \u003cem\u003eCOL3A1\u003c/em\u003e mRNA levels, induced by TGF-β1, up to 50%, while knockdown of \u003cem\u003eERBB4\u003c/em\u003e expression significantly attenuated this effect. (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ec). In contrast, NA-1, a compound that contains the pharmacophore but does not induce ERBB4 dimerization, did not affect \u003cem\u003eCOL3A1\u003c/em\u003e expression in HCFs induced by TGF-β1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003e \u003cb\u003eEF-1 reduces cardiomyocyte cell death and hypertrophy\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e\u003c/p\u003e \u003cp\u003eBecause NRG1 has been shown to decrease cardiomyocyte cell death and to attenuate cardiomyocyte hypertrophy, we next evaluated the effects of EF-1 on cardiomyocytes and again included an \u003cem\u003eERBB4\u003c/em\u003e knockdown experiment.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e Cardiotoxicity was induced in iAMs with hydrogen peroxide (H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e). We first observed that exposure of iAMs to 100 \u0026micro;M H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e increased total cell death to 62% (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ed) and that \u003cem\u003eErbb4\u003c/em\u003e siRNAs reduced \u003cem\u003eERBB4\u003c/em\u003e mRNA in iAMs levels by more than 50% (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ee). Next, EF-1 dose-dependently decreased H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e-induced cardiomyocyte cell death with more than 75% at its highest dose, an effect that was blunted by \u003cem\u003eErbb4\u003c/em\u003e knockdown (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ef). In contrast, NA-1 did not have a significant effect on cardiomyocyte survival (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ef). Finally, Ang II induced cardiomyocyte hypertrophy, as indicated by a significant increase in cross sectional area (CSA), and this effect was dose-dependently attenuated by EF-1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003eg).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eEF-1 prevents myocardial fibrosis\u003c/h2\u003e \u003cp\u003eBefore performing \u003cem\u003ein vivo\u003c/em\u003e studies, we evaluated the pharmacological stability of EF-1. EF-1 remained stable in plasma of both human and mouse for at least 6 h (Suppl. Figure\u0026nbsp;5a). Moreover, the half-life of EF-1 was \u0026gt;\u0026thinsp;6 h when incubated with human liver microsomes and 15 min when incubated with mouse microsomes (Suppl. Figure\u0026nbsp;5b). Next, we evaluated the effects of EF-1 in a mouse model of AngII-induced myocardial fibrosis.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e The experimental design is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003ea. To evaluate changes in mRNA expression of markers of fibrosis, EF-1 was administered simultaneously with AngII for 1 week in wild type mice. EF-1 not only significantly reduced \u003cem\u003eCol1a1\u003c/em\u003e and \u003cem\u003eCol3a1\u003c/em\u003e mRNA expression induced by AngII (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003eb; \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003ec), indicating anti-fibrotic and cardioprotective properties, but also significantly decreased atrial natriuretic peptide (\u003cem\u003eNppa\u003c/em\u003e) expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003ed), a cardiac stretch marker. To study effects on tissue fibrosis, EF-1 was administered simultaneously with AngII for 4 weeks in wild type mice. Masson's trichrome staining of myocardial tissues showed that EF-1 significantly prevented both interstitial and perivascular fibrosis, induced by AngII (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003ee). EF-1 did not significantly alter cardiac dimensions on ultrasound (Suppl. Table\u0026nbsp;1a). To evaluate whether the observed anti-fibrotic effects could be mediated by targets unrelated to ERBB4, we performed a similar experiment with NA-1 (containing the common pharmacophore but without induction of ERBB4 dimerization). NA-1 did not significantly prevent interstitial or perivascular AngII-induced fibrosis (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003ef). Additionally, we evaluated the effects of EF-1 in Ang II-treated transgenic mice with tamoxifen-induced deletion of \u003cem\u003eErbb4\u003c/em\u003e (\u003cem\u003eErbb4\u003c/em\u003e-null mice). \u003cem\u003eErbb4\u003c/em\u003e deletion was confirmed in the heart by western blot analysis (Suppl. Figure\u0026nbsp;5). As expected, \u003cem\u003eErbb4\u003c/em\u003e deletion resulted in a cardiomyopathy phenotype, with a significant increase in left ventricular internal diameter in diastole (LVIDd) and systole (LVIDs), and a fall in fractional shortening (FS) (Suppl. Table\u0026nbsp;1c). Since the commonly used AngII dose of 1,000 ng/kg/day induced 80% mortality in the Erbb4-null mice (data not shown), the dose of AngII was lowered to 400 ng/kg/day,\u003csup\u003e38\u003c/sup\u003e EF-1 did not significantly prevent interstitial or perivascular AngII-induced fibrosis in \u003cem\u003eErbb4\u003c/em\u003e-null mice indicating that ERBB4 is necessary for the effects of EF-1 on cardiac fibrosis (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003eg). Of note, when \u003cem\u003eiCAGCre-Erbb4\u003c/em\u003e\u003csup\u003e\u003cem\u003ef/f\u003c/em\u003e\u003c/sup\u003e mice were injected with corn oil without tamoxifen, the inhibitory effects of EF-1 on AngII-induced interstitial and perivascular fibrosis were preserved (data not shown).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eEF-1 prevented cardiomyocyte injury in DOX-treated mice\u003c/h2\u003e \u003cp\u003eBecause ERBB4 activation is known to prevent cell death, we investigated the effects of EF-1 in a mouse model of DOX-induced acute cardiac toxicity.\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e The experimental design of this experiment is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ea. Osmotic minipumps with either EF-1 or Veh were implanted in wild type and \u003cem\u003eErbb4-null\u003c/em\u003e mice. Four days after the start of the EF-1 or Veh treatment, the animals received an intraperitoneal injection of 20 mg/kg DOX followed 3 days later by the measurement of cardiac troponin I (cTnI) plasma levels. EF-1 significantly prevented the DOX-induced increase in circulating cTnI plasma levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eb). To verify whether the cardioprotective effects of EF-1 were mediated by ERBB4, we repeated the experiment using \u003cem\u003eErbb4\u003c/em\u003e-null mice and observed that EF-1 did not prevent the increase in cTnI plasma levels in these mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ec).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eEF-1 reduces cardiac remodeling and interstitial fibrosis after MI\u003c/h2\u003e \u003cp\u003eIn view of the above effects induced by EF-1, and also because NRG1 has been shown to attenuate adverse cardiac remodelling in MI models, we assessed the effects of EF-1 in a murine MI model (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ed).\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e,\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e Mice were randomized 1 week after ligating the left anterior descending artery (LAD) to implantation of osmotic minipumps with either EF-1 or Veh. Treatment lasted for 28 days, during which cardiac size and function was evaluated using echocardiography (Suppl. Table\u0026nbsp;1e). Representative echocardiographic images of all groups are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ee. In Veh-treated mice, MI significantly increased left ventricular end-diastolic volume (LVEDV; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ef) and left ventricular end-systolic volume (LVESV; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eg) and decreased left ventricular ejection fraction (EF, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eh). 4-week treatment with EF-1 significantly attenuated the increase in LVEDV and LVESV and resulted in a trend towards an increased EF (p\u0026thinsp;=\u0026thinsp;0.0858). Additionally, interstitial fibrosis in the remote myocardium of the infarcted hearts remained significantly lower in mice treated with EF-1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ei; \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ej).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eBy combining HTS and chemoinformatics, we identified a series of compounds that consist of a common pharmacophore and that induce dimerization of the ERBB4 receptor. The most potent compound, EF-1, partially activated downstream signaling pathways that are activated by NRG1, the natural ligand of ERBB4. EF-1 decreased fibroblast collagen production, decreased cardiomyocyte hypertrophy, and improved cardiomyocyte survival \u003cem\u003ein vitro\u003c/em\u003e. \u003cem\u003eIn vivo\u003c/em\u003e, EF-1 mitigated myocardial fibrosis in an AngII model of cardiac fibrosis, prevented acute cardiomyocyte injury in DOX-treated mice, and reduced cardiac dilation and cardiac fibrosis in mice that underwent MI. We showed that the \u003cem\u003ein vitro\u003c/em\u003e effects of EF-1 were ERBB4-dependent since they could be abrogated by siRNAs against \u003cem\u003eERBB4\u003c/em\u003e. Moreover, the \u003cem\u003ein vivo\u003c/em\u003e effects of EF-1 could be abrogated by transgenic deletion of \u003cem\u003eErbb4\u003c/em\u003e in mice. Finally, a compound containing the same pharmacophore as EF-1 but without inducing ERBB4 dimerization (NA-1), did not have the same \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e effects as EF-1, further supporting that the effects of EF-1 depend on its ability to induce ERBB4 dimerization.\u003c/p\u003e \u003cp\u003ePrevious studies have shown the importance of the NRG1/ERRB4 signaling pathway in cardiac development, physiology, and adaptation during disease.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e Based on these studies, rNRG1 has been developed as a potential therapy for chronic heart failure (CHF). Despite encouraging results in phase I and II clinical trials over a decade ago,\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e no results of phase III trials have been published yet. Due to its short plasma half-life, rNRG1 must be administered by continuous intravenous infusion. Moreover, in the published phase I and II trials,\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e rNRG1 was administered for 10 consecutive days without follow-up treatment. Although, short-term effects of a brief rNRG1 administration were significant, it seems unlikely that this treatment regimen will provide long-lasting benefits for CHF patients. Therefore, development of small-molecule ERBB4 agonists could result in a more efficacious therapy for CHF patients.\u003c/p\u003e \u003cp\u003eReceptor dimerization is an established mechanism for the initiation of signal transduction, seen in many cell surface receptors.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e Examples are protein\u0026ndash;tyrosine kinase receptors (including ERBB4), the tumor necrosis factor receptor family, protein-serine/threonine kinase receptors, antigen receptors, and members of the cytokine receptor superfamily.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e Discovery of small molecules with agonist activity on receptor dimerization has been recognized as a challenging endeavor.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e There are few publications on small molecules capable of disrupting specific protein\u0026ndash;protein interactions (antagonists), and the additional requirement for agonists to bind to but also induce dimerization of two receptor molecules makes this aim even more challenging.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e For instance, considerable effort has been put into the identification of small non-peptide agonists of the erythropoietin receptor, resulting in the identification of weak activators.\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e Screening for compounds activating the related thrombopoietin receptor (TPOR) resulted in identification of a potent activator, called SB394725.\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e Structural studies have shown that SB394725 interacts with the juxtamembrane residues of the transmembrane region of TPOR,\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e but whether SB394725 directly induces dimerization of TPOR is unclear.\u003c/p\u003e \u003cp\u003eThe location of the exact binding site of the hit compounds on ERBB4 is currently unknown, but most likely differs from NRG1, as our data indicate that EF-1 does not compete with NRG1 for binding to U2OS ERBB4/ERBB4 dimerization cells and because EF-1 potentiates the effect of NRG1 on homodimerization of ERBB4. This is consistent with a mechanism of action based on ago-allosteric modulation of ERBB4 receptor dimerization.\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e Allosteric activation of ERBB4 and potentiation of the effects of NRG1 could partially explain the remarkable biological effects of EF-1 \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e, despite its potency being lower than that of NRG1. Nevertheless, identification of the binding site of the small molecules could facilitate computational screening of novel small molecules and \u003cem\u003ein silico\u003c/em\u003e optimization. Binding pockets are potentially located at the domains involved in receptor dimerization, for instance domain II.\u003c/p\u003e \u003cp\u003eBoth NRG1 and EF-1 induce ERBB4 receptor homodimerization, although the potency and efficacy of EF-1 is lower than those of NRG1. Remarkably, EF-1 also induced ERBB2-ERBB4 heterodimerization, with the same efficacy as NRG1. EF-1 induces phosphorylation of key proteins in 2 canonical pathways activated by NRG1: ERK1/2 and AKT. ERK1/2 phosphorylation induced by EF-1 is transient with peak levels at 15 min, which is in line with published data on NRG1-induced ERK1/2 phosphorylation in neonatal rat ventricular myocytes.\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e AKT phosphorylation induced by EF-1, however, is much slower compared to NRG1-induced AKT phosphorylation, peaking at 2 h instead of 15 min.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Differences in potency, efficacy, signaling kinetics and the ability to induce ERBB dimerization pairs between EF-1 and NRG1 could at least partially explain the differences observed in AKT and ERK1/2 phosphorylation and in gene expression after the treatment of iAMs and HCFs with these compounds. The number of DEGs after EF-1 treatment of iAMs was significantly lower than after stimulation of the cells with the more potent NRG1 unlike the number of DEGs in HCFs, which were significantly higher after EF-1 treatment than after NRG1 stimulation. Finally, although we tested the effect of EF-1 in a number of contexts that required ERBB4 receptors including in cells transfected with ERBB4 targeting siRNA, and in \u003cem\u003eErbb4\u003c/em\u003e-null mice, and examined effects of non-active compounds that shared the same pharmacophore, off-target effects cannot completely be excluded both \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eIn summary, we showed that small molecules can act as ERBB4 agonists inducing ERBB4 dimerization and triggering ERBB4-mediated biological effects in fibroblasts and cardiomyocytes. We also showed \u003cem\u003ein vivo\u003c/em\u003e evidence that these small molecules could be a novel therapeutic strategy for treatment of CHF. As ERBB4 is also important in other diseases like fibrotic, inflammatory, and neurological disorders,\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan additionalcitationids=\"CR52 CR53 CR54\" citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e small-molecule ERBB4 agonists could also be of therapeutic relevance in other diseases.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec10\"\u003e\n \u003ch2\u003eStudy approval\u003c/h2\u003e\n \u003cp\u003eAll animal experiments were approved by the Ethical Committee of the University of Antwerp and conformed to the Guide for the Care and Use of Laboratory Animals, 8th edition published by the US National Institutes of Health in 2011, and to the European Communities Council Directive 2010/63/EU for the protection of animals used for experimental purposes. All animals were fed on a standard chow, were provided with water \u003cem\u003eat libitum\u003c/em\u003e, and were housed at a constant temperature of 22\u0026deg;C and humidity of 50% in a 12 h controlled light/dark cycle. Throughout the experimental period, mice were closely monitored for any signs of distress or adverse effects.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\"\u003e\n \u003ch2\u003eCells\u003c/h2\u003e\n \u003cp\u003ePathHunter U2OS ERBB4/ERBB4 (Eurofins, 93-0961C3), U2OS ERBB2/ERBB4 (Eurofins, 493-0960C3) and ERBB2/ERBB3 (Eurofins, 93-1042C3) dimerization cell lines were cultured according to the manufacturer\u0026rsquo;s instructions. Briefly, cells were cultured in Cell Culture Reagent 103 (Eurofins, 92-3103G) supplemented with 250 \u0026micro;g/mL Hygromycin B (Eurofins, 92\u0026thinsp;\u0026minus;\u0026thinsp;0029) and 500 \u0026micro;g/mL G418 (Eurofins, 92\u0026thinsp;\u0026minus;\u0026thinsp;0030), and were maintained at 37\u0026deg;C in humidified atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e. HCFs (Innoprot, P10454) were cultured in fibroblast medium (Innoprot, P60108) supplemented with 10% (v/v) fetal bovine serum (FBS, Innoprot), 10% (v/v) fibroblast growth supplement (Innoprot), and 1% (v/v) penicillin/streptomycin solution (Innoprot). iAMs\u003csup\u003e33\u003c/sup\u003e were cultured in Advanced DMEM F-12 (Thermo Fisher Scientific, 12634028) supplemented with 2% (v/v) heat-inactivated FBS (Thermo Fisher Scientific, 10270106), 1% (v/v) penicillin/streptomycin (10,000 units/mL and 10 mg/mL respectively, Thermo Fisher Scientific, 15140122), 1\u0026times; GlutaMAX (Thermo Fisher Scientific, 35050061), and 100 ng/mL doxycycline (Tocris, 4090) for proliferation. To induce differentiation, iAMs were transferred to medium without doxycycline.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\"\u003e\n \u003ch2\u003eHTS and chemoinformatics\u003c/h2\u003e\n \u003cp\u003eAn HTS of 10,240 compounds (Pharmacological Diversity Set, Enamine) was performed using the PathHunter U2OS ERBB4/ERBB4 dimerization cell line. U2OS ERBB4/ERBB4 dimerization cells were seeded at a density of 5\u0026times;10\u003csup\u003e3\u003c/sup\u003e cells/well in 50 \u0026micro;L Cell Plating 0 Reagent (Eurofins, 93-0563R0A) in white 384-well plates (Greiner Bio-One, 781080). Cells were treated with compound (10 \u0026micro;M), NRG1 (positive ctrl, 1 \u0026micro;M; Eurofins, 92-1031), or phosphate-buffered saline (negative ctrl, PBS; Thermo Fisher Scientific, 14040133). All wells contained dimethyl sulfoxide (DMSO; Merck Life Science, D2438) at a final concentration of 1%. The cells were subsequently incubated for 6 h at 37\u0026deg;C in a humidified atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e. Then, 25 \u0026micro;L of PathHunter Flash Detection Reagent (Eurofins, 93\u0026ndash;0247) was added to each well, and cells were incubated at room temperature (RT) in the dark for 1 h. Subsequently, luminescence was measured using the EnVision plate reader (Revvity). Spotfire (TIBCO) was used for data analysis and visualization and Collaborative Drug Discovery Vault for compound registry and data management. Data from the primary screening was analyzed via the HTS-Corrector software.\u003csup\u003e56\u003c/sup\u003e In HTS-Corrector, intraplate normalization (via median polish) was performed to correct for row, column, or edge effects. The analysis output was the normalized values for each compound. Subsequently, normalized values were used to perform inter-plate normalization, which generated B-scores as final analysis output. The top 80 compounds having the highest B-score were selected and re-tested in a confirmation screen under the same assay conditions as the primary screen. In the confirmation screen, the threshold for hit selection was defined as the average signal of the negative control plus three times the standard deviation (SD) of the negative control.\u003c/p\u003e\n \u003cp\u003eTo validate and cluster hit compounds, chemoinformatic methods were used to identify the MCSs between the biological active compounds. To estimate the relative significance of each MCS pattern for biological activity, the enrichment of each MCS pattern was calculated by comparing the occurrence of the respective MCS pattern in both the hitlist and reference set (which was the entire Enamine HTS collection consisting of 1,773,567 compounds). MCS patterns with the highest enrichment were then used to select compounds from the Enamine library for additional screening that contain the respective MCS from the Enamine library, resulting in an additional 111 compounds for follow-up screening. In parallel, we developed a random-forest machine-learning model using Morgan fingerprints of active and inactive molecules of previous screenings.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\"\u003e\n \u003ch2\u003ePathHunter dimerization assay in 96-well format\u003c/h2\u003e\n \u003cp\u003eU2OS ERBB4/ERBB4 dimerization cells were seeded at a density of 5\u0026times;10\u003csup\u003e4\u003c/sup\u003e cells/well in 100 \u0026micro;L Cell Plating 0 Reagent, in white 96-well plates (PerkinElmer, 6005680) and incubated for 24 h at 37\u0026deg;C, 5% CO\u003csub\u003e2\u003c/sub\u003e. Cells were treated for 6 h with different concentrations of the 8 hit compounds (Enamine, EF-1 \u0026ndash; EF-8) or compound NA-1, NRG1 (Peprotech, 100-03), or PBS. Next, 110 \u0026micro;L PathHunter Flash Detection Reagent was added to each well and incubated at RT for 1 h after which the luminescence signal was measured using the Luminoskan Ascent (Thermo Fisher Scientific). Dose-response curves of the compounds were performed in \u003cem\u003e2-fold\u003c/em\u003e (0.0625\u0026ndash;32 \u0026micro;M). The same experimental set-up was used for the U2OS ERBB2/ERBB4 and U2OS ERBB2/ERBB3 dimerization cell lines. For co-administration of NRG1 and EF-1, cells were pretreated for 10 min with 5 \u0026micro;L of EF-1 (1, 10, or 32 \u0026micro;M) before adding 5 \u0026micro;L of different NRG1 concentrations and incubated for 6 h at 37\u0026deg;C, 5% CO\u003csub\u003e2\u003c/sub\u003e. All wells contained DMSO at a final concentration of 0.9%.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\"\u003e\n \u003ch2\u003eFluorescence-based competition binding assay\u003c/h2\u003e\n \u003cp\u003eNRG1 was dissolved in PBS at a concentration of 1 mg/mL and labeled with Alexa Fluor 488 using the Alexa Fluor 488 microscale protein labeling kit (Thermo Fisher Scientific, A30006). A Bio-gel P-4 (Bio-Rad, 1504124) fine resin suspended in PBS and a dye:protein molar ratio of 5 was used to purify the labelled NRG1 according to the manufacturer\u0026rsquo;s instructions. Fluorescent labeling was evaluated by performing a dose-response experiment using F-NRG1 (1\u0026ndash;1000 nM) on the U2OS ERBB4/ERBB4 dimerization cell line. Next, the same cell line was used to perform a competition binding assay with F-NRG1 using flow cytometry. Cells were plated in transparent U-bottom 96-well plates (Greiner Bio-One, M9436) at 5\u0026times;10\u003csup\u003e5\u003c/sup\u003e cells/mL in 50 \u0026micro;L ice-cold buffer (PBS with 0.1% bovine serum albumin (BSA; Sigma-Aldrich, A7906) and 0.05% sodium azide (Sigma-Aldrich, S2002)). Next, cells were centrifuged at 17,968 x \u003cem\u003eg\u003c/em\u003e for 4 min at RT and the supernatants were discarded. The cell pellets were washed with 50 \u0026micro;L ice-cold buffer by gently pipetting up and down. Centrifugation and washing were repeated, and the supernatants were discarded. For the competition assay between NRG1 and F-NRG1, cells were treated with NRG1 (1\u0026ndash;100 nM) and F-NRG1 (30 nM). For the competition assay between EF-1 and F-NRG, cells were treated with EF-1 (0.1\u0026ndash;100 \u0026micro;M) and F-NRG1 (30 nM). All wells contained DMSO at a final concentration of 0.9%. After gentle mixing of each sample, the plate was incubated for 1 h at 4\u0026deg;C on a microplate shaker. Next, the plate was centrifuged at 17,968 x \u003cem\u003eg\u003c/em\u003e for 4 min at RT and the supernatants were discarded. The cells were then washed with 100 \u0026micro;L ice-cold buffer, pelleted by centrifugation and washed again. The supernatants were discarded and 100 \u0026micro;L ice-cold buffer was added to the cell pellets and pipetted up and down to generate single-cell suspensions. These suspensions were transferred to 5-mL polystyrene round-bottom tubes, kept on ice, and exposed to minimum light until flow cytometric analysis (BD Accuri C6, BD Biosciences). Unstained cells were used to set the parameters of the flow cytometer.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\"\u003e\n \u003ch2\u003eRNA sequencing\u003c/h2\u003e\n \u003cp\u003eiAMs were seeded at a density of 6\u0026times;10\u003csup\u003e6\u003c/sup\u003e cells/mL in differentiation medium in 75-cm\u003csup\u003e2\u003c/sup\u003e cell culture flasks (Greiner Bio-One, 658175) and incubated for 9 days (d0\u0026thinsp;=\u0026thinsp;day of seeding). HCF were seeded at a density of 3x10\u003csup\u003e6\u003c/sup\u003e cells/mL in fibroblast medium in 75-cm\u003csup\u003e2\u003c/sup\u003e cell culture flasks. Then, cells were incubated for 16 h at 37\u0026deg;C in the presence of EF-1 (32 \u0026micro;M), PBS, or NRG1 (0.1 \u0026micro;M). All wells contained DMSO at a final concentration of 0.9%. Next, cells were lysed using 350 \u0026micro;L buffer RLT (Qiagen) supplemented with \u0026beta;-mercaptoethanol (100:1; Merck, 444203). Total RNA was isolated using the RNeasy Micro Kit (Qiagen, 74104) according to the manufacturer\u0026rsquo;s protocol, with an extra step of DNase digestion. The concentration and quality of RNA were determined using a Qubit fluorometer (Thermo Fisher Scientific) and 2100 Agilent BioAnalyzer (Agilent Technologies), respectively. Samples with an RNA integrity number\u0026thinsp;\u0026gt;\u0026thinsp;7 were used for library preparation. Sequencing libraries were prepared by Genewiz/Azenta (Leipzig) on cDNA prepared from polyadenylated mRNA. Libraries were sequenced using a NovaSeq 6000 (Illumina) with a read length of 2\u0026times;150 bp.\u003c/p\u003e\n \u003cp\u003eGene expression was quantified at the transcript level using Salmon (v1.10.0)\u003csup\u003e57\u003c/sup\u003e, with the validatMappings and -gcBias parameters switched on, to the Rnor_6.0 or GRCh38 transcriptome. Transcript level counts were aggregated to gene level using the import in the tximport package (v1.26.1)\u003csup\u003e58\u003c/sup\u003e, setting countsFromAbundance to \u0026lsquo;lengthScaledTPM\u0026rsquo; in R (v4.1.1). DESeq2 R package (v1.38.3)\u003csup\u003e59\u003c/sup\u003e was used for differential gene expression analysis between different conditions. The batch variability of different sequencing runs was accounted for by defining \u0026ldquo;batch\u0026rdquo; as a covariate in the linear model to analyse differential gene expression. Differential gene expression heat-maps were generated by using the pheatmap R package (v1.0.12), and volcano plots by EnhancedVolcano (v1.10.0). The overlapping genes between different conditions were obtained by VennDiagram (v1.7.3)\u003csup\u003e60\u003c/sup\u003e. Functional enrichment of DEGs was determined using a hypergeometric test against the Gene Ontology database by using the ClueGO (v2.5.7)\u003csup\u003e61\u003c/sup\u003e module of Cytoscape (v3.9.1)\u003csup\u003e62\u003c/sup\u003e with Benjamini\u0026ndash;Hochberg adjusted (FDR) \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05. GSEA Preranked method was performed to identify the Hallmark pathways.\u003csup\u003e63\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003eThe RNA sequencing data generated has been deposited in NCBI\u0026rsquo;s Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo) and is accessible through GEO Series accession numbers GSE256024 for iAM data and GSE261219 for HCF data.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\"\u003e\n \u003ch2\u003eSiRNA transfection experiment\u003c/h2\u003e\n \u003cp\u003eSiRNA transfection was carried out according to the manufacturer\u0026rsquo;s instructions, using 1.25 \u0026micro;L DharmaFECT 1 transfection reagent (Horizon Discovery, T-2001) and 2.5 \u0026micro;L of 10 \u0026micro;M \u003cem\u003eERBB4\u003c/em\u003e siRNAs or non-targeting siRNAs (ON-TARGETplus Human \u003cem\u003eERBB4\u003c/em\u003e or Non-Targeting Pool, Horizon Discovery, D-00180/T-2001) per well. After the assay, cells transfected with and without \u003cem\u003eERBB4\u003c/em\u003e siRNAs, to determine knockdown efficiency, were lysed using RA1 lysis buffer (Macherey-Nagel, 740955.250) supplemented with \u0026beta;-mercaptoethanol (100:1), scraped with a cell scraper and collected in an Eppendorf tube (Greiner Bio-One, 616201) before RNA isolation for reverse transcription-quantitative polymerase chain reaction (RT-qPCR).\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIn vitro\u003c/strong\u003e \u003cstrong\u003ecollagen expression\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eHCFs were seeded at a density of 1.5\u0026times;10\u003csup\u003e5\u003c/sup\u003e cells/well in doxycycline-free medium in 12-well plates (Greiner Bio-One, 665180) and incubated overnight. Next, siRNA transfection was carried out as described above. Cells were incubated for 22 h at 37\u0026deg;C and 5% CO\u003csub\u003e2\u003c/sub\u003e. Next, medium was changed to fresh differentiation medium (without doxycycline), and cells were stimulated for 24 h with either PBS, EF-1 (4\u0026ndash;32 \u0026micro;M) or NA-1 (4\u0026ndash;32 \u0026micro;M) together with TGF-\u0026beta;1 (10 ng/mL; Peprotech, 100\u0026thinsp;\u0026minus;\u0026thinsp;21). All wells had a final concentration of 0.9% DMSO. Compounds were pre-incubated for 10 min before addition of TGF-\u0026beta;1. Next, cells were lysed using 100 \u0026micro;L RA1 lysis buffer supplemented with \u0026beta;-mercaptoethanol (100:1), scraped with a cell scraper, and collected in an Eppendorf tube before RNA isolation for RT-qPCR.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\"\u003e\n \u003ch2\u003eH\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e\u0026ndash;induced cell death assay\u003c/h2\u003e\n \u003cp\u003eiAMs were seeded at a density of 2.7x10\u003csup\u003e5\u003c/sup\u003e cells/well in 48-well plates (Greiner Bio-One, 677180) and differentiated over 9 days in medium without doxycycline (d0\u0026thinsp;=\u0026thinsp;day of seeding). Next, siRNA transfection was carried out as described above. After incubation for 24 h at 37\u0026deg;C and 5% CO\u003csub\u003e2\u003c/sub\u003e, the culture medium was refreshed. Next, the cells were pretreated for 10 min with either EF-1 (4\u0026ndash;32 \u0026micro;M), PBS, or NA-1 (4\u0026ndash;32 \u0026micro;M). Subsequently, H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e (100 \u0026micro;M; Merck, H1009) was added and the cells were incubated for 4 h at 37\u0026deg;C. All wells contained DMSO at a final concentration of 0.9%. Next, 150 \u0026micro;L of the culture medium in each well was transferred to a 96-well plate and mixed with 100 \u0026micro;L Toxilight AK detection reagent (Lonza, LT07-217). After incubation at RT for 5 min, luminescence was measured using the Luminoskan Ascent. A 100% lysis control (Lonza, LT27-239) was used to determine the percentage of total dead cells.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\"\u003e\n \u003ch2\u003eCardiomyocyte hypertrophy assay\u003c/h2\u003e\n \u003cp\u003eiAMs were seeded at a density of 10\u003csup\u003e4\u003c/sup\u003e cells/well in differentiation medium in 24-well plates (Greiner Bio-One, 662160) and differentiated over 9 days (d0\u0026thinsp;=\u0026thinsp;day of seeding). Next, the cells were pretreated for 1 h with either EF-1 (4\u0026ndash;32 \u0026micro;M) or PBS after which AngII (100 nM; Merck, A9525) was added and the cells were incubated for 24 h at 37\u0026deg;C. All wells contained DMSO at a final concentration of 0.9%. To determine CSA, cells were fixed with 4% paraformaldehyde (Thermo Fisher Scientific, 043368.9M) for 30 min at 4\u0026deg;C. Cells were washed thrice with PBS and permeabilized by incubation with 0.1% Triton X-100 (Merck, 10789704001) for 10 min at RT. iAMs were stained with Alexa Fluor 568 phalloidin (Thermo Fisher Scientific, A12380) in 1% BSA-PBS solution for 1 h at RT. After washing 3 times with PBS, 4\u0026rsquo;,6-diamidino-2-phenylindole dihydrochloride (DAPI, Merck, D9542) was added. Images were obtained by fluorescence microscopy (Celena S).\u003c/p\u003e\n \u003cp\u003eCSA was quantified using an automated algorithm, custom-made in Python. The mode \u003cem\u003em\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e (maximum of the histogram) of DAPI intensities (excluding 0) was determined and binary segmentation of nuclei was created with threshold \u003cem\u003em\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e + \u003cem\u003eSD\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e (with \u003cem\u003eSD\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e the SD of DAPI intensities). Distinct objects with an area less than 500 pixels and nuclei where the average overlapping iAM signal was less than \u003cem\u003emd\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e + 0.5 \u003cem\u003eSD\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e (with \u003cem\u003emd\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e and \u003cem\u003eSD\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e the median and SD of the Alexa Fluor 568 intensity, respectively) were removed. Binary segmentation of Alexa Fluor 568 signal was created using a threshold of \u003cem\u003em\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e + 0.5 \u003cem\u003eSD\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e (with \u003cem\u003em\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e the mode of Alexa Fluor 568 intensities excluding 0), followed by one pass of binary erosion and two passes of binary dilation, each with a 1-connected neighborhood. Any pixels with intensities less than \u003cem\u003em\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e \u0026shy; 0.5 \u003cem\u003eSD\u003c/em\u003e\u003csub\u003eA\u003c/sub\u003e were removed from the segmentation. Next, a gradient image was created from the Alexa Fluor 568 phalloidin channel. A grey erosion on the original Alexa Fluor 568 intensities (structuring element 3x3 pixels) was followed by a Gaussian blur (\u003cem\u003e\u0026sigma;\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3 pixels) and a 2D Scharr filter as implemented by \u0026lsquo;scikit-image\u0026rsquo;. Intensities of the resulting gradient image were scaled to the full 8-bit range, before another grey dilation was performed with a 5\u0026times;5 pixel structuring element to smooth the gradient. The gradient was used for watershed segmentation of the individual cells with segmented nuclei as seeds. The implementation provided by \u0026lsquo;scikit-image\u0026rsquo; was used with a \u003cem\u003ecompactness\u003c/em\u003e parameter of 0.1. All objects extending to the image borders were removed. Holes within each remaining individual object were filled, and 3 iterations of binary erosion followed by 3 iterations of binary dilation removed small protrusions. For each cell, the CSA was reported.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIn vitro\u003c/strong\u003e \u003cstrong\u003eassays to assess compound stability in plasma and in liver microsomes\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eA 5-\u0026micro;L aliquot of compound solution (10 mM in DMSO) was added to 995 \u0026micro;L of Non-Swiss Albino Mouse Plasma (Innovative Research, IMSNSAPLAK2E10mL) or Pooled Normal Human plasma (Innovative Research, IPLAK2E10ML) in sodium citrate to obtain a final concentration of 50 \u0026micro;M compound in plasma. The mixture was gently shaken for 6 h at 37\u0026deg;C. Aliquots of 100 \u0026micro;L were taken at various time points (0, 0.5, 1, 2, and 6 h), and diluted with 400 \u0026micro;L of ice-cold acetonitrile (Sigma-Aldrich, AX0156). The resulting suspensions were centrifuged at 17,968 x \u003cem\u003eg\u003c/em\u003e for 5 min. Subsequently, 50 \u0026micro;L of the supernatant was diluted with 950 \u0026micro;L of ice-cold acetonitrile and analyzed by liquid chromatography with tandem mass spectrometry (LC-MS/MS; Waters Acquity H-class UPLC system with a Bruker Daltonics Esquire 3000 plus ion trap mass spectrometer and an Agilent 1100 Series LC system). Samples were analyzed in triplicate and plotted against a standard curve (compound at 31\u0026ndash;1000 nM in plasma and diluted in ice-cold acetonitrile as described above).\u003c/p\u003e\n \u003cp\u003eA mixture of 713 \u0026micro;L milliQ water (Merck, C85358), 200 \u0026micro;L 0.5 M phosphate buffer (pH 7.4, Becton Dickinson, TBS5034), 50 \u0026micro;L NADPH regenerating system solution A (Becton Dickinson), 10 \u0026micro;L NADPH regenerating system solution B (Becton Dickinson) and 2 \u0026micro;L compound (5 mM in DMSO) was prepared and heated for 5 min at 37\u0026deg;C. A volume of 25 \u0026micro;L human and mouse liver microsomes (0.5 mg protein/mL, Corning Life Sciences, 452117 and 452220, respectively) was added to the mixture and 20 \u0026micro;L samples were withdrawn at 0, 0.25, 0.5, 1, 2, 4, 6 and 24 h. Next, 80 \u0026micro;L of ice-cold acetonitrile was added to the samples. After a 10-min incubation period on ice, the mixtures were centrifuged at 15,493 x \u003cem\u003eg\u003c/em\u003e for 5 min at 4\u0026deg;C. Finally, 75 \u0026micro;L of an acetonitrile/water (10/90) mixture was added to 25 \u0026micro;L of supernatant and the resulting samples were analysed in triplicate by LC-MS/MS.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\"\u003e\n \u003ch2\u003eMouse model of AngII-induced myocardial fibrosis\u003c/h2\u003e\n \u003cp\u003eThirteen-week-old C57BL/6N (Charles River, 027) male mice were randomized to the ctrl group (n\u0026thinsp;=\u0026thinsp;5 mice) or to the groups treated with EF-1 (EF-1 group, n\u0026thinsp;=\u0026thinsp;4 mice), AngII plus vehicle (AngII/Veh group, n\u0026thinsp;=\u0026thinsp;5 mice), or AngII plus EF-1 (AngII/EF-1 group, n\u0026thinsp;=\u0026thinsp;5 mice). AngII (1,000 ng/kg/min in PBS), Veh (DMSO/propylene glycol/50:50), and EF-1 (2 mg/kg/day in Veh) were administered for 4 weeks using subcutaneously implanted micro-osmotic pumps (Alzet, model 1004). Four weeks after implantation, cardiac ultrasound was performed, mice were euthanized, and hearts were collected. A similar set-up was used for the 1-week study (Alzet, model 1007D). In some experiments, EF-1 was replaced by NA-1 (2 mg/kg/day in Veh).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\"\u003e\n \u003ch2\u003eMouse model of acute high-dose DOX-induced cardiotoxicity\u003c/h2\u003e\n \u003cp\u003eTwelve-week-old C57BL/6N female mice were randomized to the ctrl group (n\u0026thinsp;=\u0026thinsp;9 mice), or to the groups treated with DOX plus Veh (DOX/Veh group, n\u0026thinsp;=\u0026thinsp;8 mice), or DOX plus EF-1 (DOX/EF-1 group, n\u0026thinsp;=\u0026thinsp;8 mice). EF-1 (2 mg/kg in Veh) or Veh were administered for 1 week using subcutaneous micro-osmotic pumps and started on day 1. DOX (20 mg/kg; Pfizer, 4222) was administered intraperitoneally once on day 4 to induce cardiotoxicity. On day 7, mice were euthanized, serum samples were collected via the retrobulbar sinus and hearts were excised.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec21\"\u003e\n \u003ch2\u003eMouse model of MI\u003c/h2\u003e\n \u003cp\u003eTwelve-week-old female Balb/cJ mice were randomized into either the ctrl or MI group. Mice in the MI group underwent surgical ligation of the LAD, while mice in the ctrl group underwent a sham procedure (n\u0026thinsp;=\u0026thinsp;8 mice). Briefly, all mice received an injection of 0.1 mg/kg buprenorphine (Produlab Pharma) before induction of anesthesia with 8% sevoflurane (Zoetis). Anesthesia was maintained with 4.5% sevoflurane and mice were intubated and ventilated. An incision of 15 mm was made on the left side of the thorax and the thoracic cavity was opened at the third intercostal space using blunt forceps. The LAD was permanently ligated with a 8/0 polypropylene monofilament sutures (Ethicon, F1894) after which the thoracic cavity and skin were closed with 6/0 polypropylene monofilament sutures (Ethicon, F1841). and 5/0 polyamide 6 sutures (Ethilon, F2412H), respectively. Mice received another dose of 0.1 mg/kg buprenorphine 6\u0026ndash;8 h after the initial dose. Two days after surgery, cardiac ultrasound was performed to exclude mice without successful MI (i.e. displaying hypokinesia in \u0026ge;\u0026thinsp;2 out of 5 segments). Mice with successful MI were randomized into 2 groups: an MI/EF-1 group (n\u0026thinsp;=\u0026thinsp;9 mice) and an MI/Veh group (n\u0026thinsp;=\u0026thinsp;8 mice). Seven days after MI surgery, the mice were equipped with osmotic mini-pumps containing either EF-1 (2 mg/kg/day) or Veh, as described above. Cardiac ultrasound was performed weekly. After 4 weeks of treatment, mice were euthanized, and hearts were collected.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec22\"\u003e\n \u003ch2\u003eTransgenic mouse models\u003c/h2\u003e\n \u003cp\u003eFloxed \u003cem\u003eErbb4\u003c/em\u003e mice (\u003cem\u003eErbb4\u003c/em\u003e\u003csup\u003e\u003cem\u003ef/f\u003c/em\u003e\u003c/sup\u003e; B6; 129-Erbb4tm1Fej/Mmucd, MMRRC, #010439-UCD) were crossed with \u003cem\u003eCAGGCre-ER\u0026trade;\u003c/em\u003e mice (Jackson Laboratory, 004682) containing the \u003cem\u003eTg(CAG-cre/Esr1*)5Amc\u003c/em\u003e transgene that expresses Cre recombinase under the control of a chicken beta actin promoter/enhancer coupled to the human cytomegalovirus immediate-early gene enhancer, resulting in expression of tamoxifen-inducible Cre-ERT in most cell types.\u003csup\u003e64\u003c/sup\u003e To induce Cre recombinase-mediated deletion of \u003cem\u003eErbb4\u003c/em\u003e, 11-week-old female and male i\u003cem\u003eCAGGCre-ER\u003c/em\u003e\u003csup\u003e\u003cem\u003eTM\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e/Erbb4\u003c/em\u003e\u003csup\u003e\u003cem\u003ef/f\u003c/em\u003e\u003c/sup\u003e mice were intraperitoneally injected with tamoxifen (Merck, T5648; 10 mg/kg; in corn oil) daily for 5 consecutive days. Mice were used in aforementioned studies two weeks after the start of the tamoxifen injections. As a control experiment, 11-week-old female and male i\u003cem\u003eCAGGCre-ER\u003c/em\u003e\u003csup\u003e\u003cem\u003eTM\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e/Erbb4\u003c/em\u003e\u003csup\u003e\u003cem\u003ef/f\u003c/em\u003e\u003c/sup\u003e mice were injected with corn oil without tamoxifen.\u003c/p\u003e\n \u003cdiv id=\"Sec23\"\u003e\n \u003ch2\u003eCardiac ultrasound\u003c/h2\u003e\n \u003cp\u003eEchocardiography was performed using the Vevo F2-LAZRX (FUJIFILM VisualSonics) and UHF57x probe. Mice were anesthetized with 1.5% isoflurane (Alvira, BE-V512222). Parasternal long-axis B-mode images and M-mode images were obtained and analyzed using VevoLAB software (FUJIFILM VisualSonics, Version 5.7.1). For the AngII experiments, measurements were obtained through analysis of the short-axis M-mode. For the MI experiments, measurements were obtained through analysis of the parasternal long-axis B-mode, of which the area and end-volume of the left ventricular cavity in diastole and systole were determined by tracing the endocardial border. Acquisitions, measurements and analyses were performed blinded.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec24\"\u003e\n \u003ch2\u003eSerum cTnI enzyme-linked immunoassay\u003c/h2\u003e\n \u003cp\u003eSerum samples were centrifuged for 15 min (1,000 \u0026times; \u003cem\u003eg\u003c/em\u003e) at 4\u0026deg;C and analyzed using a mouse cardiac Troponin I Type 3 ELISA kit (Novus Biologicals, NBP3\u0026ndash;00456), according to the manufacturer\u0026apos;s instructions. Briefly, a 96-well microplate was coated with an anti-mouse TNNI3/cTnI primary antibody (Capture Antibody Solution). Standard solution or samples (100 \u0026micro;L) were added to the plate and incubated for 90 min at 37\u0026deg;C. A biotinylated anti-mouse TNNI3/cTnI detection antibody was added, and the plate was incubated for 1 h at 37\u0026deg;C followed by 3 washes with Wash Buffer. An avidin\u0026ndash;horseradish peroxidase conjugate (Detection Antibody Solution) was then added, and the plate was incubated once more for 30 min at 37\u0026deg;C. After 3 washes with Wash Buffer, Substrate Reagent was added. Following incubation for 15 min at 37\u0026deg;C, the enzymatic reaction was terminated by the addition of Stop Solution. Sample optical densities at 450 nm were converted to tissue concentrations of mouse TNNI3/cTnI using a calibration curve.\u003c/p\u003e\n \u003cdiv id=\"Sec25\"\u003e\n \u003ch2\u003eWestern blot analysis\u003c/h2\u003e\n \u003cp\u003eiAMs were seeded at a density of 10\u003csup\u003e6\u003c/sup\u003e cells/well in a transparent 6-well plate (Greiner Bio-One, 657160) and treated with either EF-1 (32 \u0026micro;M) or PBS for 0, 15, 60, 120, or 240 min; all wells contained DMSO at a final concentration of 0.9%. Hearts of the aforementioned transgenic mice were removed immediately after the animals had been killed. Cells and heart tissue were collected in RIPA lysis buffer (Thermo Fisher Scientific, 89900) supplemented with protease inhibitors (Merck, 11836153001) and phosphatase inhibitors (Merck, 4906845001) and the lysates were centrifuged at 14,000 \u0026times; \u003cem\u003eg\u003c/em\u003e for 10 min. The supernatants were collected, supplemented with sample buffer, and incubated at 95\u0026deg;C for 5 min. Protein quantification was done with Pierce\u0026trade; BCA Protein Assay Kits (Thermo Fisher, 23227) according to the manufacturer\u0026rsquo;s instructions. Equal protein amounts of clarified lysates (20 \u0026micro;g) were subsequently separated by sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (165 V, 400 mA, 60 min) and transferred onto polyvinylidene fluoride membranes by electroblotting (100 V, 400 mA, 90 min). Next, the membranes were incubated for 1 h with Li-Cor blocking buffer (Li-Cor, 927-60001) at RT in Tris-buffered saline (TBS), supplemented with 0.1% Tween (BIO-RAD, 1706531). Next, primary antibodies were added (Suppl. Table 2a) in Li-Cor blocking buffer supplemented with 0.1% Tween, and the membranes were incubated overnight at 4\u0026deg;C on a shaker. Following washing with TBS-0.1% Tween, the membranes were incubated for 1 h at RT with corresponding IRDye-conjugated secondary antibodies (Suppl. Table 2b) in Li-Cor blocking buffer supplemented with 1% SDS and 0.1% Tween. The membrane was visualized with the Odessey imaging system.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec26\"\u003e\n \u003ch2\u003eRT-qPCR analysis of gene expression in fibroblasts, cardiomyocytes, and myocardial tissue\u003c/h2\u003e\n \u003cp\u003eRNA was extracted using Nucleospin RNA XS (Macherey-Nagel, 740902.50) according to the manufacturer\u0026apos;s instructions. For cDNA synthesis, total RNA was added to a mixture containing buffer with random hexamers and reverse transcriptase enzyme (TaqMan reverse transcription reagents, Applied biosystems, N8080234) after which samples were incubated for 10 min at 25\u0026deg;C, 30 min at 48\u0026deg;C, and 5 min at 95\u0026deg;C. RT-qPCR was performed on QuantStudio 3 Real-time PCR system (Applied Biosystems) using Taqman Universal PCR Master Mix (Applied Biosystems, 4304437) and Taqman primers according to the manufacturer\u0026rsquo;s instructions. Settings were as follows: 2 min at 95\u0026deg;C followed by 10 min at 95\u0026deg;C, 40 cycles (45 cycles for ERBB4) of denaturation at 95\u0026deg;C for 15 s, and 1 min at 60\u0026deg;C. All reactions were run in duplicate, and all data were normalized against housekeeping genes glyceraldehyde-3-phosphate dehydrogenase (\u003cem\u003eGAPDH)\u003c/em\u003e, beta-actin (\u003cem\u003eACTB\u003c/em\u003e), or phosphoglycerate kinase 1 (\u003cem\u003ePGK1\u003c/em\u003e). Expression levels were calculated using the comparative cycle method and expressed as fold change (FC) to appropriate controls. The following TaqMan primers were used (Thermo Fisher Scientific): GAPDH (Hs02758991_g1, Mm99999915_g1 and Rn01775763_g1), PGK1 (Hs99999906_m1), COL1A1 (Mm00801666_g1), COL3A1 (Hs00943809_m1 and Mm00802305_g1), ERBB4 (Hs00955525_m1 and Rn00572447_m1), ACTB (Mm02619580_g1) and ANP (Mm01255747_g1).\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec27\"\u003e\n \u003ch2\u003eHistology and immunostaining\u003c/h2\u003e\n \u003cp\u003eHearts were fixed in 4% paraformaldehyde, paraffin-embedded, and cut into 5-\u0026micro;m sections. Collagen distribution was visualized by Masson\u0026rsquo;s trichrome staining. Images were acquired with an Olympus BX43 microscope (Olympus Stream Motion Software) and analyzed with ImageJ 2.14.0 software. Cardiac total and perivascular fibrosis were expressed as the ratio of positively stained fibrotic area (blue) to the total area or vascular lumen area, respectively. Quantification was performed by a person blinded to the treatment protocol.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec28\"\u003e\n \u003ch2\u003eData analysis and statistics\u003c/h2\u003e\n \u003cp\u003eStatistical analysis was performed using GraphPad Prism version 10. Data was checked for normality using the Shapiro-Wilk normality test. Data were expressed as the mean of independent repeats\u0026thinsp;\u0026plusmn;\u0026thinsp;SD. Comparison between groups was performed using unpaired t-test or one-way analysis of variance (ANOVA) with Dunnett or Tukey corrections for multiple comparisons.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eACTB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eBeta-actin\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eAKT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eProtein kinase B\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eAngII\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eAngiotensin-II\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eANOVA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eAnalysis of variance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eBSA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eBovine serum albumin\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003ecDNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003ecopy DNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eCHF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eChronic heart failure\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eCol1a1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eCollagen type 1 alpha 1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eCol3a1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eCollagen type 3 alpha 1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eCSA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eCell surface area\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eCST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eCell-Signaling Technologies\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003ecTnI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eCardiac troponin I\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eCtrl\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eControl\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eDAPI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003e4\u0026apos;,6-Diamidino-2-phenylindole dihydrochloride\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eDEG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eDifferentially expressed gene\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eDMSO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eDimethyl sulfoxide\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eDNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eDeoxynucleic acid\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eDOX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eDoxorubicin\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eDRC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eDose-response curve\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eEC\u003csub\u003e50\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eHalf maximal effective concentration\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eEF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eEjection fraction\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eEmax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eMaximum response\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eERBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eErythroblastic leukemia viral oncogene homolog\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u003cem\u003eErbb4\u003csup\u003ef/f\u003c/sup\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eFloxed \u003cem\u003eErbb4\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eERK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eExtracellular signal-regulated kinase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eFBS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eFetal bovine serum\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eFC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eFold change\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eF-NRG1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eFluorescent NRG1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eFS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eFractional shortening\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eGAPDH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eGlyceraldehyde-3-phosphate dehydrogenase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eH\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eHydrogen peroxide\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eHCF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eHuman cardiac fibroblast\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eHTS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eHigh throughput screening\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eiAM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eConditionally immortalized rat atrial cardiomyocyte\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eIC\u003csub\u003e50\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eHalf maximal inhibitory concentration\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eIVSd\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eInterventricular septum thickness in diastole\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eIVSs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eInterventricular septum thickness in systole\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLAD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLeft anterior descending coronary artery\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLC-MS/MS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLiquid chromatography with tandem mass spectrometry\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLVEDV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLeft ventricular end-diastolic volume\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLVESV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLeft ventricular end-systolic volume\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLVIDd\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLeft ventricular internal diameter in diastole\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLVIDs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLeft ventricular internal diameter in systole\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLVPWd\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLeft ventricular posterior wall thickness in diastole\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eLVPWs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eLeft ventricular posterior wall thickness in systole\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eMAPK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eMitogen-activated protein kinase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eMCS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eMaximum common substructure\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eMI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eMyocardial infarction\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003emRNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eMessenger RNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eMW\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eMolecular weight\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eNon-active pharmacophore-containing compound\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eNppa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eAtrial natriuretic peptide\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eNRG1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eNeuregulin-1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003ePBS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003ePhosphate-buffered saline\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003ePGK1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003ePhosphoglycerate kinase 1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003ePI3K\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003ephosphatidylinositol\u0026nbsp;3-kinase\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eRNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eRibonucleic acid\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eRNA-seq\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eRNA sequencing\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003erNRG1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eRecombinant neuregulin-1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eRT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eRoom temperature\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eRT-qPCR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eReverse transcription-quantitative polymerase chain reaction\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eStandard deviation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eSDS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eSodium dodecyl sulfate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003esiERBB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003esilencing RNA against \u003cem\u003eERBB4\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003esiRNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eSilencing RNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eTBS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eTris-buffered saline\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eTGF-\u0026beta;1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eTransforming growth factor-\u0026beta;1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eTPOR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eThrombopoietin receptor\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"31.29139072847682%\" valign=\"bottom\"\u003e\n \u003cp\u003eVeh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"68.70860927152317%\" valign=\"bottom\"\u003e\n \u003cp\u003eVehicle\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgments\u003c/p\u003e\n\u003cp\u003eJuan Zhang and Minka Bax (Laboratory of Experimental Cardiology, Leiden University Medical Center, Leiden, the Netherlands) are gratefully acknowledged for arranging the transfer of iAMs and providing protocols for their handling. We also thank Tine Bruyns and Mandy Vermont (University of Antwerp, Antwerp, Belgium) for technical support.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis work was supported by a Geconcerteerde onderzoeksactie grant (GOA, PID36444) of the University of Antwerp; by a Senior Clinical Investigator fellowship (to VFS), a PhD fellowship (to JMTC and CC), and research grants of the Fund for Scientific Research Flanders (Application numbers 1842219N, G021019N, G021420N, 1S49323N, and 11PBU24N); VLIR/iBOF Grant 20-VLIR-iBOF-027 (to \u0026nbsp;NV, VFS, HLR, and GWDK).\u003c/p\u003e\n\u003cp\u003eDisclosures\u003c/p\u003e\n\u003cp\u003ePatent \u0026quot;MODULATORS OF ERBB4 IN THE TREATMENT OF DISEASES\u0026quot;; EP20210160742; Inventors: Vincent FM Segers, Gilles W De Keulenaer, Eline Feyen, Hans De Winter.\u003c/p\u003e\n\u003cp\u003eAuthor contributions\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;V.F.M.S. and G.W.D.K. conceived and designed research; J.MT.C, E.F., S.V.d.B, B.G, C.C, J.v.F, M.T., L.N., performed experiments; J.MT.C., E.F., B.G., Y.F., J.V.H. and E.M.W. analyzed data; J.MT.C., E.F., G.W.D.K. and V.F.M.S. interpreted results of experiments; J.MT.C. and E.F. prepared figures; J.MT.C. and E.F. prepared manuscript; J.MT.C, E.F., S.V.d.B, B.G, C.C, J.v.F, E.M.W., B.V.B, A.A.F.D.V, N.V., D.A.P., D.A., H.L.R., H.D.W., G.W.D.K. and V.F.M.S. edited and revised manuscript; J.MT.C, E.F., S.V.d.B, B.G, Y.F., C.C, J.v.F, M.T., J.V.H., L.N., E.M.W., B.V.B, A.A.F.D.V, N.V., D.A.P., D.A., H.L.R., H.D.W., G.W.D.K. and V.F.M.S. approved final version of manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eNemeth BT, Varga ZV, Wu WJ, Pacher P (2017) Trastuzumab cardiotoxicity: from clinical trials to experimental studies. Br J Pharmacol 174:3727\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeglinski M, Ludke A, Jassal DS, Singal PK (2011) Trastuzumab-induced cardiac dysfunction: A \u0026lsquo;dual-hit\u0026rsquo;. Exp Clin Cardiol 16:70\u0026ndash;74\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGassmann M et al (1995) Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor. \u003cem\u003eNature 1995 378:6555\u003c/em\u003e 378, 390\u0026ndash;394\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeyer D, Birchmeier C (1995) Multiple essential functions of neuregulin in development. \u003cem\u003eNature 1995 378:6555\u003c/em\u003e 378, 386\u0026ndash;390\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee KF et al (1995) Requirement for neuregulin receptor erbB2 in neural and cardiac development. Nature 378:394\u0026ndash;398\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOdiete O, Hill MF, Sawyer DB (2012) Neuregulin in cardiovascular development and disease. Circ Res 111:1376\u0026ndash;1385\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLemmens K, Doggen K, De Keulenaer GW (2007) Role of neuregulin-1/ErbB signaling in cardiovascular physiology and disease: Implications for therapy of heart failure. Circulation 116:954\u0026ndash;960\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Keulenaer GW et al (2019) Mechanisms of the Multitasking Endothelial Protein NRG-1 as a Compensatory Factor During Chronic Heart Failure. Circ Heart Fail 12:6288\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Y et al (2022) Neuregulin-1, a potential therapeutic target for cardiac repair. Front Pharmacol 13:945206\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFukazawa R et al (2003) Neuregulin-1 protects ventricular myocytes from anthracycline-induced apoptosis via erbB4-dependent activation of PI3-kinase/Akt. J Mol Cell Cardiol 35:1473\u0026ndash;1479\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVermeulen Z et al (2017) Inhibitory actions of the NRG-1/ErbB4 pathway in macrophages during tissue fibrosis in the heart, skin, and lung. Am J Physiol Heart Circ Physiol 313:H934\u0026ndash;H945\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRyzhov S et al (2017) ERBB signaling attenuates proinflammatory activation of nonclassical monocytes. Am J Physiol Heart Circ Physiol 312:H907\u0026ndash;H918\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBersell K, Arab S, Haring B, K\u0026uuml;hn B (2009) Neuregulin1/ErbB4 Signaling Induces Cardiomyocyte Proliferation and Repair of Heart Injury. Cell 138:257\u0026ndash;270\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD\u0026rsquo;Uva G et al (2015) ERBB2 triggers mammalian heart regeneration by promoting cardiomyocyte dedifferentiation and proliferation. Nat Cell Biol 17:627\u0026ndash;638\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalindo CL, Ryzhov S, Sawyer DB (2014) Neuregulin as a heart failure therapy and mediator of reverse remodeling. Curr Heart Fail Rep 11:40\u0026ndash;49\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFuller SJ, Sivarajah K, Sugden PH (2008) ErbB receptors, their ligands, and the consequences of their activation and inhibition in the myocardium. J Mol Cell Cardiol 44:831\u0026ndash;854\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFalls DL (2003) Neuregulins: functions, forms, and signaling strategies. Exp Cell Res 284:14\u0026ndash;30\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao Y et al (1998) Neuregulins Promote Survival and Growth of Cardiac Myocytes. J Biol Chem 273:10261\u0026ndash;10269\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu X et al (2006) Neuregulin-1/erbB-Activation Improves Cardiac Function and Survival in Models of Ischemic, Dilated, and Viral Cardiomyopathy. J Am Coll Cardiol 48:1438\u0026ndash;1447\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi B et al (2011) Therapeutic effects of neuregulin-1 in diabetic cardiomyopathy rats. Cardiovasc Diabetol 10:69\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDugaucquier L et al (2020) The role of endothelial autocrine NRG1/ERBB4 signaling in cardiac remodeling. Am J Physiol Heart Circ Physiol 319:H443\u0026ndash;H455\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJabbour A et al (2011) Parenteral administration of recombinant human neuregulin-1 to patients with stable chronic heart failure produces favourable acute and chronic haemodynamic responses. Eur J Heart Fail 13:83\u0026ndash;92\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao R et al (2010) A phase II, randomized, double-blind, multicenter, based on standard therapy, placebo-controlled study of the efficacy and safety of recombinant human neuregulin-1 in patients with chronic heart failure. J Am Coll Cardiol 55:1907\u0026ndash;1914\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLenihan DJ et al (2016) A Phase I, Single Ascending Dose Study of Cimaglermin Alfa (Neuregulin 1β3) in Patients With Systolic Dysfunction and Heart Failure. JACC Basic Transl Sci 1:576\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolbro T et al (2003) The ErbB2/ErbB3 heterodimer functions as an oncogenic unit: ErbB2 requires ErbB3 to drive breast tumor cell proliferation. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e 100, 8933\u0026ndash;8938\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArteaga CL, Engelman JA (2014) ERBB Receptors: From Oncogene Discovery to Basic Science to Mechanism-Based Cancer Therapeutics. Cancer Cell 25:282\u0026ndash;303\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin Y, Liu H, Wang X (2020) Neuregulin-1, a microvascular endothelialderived protein, protects against myocardial ischemiareperfusion injury (Review). Int J Mol Med 46:925\u0026ndash;935\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSegers VFM, Dugaucquier L, Feyen E, Shakeri H, De Keulenaer G (2020) W. The role of ErbB4 in cancer. Cell Oncol 43:335\u0026ndash;352\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJay SM et al (2013) An engineered bivalent neuregulin protects against doxorubicin-induced cardiotoxicity with reduced proneoplastic potential. Circulation 128:152\u0026ndash;161\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFang S-J et al (2010) Neuregulin-1 preconditioning protects the heart against ischemia/reperfusion injury through a PI3K/Akt-dependent mechanism. Chin Med J (Engl) 123:3597\u0026ndash;3604\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJie B et al (2012) Neuregulin-1 suppresses cardiomyocyte apoptosis by activating PI3K/Akt and inhibiting mitochondrial permeability transition pore. Mol Cell Biochem 370:35\u0026ndash;43\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGallo S, Vitacolonna A, Bonzano A, Comoglio P, Crepaldi T (2019) ERK: A key player in the pathophysiology of cardiac hypertrophy. Int J Mol Sci 20\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu J et al (2018) Generation and primary characterization of iAM-1, a versatile new line of conditionally immortalized atrial myocytes with preserved cardiomyogenic differentiation capacity. Cardiovasc Res 114:1848\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirabo A et al (2017) Neuregulin-1β induces proliferation, survival and paracrine signaling in normal human cardiac ventricular fibroblasts. J Mol Cell Cardiol 105:59\u0026ndash;69\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLemmens K, Segers VFM, Demolder M, De Keulenaer GW (2006) Role of neuregulin-1/ErbB2 signaling in endothelium-cardiomyocyte cross-talk. J Biol Chem 281:19469\u0026ndash;19477\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuramochi Y et al (2004) Cardiac endothelial cells regulate reactive oxygen species-induced cardiomyocyte apoptosis through neuregulin-1beta/erbB4 signaling. J Biol Chem 279:51141\u0026ndash;51147\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHervent A-S, Vandekerckhove L, De Keulenaer GW (2013) Neuregulin-1 antagonizes myocardial fibrosis and diastolic dysfunction in angiotensin-II treated mice. Eur Heart J 34:P2434\u0026ndash;P2434\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShakeri H et al (2021) Neuregulin-1 compensates for endothelial nitric oxide synthase deficiency. Am J Physiol Heart Circ Physiol 320:H2416\u0026ndash;H2428\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBian Y et al (2009) Neuregulin-1 attenuated doxorubicin-induced decrease in cardiac troponins. Am J Physiol Heart Circ Physiol 297:H1974\u0026ndash;H1983\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu F-F et al (2005) Heterozygous knockout of neuregulin-1 gene in mice exacerbates doxorubicin-induced heart failure. Am J Physiol Heart Circ Physiol 289:H660\u0026ndash;H666\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShiraishi M, Yamaguchi A, Suzuki K (2022) Nrg1/ErbB signaling-mediated regulation of fibrosis after myocardial infarction. FASEB J 36:e22150\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHedhli N et al (2011) Endothelium-Derived Neuregulin Protects the Heart Against Ischemic Injury. Circulation 123:2254\u0026ndash;2262\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMendes-Ferreira P, De Keulenaer GW, Leite-Moreira AF, Br\u0026aacute;s-Silva C (2013) Therapeutic potential of neuregulin-1 in cardiovascular disease. Drug Discov Today 18:836\u0026ndash;842\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVermeulen Z, Segers VFM, De Keulenaer GW (2016) ErbB2 signaling at the crossing between heart failure and cancer. Basic Res Cardiol 111:60\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoger DL, Goldberg J (2001) Cytokine receptor dimerization and activation: prospects for small molecule agonists. Bioorg Med Chem 9:557\u0026ndash;562\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD\u0026ouml;mling A, Beck B, Baumbach W, Larbig G (2007) Towards erythropoietin mimicking small molecules. Bioorg Med Chem Lett 17:379\u0026ndash;384\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErickson-Miller CL et al (2005) Discovery and characterization of a selective, nonpeptidyl thrombopoietin receptor agonist. Exp Hematol 33:85\u0026ndash;93\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim M-J et al (2007) NMR structural studies of interactions of a small, nonpeptidyl Tpo mimic with the thrombopoietin receptor extracellular juxtamembrane and transmembrane domains. J Biol Chem 282:14253\u0026ndash;14261\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwartz TW, Holst B (2007) Allosteric enhancers, allosteric agonists and ago-allosteric modulators: where do they bind and how do they act? Trends Pharmacol Sci 28:366\u0026ndash;373\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePentassuglia L et al (2016) Neuregulin-1β promotes glucose uptake via PI3K/Akt in neonatal rat cardiomyocytes. Am J Physiol Endocrinol Metab 310:E782\u0026ndash;E794\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeng F, Miyazawa T, Kloepfer LA, Harris RC (2018) ErbB4 deletion accelerates renal fibrosis following renal injury. Am J Physiol Ren Physiol 314:F773\u0026ndash;F787\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang Y, Shi J, Zhou J, He C, Gu R (2023) ErbB4 promotes M2 activation of macrophages in idiopathic pulmonary fibrosis. Open Life Sci 18\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalsh T et al (2008) Rare Structural Variants Disrupt Multiple Genes in Neurodevelopmental Pathways in Schizophrenia. Sci (1979) 320:539\u0026ndash;543\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIwakura Y, Nawa H (2013) ErbB1-4-dependent EGF/neuregulin signals and their cross talk in the central nervous system: pathological implications in schizophrenia and Parkinson\u0026rsquo;s disease. Front Cell Neurosci 7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBanerjee A, MacDonald ML, Borgmann-Winter KE, Hahn C-G (2010) Neuregulin 1\u0026ndash;erbB4 pathway in schizophrenia: From genes to an interactome. Brain Res Bull 83:132\u0026ndash;139\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMakarenkov V et al (2006) HTS-Corrector: software for the statistical analysis and correction of experimental high-throughput screening data. Bioinformatics 22:1408\u0026ndash;1409\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatro R, Duggal G, Love MI, Irizarry RA, Kingsford C (2017) Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods 14:417\u0026ndash;419\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoneson C, Love MI, Robinson MD (2016) Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Res 4, 1521\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLove MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15:550\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen H, Boutros PC (2011) VennDiagram: a package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinformatics 12:35\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBindea G et al (2009) ClueGO: a Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks. Bioinformatics 25:1091\u0026ndash;1093\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShannon P et al (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13:2498\u0026ndash;2504\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSubramanian A et al (2005) From the Cover: Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102:15545\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHayashi S, McMahon AP (2002) Efficient Recombination in Diverse Tissues by a Tamoxifen-Inducible Form of Cre: A Tool for Temporally Regulated Gene Activation/Inactivation in the Mouse. Dev Biol 244:305\u0026ndash;318\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-4175488/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4175488/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Heart failure is a common and deadly disease, requiring new therapeutic approaches. The neuregulin-1 (NRG1)/erythroblastic leukemia viral oncogene homolog 4 (ERBB4) pathway is an interesting target because of its cardioprotective effects. The therapeutic use of recombinant NRG1 has been difficult, because it requires intravenous administration and is non-selective for the ERBB4 receptor. Moreover, development of small-molecule agonists of receptor dimers is generally considered to be challenging. Here, we hypothesized that small-molecule-induced activation of ERBB4 is feasible and can protect against myocardial cell death and fibrosis. To this end, we screened 10,240 compounds for their ability to induce homodimerization of ERBB4. We identified a series of 8 structurally similar compounds (named EF-1 – EF-8) that concentration-dependently induced ERBB4 dimerization, with EF-1 being the most potent. EF-1 decreased in an ERBB4-dependent manner cell death and hypertrophy in cultured atrial cardiomyocytes and collagen production in cultured human cardiac fibroblasts. EF-1 also inhibited angiotensin-II (AngII)-induced myocardial fibrosis in wild-type mice, but not in Erbb4-null mice. Additionally, EF-1 decreased troponin release in wild-type mice treated with doxorubicin (DOX), but not in Erbb4-null mice. Finally, EF-1 improved cardiac function in a mouse model of myocardial infarction (MI). In conclusion, we show that small-molecule-induced ERBB4 activation is possible, displaying anti-fibrotic and cardiomyocyte protective effects in the heart. This study can be the start for the development of small-molecule ERBB4 agonists as a novel class of drugs to treat heart failure.","manuscriptTitle":"Small molecule-induced ERBB4 activation to treat heart failure","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-08 17:48:46","doi":"10.21203/rs.3.rs-4175488/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"1b4111ca-5b6e-48d7-908f-17475434e267","owner":[],"postedDate":"April 8th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":30019141,"name":"Biological sciences/Drug discovery/Drug screening/High-throughput screening"},{"id":30019142,"name":"Biological sciences/Physiology/Cardiovascular biology/Cardiovascular diseases/Heart failure"}],"tags":[],"updatedAt":"2025-01-11T08:06:20+00:00","versionOfRecord":{"articleIdentity":"rs-4175488","link":"https://doi.org/10.1038/s41467-024-54908-5","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2025-01-10 05:00:00","publishedOnDateReadable":"January 10th, 2025"},"versionCreatedAt":"2024-04-08 17:48:46","video":"","vorDoi":"10.1038/s41467-024-54908-5","vorDoiUrl":"https://doi.org/10.1038/s41467-024-54908-5","workflowStages":[]},"version":"v1","identity":"rs-4175488","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4175488","identity":"rs-4175488","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}