C3orf58, a novel paracrine protein, stimulates cardiomyocyte cell-cycle progression through the PI3K-AKT-CDK7 pathway

Abstract

Rationale: The regenerative capacity of the heart is markedly diminished shortly after birth, coinciding with overall withdrawal of cardiomyocytes from cell cycle. Consequently, the adult mammalian heart has limited capacity to regenerate after injury. The discovery of factors that can induce cardiomyocyte proliferation is, therefore, of high interest and has been the focus of extensive investigation throughout the past years.

Objective: We have recently identified C3orf58 as a novel hypoxia and Akt induced stem cell factor (HASF) secreted from mesenchymal stem cells, which can promote cardiac repair through cytoprotective mechanisms. Here, we tested the hypothesis that HASF can also contribute to cardiac regeneration by stimulating cardiomyocyte division and proliferation.

Methods and results: Neonatal ventricular cardiomyocytes were stimulated in culture for 7 days with purified recombinant HASF protein. Compared with control untreated cells, HASF-treated neonatal cardiomyocytes exhibited 60% increase in DNA synthesis as measured by bromodeoxyuridine incorporation. These results were confirmed by immunofluorescence confocal microscopy showing a 50% to 100% increase in the number of cardiomyocytes in the mitotic and cytokinesis phases. Importantly, in vivo cardiac overexpression of HASF in a transgenic mouse model resulted in enhanced level of DNA synthesis and cytokinesis in neonatal and adult cardiomyocytes. These proliferative effects were modulated by a phosphoinositide 3-kinase-protein kinase B-cycle-dependent kinase 7 pathway as revealed by the use of phosphoinositide 3-kinase -pathway-specific inhibitors and silencing of the Cdk7 gene.

Conclusions: Our studies support the hypothesis that HASF induces cardiomyocyte proliferation via a phosphoinositide 3-kinase-protein kinase B-cycle-dependent kinase 7 pathway. The implications of this finding may be significant for cardiac regeneration biology and therapeutics.

Keywords: myocytes, cardiac; paracrine; proliferation; regeneration.

Figure 1. HASF increases DNA synthesis in neonatal cardiomyocytes in vitro

A) BrdU uptake in neonatal cardiomyocytes as measured by ELISA. FGF1 (5.8 nM) was used as positive control. B) Representative images (20×) of BrdU⁺ cardiomyocytes without and with HASF treatment visualized by confocal microscopy. Magnification of insets is shown on the right. C) Quantification of BrdU⁺cTnT⁺/cTnT⁺ neonatal cardiomyocytes after 7 days of HASF treatment. cTnT (red), BrdU (green) and nuclear DAPI (blue), Unt; untreated. *P<0.05 vs. untreated. NS: non-significant. Representative data from 3–4 sets of experiments are shown.

Figure 2. HASF induces cell cycle re-entry and cytokinesis in neonatal cardiomyocyte in vitro

A) Representative FACS gated plots and (B) quantitative analysis of BrdU⁺7AAD⁺ control or HASF treated neonatal cardiomyocytes. C) Representative image of Ki67⁺ stained cardiomyocytes. D) Quantification of Ki67⁺ cTnT⁺/cTnT⁺ neonatal cardiomyocytes after 7 days of HASF treatment as evaluated by confocal microscopy. 1500–2000 cells were counted from each experimental group. FGF1(5.8nM) was used as positive control, *P<0.05 vs. untreated. E) Upper panel, representative images of neonatal cardiomyocytes showing Aurora B in metaphase, anaphase and cytokinesis. Lower panel, representative images of untreated and HASF treated neonatal cardiomyocytes stained for Aurora B and visualized by confocal microscopy. Magnified inset in untreated cells shows condensed nuclear AuroraB. Magnified inset in HASF-treated shows AuroraB staining during cytokinesis (anaphase). Aurora B (red), cTnT (green) and nuclear DAPI (Blue). White arrows point to AuroraB. 4000–8000 cells were counted from each experimental group. F) Bar graph depicts quantification of Aurora B staining in dividing (cytokinesis) neonatal cardiomyocytes. *P<0.05 vs. untreated. For all images: image magnifications 40×, scale bar: 50 µm. Unt; untreated. Representative data of 3–4 sets of experiments are shown.

Figure 3. HASF increases DNA synthesis and proliferation in neonatal cardiomyocyte in vivo

A) Representative tiled images of WT and HASF transgenic (Tg) 3 day old neonatal mice heart tissue stained for BrdU (red) and cTnT (green) and laminin (white) visualized by confocal microscopy. B) Representative tiled image of 3 day old neonatal mice heart tissue stained for H3P (red), cTnT (green) and nuclei (blue) visualized by confocal microscopy. C and D) Quantification of A and B respectively. 30–60 HPF images/heart were captured and analyzed. Tiled images magnification is 40× and scale bar=50 µm with magnified insets showing nuclear BrdU (A) and H3P (B) staining. n=4, *P<0.05 vs. WT.

Figure 4. HASF induces DNA synthesis in adult cardiomyocyte in vivo

A) Representative image of adult cardiac tissue in the peri-infarct area 2 weeks post MI visualized by confocal microscopy (40×) showing proliferating cTnT+ cardiomyocytes (white arrows) and proliferating cTnTNeg non-cardiomyocyte (yellow arrows). An example of a proliferating cardiomyocyte (boxed area) is presented in the magnified inset at the bottom. EdU (red), cTnT (green) and laminin (blue). B) Quantification of total EdU⁺ and/or BrdU⁺ proliferating cTnT⁺ cardiomyocytes. MI animals N=7, sham animals N=2–3. ^‡P≤0.05 for ^‡Tg vs. WT for effect of treatment, and * P≤0.05 for Tg-MI vs. WT-MI as analyzed with 2-way Anova and Bonferonni post-hoc test. 48–60 HPF images/heart were analyzed. Scale bar: 50 µm.

Figure 5. The proliferative effect of HASF in is mediated via the PI3K-AKT pathway

A) Phosphorylation of AKT (T308 and S473) by HASF treatment at various time points in treated neonatal cardiomyocytes as analyzed by Western-blot. B) This effect is blocked by the presence of PI3K inhibitors, LY294002 (LY, 50µM) and Wortmannin (Wort. 1µM). C) BrdU uptake was measured in neonatal cardiomyocytes in the presence of PI3K inhibitors LY294002 (10µM), Wortmannin (0.2 µM) by ELISA. Insulin (10µM) was used as a positive control. *P<0.05 vs. untreated. Unt; untreated. Data are representative of 3–4 sets of experiments.

Figure 6. HASF-mediated proliferation in neonatal cardiomyocytes is modulated by CDK7

A) Quantification of RNA expression level of cyclin dependent kinases (CDKs) analyzed by qRT-PCR. B) Protein expression level of CDK7 and cyclinH after 7-days of HASF treatment in neonatal cardiomyocytes as shown by western-blot. The HASF effect is blocked in the presence of (C) PI3K inhibitor (LY294002, 10 µM). D) DNA synthesis assessed by Elisa for BrdU uptake in neonatal cardiomyocytes subjected to Cdk7 gene suppression with siRNA treatment. Unt: untreated, NonTarget: non targeting negative siRNA control and siCDK7: siRNA against the Cdk7 gene.