DNA Damage Response Mediates Pressure Overload-Induced Cardiomyocyte Hypertrophy

Cardiac hypertrophy is a common physiologic response to cardiac stress that is characterized by increased thickness of the heart muscle. Maladaptive cardiac hypertrophy is often associated with diseases such as hypertension, valvular disease and ischemic heart disease, all of which can result in heart failure. Despite the large body of literature outlining mechanisms of cardiac hypertrophy, novel mechanisms continue to be identified, highlighting the complexity of this phenotype.

For a brief period of time after birth, mammalian cardiomyocytes possess the capacity to recover heart function after substantial cardiac injury through proliferation of pre-existing cardiomyocytes, after which most cardiomyocytes permanently exit the cell cycle, and cardiac growth occurs through hypertrophy rather than hyperplasia. Our group previously showed that DNA damage response (DDR) is an important mediator of postnatal cell cycle arrest in cardiomyocytes. However, whether DDR also regulates hypertrophic growth of cardiomyocytes remains unknown.¹

Therefore, we conducted the current study to examine the role of DDR in regulation of cardiomyocyte hypertrophy in response to pressure overload. First, we used angiotensin II (Ang II)-infused mice as a model for pressure overload-induced cardiomyocyte hypertrophy. All animal studies were conducted according to institutional guidelines following approval of the corresponding animal protocols. Infusion of Ang II for 2 weeks (Angiotensin II, Sigma; 1000ng/kg/minute, alzet osmotic pump) in adult C57/B6 mice caused cardiac hypertrophy and significantly induced DNA double-strand breaks as indicated by increased phosphorylated histone H2AX (Ser139) (γ-H2AX) foci. This resulted in a significant increase in the number of phosphorylated ATM (Ser1987) foci, the primary kinase in the DDR pathway (Figure A, Top left and middle panels). Similar findings were observed in a second mouse model of pressure overload achieved through transverse aortic constriction (TAC) (Figure A, Bottom left and middle panels-Bar graphs). Importantly, we found that TAC resulted in a marked increase in phosphorylated DNA-PKcs (Thr2609) in cardiomyocyte nuclei (Figure A, Right panels). While activation of ATM can occur in a variety of types of DNA damage, activation of DNA-PKcs, is specifically induced by DNA double strand breaks (DSBs)². These results indicate that pressure overload induces DSBs in cardiomyocytes.

Figure. Inhibition of ATM kinase signaling repressed pressure overload-induced cardiomyocyte hypertrophy.

A. Left panels: Representative confocal images with anti-γ-H2AX and anti-cardiac troponin T antibodies show significantly increased number of γ-H2AX foci in Ang II cardiomyocyte nuclei. Middle Panels: Representative confocal images with anti-pATM and anti-a-actinin antibodies show a significantly increased number of pATM foci in Ang II cardiomyocyte nuclei. Right Panels: Representative confocal images with anti pDNA-PKcs (phosphorylated DNA-dependent protein kinase, catalytic subunit) (Thr2609) and anti-a-actinin antibodies show a significantly increased number of pDNA-PKcs (phosphorylated DNA-dependent protein kinase, catalytic subunit) foci in TAC cardiomyocyte nuclei. Each graph shows the quantification of γ-H2AX, pATM and pDNA-PKcs (phosphorylated DNA-dependent protein kinase, catalytic subunit) foci in cardiomyocyte nuclei in both Ang II mice and TAC-operated mice (15 cells were quantified each, n=3 each). B. KU60019 injected Ang II mice show lower heart weight to tibial length ratio and smaller cardiomyocyte size as assessed by wheat germ agglutinin (WGA) staining compared to DMSO injected Ang II mice (2-month-olds male C57/B6J, n=3 each). C. KU60019 injected TAC-operated mice show a lower heart weight to tibial length ratio and smaller cardiomyocyte size as assessed by WGA staining comparison to DMSO injected TAC-operated mice (2-month-olds male C57/B6J, n=3 each). D. ATM KO mice have a lower heart weight to tibial length ratio and smaller cardiomyocyte size compared to control mice after three weeks of TAC operation (2-month-olds male, n=8 for sham, n=6 for aMHC-MerCreMer, n=8 for aMHC-MerCreMer;ATM^f/f). E. Western blot and quantitative analysis show a reduction in markers of cardiac hypertrophy in the ATM KO than control (n=3 each). GAPDH is used to normalize sample loading. F. Expression of Cain is significantly increased in the ATM KO with TAC by immunostaining images of anti-Cain antibody. G. Schematic of the suggested signaling pathway. Activated ATM caused by DNA DSB accelerates degradation of Cain resulting in Calcineurin activation. ATM also phosphorylates 4E-BP1 to promote dissociation of the 4E-BP1/eIF-4E (eukaryotic translation initiation factor 4E) complex. Free eIF-4E (eukaryotic translation initiation factor 4E) initiates translation and increases protein synthesis.

Scale bars represent 10um for low magnification views and 5um for high magnification views in Figure A and 1mm for H and E and 50um for WGA staining in Figure B, C and D and 10um in Figure F. Data are presented as mean ± s.e.m. Student’s t-test was used to determine statistical significance. *P<0.05, **P<0.01.

To examine whether DDR mediates cardiomyocyte hypertrophy, we performed daily injections of KU60019 (20mg/Kg/day), a selective inhibitor of the ATM kinase, during angiotensin II infusion for 2 weeks. This resulted in a significant decrease in cardiac mass, and cardiomyocyte size in the treated group as compared with the control group (Figure B). Similar results were also obtained using KU60019 in a TAC model of pressure overload-induced hypertrophy (Figure C).

In addition, to examine whether genetic deletion of ATM in cardiomyocytes similarly inhibits pressure overload-induced cardiomyocyte hypertrophy, we generated a cardiomyocyte-specific, conditional ATM knockout mouse. The ATM KO (ɑMHC-MerCreMer;ATM^f/f) was produced by crossbreeding between A1cf^{Tg(Myh6-Cre/Esr1*)1Jmk}/J (ɑMHC-MerCreMer) and ATM^tm2.1Fwa/J (ATM^f/f), (obtained from the Jackson Laboratory). Similar to the previous results with pharmacologic ATM inhibition, we found that the genetic deletion of ATM prevented TAC-induced cardiomyocyte hypertrophy (Figure D).

Quantitative analysis of cardiac hypertrophy pathways by Western blot showed a significant reduction of both calcineurin and Rcan 1.4 (a marker of calcineurin activation) levels in the ATM KO hearts. We also found that the abundance of phosphorylated 4E-BP1 (also known as Phas-1) was significantly decreased in the ATM KO hearts. Phosphorylation of 4E-BP1 (an inhibitor of mRNA translation) results in its inhibition and induction of cardiomyocyte hypertrophy (Figure E). We did not observe significant differences in other hypertrophy pathways tested (Figure E). Interestingly, we found that the abundance of cain, a calcineurin inhibitor protein, was significantly increased in the ATM KO hearts (Figure F).

To further examine how ATM regulates cain and 4E-BP1, we explored whether they include conserved SCD (S/T-Q cluster domains), which are putative targets for ATM. Intriguingly, we found that both Cain and 4E-BP1 have been previously shown to be direct targets of ATM^{3, 4} While our results implicate calcineurin and 4E-BP1 as mediators of cardiomyocyte hypertrophy downstream of ATM, there may be other ATM and DDR targets involved in regulation of cardiomyocyte hypertrophy following DNA damage.

It is important to note that while a recent elegant study showed that DNA single strand breaks (SSBs) occur during pressure overload-induced cardiomyocyte hypertrophy and heart failure, the investigators reported no induction of DSB. This discrepancy may have occurred due to the lower sensitivity of COMET assays used in that study for detection of DSB.⁵ As outlined earlier, the combination of ATM and DNA-PKcs activation in cardiomyocytes, as we report here, is indicative of DSB. Collectively, there is a growing body of evidence suggesting that DDR is an important regulator of both physiological and pathological cardiomyocyte growth.

In summary, we demonstrate that pressure overload induces DNA DSBs in cardiomyocytes, resulting in activation of ATM. Importantly; our results suggest that disruption of DDR through pharmacologic or genetic loss of ATM function can modulate pressure overload-induced cardiomyocyte hypertrophy.