henhaobio
Sign in

bioRxivBiochemistryPreprint

The epidermal growth factor (EGF) receptor family network in cardiac hypertrophy and heart failure

Stephen J. FulleriD, Susanna T.E. CooperiD, Joshua J. CulliD, Nikola ADAMCZYKiD, Charlotte Tapsell, Riley Pokora, John Spilletts, Philip Richard DashiD, Peter H. SugdeniD, Angela ClerkiD

PDFDOIData
In-depth full-text digest available · Jump to digest ↓

Abstract

The epidermal growth factor receptor (EGFR) network comprises 4 receptors (EGFR, ERBB2, ERBB3, ERBB4) and multiple ligands (EGFR-Ls; for example, AREG/EGF/TGFA that bind to EGFR, EREG/HBEGF that bind to EGFR/ERBB3, and NRG1 that binds to ERBB3/4). NRG1 and ERBB2/ERBB4 are cardioprotective, whilst other EGFR-Ls are implicated in cardiomyocyte hypertrophy. We sought to understand the importance of this network in human heart failure and development of cardiac hypertrophy. By mining data from the Human Protein Atlas, we established that the EGFR family and EGFR-Ls are expressed in different cardiac cells. Furthermore, EGFR-L mRNAs were modulated in failing vs non-failing human hearts implicating them in end-stage disease. In rat cardiomyocytes, EGFR-Ls all activated extracellular signal-regulated kinases 1/2 (ERK1/2) and Akt. However, NRG1 preferentially activated Akt whereas AREG/EREG/HBEGF/TGFA preferentially activated ERK1/2. EGF and NRG1 differentially regulated early gene expression with the response to NRG1 having a greater requirement for ERBB2. To determine if the network is involved in developing cardiac hypertrophy, mice were treated with angiotensin II (AngII) without/with afatinib (to inhibit all EGFR family members). Expression of EGFR-L mRNAs was modulated by AngII over 3-7d. At 7d, echocardiography and segmental strain analysis identified greater cardiac dysfunction in hearts from mice treated with afatinib/AngII compared with AngII alone, associated with inhibition of cardiomyocyte hypertrophy and enhanced cardiac fibrosis. In conclusion, the EGFR network plays an important role in cardiomyocyte and cardiac hypertrophy and is dysregulated in human failing hearts. Further insight into the network may enable identification of novel therapies for cardiac disorders.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (26 Oct 2026) · CC BY 4.0

EGFR-family ligand mRNAs shift in failing human hearts; NRG1 and EGF drive distinct, differently ERBB2-dependent transcriptional programmes in rat cardiomyocytes; afatinib blunts mouse angiotensin II hypertrophy but worsens fibrosis.

EGFR-family inhibition in cancer causes cardiac dysfunction in some patients, but the paper notes that expression profiles of the family in individual cardiac cell types were unknown. It adds ligand-level mRNA profiling in human cardiac cells and failing hearts, ligand-specific and ERBB2-dependent signalling in rat cardiomyocytes, and an EGFR-family inhibitor test in mouse hypertrophy.

Key results

  1. ShownIn Human Protein Atlas single-cell heart data, cardiomyocytes expressed the receptor mRNAs in the order ERBB4 > ERBB2 > EGFR > ERBB3, while EGFR mRNA was enriched in fibroblasts, epicardial cells and adipocytes.

    ERBB4 mRNA was about 10-fold higher than ERBB2 in cardiomyocytes; ERBB2 was at a level similar to EDNRA. NRG1 was almost exclusive to endothelial cells, NRG2 to cardiomyocytes, and AREG was high in epicardial cells. Data cover the 17 cell types reported in the Figure 1 legend (labelled c-0 to c-17). · The EGFR family and their ligands in…; Figure legends; mRNA expression of EGFR family and ligands in…

  2. SuggestedLigand mRNAs, but not receptor mRNAs, differed between failing and non-failing human left ventricle: HBEGF and NRG1 were higher, EREG and TGFA lower in failing hearts.

    qPCR with Mann-Whitney tests; EDNRA was up in failing hearts as expected; EGFR-family receptor mRNAs did not change significantly. Sample numbers are not given in the text. · The EGFR family and their ligands in…; Statistical analysis and presentation of data

  3. ShownIn rat neonatal cardiomyocytes, NRG1 favoured Akt and was far more ERBB2-dependent than EGF, which acted mainly through ERK1/2 with little ERBB2 requirement.

    Akt phosphorylation 26.7 ± 3.2 fold with NRG1 vs 7.6 ± 2.3 fold with EGF (n=4). Tucatinib (ERBB2 inhibitor) cut NRG1-induced Akt by 73.4% and ERK1/2 by 50.4%, with no significant effect on EGF-induced ERK1/2. NRG1, not EGF, raised Jun, Fos, Atf3 and Myc mRNAs. · EGF family receptor signalling to gene…; Figure legends

  4. ShownRNA-seq showed EGF and NRG1 produce qualitatively different transcriptomes, and tucatinib reversed far more of the NRG1 than the EGF response.

    Tucatinib significantly inhibited 100 of 815 EGF-regulated DEGs (12.3%) vs 455 of 995 NRG1-regulated DEGs (45.7%); n=4 per group, FDR<0.05. KEGG analysis: NRG1 more on AMPK/mTOR, EGF more on TNF, IL17, JAK-STAT. PCA PC1 and PC2 explained 27.5% and 17.4% of variance. · EGF family receptor signalling to gene…; Figure legends; RNA-sequencing (RNAseq)

  5. ShownIn mice given angiotensin II for 7 d, afatinib reduced the rise in predicted LV mass and in cardiomyocyte cross-sectional area, increased cardiac fibrosis further (AngII-induced fibrosis was mainly perivascular), and lowered strain and displacement in basal LV segments.

    Echo n=7 per group. AngII alone did not significantly change ejection fraction, stroke volume, cardiac output or global longitudinal strain; afatinib plus AngII gave a significant GLS decline at 3 d. Fibrosis-gene mRNAs rose with AngII but were unchanged by afatinib; Myh7 was higher with afatinib. · Afatinib inhibits cardiac hypertrophy induced by…; Figure legends; Echocardiography; Histological staining

Methods, data and reproducibility

Combines mining of public human heart single-cell data, qPCR on human left ventricle, signalling and RNA-seq in ligand-treated rat neonatal cardiomyocytes with afatinib or tucatinib, and a mouse angiotensin II hypertrophy model with afatinib, assessed by echocardiographic strain and histology.

  • Human Protein Atlas v25.0 heart single-cell clusters (17 cell types) · human · n = 17 cell clusters · GSE183852
  • Human left ventricle, failing vs non-failing hearts (University of Pittsburgh), qPCR · human
  • Rat neonatal cardiomyocyte and non-myocyte cultures: signalling, qPCR · rat (Sprague-Dawley) · n = n=4 independent preparations
  • Rat neonatal cardiomyocyte RNA-seq (vehicle, tucatinib, EGF, NRG1, EGF or NRG1 with tucatinib) · rat · n = n=4 per group · E-MTAB-15921
  • Male C57Bl/6J mice: AngII with or without afatinib; echocardiography, histology, qPCR · mouse · n = 30 mice total; echo n=7 per group
Methods & tools
Human Protein Atlas, Hisat2 v2.0.5, StringTie v1.3.3b, featureCounts v1.5.0-p3, DESeq2 1.20.0, clusterProfiler, ClustVis, GraphPad Prism 10.6.1, VevoStrain, ImageJ, afatinib, tucatinib
Reagents & environment
Software versions stated for the RNA-seq pipeline (Hisat2 v2.0.5, StringTie v1.3.3b, featureCounts v1.5.0-p3, DESeq2 1.20.0) and GraphPad Prism 10.6.1; no containers or requirement files.
Reproducibility gaps
No analysis code is given; RNA-seq processing was done by a contract provider. Other data are stated to be available from the authors on request; supplementary tables and spreadsheets are cited but not in the text.
Limitations (5)
  • Afatinib was given at a high dose chosen to block all EGFR-family receptors, and the study was not designed to assess cardiotoxicity of the drug as used in cancer therapy. · stated by the authors
  • Neonatal rat cardiomyocyte cultures are mixed with non-myocytes and may express EGFR differently from adult cells; adult cardiomyocytes could not be used because they need survival supplements that activate Akt. · stated by the authors
  • Receptor mRNA did not change in failing hearts, whereas other studies report protein-level changes; the authors suggest post-transcriptional regulation or differing patient populations. · stated by the authors
  • The Methods say RNA-seq differential expression used two biological replicates per condition, but the Results and Figure 4 state n=4 per group; the human sample size is not reported. · noted in this summary
  • The mouse inhibitor study used a single afatinib dose over 7 d and no receptor-selective inhibitor, so the contribution of individual EGFR-family members in vivo is not resolved. · noted in this summary

Read this if you work on cardio-oncology or cardiac signalling and want ligand-level EGFR-family expression data, EGF vs NRG1 ERBB2 dependence, or an EGFR-family inhibition result in mouse hypertrophy.

Citations
1

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISVaparanta '22Yin '17Clerk '21Alharbi '22Dai '21Luanpaisano '25
71% semantic matchCell Biology
K. M. Vaparanta, A. Jokilammi, I. Paatero+7 more·bioRxiv·Oct 5, 2022