Editor's evaluation: Myofibroblast senescence promotes arrhythmogenic remodeling in the aged infarcted rabbit heart

Abstract

Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Methods Data availability References Decision letter Author response Article and author information Metrics Abstract Progressive tissue remodeling after myocardial infarction (MI) promotes cardiac arrhythmias. This process is well studied in young animals, but little is known about pro-arrhythmic changes in aged animals. Senescent cells accumulate with age and accelerate age-associated diseases. Senescent cells interfere with cardiac function and outcome post-MI with age, but studies have not been performed in larger animals, and the mechanisms are unknown. Specifically, age-associated changes in timecourse of senescence and related changes in inflammation and fibrosis are not well understood. Additionally, the cellular and systemic role of senescence and its inflammatory milieu in influencing arrhythmogenesis with age is not clear, particularly in large animal models with cardiac electrophysiology more similar to humans than previously studied animal models. Here, we investigated the role of senescence in regulating inflammation, fibrosis, and arrhythmogenesis in young and aged infarcted rabbits. Aged rabbits exhibited increased peri-procedural mortality and arrhythmogenic electrophysiological remodeling at the infarct border zone (IBZ) compared to young rabbits. Studies of the aged infarct zone revealed persistent myofibroblast senescence and increased inflammatory signaling over a 12-week timecourse. Senescent IBZ myofibroblasts in aged rabbits appear to be coupled to myocytes, and our computational modeling showed that senescent myofibroblast-cardiomyocyte coupling prolongs action potential duration (APD) and facilitates conduction block permissive of arrhythmias. Aged infarcted human ventricles show levels of senescence consistent with aged rabbits, and senescent myofibroblasts also couple to IBZ myocytes. Our findings suggest that therapeutic interventions targeting senescent cells may mitigate arrhythmias post-MI with age. Editor's evaluation This study describes important results and convincing evidence linking myofibroblast senescence in the aged heart with a pro-arrhythmogenic phenotype. This is in turn related to higher mortality after myocardial infarction in the aged rabbit heart. These constitute important empiric as opposed to detailed findings. They nevertheless will be of interest to clinician scientists studying cardiac function and disease. https://doi.org/10.7554/eLife.84088.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Sudden cardiac death (SCD) remains the leading cause of death worldwide and is responsible for up to 20% of all deaths in the USA (Benjamin et al., 2019; Deo and Albert, 2012). The incidence of SCD increases exponentially with age, which poses a growing problem as the portion of the US population 65 years or older is estimated to increase to 15% by 2030 (Benjamin et al., 2019; Partridge et al., 2018). The vast majority of SCDs are caused by ventricular tachycardia/ventricular fibrillation (VT/VF) subsequent to myocardial infarction (MI) (Kolettis, 2013). Immediate clinical reperfusion interventions can limit the extent of acute ischemic injury and acute arrhythmic mortality post-MI, however progressive tissue remodeling in the days-to-years post-MI can establish a substrate and trigger for potentially lethal arrhythmias (Mendonca Costa et al., 2018). This tissue remodeling process is well described in young animals (van den Borne et al., 2010), but less is known about what changes to this process occur with age and how these changes increase risk of arrhythmia initiation and propagation. We have previously characterized the efficacy of the New Zealand White rabbit as a model of the aging human heart. We showed that the aging rabbit heart recapitulates cardiac dysfunctions observed in aging humans, including aortic stiffening, reduced ventricular compliance, increased interstitial fibrosis, abnormal conduction in the His-Purkinje system, and lower thresholds for induced ventricular arrhythmias (Cooper et al., 2012). We also demonstrated defects in autophagy and mitochondrial function in isolated aged rabbit myocytes resulting in increased mitochondrial reactive oxygen species which oxidized the ryanodine receptor and promoted spontaneous Ca2+ release (Cooper et al., 2013; Murphy et al., 2019). To better investigate the interaction between post-MI tissue remodeling and arrhythmogenesis with age, we have established a minimally invasive surgical procedure to reproducibly infarct the rabbit heart regardless of age or variable coronary anatomy, whereby an embolic coil occludes the left coronary artery (Morrissey et al., 2017). Infarction by this method results in electrophysiological remodeling characteristic of human MI, including functional reentry anchored at the infarct border zone (IBZ), reduction in ejection fraction, and lower thresholds for monomorphic VT induced by programmed stimulation (Morrissey et al., 2017; Ziv et al., 2012). Cellular senescence is a stress response characterized by irreversible proliferation arrest, resistance to apoptosis, and secretion of a collection of inflammatory cytokines, growth factors, and proteases termed the senescence-associated secretory phenotype (SASP) (Coppé et al., 2010; Rodier et al., 2009; Campisi and Robert, 2014). Transient senescence plays beneficial roles in young animals in wound healing by limiting fibrosis, in tumor suppression, and in development (Wang et al., 2011; Storer et al., 2013; Demaria et al., 2014). However, senescent cells accumulate with age in many tissues and accelerate many aging-associated pathologies in a paracrine manner via SASP-mediated chronic inflammatory signaling and/or in a juxtacrine manner via transfer of intracellular proteins to neighboring cells (Prata et al., 2018; Vicente et al., 2016; Burton and Krizhanovsky, 2014; Shibamoto et al., 2019; Kirschner et al., 2020; Biran et al., 2015). Genetic or pharmacological elimination of senescent cells with age slows the onset and progression of many aging pathologies, including atherosclerosis, diabetes, idiopathic pulmonary fibrosis, osteoarthritis, and neurodegenerative diseases (Baker et al., 2016; Baker et al., 2011; Childs et al., 2015; Zhu et al., 2015; Chang et al., 2016; Kim and Kim, 2019; Kirkland et al., 2017; van Deursen, 2019; Kirkland and Tchkonia, 2017). In the adult infarcted wild-type mouse heart, senescent cells arise in the scar and IBZ around day 3 post-MI and are largely cleared by day 7 (Shibamoto et al., 2019; Zhu et al., 2013). Most of these senescent cells are cardiac myofibroblasts, the main cell type responsible for secreting extracellular matrix components forming the scar. This observation is consistent with epidermal wound studies in mice, suggesting a role for transient myofibroblast senescence as a normal part of the wound healing process in limiting fibrosis (Demaria et al., 2014). Infarction of adult constitutive p53 KO mice, in which senescence induction is impeded, results in increased fibrosis and decreased inflammation in the scar 7 days post-MI (Zhu et al., 2013). In a cell culture model of isolated mouse cardiac fibroblasts, senescent cardiac fibroblasts limit the proliferation of non-senescent cardiac fibroblasts in a juxtacrine but not paracrine fashion (Shibamoto et al., 2019). These results suggest a beneficial role for transient myofibroblast senescence in limiting fibrosis post-MI. Conversely, insights toward a pathological role of chronic cardiac senescence come from aging mouse models of MI. Treatment of aged mice with the senolytic drug navitoclax (i.e. a drug that specifically eliminates senescent cells) before or after MI significantly reduces senescence burden, improves cardiac function, mitigates cardiac remodeling, and reduces scar size post-MI, although the effects on arrhythmias has not been studied to our knowledge (Dookun et al., 2020). These results suggest that acute induction of senescence followed by timely clearance limits excess scarring, but persistence of cardiac senescence as seen with age can promote cardiac fibrogenesis and worsen cardiac function and might lead to increased risk of arrhythmias. Although these findings in mice are important, the mechanisms underlying a relationship between age-associated senescence and arrhythmogenesis is not well understood, particularly in an animal model like the rabbit whose cardiac electrophysiology is more functionally relevant to humans than that of mice. Potential pro-arrhythmic effects of senescent cells could occur through a paracrine fashion via SASP components. Indeed, treatment of isolated rat and mouse cardiomyocytes with exogenous SASP components including IL-6, IL-1β, and TNF-α have pro-arrhythmic effects in ion channel remodeling (Francis Stuart et al., 2016; Aromolaran et al., 2018). Senescent myofibroblasts might be able to interfere with cardiomyocyte electrophysiology from relatively long distances through chronic inflammatory signaling. Alternatively, the direct cell-cell coupling via gap junctions of senescent myofibroblasts to cardiomyocytes at the IBZ might alter their electrophysiology more than coupling with non-senescent myofibroblasts. Such a process would establish regional heterogeneities in ion channel activity, action potential duration (APD), and other electrophysiological factors anchored at the IBZ that would permit reentrant current and therefore VT/VF. We hypothesized that aged infarcted rabbits experience a persistence of senescent myofibroblasts in the scar and IBZ compared to young, and that these senescent myofibroblasts act in a paracrine or juxtacrine fashion to induce pro-arrhythmic remodeling in IBZ cardiomyocytes. Here, we demonstrate that aged rabbits compared to young exhibit increased peri-procedural deaths mostly due to VT/VF, consistent with prolongation of APD and alternans at the IBZ associated with higher frequency VF. We observed no difference in the size of the scar or IBZ geometry over the first 12 weeks post-MI between young and aged rabbits. However, whereas senescent cells in the young rabbit scar were largely cleared by 3 weeks post-MI, senescence of mostly myofibroblasts remained high in the aged rabbits up to 12 weeks post-MI and correlated with increased local expression of inflammatory cytokines. In ventricular tissue samples from aged infarcted human patients, we observed elevated levels of senescence markers, and most senescent cells appeared to be myofibroblasts. Using primary adult rabbit cardiac fibroblasts, we induced senescence via treatment with the drug etoposide. With this method, we did not observe any effect of exogenous or co-cultured conditioned media from senescent cardiac fibroblasts on the IKr current or APD of treated rabbit myocytes compared to conditioned media from proliferating cardiac fibroblasts. However, in both aged rabbit and human cardiac tissue, we observed senescent cardiac myofibroblasts in close proximity to IBZ myocytes, with the gap junction protein Cx43 appearing to couple the cells. Previous studies in rabbits have demonstrated a role for Cx43 in the coupling of myocytes and fibroblasts as evidenced by dye transfer and optogenetic studies (Camelliti et al., 2004a; Camelliti et al., 2004b; Schultz et al., 2019). Based on our electrophysiological and volumetric measurements of senescent and non-senescent fibroblasts, we performed computational modeling which suggested that coupling of myocytes with senescent cardiac myofibroblasts would result in prolongation of APD and lower thresholds for conduction block. Altogether, our results indicate a pathological role for the persistence of senescent cardiac myofibroblasts in directly exacerbating arrhythmogenesis at least via a direct cell-cell interaction through gap junctions with age post-MI. These findings suggest that targeted elimination of senescent cells post-MI could be an effective therapeutic method to mitigate senescence burden and combat arrhythmias in aged infarcted individuals. Results Aged infarcted rabbits exhibit increased incidence of peri-procedural arrhythmias A total of 73 young (≤1 year) and 99 aged (≥4 years) female New Zealand White rabbits were subjected to minimally invasive coil embolization of the left coronary artery to induce MI of the apical left ventricular free wall (Figure 1). The average age for young rabbits was 7.8 months and for aged rabbits was 4.8 years. Aged rabbits exhibited significantly higher peri-procedural mortality compared to young rabbits (36% of aged rabbits versus 19% of young rabbits, p=0.0120), with most young and aged peri-procedural deaths attributed to VT/VF. We observed no significant difference in SCD in the acute (3–72 hr) or chronic (≥72 hr) periods post-MI between young and aged rabbits. Aged rabbits exhibited a significantly higher incidence of peri-procedural lethal or nonlethal VT/VF (44% of aged rabbits versus 23% of young rabbits, p=0.0040) (Figure 1C). Kaplan-Meyer analysis shows aged rabbits had a significantly decreased overall survival post-MI compared to young (p=0.0285), and all mortality of young and aged rabbits occurred within the first 48 hr post-MI (Figure 1D). Of rabbits that initiated peri-procedural VT/VF, we did not find any significant difference in the number or success rate of defibrillation attempts to restore sinus rhythm (p=0.38 and p=0.54, respectively, data not shown). Overall, our findings indicate the aged rabbit heart is more susceptible to acute, potentially lethal ischemic arrhythmias compared to young rabbits. These trends correlate with the age-associated increase in mortality of out-of-hospital first-time acute MI in humans (Benjamin et al., 2019). Figure 1 Download asset Open asset Aged rabbits exhibit increased incidence of peri-procedural arrhythmias. (A) Survival table of young and aged infarcted rabbits. Procedural deaths were defined as death occurring within the first 3 hr of surgery. (B) Histogram of rabbits included in the study by age at the time of surgery. (C) Incidence of procedural ventricular tachycardia/ventricular fibrillation (VT/VF) in young and aged infarcted rabbits. Numbers inside bars are number of rabbits. *p<0.05, two-tailed exact test. (D) Survival curves of young and aged rabbits post-MI (p<0.05, log rank test). Figure 1—source data 1 Raw data pertaining to rabbits used in the study that was used to create Figure 1. https://cdn.elifesciences.org/articles/84088/elife-84088-fig1-data1-v2.xlsx Download elife-84088-fig1-data1-v2.xlsx Progression of infarct size, IBZ geometry, and fibrosis is consistent between young and aged rabbits Long-term progressive electrophysiological tissue remodeling at the IBZ can facilitate the initiation and propagation of arrhythmias post-MI even after the ischemic event, increasing risk of SCD (Axford-Gatley and Wilson, 1988). Larger scars and increased IBZ fibrosis can establish a source-sink mismatch and regional heterogeneities in APD and conduction velocity which underlie VT/VF post-MI (Richardson et al., 2015; Neuschl et al., 2018). Additionally, progressive infarct expansion can result in declining cardiac function and increased risk of ventricular wall rupture or VT/VF (Richardson and Holmes, 2015; Zhang et al., 2014). However, the cellular mechanisms underlying age-associated differences in post-MI tissue remodeling and their arrhythmogenic consequences are not well understood. We previously demonstrated that coil embolization of the young and aged rabbit left coronary artery resulted in similar scar size 3 weeks post-MI (Morrissey et al., 2017). Here, we further investigated potential age-associated changes in the dynamics of scar size, shape, and fibrosis throughout a 12-week timecourse post-MI. Cardiac tissue was harvested from young and aged infarcted rabbits at 1, 2, and 3 weeks post-MI to encompass the period following scar formation, and at 12 weeks post-MI to interrogate potential long-term changes. To assess scar size, we measured the percent infarcted area from epicardial left ventricular free wall dissections via gross anatomic photos. We found no significant difference in the percent infarcted area of the left ventricle between young and aged rabbits at any timepoint (Figure 2A). We then compared the functional area of the IBZ between young and aged infarcted rabbits as represented by protrusions of the scar into the surviving myocardia, since IBZ geometry can affect regional heterogeneities in conduction velocity and APD. We utilized hematoxylin and eosin (H&E)-stained frozen left ventricle sections from young and aged infarcted rabbit hearts, laid flat along the circumferential-longitudinal plane prior to freezing. To better characterize the 3D structure of the IBZ from each heart, at least five slides over at least a 500 μm endocardial-to-epicardial span per rabbit were analyzed. We found no significant difference in either the number or length of protrusions between young and aged rabbits at any timepoint post-MI or over time for each age group (Figure 2B). To assess potential age-associated changes in fibrosis post-MI, we analyzed frozen left ventricular sections stained with Masson’s trichrome. We observed no difference at any timepoint between young and aged rabbits post-MI in either percent fibrotic area of the entire left ventricular free wall or in percent interstitial fibrotic area in the remote zone (RZ) (Figure 2C–D). Overall, we observed no significant age-associated differences in the dynamics of scar size, IBZ geometry, or fibrosis in the first 12 weeks post-MI, suggesting the physical properties of the scar are not sufficient to explain age-associated differences in arrhythmogenesis post-MI. Figure 2 Download asset Open asset Progression of infarct size, infarct border zone (IBZ) geometry, and fibrosis is consistent between young and aged rabbits. (A) Left: Whole rabbit heart, with infarct zone outlined. Middle: Dissected left ventricular free wall. Right: Quantification of percent infarcted area of the left ventricular free wall. (B) Left: Representative hematoxylin and eosin (H&E)-stained images of IBZ (left portion of images) protrusions into surviving myocardia (right portion of images). Right: Quantification of number (left) and length (right) of protrusions. (C) Left: Representative Masson’s trichrome-stained left ventricular section. Right: Quantification of percent fibrotic area. (D) Left: Representative Masson’s trichrome-stained remote zone images showing interstitial fibrosis. Right: Quantification of % fibrotic interstitial area. Dots represent average data for each rabbit, error bars SEM. Figure 2—source data 1 Raw data used to create Figure 2. Each tab contains raw data for the corresponding panel of Figure 2. https://cdn.elifesciences.org/articles/84088/elife-84088-fig2-data1-v2.xlsx Download elife-84088-fig2-data1-v2.xlsx Electrophysiological remodeling in the IBZ of aged rabbit hearts Although aging is associated with changes in numerous biological processes potentially impacting infarct healing, less is known about the role of age-associated changes in post-MI electrophysiological tissue remodeling and how such changes might contribute to differential risk of lethal arrhythmias. We hypothesized that more severe electrophysiological remodeling in the aged IBZ compared to young establishes a greater arrhythmic substrate in aged rabbit hearts. To test this hypothesis, we investigated action potential (AP) dynamics in young (n=17) and aged (n=5) rabbit hearts ex vivo at 3 weeks post-MI using optical mapping. The epicardial APDs were recorded using the fluorescent voltage-sensitive dye di-4-ANEPPS and averaged from multiple randomized 4 cm2 regions of the scar, IBZ, and RZ during 350 ms cycle length stimulation (Figure 3A). Representative AP traces and APD maps from young and aged rabbit hearts are shown in Figure 3B. Figure 3 Download asset Open asset The border zone of aged rabbits shows action potential duration (APD) prolongation, APD alternans, and faster ventricular fibrillation (VF) frequency. (A) Photograph of a rabbit heart showing the infarct zone (IZ), infarct border zone (IBZ), and remote zone (RZ). White line indicates a representative plane along which alternans images in (C) were recorded. (B) Top: Representative action potential (AP) traces from the RZ and IBZ of young and aged rabbits. Middle: Representative APD maps from young and aged rabbits, showing APD at different points along the border zone. Bottom: Quantification of APD and conduction velocity in the RZ and IBZ of young and aged rabbits. *p<0.05, two-tailed exact test. (C) Representative AP trace of an aged infarcted rabbit at 3 weeks post-MI showing alternans in the IBZ. (D) Top: Representative traces showing VF in young and aged rabbits after electrical induction. Bottom: Representative VF frequency maps of IBZ of young and aged rabbits. Right: Quantification of VF frequency and cycle length at which VF was induced from young and aged rabbits. *p<0.05, two-tailed exact test. Dots represent average data for each rabbit. Error bars: SEM. Figure 3—source data 1 Raw data used to create Figure 3. Tabs contain raw data corresponding to panel B and panel D. https://cdn.elifesciences.org/articles/84088/elife-84088-fig3-data1-v2.xlsx Download elife-84088-fig3-data1-v2.xlsx In aged rabbits, APD at the IBZ was significantly longer than in the RZ (183.1±6.64 ms in IBZ and 169.4±7.90 ms in RZ, p<0.05), whereas in young rabbits, APD in the IBZ was significantly shorter compared to the RZ (178.9±6.99 ms in IBZ and 202.2±6.32 ms in RZ, p<0.05). Compared to young rabbits, aged rabbits exhibited greater APD heterogeneity along the IBZ (Figure 3B). In both young and aged rabbits, conduction velocity was significantly slower in the IBZ compared to the RZ (0.32±0.10 m/s in IBZ and 0.96±0.30 m/s in RZ of aged hearts, 0.37±0.05 m/s in IBZ and 0.90±0.13 m/s in RZ of young hearts), consistent with previous findings (Mendonca Costa et al., 2018). We observed that the IBZ of aged rabbit hearts can support spatially discordant APD alternans at relatively slow heart rates, which has been associated with an increased risk for arrhythmias (Figure 3C; Liu et al., 2018). These results were not observed in the young hearts under the same conditions. To further study age-associated changes in arrhythmogenic substrate, we induced VF by burst electrical stimulations. Representative VF traces and VF frequency maps are shown in Figure 3D. Aged rabbits exhibited significantly higher dominant VF frequency compared to young, which may be related to the underlying complexity of arrhythmia. Additionally, the aged rabbit hearts displayed greater spatial variation of VF frequencies across the RZ compared to young, whereas in young rabbits slower-frequency regions tended to cluster near the IBZ. Altogether, these findings suggest that age-associated changes in electrical remodeling 3 weeks post-MI increased the heterogeneity of rate-dependent APD dynamic properties between the IBZ and RZ, notably the prolongation in APD in the IBZ of aged rabbits, which demonstrates a more arrhythmogenic substrate. Senescence of myofibroblasts is elevated and persistent in the aged rabbit heart post-MI and correlates with increased inflammation To investigate potential cellular mechanisms underlying the increase in arrhythmogenic substrate in aged infarcted rabbits, we next investigated the dynamics of cellular senescence in young and aged rabbit hearts over time post-MI. Whereas in the proliferative phase post-MI, the scar is rich with myofibroblasts secreting ECM components, in the maturation phase, the fibrogenic activity of myofibroblasts is resolved at least partially via induction of senescence (Zhu et al., 2013; Fu et al., 2018; Kanisicak et al., 2016; Prabhu and Frangogiannis, 2016; Jun and Lau, 2018). To interrogate the dynamics of senescence in young and aged infarcted rabbit hearts, we first performed senescence-associated β-galactosidase (SA-β-gal) staining of frozen left ventricular sections from young and aged rabbits over the first 12 weeks post-MI (Figure 4A–C). In the aged rabbit scar, the percent of SA-β-gal+ cells in the scar remained elevated over time even at 12 weeks post-MI. Conversely in the young rabbit scar, the percent of SA-β-gal+ cells was relatively high for the first 2 weeks and largely resolved by the third week post-MI, resulting in significantly decreased senescence at 3 and 12 weeks post-MI compared to aged rabbits (Figure 4C). In the IBZ, the same trend was observed in young and aged rabbits but did not reach statistical significance. The young and aged rabbit RZ displayed relatively very few SA-β-gal+ cells. These RZ data are consistent with our previous assessments of baseline levels of senescence in the non-infarcted young and aged rabbit myocardia, in which we observed similarly low levels of SA-β-gal (not shown). To determine if the observed age-associated difference in senescence could be explained as an artifact of the MI procedure, we compared SA-β-gal histology between young and aged rabbits 2 weeks after sham infarction, a timepoint at which SA-β-gal signal was elevated in the young and aged infarcted rabbit scar. In the young and aged sham-infarcted rabbits, we observed low levels of SA-β-gal+ cells, with no significant difference between them (Figure 4—figure supplement 1). Figure 4 with 1 supplement see all Download asset Open asset Senescence of myofibroblasts is elevated and persistent in the aged rabbit heart post-MI and correlates with increased inflammation. (A) Representative senescence-associated β-galactosidase (SA-β-gal)-stained image showing examples of infarct zone, border zone, and remote zone (RZ). (B) Representative SA-β-gal-stained images showing infarct zone, border zone, and RZ from young and aged rabbits at 3 weeks post-MI. (C) Quantification of percent SA-β-gal+ cells from the scar (top), infarct border zone (IBZ) (middle), and RZ (bottom) in young and aged rabbits at 1, 2, 3, and 12 weeks post-MI. (D) Representative confocal images of αSMA/γH2AX double immunofluorescence staining (top row) and CD31/γH2AX double immunofluorescence staining (bottom row) from young (left) and aged (right) in the infarct zone of rabbits at 12 weeks post-MI. White indicates autofluorescence and was used to avoid false positive fluorescence signal. (E) Quantification of % of nuclei with three or more γH2AX foci. (F) Quantification of percent αSMA+ cells (left) and the percent of γH2AX+ cells that are αSMA+ (right). (G) Quantification of percent CD31+ cells (left) and the percent of γH2AX+ cells that are CD31+ (right). (H) Quantification of expression of senescence and senescence-associated secretory phenotype (SASP) genes via RT-qPCR from young and aged rabbits 3 weeks post-MI. N=3 rabbits per condition. Dots represent average data for each rabbit, error bars SEM. Two-tailed exact test: *p<0.05 compared to young, # p<0.05 compared to respective RZ. Figure 4—source data 1 Raw data used to create Figure 4. Tabs contain data corresponding to panel A, panels E and F, panels E and G, and panel H, respectively. https://cdn.elifesciences.org/articles/84088/elife-84088-fig4-data1-v2.xlsx Download elife-84088-fig4-data1-v2.xlsx To further characterize senescence and determine the cellular identity of senescent cells in the young and aged rabbit scar, we performed immunofluorescence staining on frozen left ventricular sections for the DNA damage marker γH2AX as a marker of senescence (Ito et al., 2018) and either the myofibroblast marker αSMA (Tallquist, 2020) or the endothelial cell marker CD31 (Pusztaszeri et al., 2006; Figure 4D). To account for transient DNA damage in otherwise healthy cells represented by low numbers of γH2AX nuclear foci, only nuclei exhibiting three or more γH2AX foci were scored as γH2AX positive (‘γH2AX+)’. These results corroborated our SA-β-gal data as the percent of γH2AX+ nuclei in the aged rabbits remained elevated over time, whereas the percent of γH2AX+ nuclei in the young rabbit scar was resolved by 3 weeks post-MI, resulting in significant increases in the percent of γH2AX+ nuclei at 3 and 12 weeks post-MI in aged rabbits compared to young (Figure 4E). Specifically, in aged rabbits 12 weeks post-MI, we observed 9.9 ± 3% of cells were γH2AX whereas in young rabbits 12 weeks post-MI, only 0.6 ± 0.2% of cells were γH2AX+. From our αSMA immunostaining, in both the aged and young rabbit scar, we observed initially high percentages of αSMA+ cells followed by a decline over time post-MI with no significant difference between young and aged at any timepoint (Figure 4F, left). At 2 weeks post-MI, 39.5 ± 17.5% of cells were αSMA+ in aged rabbits and 36.5 ± 8.1% of cells were αSMA+ in young rabbits. This observation is consistent with previous mouse genetic lineage tracing reports showing fibroblasts that became myofibroblasts remained in the scar but lost αSMA expression by 2 weeks post-MI (Fu et al., 2018). From our double immunofluorescence staining, we observed that the majority of γH2AX+ cells were αSMA+ at 1 week post-MI in aged and young rabbits, consistent with the high levels of αSMA signal (Figure 4F, right). Interestingly, although at 2 weeks post-MI we observed an ~50% decrease in the percent of αSMA+ cells, we observed that ~80% of γH2AX+ cells were αSMA+, indicating a propensity toward senescence of myofibroblasts. Specifically, at 2 weeks post-MI, 66.9 ± 34.3% of γH2AX+ cells were αSMA+ in aged rabbits, and 66.2 ± 45.0% of γH2AX+ cells were αSMA+ in young rabbits. At 3 and