Cdk1 Deficiency Extends the Postnatal Window of Cardiomyocyte Proliferation and Restores Cardiac Function after Myocardial Infarction

Abstract

Cyclin-dependent kinase 1 (Cdk1) is a master regulator of the G2-M transition between DNA replication and cell division. This study investigates the regulation of cardiomyocyte (CM) proliferation during the early neonatal period and following ischemic injury in adult mice. We analyzed cell cycle dynamics with the assessment of DNA synthesis, and cytokinesis in murine hearts during the first 15 days after birth. A distinct proliferative block was observed at 1 day, followed by a second wave of DNA synthesis at 4 days, leading to CM binucleation (CMBN) by day 5. Genome-wide mRNA profiling revealed the differential expression of cell cycle regulatory genes during this period, with a downregulation of factors involved in cell division and mitosis. The loss of Cdk1 impaired CMBN but extended the neonatal CM proliferation window until day 10 post-birth. In adult hearts, the cardiac-specific ablation of Cdk1 triggered CM proliferation post-myocardial-infarction (MI) in specific zones, driven by the activation of EGFR1 signaling and suppression of the anti-proliferative p38 and p53 signaling. This was accompanied by restoration of fractional shortening, mitochondrial function, and decreased reactive oxygen species. Additionally, cardiac hypertrophy was mitigated, and survival rates post-MI were increased in Cdk1-knockout mice. These findings reveal a novel role of Cdk1 in regulating cell cycle exit and re-entry in differentiated CMs and offer insights into potential strategies for cardiac repair.

1. Introduction

Heart failure is the leading cause of morbidity and mortality in North America and the second leading cause of extended hospital stays [1]. The most common etiology of heart failure occurs following myocardial infarction (MI), underscoring the heart’s particular vulnerability to ischemic injury [2]. After an ischemic insult, the damaged cardiomyocytes (CMs) are replaced with fibrotic tissue [3]. This irreversible loss of CMs, coupled with the poor proliferative capacity of the surviving CMs, contributes to the development of heart failure [2]. Current treatment strategies primarily address the maladaptive neurohormonal changes but do not necessarily improve cardiac function [4]. Therefore, developing innovative approaches to restore cardiac function will not only improve patients’ quality of life but also reduce associated healthcare costs.

The regenerative capacity of the mammalian heart is lost shortly after birth when CMs transition from a proliferative to a quiescent state [5]. During the embryonic stage and up until birth, CMs undergo DNA replication, the G2/M phase, and cytokinesis, leading to CM proliferation [6,7,8,9]. Around postnatal day 5, CMs complete a final round of partial cell cycle activity, resulting in CMBN [9]. Thereafter, CMs exit the cell cycle and growth predominantly occurs through hypertrophy [6,7,8,10,11,12,13,14,15,16,17]. This transition is associated with changes in the levels of cell cycle regulators, where the expression of cell cycle inhibitors increases and cell-cycle-activating cyclin/Cdk complexes decrease [18,19,20,21,22,23,24,25,26]. Additionally, CM proliferation is tightly regulated by numerous intricate molecular pathways, including the Hippo/YAP mitogenic signaling pathway [11,27,28,29]. The progression through the cell cycle is positively regulated by Cdks and cyclins, their associated regulatory subunits [30], in a carefully orchestrated manner [31]. However, the factors controlling CM cell cycle progression during the postnatal period remain unknown.

Cdk1, also known as Cdc2, plays a central role in progressing the cell cycle. Cdk1 regulates essential processes during the G2/M phase transition and mitosis, ensuring accurate chromosome segregation and proliferation [32]. Cdk1 is activated by forming a complex with B-type cyclins in the late G2 phase [33]. In contrast, Wee1 kinase can phosphorylate and inhibit Cdk1, thereby delaying entry into mitosis [34]. Active Cdk1-cyclin B governs entry into mitosis by phosphorylating target proteins involved in chromatin condensation, nuclear envelope breakdown, and microtubule formation [33,35]. By controlling these events, Cdk1 ensures the proper execution of mitosis, safeguards the genome integrity, and maintains cell viability [36].

Based on these findings, we hypothesized that the loss of Cdk1 in murine CMs would result in proliferative abnormalities in both neonatal and adult hearts. Here, we report the unexpected finding that Cdk1 is required for normal cell cycle arrest and CMBN during the early postnatal period but is redundant for CM division. Furthermore, cardiac-specific Cdk1 deletion in adult mice promoted cardiac repair through enhanced CM proliferation post-MI, driven by the activation of proliferative pathways such as EGFR1 and the suppression of inhibitory pathways like p38 and p53. This highlights Cdk1’s critical role in regulating the balance between proliferation and differentiation in the heart.

2. Results

2.1. Cell Cycle Regulation of Cardiomyocytes during the Early Neonatal Period

We began our analysis with the assessment of heart weight/body weight (HBW) ratios in neonatal mice during the first 7 d after birth. We found that HBW did not significantly change over this period (Figure 1A). However, with the observation that mRNAs involved in DNA replication and mitosis showed a biphasic periodicity, we investigated DNA synthesis in CMs in the first 7 d after birth. Neonatal C57BL/6J wild-type (wt) mice were injected subcutaneously with a single dose of 5-ethynyl-2′-deoxyuridine (EDU), a thymidine analogue [37]. Cardiac specimens were prepared for indirect immunofluorescence microscopy employing an EDU assay, and anti-phospho-Histone Pi.H3-Ser28 (H3.Pi-S28) antibodies (Figure 1B) [38] to identify CMs in the S and M phases in the LV region, in conjunction with anti-cardiac actinin and 4′,6-diamidino-2-phenylindole (Dapi). We also co-stained neonatal cardiac sections with an antibody to pericentriolar material 1 (Pcm1), a CM-specific peri-nuclear marker [39,40], and EDU to specifically identify CMs in the S phase (Figure 1C). Using this approach, the CM labeling index was monitored from embryonic day 21 (0 d), neonatal 12 h, and 1 d through 7 d (Figure 1C). Exceedingly high labeling indexes were detected at 0 d and 12 h (Figure 1D). In contrast, levels of EDU/Pcm-1 double-positive CM nuclei dropped dramatically at 1 d (Figure 1D,E). This proliferative block was corroborated by immunofluorescence analysis of Pi.H3-Ser28-positive CMs in the M phase (Figure 1E), and aurora kinase B (AurkB)-positive CMs in cytokinesis (Figure 1F,G) [41]. Re-initiation of DNA synthesis occurred at 4 d, with a peak labeling index occurring at 5 d (49 ± 4.3%; p < 0.001 vs. 2 d). The number of EDU-positive CM nuclei decreased again markedly by 6 d (5.1 ± 2.1%; p < 0.01 vs. 2 d) (Figure 1D). There were no sex-dependent differences noted.

To ascertain that this second wave of DNA synthesis represented CMBN, we analyzed this morphological developmental change in 2-day- versus 5-day-old CMs by 3D reconstruction of confocal immunofluorescence micrographs of Dapi-stained LV cardiac sections [42]. CMBN was virtually absent in 1-day-old cardiomyocytes (Figure 1H). However, by day 5, 57 ± 9.1% of cardiomyocytes had undergone binucleation (p < 0.001 vs. 1 d). Combined, all these findings demonstrate that CM proliferation ceases in a sharply defined time window after birth and is followed by a distinct second phase of CMBN.

2.2. Differential mRNA Expression of Cell Cycle Factors in the Early Postnatal Period

To elucidate the molecular mechanisms responsible for the cessation of proliferation, DNA binucleation, and cell cycle exit, we performed genome-wide mRNA microarray profiling of LV tissue samples [43]. Unsupervised hierarchical cluster analysis at a high confidence threshold revealed that approximately 1900 individual transcripts changed significantly in murine hearts at 1 d through 15 d, relative to 0 d (Figure 2A). Three-dimensional principal component analysis (3D-PCA) was performed to compare gene expression profiles across postnatal days, revealing clustering between the 1 d, 3 d, 7 d, and 10 d samples, and a closer association between the 0 d and 5 d samples (Figure 2B). This pattern is consistent with our microscopic results, which show CM proliferation at 0 d and CM binucleation at 5 d (Figure 1D). Both processes are largely absent in the other samples. While the clustering at 0 d and 5 d may not be as tight as the other groups, the 3D-PCA clearly indicates that these samples are visually distinct from the other time points, reflecting their unique biological characteristics. Moreover, the similarity between 0 d and 5 d can be explained by the fact that CM division and binucleation are regulated by a similar set of mRNAs, which may contribute to their closer positioning in the plot (Figure 2E).

Figure 1D shows that CM division was virtually absent between 1 and 3 days after birth. To identify gene clusters with similar biological functions in CM transcriptomes undergoing binucleation, we performed a Gene Set Enrichment Analysis (GSEA) on the transcriptomes of resting CMs at 3 d postnatal compared to 0 d, which represent proliferating CMs (Figure 2C). The top enriched gene sets in the 3 d vs. 0 d comparison include pathways associated with cell cycle control, DNA replication, and regulatory checkpoints (e.g., V$E2F1DP1RB_01, DNA_REPLICATION, CELL_CYCLE, and CELL_CYCLE_CHECKPOINT_GO). These data indicate that although CMs are largely resting at 3 d, there are still significant transcript levels of genes regulating DNA replication and the cell cycle. This may represent a transitional phase where the majority of CMs have exited the active proliferative phase (as seen at 0 d) but retain a baseline readiness for cell cycle re-entry. Gene sets such as “CELL_CYCLE_CHECKPOINT_GO” and “RB_P130_DN” support this negative control, likely through checkpoint mechanisms that prevent uncontrolled proliferation. Thus, the transcriptomic profile at 3 d likely reflects a transitional state where CMs are resting but remain poised to re-enter the cell cycle if needed.

Figure 2D shows the GSEA results comparing transcriptomes at 5 d vs. 0 d, where 5 d represents CMs undergoing binucleation. The top enriched GSEA gene sets in the 5 d vs. 0 d comparison include “E2F1_DNA_UP,” “PROLIFERATION_GENES,” “MITOTIC_CELL_CYCLE_CHECK POINT,” and “CELL_CYCLE_KEGG.” The GSEA signature at 5 d emphasizes that, even as binucleation occurs, the cell cycle is tightly regulated to ensure that CMs undergo binucleation rather than complete cytokinesis. While CMs have exited the typical division cycle, they retain essential regulatory mechanisms that guide the process of binucleation, a key step in the maturation of CMs.

Combined, at 3 d, the transcriptome reflects a resting phase with active cell cycle suppression, while at 5 d, the focus shifts towards binucleation, which involves the reactivation of distinct proliferative processes, yet these are tightly controlled to avoid cell division. Our view is further corroborated by a Gene Ontology (GO) analysis, which shows that the highest-ranked GO terms include transcripts involved in the regulation of nuclear division, negative regulation of cell division, and regulation of the mitotic cell cycle (Figure 2E).

Next, we validated the transcriptomic results of well-characterized cell cycle regulatory genes by qRT-qPCR analysis (Figure 2F) [42]. We found that the mRNA levels of a set of factors involved in the transition from the S phase into mitosis (AurkB, Skp2), G1-Cdks (Cdk1, Cdk2, Cdk4), G1/G2-cyclins (cyclins A, B, E, F), and key cell cycle activators (e.g., Cdc25A) and inhibitors (Cdkn2b, Cdnkn3) were significantly downregulated by postnatal day 7–10, coinciding with terminal cell cycle exit (Figure 2F). These analyses revealed that distinct cell cycle factors are differentially regulated in the immediate postnatal period. Our transcriptomic findings were confirmed by Western blotting of CM cell cycle inhibitors, cell cycle regulators, Mapk signaling pathways, and cardiac-specific genes (Figure 2G) [42]. Among all these factors, Cdk1 was amongst the top ten enriched cell cycle genes that were selectively upregulated at 1 d and 5 d post-birth (Figure 2F). Thus, we decided to examine the potential role of Cdk1 in the regulation of neonatal CM proliferation in vivo.

2.3. Ablation of Cdk1 Expands the Proliferative Time Window in Neonatal Cardiomyocytes

Factors that control cell cycle exit and binucleation in neonatal CMs are still not well characterized [44,45]. To investigate the mitotic regulators during this transition, we generated cardiac-specific Cdk1-mutant mice (Figure 3A). Transgenic mice homozygous for the loxP-flanked (floxed) allele of Cdk1^fl/fl (controls) [46] were bred with αMyh6-Cre^+/+ transgenic mice [47] to obtain Cdk1^fl/fl;αMyh6 (Cdk1KOc) mice (Figure 3B). These mice had the Cdk1 gene deleted specifically in CMs at embryonic day E14 post-coitum (p.c.), as analyzed by Western blotting and RT-qPCR (Figure 3C,D).

Cardiac function was not altered in Cdk1KOc mice versus controls, as assessed by echocardiographic measurement of fractional shortening (FS; p > 0.01) (Figure 3E). In growing animals, heart weight/tibia length (HTL) is a more reliable indicator of cardiac hypertrophy, especially in growing animals, as body weight can fluctuate due to factors unrelated to heart size. We found normal HTL ratios at 15 days post-birth for all four experimental groups (Figure 3F), demonstrating that the Cdk1KOc mice do not have a gross cardiac phenotype during the adolescent period. HTL ratios were also not significantly different in Cdk1KOc mice at 3 months of age relative to controls (p > 0.01) (Figure 3G). Intriguingly, we observed smaller CMs in Cdk1KOc animals compared to controls at 15 d (Figure 3H) and at 3 months post-birth (Figure 3H), suggesting that Cdk1 may potentially regulate CM cell division during the early neonatal stage.

2.4. Loss of Cdk1 Prolongs the Proliferative Window of Cardiomyocytes during the Postnatal Stage

Initially, we investigated whether genetic silencing of Cdk1 affects DNA synthesis in Cdk1KOc hearts in the early postnatal period. For analysis by confocal immunofluorescence microscopy, mice were injected subcutaneously with EDU, and hearts were excised, fixed, and sectioned. Subsequently, specimens were co-stained with antibodies to α-actinin, EDU, Pi.H3-Ser28, wheat germ agglutinin (WGA), and Dapi to visualize nuclear DNA. We detected a very high EDU-labeling index in CMs from control mice at 1 d that dropped dramatically at 2 d (p < 0.001 vs. 0 d) (Figure 4A, Figures S1 and S2). This was followed by a re-induction of DNA synthesis that peaked at 5 d. A progressive decrease in the number of CMs in the S phase was observed over 7 d and 10 d, until DNA synthesis in this cell type ceased completely by 15 d. Intriguingly, the numbers of EDU-positive CMs in Cdk1KOc mice remained markedly higher on 1 d, 2 d, and 5 d when compared to the controls (p < 0.001). Subsequently, DNA synthesis in Cdk1-deficient CMs progressively decreased over 7 d–10 d and was undetectable at 15 d (Figure 4A).

To analyze the distribution of CMs in mitosis, cardiac sections were co-stained with antibodies to H3.Pi-S28, an M-phase-specific nuclear marker, in combination with α-actinin immunostaining to identify CMs. Microscopic inspection showed dramatically elevated numbers of H3.Pi-S28-positive CMs in control animals at 0 d and 5 d (Figure 4B and Figure S3). This effect was very much lower in control CMs at 7 d and undetectable at 10 d and 15 d (p < 0.01 vs. 0 d). Again, the number of CMs in mitosis remained significantly higher in Cdk1KOc mice at between 1 d to 2 d and at 7 d post-birth in comparison to the control animals (p < 0.01). In this strain, H3.Pi-S28-positive CMs were never observed at 15 d (Figure 4B). Additionally, Cdk1KOc hearts at 4 w and 6 w of age failed to show any signs of cell cycle activity (Figure S4). Inspection of non-cardiomyocytes (NCMs) during the first week after birth revealed no significant differences in the number of cycling NCMs in Cdk1KOc compared to control hearts (Figure S5). Importantly, all these cell cycle events in CMs are sex-independent processes, since no difference was noted in the number of H3.Pi-S28-positive CMs in female mice in comparison to male mice (Figure S6).

Reportedly, CMs undergo a binucleation process at 5 d postnatally, implying that in these cells, the M phase can occur independently of cytokinesis [8]. Therefore, we analyzed whether Cdk1-mutant CMs completed cytokinesis by forming two daughter cells. Cytokinesis and polyploidization were distinguished by immunostaining with anti-AurkB antibodies to detect midbody structures between daughter CMs in fixed cardiac specimens by confocal microscopy. In the presence of Cdk1, CM cytokinesis was detected only at 0 d and 1 d post-birth based on the analysis of AurkB-positive midbody structures between CMs (p < 0.01 vs. 2 d) (Figure 4C). The second phase of DNA synthesis at 5 d was associated only with CMBN in control hearts, as indicated by the absence of AurkB-positive CM (p < 0.01 vs. 1d). In contrast, the ablation of Cdk1 induced ongoing CM cytokinesis from 1 d to 7 d (p < 0.001 vs. 1 d–7 d controls) (Figure 4C). In particular, there was a significant increase in the percentage of binucleated CMs and a decrease in mononucleated CMs in control hearts at 7 d when compared to hearts from Cdk1KOc mice (p < 0.001) (Figure 4D). Thus, the loss of Cdk1 not only extended the normal window of postnatal CM proliferation by 6 d but also markedly impaired CMBN. This effect was dependent on the absence of Cdk1 and never observed in controls. Intriguingly, the high numbers of dividing CMs in the hearts of Cdk1KOc mice at 0 d are compatible with the concept that Cdk1 is not absolutely required for prenatal CM proliferation.

2.5. Loss of Cdk1 Triggers Cell Cycle Re-Entry and Proliferation of Adult Cardiomyocytes Post-MI

To investigate the potential role of Cdk1 loss in CM proliferation and cardiac remodeling in a clinically relevant model of ischemic stress, 10–12-week-old male and female Cdk1KOc and control mice were subjected to MI by permanent ligation of the left anterior descending artery. Next, RNA-seq analysis on total left ventricular (LV) tissues from Cdk1KOc and control mice was performed at 4 days post-MI for genome-wide mRNA profiling. Our previous study identified the 4-day-post-MI time point as a critical ‘window’ that drives the proliferative cardiac phenotype in vivo [42]. Three-dimensional principal component analysis (3D-PCA) revealed distinct clustering of Cdk1KOc and control samples, both with and without MI, as analyzed using Strand-NGS software (version 4.1; Strand Life Sciences, Hebbal, Bangalore, India, Karnataka) (Figure 5A). Unsupervised hierarchical clustering at a high-confidence threshold identified 1433 transcripts that significantly differed between Cdk1KOc and control mice post-MI compared to sham-operated controls (Figure 5B,C). This finding was further supported by a Volcano plot analysis, which showed that Cdk1 ablation led to the destabilization of mRNAs in control versus Cdk1KOc mice (Figure 5D).

To explore gene clusters with similar biological functions in mutant and wild-type transcriptomes, we performed a Gene Set Enrichment Analysis (GSEA) at 4 days post-MI treatment. We found that transcripts related to ventricular remodeling, mitochondrial function, cardiac contraction, and mitosis were among the most enriched GSEA terms (Figure 5E). Notably, transcripts for cell-cycle-promoting factors, such as Mcm2, Mcm3, and Tert, were selectively enriched at 4 days post-MI (Figure 5F). At 4 d post-MI, the area at risk (AR) in Cdk1KOc and control groups was of equal size and involved greater than 80% of the cross-sectional area of the left ventricle, measured at the level of the papillary muscles (Figure S7). We also noted a decreased sarcomeric α-actinin expression and lower extracellular matrix content, as evaluated by wheat germ agglutinin (WGA) staining. The border zone (BZ) showed α-actinin expression comparable to that of the area at risk, with a markedly higher level of WGA reactivity. This is in stark contrast to the remote area (RA) where both α-actinin and WGA are observed at levels comparable to sham Cdk1KOc and sham controls (Figures S7 and S8).

To further define the role of Cdk1 loss in cardiac remodeling after an MI, DNA synthesis was assessed using EdU labeling in adult mice, followed by confocal immunofluorescence microscopy. Intriguingly, we observed the following numbers of S-phase CMs in the LV region of Cdk1KOc mice at 4 d post-MI: (1) AR 61 ± 10.9 nuclei/mm², (2) BZ 33 ± 7.6 CM nuclei/mm², (3) RA 1.2 ± 3.4 CM nuclei/mm², and (4) total number (TN) 98 ± 11.3 CM nuclei/mm² (p < 0.01) (Figure 5G,H, Figures S7 and S8). These results clearly demonstrate that the induction of DNA synthesis in CMs lacking Cdk1 requires an ischemic stress.

Assessment of numbers of CMs in the LV region of Cdk1KOc mice in the M phase by immunofluorescence microscopy employing anti-phospho-Histone 3 phosphorylated at H3.Pi-S28 showed (1) AR 19.8 ± 7.6 CM nuclei/mm², (2) BZ 9.7 ± 3.2 CM nuclei/mm², (3) RA zero CM nuclei/mm², and (4) TN 30 ± 8.6 CM nuclei/mm² (p < 0.01) (Figure 5I–K and Figure S9). In all these experiments, EDU-positive and H3.Pi-S28-positive CMs were absent from the control mice. In addition, the induction of CMs in the M phase post-MI was restricted to CMs residing in the BZ and was absent from the right ventricle.

Next, we analyzed whether CMs in the LV region of Cdk1KOc mice completed cytokinesis by dividing into two daughter cells by detecting midbody structures in the final phase of daughter cell separation employing anti-AurkB antibodies. MI in Cdk1KOc animals induced cytokinesis in 15.2 ± 4.6 CM/mm² in the AR and minimally in the BZ (p < 0.001) (Figure 5L–N and Figure S10) that was not observed in the controls. This supports the notion that Cdk1 deletion restricts proliferation to a specific subpopulation located in an ischemic milieu. Of note, CM proliferation in hearts from sham Cdk1KOc and sham control mice was never observed.

Infarct ‘wound healing’ is characterized by the initiation of non-CM (NCM) proliferation that mainly comprises endothelial cells and fibroblasts [48]. We noticed a significant increase in the numbers of NCMs in the S and M phases in the LV region of Cdk1KOc and control mice at 4 d following MI (p < 0.001 vs. sham Cdk1KOc and sham controls) (Figure 5G,J and Figure S7). The number of proliferating EDU-positive NCMs was markedly lower in the hearts of Cdk1KOc mice compared to the controls (150 ± 6.1 vs. 351 ± 9.6 NCM nuclei/mm²). All these findings support the view that Cdk1 ablation induces adult CM proliferation after ischemic injury. We and others have previously shown that cell cycle re-entry in adult CMs is tightly regulated by Cdk2 [42]. At 4 d post-MI, sham control hearts lacked an induction of Cdk2 protein relative to Cdk1KOc post-MI (Figure 5O). Elevated Cdk2 protein levels were accompanied by the sustained induction of proliferative cyclins A, B, D2, and E (Figure 5O). Intriguingly, ablation of Cdk1 also led to extremely low protein levels of Cdk2-inhibitory p21 and p19, cdk4/6 inhibitors.

We have established that adult Cdk1KOc mice have developed a hypertrophic heart (Figure 3I) that mainly consists of mononucleated CMs (78 ± 7.6%; p < 0.001 vs. control) (Figure 4D). Thus, we examined whether Cdk1 deficiency can further influence the ploidy of CMs in the LV region post-MI. We observed a small but significant increase (11%) in mononucleated CMs, and a decrease (−75.4%) in binucleated CMs in the AR/BZ areas in Cdk1KOc mice post-MI (p < 0.001) (Figure 5P). This data provides additional evidence that genetic Cdk1 ablation leads to CM proliferation rather than polyploidy.

2.6. Activation of EGFR1 Signaling Promotes Cardiomyocyte Proliferation in Cdk1KOc Mice Post-MI

RNA-seq analysis of LV tissue samples at 4 d post-MI identified differential pathway activation between Cdk1KOc and control mice, as analyzed by employing the pathway analysis module of Strand-NGS (Figure 5Q). The EGFR1 signaling pathway was significantly upregulated in the Cdk1KOc MI group, as indicated by the enrichment of 88 out of 159 genes within this pathway (p < 0.001). This activation is associated with enhanced CM proliferation in the Cdk1KOc mice post-MI. In contrast, the p53- and p38-MAPK signaling pathways were predominantly activated in the control group. The activation of these pathways is known to inhibit CM division, suggesting that control mice respond to MI with signals that block cell cycle re-entry [49,50]. In contrast, the absence of Cdk1 allows for the activation of proliferative pathways, such as EGFR1, potentially facilitating CM proliferation (Figure 5Q) [51]. The heatmap in Figure 5R further corroborates the pathway analysis by displaying the expression profiles of key genes involved in the EGFR1, p38, and p53 signaling pathways across the four different experimental groups. Notably, key genes involved in the EGFR1 pathway, such as ErbB4, Erk1, and Sos2, show marked upregulation in the Cdk1KOc MI group when compared to the controls, aligning with the observed pathway activation. Conversely, key genes associated with the p38-MAPK and p53 pathways, including p38α, Cdkn1, and Gadd45g, are upregulated in the control group post-MI, which is consistent with the inhibition of CM proliferation in this cohort (Figure 5R). Western blot analysis confirms the activation of these pathways at the protein level (Figure 5S). In Cdk1KOc mice post-MI, there is an increase in phosphorylated ERK1/2 (Pi-ERK1/2), a key downstream effector of EGFR1 signaling, supporting the role of this pathway in promoting CM proliferation. In contrast, control mice show increased levels of p53 and phosphorylated p38α (Pi-38α), further validating the RNA-seq findings that these pathways are involved in blocking CM division following an MI. The presence of these protein markers in the Cdk1KOc and control groups provides strong evidence for the differential activation of proliferative vs. inhibitory pathways depending on the genetic background of the experimental strains.

2.7. Cdk1 Deficiency Mitigates Hypertrophy, Boosts Survival, and Improves Heart Function Post-MI

Next, we assessed the physiological consequences of Cdk1 loss by echocardiographic measurements of fractional shortening (FS), left ventricular end-diastolic diameter (LVEDD), and left ventricular end-systolic diameter (LVESD) across Cdk1KOc and controls at 21 d post-MI (Figure 6A). FS, an important indicator of systolic function, was similar between sham-operated controls and Cdk1KOc mice. However, FS in Cdk1KOc mice increased by 49% relative to the controls (Cdk1KOc post-MI: 29 ± 2%, vs. controls post-MI: 20 ± 3%; p < 0.01) (Figure 6A). These data demonstrate that Cdk1 ablation preserved contractile function post-MI, which is crucial for limiting the progression to heart failure. We observed that both LVEDD and LVESD, parameters reflecting LV size and remodeling, were significantly improved in Cdk1KOc mice against the controls post-MI. LVEDD was significantly lower in Cdk1KOc mice (4.8 ± 0.4 mm) compared to the controls (5.3 ± 0.5 mm; p < 0.01), indicating less ventricular dilatation (Figure 6A). Similarly, LVESD was reduced in the Cdk1KOc group (3.0 ± 0.2 mm) next to the controls (3.8 ± 0.3 mm, p < 0.01). These findings show that Cdk1 ablation attenuates adverse remodeling post-MI, implying that abrogation of Cdk1 function has cardioprotective effects.

Western blot analysis of LV tissue samples in Figure 6B shows several key cardiac proteins, including Serca2a and RyR2, along with their phosphorylated activated forms. The Cdk1KOc cohort demonstrates altered protein expression post-MI that correlates with improved calcium handling and contractility when compared to the controls. Figure 6C shows levels of sarcomeric genes, which are crucial for cardiac muscle contraction, across different experimental groups. Notably, key genes such as Ttn, Myh7b, and Tnnt1 are significantly upregulated in the Cdk1KOc MI group relative to the controls. This elevated expression of sarcomeric genes in Cdk1KOc mice post-MI suggests enhanced contractile machinery, contributing to improved cardiac function. The heatmap in Figure 6D shows the differential expression of genes involved in the cardiac cycle. Key genes such as Kcnj3, Scn4b, and Prkca, which are involved in cardiac rhythm and contraction regulation, are specifically upregulated in Cdk1KOc mice post-MI. This upregulation implies that the Cdk1 knockout promotes a more efficient cardiac cycle post-MI, likely contributing to the observed preservation of heart function in these mice.

Figure 6E illustrates the significant upregulation of transcript levels of various cardiac transcription factors, highlighting Gata4, Srf, and Nkx2-5 in Cdk1KOc mice post-MI compared to the controls. These transcription factors are known to regulate genes critical for cardiac function and remodeling, and their increased expression in the knockout mice may play a pivotal role in the observed enhancement of cardiac function post-MI. Collectively, these findings indicate that the upregulation of sarcomeric and cardiac cycle genes, along with key transcription factors, contributes to the superior cardiac function observed in Cdk1KOc mice post-MI.

Increases in HBW ratios in Cdk1KOc mice were significantly less post-MI versus MI/controls (Figure 6F). Moreover, lung weight/body weight (LBW) ratios, an index of pulmonary congestion, were also less elevated in Cdk1KOc mice as opposed to controls post-MI (Figure 6G). Importantly, the Mantel–Cox test showed significantly improved survival in Cdk1KOc mice at 21 d post-MI relative to the controls (p < 0.001) (Figure 6H). RT-qPCR analysis of canonical hypertrophic marker genes in LV tissue samples revealed decreased transcript levels ANP, BNP, and β-MHC in Cdk1KOc mice post-MI against controls (p < 0.01) at 3 d and 21 d (Figure 6I,J)). Moreover, the CM cross-sectional area (Figure 6K) and infarct size (Figure 6L,M), as measured at the peri-infarct zone, were significantly lower in Cdk1KOc mice in contrast to the controls post-MI (p < 0.01).

Inflammation typically precedes angiogenesis in the sequence of biological responses following MI [3,4,52]. The heatmap in Figure 6N shows the expression levels of inflammation-related genes across different experimental groups. In LV tissue samples of Cdk1KOc mice post-MI, several key pro-inflammatory genes (Stat1, Il6ra, Nfkb1, Tgfbr1) are significantly upregulated against the control cohort. This indicates a heightened inflammatory response in Cdk1KOc mice following MI. Notably, genes involved in promoting angiogenesis (Vegfa, Flt1, Pecam1, Angpt1) are upregulated in the Cdk1KOc mice relative to the controls post-MI (Figure 6O). This suggests that Cdk1KOc mice may have enhanced angiogenic responses post-MI, potentially contributing to better tissue repair and remodeling. Additionally, immunofluorescence images of fixed Cdk1KOc-derived cardiac tissue samples exhibit a significantly higher vWF expression post-MI (p < 0.01) versus controls, indicating increased endothelial cell activity and angiogenesis (Figure 6P). All these findings indicate a higher degree of coronary angiogenesis in Cdk1KOc mice post-MI as opposed to their wt counterparts.

Our heatmap analysis reveals that key apoptotic genes (Bax, Casp3, Fas, and Bcl2) are upregulated in LV tissue samples of the controls in comparison to Cdk1KOc animals post-MI (Figure 6Q). Quantification of apoptotic CM through TUNEL staining and fluorescence microscopic analysis demonstrates a significantly higher number of apoptotic CMs in the controls versus Cdk1KOc mice post-MI (p < 0.01) (Figure 6R), suggesting that Cdk1 deletion protects against CM apoptosis post-MI.

2.8. Cdk1 Loss Preserves Mitochondrial Energetics by Protecting against Ischemic Oxidative Stress

We have previously shown that MI-induced oxidative stress decreases in ATP levels negatively impact the contractile function and survival of CMs [42]. Thus, we analyzed gene expression of mt-DNA replication and mt-dynamics in LV samples of Cdk1KOc mice post-MI. The heatmap in Figure 7A shows that the transcript levels of key genes involved in mt-DNA replication such as Tfam, PrimPol, and Tfb2m are upregulated in Cdk1KOc mice post-MI. Notably, the expression of genes related to mt fission and fusion like Fis1, Mfn2, and Mff are also significantly higher in the Cdk1KOc mice post-MI (Figure 7B). Next, the mt-DNA copy number was quantified by qPCR using mt-Cytb that were normalized to the nuclear-encoded gene B2m. In Cdk1KOc mice post-MI, we found a significant restoration of the mt-DNA copy number (p < 0.01) (Figure 7C) in contrast to the controls. This restoration was accompanied by an increase in mitochondrial respiratory capacity (p < 0.01) (Figure 7D) and notably higher ATP/ADP ratios in the myocardium of Cdk1KOc mice vs. controls (p < 0.01) (Figure 7E–G).

Moreover, we also observed significantly decreased levels of cytotoxic 4-HAE, an indicator of ROS-dependent lipid peroxidation [27] (Figure 7H,I) with upregulated glutathione/oxidized glutathione (GSH/GSSG) ratios in LV samples of Cdk1KOc mice post-MI (Figure 7J–L). The heatmap in Figure 7M illustrates transcript levels of genes involved in the pentose phosphate pathway (PPP), which is critical for generating glutathione, a key molecule for producing NADPH to combat oxidative stress [53]. In Cdk1KOc mice post-MI, there is an upregulation of genes such as Tkt and G6pdx, indicating increased PPP activity compared to controls. This upregulation correlates with higher NADPH/NADP+ ratios in Cdk1KOc hearts subjected to MI (Figure 7N) and implies an enhanced capacity to neutralize oxidative stress (p < 0.01). This view agrees with our Western blot analysis showing elevated levels of antioxidative enzymes such as Sod2, Nqo1, and Pink1 in Cdk1KOc mice post-MI, underscoring the improved oxidative stress response associated with Cdk1 knockout (Figure 7O).

3. Discussion

Cdk1 has a distinct role in embryonic mammalian proliferation that is not fulfilled by Cdk2/4/6 [54]. Yet, there is a paucity of experimental evidence regarding the role of Cdk1 in regulating CM proliferation during the neonatal phase and post-MI. Our data demonstrate that Cdk1 activity is crucial for CM cell cycle arrest at 1 d and CMBN at 5 d after birth. Notably, the ablation of Cdk1 causes CMs to remain in a proliferative state until 10 d. Additionally, the genetic ablation of Cdk1 promotes CM proliferation and aids in cardiac repair after MI in the murine adult heart. This study is the first to reveal Cdk1’s dual roles in CMM.

It was surprising to discover that Cdk1 intrinsically inhibits the proliferation of both neonatal and adult CMs post-MI. This finding contradicts the prevailing belief that Cdk1 is necessary for cell proliferation. While germline Cdk1KO mice do not progress beyond the two-cell embryonic stage [54], it is interesting that Cdk1 can offset the lack of individual Cdks in Cdk2/4/6 triple-KO embryos, allowing them to develop normally [55,56]. The triple-KO embryos maintain Cdk1 expression, supporting regular cell proliferation until mid-gestation. Interestingly, the introduction of Cdk2 into the Cdk1 gene locus in mice did not prevent early embryonic death. In our study, Cdk1KOc mice were born at expected Mendelian ratios and exhibited hypertrophic ventricular walls but maintained normal cardiac function.

The rate of CM proliferation in mouse hearts varies between embryonic and postnatal stages. The four-chambered mammalian mouse heart is completely developed by embryonic day E12.5 [57]. In mice, CM proliferation peaks around E10-12 and then gradually diminishes until birth. This pattern might shed light on our observation that Cdk1 is not essential for CM proliferation after day E14, which is after the initiation of α-MHC Cre-dependent homologous recombination, and during the early postnatal phase. Rather, we hypothesize that Cdk1 is primarily required during the initial phases of heart formation in embryogenesis. Our data further corroborate the notion that Cdk2/4/6, either individually or in conjunction, can offset the absence of Cdk1 activity during the cell cycle progression of fetal and neonatal CMs. Comparable to Cdk1KOc mice, transgenic mice with cardiac-specific overexpression of Cdk2 in adults displayed an increase in the number of smaller, mononucleated CMs, suggesting enhanced CM division [58]. Hearts with Cdk2 overexpression exhibited normal FS, even with elevated levels of the canonical hypertrophic marker genes ANF and β-MHC. In contrast, the triple-KO mice lacking cyclin D1/2/3 present hypoplastic ventricular walls accompanied by severe anemia. These animals typically die around mid- to late gestation around embryonic day E16.5 [59].

Our study revealed that the loss of Cdk1 post-MI leads to the activation of the EGFR1 signaling pathway, which promotes CM proliferation [51] and the suppression of the p38 [50] and p53 pathways [49]. This is associated with blocking CM division, thereby allowing for an extended proliferative window in Cdk1KOc mice. This shift in signaling pathways is crucial for enhancing cardiac repair and preserving cardiac function post-MI. Furthermore, Cdk1 ablation was found to significantly enhance mitochondrial function and reduce the production of ROS, as evidenced by the upregulation of mitochondrial biogenesis and antioxidative enzymes in the Cdk1KOc mice. This improvement in mitochondrial energetics likely contributes to the preserved cardiac function observed in these mice.

We also demonstrate that Cdk1 is important for the regulation of the CM division cycle early after birth. Unexpectedly, the genetic deletion of Cdk1 in neonatal mouse CMs effectively disabled the intrinsic proliferative block at 1 d post-birth, extending the postnatal proliferative window to 10 days. However, there are still gaps in our understanding of the mechanisms that govern CMBN [60,61]. In inbred C57BL/6 mice, commonly employed for cardiological studies, 8-10% of CMs are mononucleated [60]. Notably, non-epigenetic mechanisms such as certain polymorphisms located in X-linked and autosomal genes can influence the outcome of mitosis and karyokinesis in a nonautonomous-CM fashion [60]. Our findings found no evidence to support this view during adolescent murine growth, as the HTW ratio of Cdk1KOc mice at 15 d and 3 m did not significantly alter from the controls (Figure 3F,G). In addition, a proliferative burst at 15 d post-birth has been suggested to give rise to binucleated mouse CMs [62]. This event is mediated by the thyroid hormone–IGF-1-Akt signaling pathway. However, several independent studies, including our present work, were unable to confirm this preadolescent event [63,64].

4. Conclusions

In this study, we investigated the regulation of the cell division cycle in CMs during the early neonatal period and its impact on heart regeneration following an MI. Our results revealed a tightly controlled sequence of events: CM proliferation ceases shortly after birth, followed by a distinct phase of CMBN. Cdk1 plays a pivotal role in regulating both CMBN and proliferation during this period. The loss of Cdk1 extended the neonatal proliferative window, increasing the number of dividing CMs. In adult mice, Cdk1 ablation post-MI activated the EGFR1 signaling pathway, promoting CM proliferation, while downregulating the p38 and p53 pathways, which are typically antiproliferative. This was associated with restoration of FS, enhanced mt function, and a reduction in oxidative stress post-MI.

Cdk1 emerges as a potential key regulator of heart regeneration post-MI. Targeting Cdk1-related pathways holds significant therapeutic potential for enhancing heart regeneration after an MI. Future research should explore the interplay between Cdk1 and other cell cycle regulators to gain a more comprehensive understanding of the balance between CMBN and CM proliferation. In conclusion, the loss of Cdk1 is a potential key regulator of heart regeneration post-infarction. Targeting Cdk1-related pathways could hold therapeutic potential for improving heart regeneration after an MI to reduce HF risks and improve quality of life.

5. Materials and Methods

5.1. Generation of Cardiac-Specific Cdk1-Knockout Mice

All animal usage was performed in accordance with the institutional animal care guidelines of University Health Network (AUP 1379; Canadian Council in Animal Care). As outlined in the ensemble website (http://useast.ensembl.org/index.html), the murine Cdk1-202 mRNA (strain reference CL57BL6) is located on chromosome 10 on the reverse strand. This transcript has 9 exons, of which 7 are coding exons. The length of the Cdk1-202 transcript is 1253 bp that codes for 297 residues and has a molecular weight of 34.1 kDa. To examine the potential role of Cdk1 in mice, a constitutive cardiac-specific Cdk1 knockout (Cdk1KOc) was generated. This was accomplished by crossing Cdk1^f/f mice (129S(B6N)-Cdk1 ^tm1Eddy/J; Jackson laboratory, Bar Harbor, ME, USA) [46] with transgenic mice carrying the a-MHC Cre recombinase construct (Jackson laboratory, B6.FVB-Tg(Myh6-cre)2182Mds/J) [47]. During generation of the Cdk1^f/f strain, mice were crossed to mice expressing FLP1 recombinase to excise the neomycin selection cassette. The cardiac-muscle-specific α-myosin heavy chain 6 promoter drives the expression of Cre recombinase at birth and results in a recombination efficiency of more than 90% [47]. Cre recombinase fuses together the flanking loxP site of the engineered Cdk1^f/f gene at either side of exon 3, thereby deleting the translational initiation site. Cdk1KOc had exon 3 from Cdk1 deleted in CM from day E14, which leads to a reading frameshift mutation that induces a premature translational stop of the Cdk1 mRNA. These mice did not express functional forms of Cdk1.

5.2. Isolation of DNA and Genotyping

DNA isolated from fresh tail snips (<5 mm) was used for genotyping. Samples were incubated in 300 μL of 50 mM NaOH for 2 h at 80 °C while rocking, and neutralized with 25 μL 1.0 M HCl in 700 μL H₂O. Samples were vigorously vortexed, centrifuged for 13,000 rpm at 10 min, and stored at 4 °C. A total of 0.5 μL of DNA sample was used per PCR reaction with the following primers: Cdk1 forward 5′-CCAGGGTGA CCTTGTGCT-3′; Cdk1 reverse 5′-AGCCTGCCTCCACTTCCA-3′; Cdk1 Post-Cre forward 5′-GCACTCGGCCTCTAAGCTC-3′; Cdk1 Post-Cre reverse 5′-TCCACTTGGGAAAGGTGTTC-3′; Myh6-Cre forward 5′-ATGACAGACAGATCCCTCCT ATCTCC-3′; Myh6-Cre reverse 5′-CTCATCA CTCGTGCATCATCGAC-3′. PCR analysis was carried out with the Applied Biosystems ProFlex PCR system by Life Technologies (4484073; Thermo Fisher Scientific, Mississauga, ON, CA); Cdk1 wild-type allele, 299 bp; Cdk1 floxed allele, 320 bp; Cdk1 Post-Cre, 838 bp (Figure 3B). Myh6-Cre transgene, 300 bp; and Myh6-Cre internal positive control, 200 bp.

5.3. Coronary Artery Ligation

To create an MI, permanent ligation of the left descending coronary artery was performed in in 10–12-week-old male and female Cdk1KOc and control mice. At the time of surgery and at 24 h post-MI, a single subcutaneous dose of buprenorphine (30 mg/g body weight) (B9275; Sigma-Aldrich, Oakville, ON, CA) was given as described previously [42].

5.4. Echocardiography

Echocardiography was performed in anesthetized mice pre-MI and at 21 d post-MI.

5.5. DNA Synthesis, TUNEL, ROS Assays, and Three-Dimensional Immunofluorescence Microscopy

For EDU-labeling experiments, EDU (160 mg/kg body weight cumulative dosage) was intraperitoneally injected twice at 4 d post-MI. Hearts were sampled 2 h after the second injection. For detection of incorporated EDU, the Click-iT™ EdU Cell Proliferation Kit for Imaging (Alexa Fluor™ 488 dye; C10337; Thermo Fisher) was employed according to the manufacturer’s specifications.

Detection of fragmented genomic DNA (apoptosis assay) was performed by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling according to the manufacturer’s instructions (no. 1684795910; Roche, Mississauga, ON, CA) [53]. For immunofluorescence measurements of cell cycle distribution, and CMBN, 3 independent fields of approximately of 0.25 mm² per left ventricular (LV) sample were counted on 3 consecutive sections. Then, 3D-reconstruction of digital microphotographs were performed using Imaris Imaging software, version 10.0.

5.6. Primary Mouse Neonatal Ventricular Cardiomyocyte Isolation

Hearts from 1-day-old neonatal Cdk1KOc mice and controls were isolated as described [65].

5.7. Preparation of Protein Extracts from Left Ventricular Heart Tissue Samples

Only left ventricular (LV) specimens were used from neonatal hearts derived from Cdk1KOc mice and controls, and they were snap frozen and stored at −80 °C. Preparation of total protein extracts was performed as described previously [42].

5.8. Western Blotting

Western blotting was carried out as described previously [42]. The antibodies employed in this study are summarized in the Supplementary Table S1.

5.9. Total RNA Isolation, Reverse Transcription, and Quantitative Real-Time PCR Assays

Total RNA isolation, reverse transcription, and quantitative real-time PCR assays were performed as described previously [42]. The following oligonucleotide primers were used:

α-smooth muscle actin (Acta2) forward 5′-CTGACAGAGGCACCACTGAA-3′, reverse 5′-CAT CTCCAGAGTCCAGCACA-3′. β-Actin forward 5′-GGCTGTATTCCCCTCCATCG-3′, reverse 5′-CCAGTTGGTAACAATGCCATGT-3′. ANP forward 5′-GCTTCCAGGCCATATTGGAG-3′, reverse 5′-GGGGGCATGACCTCATCTT-3. BNP forward 5′- GAGGTCACTCCTATCC TCTGG-3′, reverse 5′-GCCATTTCCTCCGACTTTTCTC-3′. α-MHC forward 5′-GCCCAGTA CCTCCGAAAGTC-3′, reverse 5′-GCCTTAACATACTCCTCCTT GTC -3′. β-MHC forward 5′-ACTGTCAACACTAAG AGGGTCA-3′, reverse 5′-TTGGATGATTTGATCTTCCAGGG-3′.

5.10. Cell Cycle RT-qPCR Array

RNA was purified from the hearts of neonatal wild-type mice as described previously [28]. Equal amounts of RNA (400 ng/sample) were converted into cDNA using the First Strand cDNA Synthesis kit (Qiagen, Toronto, ON, CA). The Cell Cycle PCR Array was obtained from Qiagen (330171) and used to profile the expression of 87 cell cycle genes, 5 housekeeping genes, controls for genomic DNA contamination, and the efficiency of both the reverse transcriptase and PCR reaction as described previously [42].

5.11. Gene Expression Analysis and Bioinformatics

Affymetrix Mouse Gene 2.0 ST expression arrays were processed at the Centre for Applied Genomics (Toronto, ON, Canada). Processing of probe level data and subsequent data analyses were performed using GeneSpring (Version 13.2; Agilent Technologies Inc., Santa Clara, CA, USA) as described previously [53]. Genome-wide data from the gene expression microarrays were normalized, filtered on expression (in the range of −5.2 to 3.5), filtered on Error-CV < 50.0 percent, and filtered for genes with significant differences in expression levels (log₂ fold change ±1.3; p < 0.05), followed by determination of statistical significance according to a t test with Benjamini–Hochberg correction (log₂ fold change ±1.3; p < 0.01) in wild-type neonatal mice (0 d through 15 d) in comparison to 0 d employing GeneSpring; n = 6 to 20 per time point. GSEA and GO term analysis of differentially expressed genes were carried out with GeneSpring. Microarray data were submitted to the ArrayExpress database (http://www.ebi.ac.uk/arrayexpress. Access date 11 January 2019). Accession number: E-MTAB-8168.

RNA-seq samples were processed at the Princess Margaret Genomics Centre, 101 College St., 9-601, Toronto, ON, M5G 1L7, Canada. Briefly, whole-exome RNA-seq was performed using paired-end 150 nucleotide sequencing with 40 million reads. Processing of probe-level data and subsequent data analyses were performed using Strand-NGS software. Genome-wide data were normalized and filtered for genes with significant differences in expression levels (log₂ fold change ±2.0; p < 0.01), followed by determination of statistical significance according to a t test with Benjamini–Hochberg correction (log₂ fold change ±2.0; p < 0.01) employing Strand-NGS. The 3D-PCA analysis, Volcano plot, Heatmaps, GSEA, GO, and Pathway analysis of differentially expressed genes were carried out with Strand-NGS. RNA-seq data were submitted to the ArrayExpress database (http://www.ebi.ac.uk/arrayexpress, accessed on 24 September 2024). Accession numbers: E-MTAB-14505 for samples collected 4 days post-MI and E-MTAB-14508 for samples collected 3 weeks post-MI, respectively.

5.12. Isolation of Mitochondria and Detection of Oxidative Damage, Antioxidants, and ATP Levels

Isolation of mitochondria and the detection of oxidative damage, antioxidants and ATP levels was carried out as described previously [65].

5.13. Statistical Analyses

Statistical analyses were carried out using GraphPad InStat (version 3.1) and GraphPad Prism (GraphPad Software, version 10.3.1; La Jolla, CA, USA). Data are reported as means ± s.e.m. Two-tailed p-values of < 0.05 were considered as significant. When groups passed the normality test, we performed data evaluation between 2 groups by an unpaired Student’s t-test. The statistical significance of 4 groups was calculated using a one-way analysis of variance (ANOVA) and Tukey–Kramer multiple comparison post-test.

We calculated experimental power with GraphPad Statmate (GraphPad Software, version 1, La Jolla, CA, USA). For example, for the FS data presented in Figure 6L, we used the value of 1 for the standard deviation (S.D.) for Cdk1KOc mice versus controls post-MI, together with a significance level alpha = 0.01 (two-tailed) and 80% power. Therefore, at least n = 4 per group was required in each experiment and that was increased to n = 6 per group.

Survival analysis was performed using a conservative log-rank test (Mantel–Cox). For calculation of experimental power, the expected proportion of being event-free or surviving was 20% for the Pkm2KOi mice versus controls. Thus, a sample size of n = 20 for Cdk1KOc mice and 24 for the controls post-MI in each group provided 80% power to detect an increase in survival proportion of 0.115 with a significance level alpha of 0.01 (two-tailed). For the proliferation data (Figure 5), we used an S.D. value of 3.5 and n = 6 per group in each experiment for Cdk1KOc mice versus the controls post-MI. This resulted in an 80% power to detect a difference between means of 5.78 with a significance level (alpha) of 0.01 (two-tailed).