Depletion of Endothelial PHD2 and 3 Promotes Cardiomyocyte Proliferation and Prevents Ventricular Failure Induced by Myocardial Infarction

Maintaining oxygen homeostasis is essential for the survival of aerobic living creatures. Inadequate supply of oxygen plays central roles in the pathogenesis of myocardial infarction (MI), stroke and cancers. On the other hand, moderate hypoxia can also trigger a series of adaptive responses, such as erythropoiesis and angiogenesis. The evolutionarily conserved prolyl hydroxylase domain protein (PHD)-HIF pathway plays pivotal roles in these processes¹. Interestingly, hypoxia can also induce cardiomyocyte (CM) proliferation and heart regeneration in adult mice². However, the underlying mechanisms remain elusive. From an anatomical perspective, heart is a highly organized pluricellular organ and composed of four major cell types including CMs, endothelial cells (ECs), smooth muscle cells and fibroblasts. In particular, vascular ECs, which act as the “first-responder” to environmental cues such as oxygen and nutrients, not only control vascular permeability but also maintain cardiac function by releasing various secreted factors. However, whether the interaction between ECs and CMs is involved in CM proliferation remains unexplored.

In order to understand the role of endothelial PHDs in CM proliferation, we conditionally depleted PHD2 and 3 in ECs (eKO) by treating PHD2/3^flox/flox; Cdh5-CreER^+/− mice with tamoxifen at neonatal (P2–12) or adult (8–12 weeks) stages. All protocols were approved by the Baylor College of Medicine Institutional Animal Care and Use Committee. With both depletion regimes, eKO mice showed cardiomegaly with a drastic increase in heart weight/tibia length ratio (Figure A, B). Significant increases in ejection fraction and left ventricular wall thickness were also observed in eKO mice (Figure A, B). Mice with endothelial depletion of PHD2 developed pulmonary arterial hypertension and right ventricular hypertrophy at about 2 months³. However, we didn’t observe such changes in adult eKO mice within 4 weeks tamoxifen treatment (data not shown). To examine whether CM proliferation contributes to the cardiac growth in eKO mice, we estimated the CM number following collagenase digestion of the whole hearts. The total CM number per heart was significantly increased in adult eKO mice (Figure C). In addition, we observed significant increases in Ki67 and aurora B kinase positive CMs in adult eKO mice (Figure D, E). We then sequenced mRNAs of the whole hearts. Using the DESeq2 methodology, with a selection criteria comprising of FDR-adjusted P values of <0.05 and linear fold change >=2x or <=0.5x, there were genes either upregulated (587) or downregulated (336) in eKO hearts compared to control hearts (data not shown). Gene set enrichment analysis indicated that genes involved in CM proliferation and cell cycle regulation were significantly upregulated in eKO hearts (Figure F). Interestingly, YAP/TAZ target genes, which play crucial roles in CM proliferation⁴, were also significantly upregulated in eKO hearts (Figure F). We confirmed the upregulation of YAP target genes involved in cell cycle control such as Ccne2 and Bub1 by RT-PCR (data not shown). Consistently, we observed a dramatic decrease in YAP phosphorylation and increase of active YAP accumulation in CM nuclei in eKO hearts (Figure G, H). Taken together, our data suggest that CM proliferation contributes to the heart growth in eKO mice.

To determine whether increased CM proliferation can prevent MI-induced heart failure, we permanently ligated the proximal left anterior descending (LAD) coronary artery to induce MI. Echocardiography at 3 weeks after MI revealed that eKO mice demonstrated substantially improved cardiac function with an average EF of 77.4%, which were similar to those of eKO shams and significantly higher than those of WT MI mice (Figure I). In addition, left ventricular dilation was completely blocked in eKO mice (Figure J). Finally, histological analysis revealed that eKO hearts had significantly smaller fibrotic scars than control hearts (Figure K), suggesting that increased CM proliferation enhances cardiac function and prevents ventricular failure in eKO mice after MI.

Here we report that endothelium-specific depletion of PHD2/3 stimulates CM proliferation in both neonatal and adult mice. Our data reveal an unexpected role of PHDs-mediated endothelial mechanism in CM proliferation. Current cardiac regeneration therapies have been focused on cell or gene-based approaches. However, the engraftment of transplanted cells or safe and efficient delivery of proper genes to the specific cell types in the heart remains challenging. The EC-mediated paracrine effect on CM proliferation provides us a novel avenue for seeking a safe and pharmacological treatment for heart failure and acute MI. PHDs belong to the superfamily of Fe²⁺, 2-oxoglutarate-dependent dioxygenases and are widely considered as cellular oxygen sensors¹. PHD inhibition would be an attractive therapeutic approach for heart regeneration. However, global depletion of PHD2 and 3 leads to dilated cardiomyopathy and exacerbates cardiac injury induced by chronic adrenergic stress⁵. In addition, PHD-HIF pathway plays key roles in hypoxia-induced pulmonary hypertension³. Therefore, tissue or cell type-specific inhibition of PHD needs to be considered prior to clinical applications. Nevertheless, identification of the active components secreted from ECs for heart regeneration would be of enormous importance for future studies.