Neonatal Heart Regeneration: A Comprehensive Literature Review

Abstract

Background:

The adult mammalian heart is incapable of meaningful functional recovery after injury, and thus promoting heart regeneration is one of the most important therapeutic targets in cardiovascular medicine. In contrast to the adult mammalian heart, the neonatal mammalian heart is capable of regeneration after various types of injury. Since the first report in 2011, a number of groups have reported their findings on neonatal heart regeneration. The current review provides a comprehensive analysis of heart regeneration studies in neonatal mammals conducted to date, outlines lessons learned and poses unanswered questions.

Methods:

We performed a PubMed search using the keywords “neonatal” and “heart” and “regeneration.” In addition, we assessed all publications that cited the first neonatal heart regeneration reports: Porrello et al, Science, Feb 2011 for apical resection injury; Porrello et al, PNAS, Dec 2012 for coronary ligation injury; and Mahmoud et al, Nature Methods, Jan 2014 for surgical methodology. Publications were examined for surgical models used, timing of surgery, and postinjury assessment including anatomical, histological and functional assessment, as well as conclusions drawn.

Results:

We found 30 publications that performed neonatal apical resection, 19 publications that performed neonatal myocardial infarction by coronary artery ligation, and 6 publications that performed cryoinjury using liquid nitrogen-cooled metal probes. Both apical resection and ischemic infarction injury in neonatal mice result in a robust regenerative response, mediated by cardiomyocyte proliferation. On the other hand, several reports have demonstrated that cryoinjury is associated with incomplete heart regeneration in neonatal mice. Not surprisingly, several studies suggest that injury size, as well as surgical and histological techniques can strongly influence the observed regenerative response and final conclusions. Studies have utilized these neonatal cardiac injury models to identify factors that either inhibit or stimulate heart regeneration.

Conclusions:

Overall, there is consensus that both apical resection and coronary ligation injuries during the first two days of life result in heart regeneration in neonatal mammals, whereas cryoinjury was not associated with a similar regenerative response. This regenerative response is mediated by proliferation of pre-existing cardiomyocytes, and is modifiable by injury size and surgical technique, as well as metabolic, immunologic, genetic and environmental factors.

Introduction

Cardiac regeneration in vertebrates was first identified in zebrafish, with surgical resection of the adult heart (20% of its ventricle) able to completely regenerate within 60 days without fibrotic scarring.¹ Amputation of embryonic zebrafish hearts was also reported to induce a regenerative response.² Subsequent studies demonstrated that cardiac regeneration in zebrafish is primarily mediated by proliferation of pre-existing cardiomyocytes.^{3, 4} Not unlike zebrafish cardiomyocytes, mammalian neonatal cardiomyocytes have been known to retain mitotic capacity during the first few days of life, and have been used extensively to study cardiomyocyte cell cycle kinetics in vitro.

Based on these characteristics of neonatal cardiomyocytes, we examined the regenerative capacity of the neonatal heart, using an injury similar to the zebrafish model. Resection of 15% of the left ventricle at P1 (1 day postpartum) resulted in a robust regenerative response characterized by marked induction of cardiomyocyte proliferation, with replacement of the lost myocardium 21 days post resection.⁵ This remarkable regenerative response was mediated by proliferation of pre-existing cardiomyocytes.

Genetic fate mapping was performed using an inducible, cardiomyocyte specific Myh6-MerCreMer mouse model, crossed with a LacZ reporter. A single large dose of tamoxifen successfully labelled 70–80% of cardiomyocytes when administered at P0, and was followed by apical resection at P2. These studies revealed that the majority of newly formed cardiomyocytes were derived from pre-existing cardiomyocytes (Myh6-expressing cells). However, minor contribution of non-cardiomyocytes (in the unlabelled fraction) could not be ruled out. In contrast, mice at P7 (7 days postpartum), which underwent apical resection, failed to regenerate and developed apical fibrosis. Loss of the endogenous cardiac regenerative potential within the first week of life coincides with cell cycle withdrawal of cardiomyocytes, and switch from hyperplasic to hypertrophic growth in rodents.⁶ Interestingly within this cardiac regeneration window, Field and colleagues had previously showed that P4 mouse cardiomyocytes undergo a burst of DNA synthesis without cytokinesis leading to binucleation.⁷

Since the first report, numerous groups have studied neonatal heart regeneration, either using the apical resection model or other types of injury such as ischemic myocardial infarction and cryoinjury. In total, there are over 40 different reports reproducing the phenomenon of neonatal heart regeneration, and induction of cardiomyocyte proliferation following injury of the neonatal heart

Here we review all studies that used neonatal mouse models to study cardiac regeneration. We also discuss technical considerations when conducting cardiac injury in neonatal mice. In the following sections, we will discuss these studies and outline important considerations for performing and interpreting neonatal cardiac injury models.

Neonatal Mouse Apical Resection Model

The neonatal mouse apical resection model for the purpose of studying cardiac regeneration has been reproduced by multiple groups, as listed in Table 1^5,8–27, with the exception of the Andersen, Sheikh and colleagues^22–24 who failed to observe any regeneration or cardiomyocyte proliferation. Below is an outline of all the studies that used apical resection in neonatal mice, and a detailed analysis of the discrepant results by the Andersen and Sheikh groups.

Table 1. Studies that performed apical resection in P1 mice.

Reference	Authors	Year	Journal	Model	Cardiac Regeneration?
Transient Regenerative Potential of the Neonatal Mouse Heart. 5	Porrello, E. R., Mahmoud, A. I., Simpson, E., Hill, J. A., Richardson, J. A., Olson, E. N. & Sadek, H. A.	2011	Science	Apical Resection	Yes
Surgical models for cardiac regeneration in neonatal mice. 8	Mahmoud, A. I., Porrello, E. R., Kimura, W., Olson, E. N. & Sadek, H. A.	2014	Nat. Protocols	Apical Resection	Yes
Do Neonatal Mouse Hearts Regenerate following Heart Apex Resection? 22	Andersen, Ditte C., Ganesalingam, S., Jensen, Charlotte H. & Sheikh, Soren P.	2014	Stem Cell Reports	Apical Resection	No
Multi- Investigator Letter on Reproducibility of Neonatal Heart Regeneration following Apical Resection. 9	Sadek, Hesham A., Martin, James F., Takeuchi, Jun K., Leor, J., Nie, Y., Giacca, M. & Lee, Richard T.	2014	Stem Cell Reports	Apical Resection	Yes
A systematic analysis of neonatal mouse heart regeneration after apical resection. 21	Bryant, D. M., O’meara, C. C., Ho, N. N., Gannon, J., Cai, L. & Lee, R. T.	2015	Journal of Molecular and Cellular Cardiology	Apical Resection	Yes
The Type of Injury Dictates the Mode of Repair in Neonatal and Adult Heart. 11	Konfino, T., Landa, N., Ben‐Mordechai, T. & Leor, J.	2015	Journal of the American Heart Association	Apical Resection	Yes
Cardiac myosin binding protein C regulates postnatal myocyte cytokinesis. 12	Jiang, J., Burgon, P. G., Wakimoto, H., Onoue, K., Gorham, J. M., O’meara, C. C., Fomovsky, G., Mcconnell, B. K., Lee, R. T., Seidman, J. G. & Seidman, C. E.	2015	Proceedings of the National Academy of Sciences	Apical Resection	Yes
Nerves Regulate Cardiomyocyte Proliferation and Heart Regeneration. 10	Mahmoud, Ahmed I., O’meara, Caitlin C., Gemberling, M., Zhao, L., Bryant, Donald M., Zheng, R., Gannon, Joseph B., Cai, L., Choi, W.-Y., Egnaczyk, Gregory F., Burns, Caroline E., Burns, C. G., Macrae, Calum A., Poss, Kenneth D. & Lee, Richard T.	2015	Development al Cell	Apical Resection	Yes
Acute inflammation stimulates a regenerative response in the neonatal mouse heart. 13	Han, C., Nie, Y., Lian, H., Liu, R., He, F., Huang, H. & Hu, S.	2015	Cell Res	Apical Resection	Yes
Persistent scarring and dilated cardiomyopathy suggest incomplete regeneration of the apex resected neonatal mouse myocardium — A 180 days follow up study. 23	Andersen, D. C., Jensen, C. H., Baun, C., Hvidsten, S., Zebrowski, D. C., Engel, F. B. & Sheikh, S. P.	2016	Journal of Molecular and Cellular Cardiology	Apical Resection	No
GATA4 regulates Fgf16 to promote heart repair after injury.20	Yu, W., Huang, X., Tian, X., Zhang, H., He, L., Wang, Y., Nie, Y., Hu, S., Lin, Z., Zhou, B., Pu, W., Lui, K. O. & Zhou, B.	2016	Development	Apical Resection	Yes
Modulation of tissue repair by regeneration enhancer elements. 15	Kang, J., Hu, J., Karra, R., Dickson, A. L., Tornini, V. A., Nachtrab, G., Gemberling, M., Goldman, J. A., Black, B. L. & Poss, K. D.	2016	Nature	Apical Resection	Yes
Apical Resection Mouse Model to Study Early Mammalian Heart Regeneration. 16	Xiong, J. & Hou, J.	2016	J Vis Exp	Apical Resection	Yes
Bmi1+ cardiac progenitor cells contribute to myocardial repair following acute injury. 17	Valiente-Alandi, I., Albo- Castellanos, C., Herrero, D., Sanchez, I. & Bernad, A.	2016	Stem Cell Research & Therapy	Apical Resection	Yes
Pitx2 promotes heart repair by activating the antioxidant response after cardiac injury. 14	Tao, G., Kahr, P. C., Morikawa, Y., Zhang, M., Rahmani, M., Heallen, T. R., Li, L., Sun, Z., Olson, E. N., Amendt, B. A. & Martin, J. F.	2016	Nature	Apical Resection	Yes
Peptidomics Analysis of Transient Regeneration in the Neonatal Mouse Heart. 18	Fan, Y., Zhang, Q., Li, H., Cheng, Z., Li, X., Chen, Y., Shen, Y., Song, G. & Qian, L.	2017	Journal of Cellular Biochemistry	Apical Resection	Yes
The extracellular matrix protein Agrin promotes heart regeneration in mice. 19	Bassat, E., Mutlak, Y. E., Genzelinakh, A., Shadrin, I. Y., Baruch-Umansky, K., Yifa, O., Kain, D., Rajchman, D., Leach, J., Bassat, D. R., Udi, Y., Sarig, R., Sagi, I., Martin, J. F., Bursac, N., Cohen, S. & Tzahor, E.	2017	Nature	Apical Resection	Yes
Cardiac injury of the newborn mammalian heart accelerates cardiomyocyte terminal differentiation. 24	Zebrowski, D. C., Jensen, C. H., Becker, R., Ferrazzi, F., Baun, C., Hvidsten, S., Sheikh, S. P., Polizzotti, B. D., Andersen, D. C. & Engel, F. B.	2017	Scientific Reports	Apical Resection (primarily in rats and some mice)	No
Neonatal Apex Resection Triggers Cardiomyocyte Proliferation, Neovascularization and Functional Recovery Despite Local Fibrosis.25	Sampaio-Pinto V, Rodrigues SC, Laundos TL, Silva ED, Vasques-Nóvoa F, Silva AC, Cerqueira RJ, Resende TP, Pianca N, Leite-Moreira A, D’Uva G, Thorsteinsdóttir S, Pinto- do-Ó P & Nascimento DS	2018	Stem Cell Reports	Apical Resection	Yes
Angiogenesis precedes cardiomyocyte migration in regenerating mammalian hearts.26	Ingason AB, Goldstone AB, Paulsen MJ, Thakore AD, Truong VN, Edwards BB, Eskandari A, Bollig T, Steele AN & Woo YJ.	2018	The Journal of Thoracic and Cardiovascular Surgery	Apical Resection	Yes
Ezh2 is not required for cardiac regeneration in neonatal mice.27	Ahmed A, Wang T and Delgado-Olguin P.	2018	PLOS ONE	Apical Resection	Yes

Examining this regenerative phenomenon has provided numerous insights into the endogenous cardiac regenerative capacity in mammals. Seidman and colleagues performed apical resection on wildtype and Mybpc^t/t (myosin binding protein C deficient) mice at P1 and P10.¹² Compared with wildtype mice, Mybpc deficient mice had decreased cardiac regenerative capacity and increased susceptibility to necrosis. Lee and colleagues also apically resected P1 mouse hearts, and subsequently inhibited cholinergic transmission by atropine.¹⁰ They showed that P1 resected control mice exhibit robust cardiomyocyte proliferation, but atropine treatment in post resected mice decreased cardiomyocyte proliferation, suggesting that cholinergic nerve signaling plays a role in regulating mammalian heart regeneration. To provide additional clarification, our group published a Nature Protocols report outlining the apical resection method,⁸ which was later also independently demonstrated into a JoVE video tutorial.¹⁶ Moreover, Lee, Leor and colleagues,^11,21 reproduced apical resection studies at P1 and found that the cardiac regeneration data is consistent with Sadek and colleagues,⁵ with induction of cardiomyocyte proliferation and regeneration of the apex, although the Lee group did observe residual collagen deposition.

In a subsequent report, Han and colleagues demonstrated that P1 mice that underwent apical resection and were subjected to inhibition of an acute inflammatory response (with dexamethasone and anti-Gri1 antibodies) failed to regenerate lost myocardium and exhibited fibrosis by 21 days post resection, compared with control mice which completely regenerated.¹³ In addition, P1 interleukin 6-deficient mice that underwent apical resection did not regenerate and had less cardiomyocyte proliferation than apically resected wildtype mice.¹³ Another factor, which was shown to modulate cardiac regeneration following apical resection, is Gata4. Yu and colleagues demonstrated that wildtype mice subjected to apical resection at P1 regenerate, whereas Gata4-mutant mice failed to regenerate after apical resection.²⁰

In a study by Martin and colleagues, apical resection of P1 mouse hearts with inactivated Pitx2 cardiomyocytes failed to regenerate and did not show induction of cardiomyocyte proliferation, compared with control mouse hearts which completely regenerated and exhibited robust cardiomyocyte proliferation within 21 days post resection.¹⁴ Furthermore, apex resection of P8 mouse hearts did not regenerate, but P8 mice with cardiac overexpression of Pitx2 that underwent apical resection had less scarring and improved heart function at 28 days postresection compared with control mice. Hippo deficient mice at P8 can undergo regeneration after apical resection, but Hippo deficient mice, which are also deficient in Pitx2, fail to regenerate following apical resection at P8. Further characterization demonstrated that Pitx2 interacts with Yap, and that Pitx2 promotes regeneration in part by inhibition of reactive oxygen species.

Poss and colleagues found that LEN (lepb-linked enhancer) is induced during various tissue regeneration models in zebrafish.¹⁵ Apical resection was performed in LEN-hsp68::lacZ mice at P1, and lacZ expression was detected 3 days post injury in a regeneration model suggesting that LEN is also induced in mice during cardiac regeneration. In a separate study, cardiac regeneration was observed at 21 days post resection from mice apically resected at P1, and did not detect a significant contribution of Bmi1+ progenitor cells in the regenerated myocardium.¹⁷ In addition, a recent manuscript by Qian and colleagues¹⁸ used the apical resection protocol from Sadek and Olson labs^{5, 8} and observed cardiac regeneration in neonatal mice, and found that the hearts of P1 mice regenerated whilst P7 mice did not regenerate. Tzahor and colleagues performed apical resection on P1 mice and demonstrated that deletion of Agrin (an extracellular matrix protein) impaired cardiac regeneration and exhibited significant fibrosis compared with control hearts (which regenerated), suggesting that Agrin promotes cardiac regeneration.¹⁹

Additionally, Sampaio-Pinto and colleagues conducted an elegant comprehensive apical resection on P1 hearts; with hearts assessed for up to 180 days post apical resection. They reported a significant increase in cardiomyocyte proliferation and neovascularization, with restoration of cardiac function. Interestingly, they showed that collagen volume decreased over time, but was not completely cleared. In summary, their results are mostly consistent with Porrello et al (2011)⁵ and conclude that “apex resection triggers both regenerative and reparative mechanisms, endorsing this injury model for studies aimed at promoting cardiomyocyte proliferation and/or downplaying fibrosis”,²⁵ although it is important to note here that the definition of persistent collagen in that report was not clearly defined. For example, in some of the images provided in this manuscript collagen deposition appeared to be localized only to the epicardium (Figure 1A2), whereas in other images it appeared to involve the apical myocardium (Figure 1A3 and A4). As discussed later, epicardial collagen deposition in the setting of apical myocardial regeneration should not be considered a sign of incomplete regeneration. More recently, Woo and colleagues²⁶ conducted another elegant apical resection study of P1 mice and observed that initial fibrosis is replaced with cardiomyocytes over time, and by 30 days postamputation the heart displayed “minimal fibrosis” similar to levels in the sham controls. Intriguingly, they observed coronary vessel migration toward the apical thrombus prior to migration of cardiomyocytes, highlighting an important role of angiogenesis in cardiac regeneration. This observation underscores an important and rarely studied aspect of neonatal heart regeneration, which is mechanisms of coronary vessel regeneration.²⁶ Another recent study²⁷ confirmed that apical resection and myocardial infarction in neonatal mice induce an intrinsic cardiac regenerative response within the first week of life.⁵ Collectively, these studies consistently support the notion that cardiac injury by apical resection in P1 hearts results in cardiac regeneration.

Figure 1. Models of Neonatal Heart Regeneration.

Although cardiac regeneration after neonatal mouse apical resection has been reproduced by the groups outlined above,^{5, 8–21, 25–33} Andersen and colleagues were unable to observe cardiac regeneration following apical resection surgery.^{22, 23} In response to this discrepancy, a letter by multiple independent principal investigators identified a number of potential discrepancies as possible reasons for why Andersen and colleagues were unable to observe cardiac regeneration.⁹

Consensus in Neonatal Heart Regeneration Post Apical Resection

Unanimous consensus in complex biological systems is highly unlikely, because of our lack of understanding of confounding factors such as strain and background differences, environmental, dietary and circadian factors, in addition to surgical, pharmacological, and operator variability. So although it is quite understandable, and expected, that experimental variations will occur, for example the residual collagen deposition noted by the Sampaio-Pinto²⁵ and Lee¹⁰ groups, the Andersen and Sheikh group received some attention with a claim that neonatal heart regeneration, and even induction of cardiomyocyte proliferation, were not observed. In addressing this discrepancy, we opted to let the literature decide; 4 years and more than 20 reports later, we feel that it is time to address this issue.

To better understand the basis for discrepant findings by the Andersen and Sheikh groups, we closely examined their neonatal regeneration reports. In their most recent report, Andersen and Sheikh and colleagues claimed that apical resection in rats and mice increased the rate of binucleation, but not proliferation of cardiomyocytes.²⁴ In apically resected P1 rats and mice they observed a significant induction of cardiomyocyte proliferation with more than twice the number of phospho-histone H3 (mitotic) positive cardiomyocytes compared with sham hearts, which was not observed at P6 (a phenomenon which was observed in the initial report by Porrello and colleagues). Subsequently, using anillin and aurora b kinase as markers of cytokinesis, they concluded that following apical resection, cardiomyocytes undergo binucleation without cytokinesis, which is supportive of their earlier observation that neonatal heart regeneration does not occur.²⁴

These results are puzzling for several reasons; not only do the results contradict the consensus in the literature, but they also contradict earlier claims by the same group. Andersen and colleagues initially concluded in 2014 that their data “do not support enhanced cardiomyocyte proliferation 1–7 days following AR” (apical resection) based on EdU labelling ²² where they observed no increase in EdU labelling of cardiomyocytes. In contrast, in their latest report, based on phospho-histone H3 staining, Andersen and colleagues now claim that “AR resulted in a statistically significant and nearly 2-fold increase in the mean number of cardiomyocytes in mitosis at P3 (~5% to ~13% per field),” but that this was not associated with cytokinesis and assumed to have undergone binucleation instead ²⁴. How is it possible to observe no increase in EdU incorporation (in their first report) in the same cell population that now has more than 2-fold induction in mitosis markers? Histone H3 is a core histone protein forming a major constituent of chromatin. Serine-10 and serine-28 of histone H3 are phosphorylated during mitosis, marking the G2/M transition and M phase of cell cycle. Specifically, phospho-Histone H3 marks early prophase through metaphase, anaphase and telophase stages of cell cycle.³⁴ Because S-phase, where DNA replication occurs (and thus EdU labelling) precedes mitosis, it is difficult to understand how cardiomyocytes could enter M-phase, without having undergone DNA replication in S-phase. Although it is unclear why the Andersen and Sheikh group did not observe cardiomyocyte proliferation in their reports, their methods for assessment of cardiomyocyte proliferation may have played a role. Certainly, localization of mitotic events to cardiomyocytes is difficult and requires substantial training, examining multiple indices of proliferation and utilizing z-stacking confocal microscopy. To decrease the incidence of discrepancies resulting from errors in assessment of cardiomyocyte mitosis, the field as a whole would benefit from standardization of an array of methods used for cardiomyocyte mitosis assessment, such as using genetic fate mapping models, flow cytometry, and cardiomyocyte nuclear markers in conjunction with the methods currently used.

Apart from the lack of internal consistency between reports by the Andersen and Sheikh group as outlined earlier, it is important to attempt to understand why this specific group is the only group that consistently failed to observe regeneration. Although it is impossible to determine the specific differences in surgical technique and data analysis, it is clear that there are several fundamental differences between reports by the Andersen and Sheikh group, and other investigators. First, reports by the Andersen and Sheikh group resected a much larger segment of the myocardium compared to our group and others. In their 2014 report, Andersen et al noted that they resected the same amount of myocardium as the initial apical resection report, however their published methods indicate that they used heart weight/body weight and ventricular “length” to examine the degree of resection. In contrast, our group and others used ventricular weight and surface area for quantification of degree of resection. This critical difference in methodology is clearly reflected in their Figure 1E where ventricular weights are presented.²² The first bar graph demonstrates that the difference between ventricular weights in sham versus resected hearts is more than 30%. This indicates that according to the original and subsequent reports, the Andersen and Sheikh group resected more than 2 times the amount of left ventricular myocardium. Although possibly unintentional, these results suggest that there is a limit to the amount of myocardium that can be regenerated following injury of the immature mammalian heart. The injury size concept is supported by other groups who confirmed the reproducibility of the apical resection model in neonatal mice, and based on their observation speculated that the conflicting reports by the Andersen and Sheikh group may be attributable to resection of “much more myocardium” than the 15% reported by Sadek and colleagues.¹³ Similarly, the Lee group confirmed the correlation between injury size, and the ability of the neonatal heart to regenerate after apical resection.²¹

Another important technical difference that may have influenced the degree of resection, and caused cardiac injury, is what is known as retraction, or mechanically fixing the apex of the heart prior to resection. This technique was used by the Andersen and Sheikh group,²² and has been subsequently demonstrated by the Lee group to result in substantial cardiac injury and persistent fibrosis, even in the absence of apical resection.²¹ Finally, the very definition of what constitutes regeneration may have resulted in some of the discrepant results. For example, in Figure S3 the initial report by the Andersen group,²² several sections show minimal epicardial or interstitial collagen deposition, with complete anatomic regeneration and recovery of the apical structure. In our view, the mere existence of collagen in a regenerated myocardium does not indicate lack of regeneration. Specifically, our group and others have demonstrated that extracellular matrix (ECM) deposition is a part of the normal regenerative response in the neonatal heart (Porrello et al, Science 2011 – Figure 1 I-L).⁵ Therefore, whereas the absence of myocardium in conjunction with persistence of excessive fibrosis suggests lack of regeneration, the mere presence of any epicardial or interstitial collagen in our view may suggest incomplete clearance of ECM, remnant ECM that has been pushed towards the epicardial surface as the myocardium grows, or adhesion areas resulting from attachment of the injured mouse heart to the chest wall³⁵, and not necessarily incomplete regeneration. In conclusion, despite the technically challenging nature of the apical resection surgery, there is an overwhelming consensus by multiple groups indicating that apical resection of the neonatal mammalian heart induces cardiomyocyte proliferation, and results in heart regeneration.

Neonatal Mouse Ischemic Myocardial Infarction Model

Although the apical resection model continues to be useful for studies of heart regeneration, induction of ischemic myocardial infarction by ligation of the left anterior descending (LAD) coronary artery at P1 was first developed by our group, and has become an important, clinically relevant model of neonatal cardiac injury.³⁶ LAD coronary ligation at P1 leads to robust cardiac regeneration mediated by proliferation of pre-existing cardiomyocytes and clearance of fibrotic scar by 21 days postinjury. In contrast, LAD coronary artery ligation at P7 and P14 did not regenerate and exhibited fibrotic scarring consistent with cardiac injury by apical resection.⁵ This study also revealed that miR-195 inhibits cardiac regeneration and is downregulated after cardiac injury at P1.³⁶ Shortly after the online publication of this article, Haubner and colleagues also independently developed and conducted LAD artery ligation in P1 mice which resulted in complete cardiac regeneration mediated by proliferation of cardiomyocytes with no scarring 7 days postinjury, and continued to have normal heart function at 3 months.³⁷ In contrast, LAD coronary artery ligation at P7 led to scarring similar to the adult heart response and did not regenerate.

This LAD coronary artery ligation procedure was later published in a Nature Protocols report, along with the apical resection models⁸. In one of the earlier studies, our group demonstrated that Meis1 expression decreases during the regenerative phase following LAD coronary ligation at P1, and overexpression of Meis1 in cardiomyocytes inhibits cardiac regeneration after LAD coronary artery ligation at P1.³⁸ In contrast, inhibition of Meis1 in cardiomyocytes extends the cardiomyocyte proliferation window beyond P7. In another study, Olson and colleagues showed that depletion of macrophages at P1 inhibited cardiac regeneration compared with control mouse hearts, which completely regenerated.²⁸

Leor and colleagues were able to observe complete cardiac regeneration after apical resection surgery, however LAD coronary ligation of P1 mice led to incomplete regeneration of the heart.¹¹ Notably, the scar was observed around the site of the LAD ligature, rather than involving the entire LAD territory, which suggests that at least partial regeneration has occurred around the permanent ligature. Lee and colleagues showed that LAD ligation in mice which had undergone vagotomy resulted in fibrosis, no regrowth and decreased cardiomyocyte proliferation, compared with mice that underwent LAD coronary artery ligation only at P1 which completely regenerated by 21 days post injury.¹⁰

To help clarify the procedure further, another independent group³⁹ generated a video protocol of LAD coronary artery ligation in P1 mice. This group was able to consistently observe complete cardiac regeneration by 21 days post-myocardial infarction in P1 mice.³⁹

Similarly, Ai and colleagues conducted a study inducing myocardial infarction by ligation of the LAD coronary artery on P2 mice and found that EZH1 depletion impaired cardiac regeneration (decreased cardiomyocyte proliferation and left ventricular systolic function) compared with control mice which regenerated 3 weeks post MI.⁴⁰ Furthermore, overexpression of EZH1 extended the cardiac regenerative window beyond P7, because P10 mice with overexpression of EZH1 exhibited less fibrosis, increased cardiomyocyte proliferation and improved cardiac function compared with control P10 mice which were unable to regenerate.⁴⁰

Furthermore, Sereti and colleagues outlined a detailed protocol with accompanying videos in a recent publication.⁴¹ This LAD coronary artery ligation model induces infarct size of approximately 10–15% in the left ventricle. Cardiomyocyte death (assessed by TUNEL at 1 day postinjury), regeneration, and minimal fibrosis (3 weeks after cardiac injury) were observed in mice that underwent LAD coronary artery ligation at P0.5 (10–12 hours after birth). In contrast and consistent with our previous studies, P7 mice that underwent LAD coronary artery ligation did not regenerate and developed extensive cardiac fibrosis. Key potential pitfalls from LAD coronary artery ligation to study cardiac regeneration may likely stem from the critical need for experienced operators because the LAD coronary artery is sometimes difficult to visualize. In addition, microsurgery on the small heart size during the neonatal stage requires intensive training and skill in order to be able to generate reproducible and reliable results. A list of publications that conducted LAD coronary artery ligation on P1 mice are listed in Table 2.

Table 2. Studies that performed myocardial infarction in P1 mice.

. .

Reference	Authors	Year	Journal	Model	Cardiac Regeneration?
Complete cardiac regeneration in a mouse model of myocardial infarction. 37	Haubner, B. J., Adamowicz-Brice, M., Khadayate, S., Tiefenthaler, V., Metzler, B., Aitman, T. & Penninger, J. M.	2012	Aging (Albany NY)	LAD coronary artery ligation	Yes
Regulation of neonatal and adult mammalian heart regeneration by the miR-15 family.36	Porrello, E. R., Mahmoud, A. I., Simpson, E., Johnson, B. A., Grinsfelder, D., Canseco, D., Mammen, P. P., Rothermel, B. A., Olson, E. N. & Sadek, H. A.	2013	Proceedings of the National Academy of Sciences	LAD coronary artery ligation	Yes
Meis1 regulates postnatal cardiomyocyte cell cycle arrest. 38	Mahmoud, A. I., Kocabas, F., Muralidhar, S. A., Kimura, W., Koura, A. S., Thet, S., Porrello, E. R. & Sadek, H. A.	2013	Nature	LAD coronary artery ligation	Yes
Surgical models for cardiac regeneration in neonatal mice. 8	Mahmoud, A. I., Porrello, E. R., Kimura, W., Olson, E. N. & Sadek, H. A.	2014	Nat. Protocols	LAD coronary artery ligation	Yes
Macrophages are required for neonatal heart regeneration.28	Aurora, A. B., Porrello, E. R., Tan, W., Mahmoud, A. I., Hill, J. A., Bassel-Duby, R., Sadek, H. A. & Olson, E. N.	2014	The Journal of Clinical Investigation	LAD coronary artery ligation	Yes
The Type of Injury Dictates the Mode of Repair in Neonatal and Adult Heart. 11	Konfino, T., Landa, N., Ben‐Mordechai, T. & Leor, J.	2015	Journal of the American Heart Association	LAD coronary artery ligation	Incomplete
Nerves Regulate Cardiomyocyte Proliferation and Heart Regeneration. 10	Mahmoud, Ahmed I., O’meara, Caitlin C., Gemberling, M., Zhao, L., Bryant, Donald M., Zheng, R., Gannon, Joseph B., Cai, L., Choi, W.-Y., Egnaczyk, Gregory F., Burns, Caroline E., Burns, C. G., Macrae, Calum A., Poss, Kenneth D. & Lee, Richard T.	2015	Developmental Cell	LAD coronary artery ligation	Yes
Myocardial Infarction in Neonatal Mice, A Model of Cardiac Regeneration. 39	Blom, J. N., Lu, X., Arnold, P. & Feng, Q.	2016	J Vis Exp	LAD coronary artery ligation	Yes
A reproducible protocol for neonatal ischemic injury and cardiac regeneration in neonatal mice. 41	Haubner, B. J., Schuetz, T. & Penninger, J. M.	2016	Basic Research in Cardiology	LAD coronary artery ligation	Yes
Divergent Requirements for EZH1 in Heart Development Versus Regeneration. 40	Ai, S., Yu, X., Li, Y., Peng, Y., Li, C., Yue, Y., Tao, G., Li, C., Pu, W. T. & He, A.	2017	Circulation Research	LAD coronary artery ligation (P2 mice)	Yes
Analysis of cardiomyocyte clonal expansion during mouse heart development and injury.41	Sereti K-I, Nguyen NB, Kamran P, Zhao P, Ranjbarvaziri S, Park S, Sabri S, Engel JL, Sung K, Kulkarni RP, Ding Y, Hsiai TK, Plath K, Ernst J, Sahoo D, Mikkola HKA, Iruela- Arispe ML and Ardehali R.	2018	Nature Communications	LAD coronary artery ligation	Yes
Ezh2 is not required for cardiac regeneration in neonatal mice.27	Ahmed A, Wang T and Delgado-Olguin P.	2018	PLOS ONE	LAD coronary artery ligation	Yes

Additional Neonatal Cardiac Regeneration Studies

Apical resection studies beyond P1 have also been examined. Heallen and colleagues performed apical resection at P8 (outside the cardiac regenerative window) in control and hippo-deficient hearts.²⁹ Control hearts exhibited scarring at 21 days postresection, but Hippo-deficient hearts regenerated myocardium and exhibited smaller scar size. This regeneration was mediated by proliferation from pre-existing cardiomyocytes. Furthermore, apical resection performed in P8 Hippo-deficient mouse hearts resulted in further overgrowth of cardiomyocytes in the resected zone (4 days postresection) compared with control mice which did not exhibit this trait.³¹ In the latest study by Martin and colleagues, apex resection in P8 mice did not regenerate in control and Mdx mice (which lack functional dystrophin) by P28.⁴³ In contrast, robust cardiac regeneration was observed in Salvador (Salv) conditional knockout mice, and Salv;Mdx double knockout hearts exhibited excessive cardiomyocyte proliferation at the resected area such that an additional apex was frequently observed, which suggests that the Hippo pathway and dystrophin glycoprotein complex restrict cardiomyocyte proliferation.⁴³

Rui and colleagues confirmed observations of cardiac regeneration from apical resection on neonatal mice.³³ Daily injections of thymosin β4 from P1-P7, followed by apical resection performed at P7 with alternate days of thymosin β4 injections until P19 exhibited cardiac regeneration and minimal fibrosis, compared with control mice which did not regenerate and displayed significant cardiac fibrosis. Cardiac regeneration from a contribution of Wt1+ epicardium-derived cells and islet1 cells was implicated.

Zhang and colleagues assessed the role of RE1 silencing transcription factor (REST) in cardiomyocyte proliferation. ⁴⁴ Using an adapted protocol of the apical resection model from Sadek and colleagues, ⁵ approximately 10% of hearts were amputated in neonatal mice at P4. In control mice, apically resected hearts regenerated by P28. In contrast, mice with deficient REST in cardiomyocytes failed to regenerate and exhibited less cardiomyocyte proliferation and increased fibrosis. The authors concluded that deficiency of REST decreases cardiomyocyte proliferation and impairs cardiac regeneration after neonatal cardiac injury. ⁴⁴

O’Meara and colleagues performed apical resection at P1 and performed RNA sequencing on hearts collected at 1 and 7 days postresection and identified a transcriptional signature of the regenerating heart which seems to revert to a cardiac differentiation state.³⁰ White and colleagues also identified a contribution of the nervous system in cardiac regeneration, showing that apically resected P2 mouse hearts completely regenerated within 21 days, but mice exposed to chemical sympathectomy (with 6-hydroxydopamine) developed fibrosis and failed to regenerate its myocardium.³² Moreover, apex resection (18%) performed on rats at P1 stimulates a nearly complete cardiac regenerative response by 21 days post resection and was associated with low level fibrosis, but apex resection (16%) at P7 failed to regenerate.⁴⁵ This demonstrated that cardiac regeneration can also be induced in neonatal rats.

Other LAD coronary ligation studies beyond P1 have been described. Xin and colleagues induced LAD coronary ligation in P2 wildtype mice which were completely regenerated by P28, but cardiac regeneration was impaired in YAP deficient mouse hearts and led to fibrosis.⁴⁶ LAD coronary ligation of wildtype mice at P7 assessed at P28 exhibited extensive fibrosis and loss of myocardium. In contrast, forced expression of YAP in hearts (aMHC-YapS112A Tg mice) which underwent LAD coronary artery ligation at P7 displayed less fibrosis and increased myocardium secondary to cardiomyocyte proliferation.⁴⁶ Heallen and colleagues performed LAD coronary ligation at P8 on Hippo-deficient mouse hearts, and found that these hearts exhibited less scarring and increased myocardium compared to controls at P21.²⁹

Ischemia-reperfusion injury performed at P21 on mice injected with N-acetylcysteine (a reactive oxygen species scavenger) from P1–21 showed increased cardiomyocyte proliferation, reduced fibrotic scar and improved systolic function compared with control mice, suggesting that the cardiac regenerative window can be extended beyond P7 by scavenging of reactive oxygen species.⁴⁷

The colleagues of Husain and Graham performed LAD coronary artery ligation on P2, P15 and P21 mice: P2 mouse hearts regenerated, P21 hearts did not regenerate and P15 hearts exhibited cardiac regeneration which was an intermediate between LAD coronary artery ligation performed at P2 and P21.⁴⁸ This study suggested that a burst of cardiomyocyte proliferation exists at P15 (preadolescence), although this was not observed by other groups ^49–51.

Yang and colleagues induced myocardial infarction by LAD ligation in P1 mice.⁵² Control mice regenerated and had low expression of miR-34a. However, P1 mice that received injections of miR-34a mimic postinfarction showed decreased cardiomyocyte proliferation and heart regeneration. Chen et al also conducted LAD ligation surgeries on neonatal mice and observed complete cardiac regeneration with minimal fibrosis at P21,⁵³ whereas neonatal periostin knockout mice exhibited cardiac fibrosis.

A series of studies using cryoinjury in neonatal mouse hearts were also conducted (Table 3). Jesty and colleagues showed that cryoinjury by a liquid nitrogen cooled-probe at P1-P3 induced cardiac regeneration requiring 3 months for clearance of fibrotic scar. The authors concluded that this regenerative response was mediated by C-kit cells as well as increase in cardiomyocyte proliferation,⁵⁴ although the use of a C-kit reporter (rather than a fate mapping approach) in this study makes it impossible to accurately quantify the contribution of C-kit cells. In addition, Rubin and colleagues showed that whereas mild cryoinjury induced complete regeneration, severe cryoinjury at P1 did not induce cardiac regeneration.⁵⁵ Furthermore, cryoinjury on P1 mouse hearts did not induce significant cardiomyocyte proliferation or regeneration and lead to scarring.⁵⁶ Aix and colleagues used a cryoinjury model and examined the relationship between telomerase and cardiac regeneration.⁵⁷ Cryoinjury in wild type and generation 3 (G3) Terc^−/− (telomerase deficient) mice (with premature telomere shortening) at P1 led to significant cardiac fibrosis by 7 days post injury in both groups. However, by 28 days postinjury, fibrotic scars in wildtype hearts had decreased, whereas fibrosis was unchanged in G3 Terc^−/− mouse hearts. Whereas cryoinjury to P1 wildtype mouse hearts stimulated cardiomyocyte proliferation, this did not occur in G3 Terc^−/− mouse hearts, which instead exhibited cardiomyocyte hypertrophy. The contribution of cell cycle arrest protein p21 in neonatal cardiac regeneration was also examined. Malek Mohammadi and colleagues⁵⁸ used a cardiac cryoinjury model in neonatal mice and observed cardiac regeneration, with Gata4 enhancing this process. Cardiac regeneration post cryoinjury was incomplete (small scar present at 60 days post injury), which is consistent with cardiac cryoinjury models. Finally, Xiao and colleagues showed that cardiac injury in P5 mice by cryoinjury, apical resection and left coronary artery ligation all regenerated 1 month later.⁵⁹ These studies are listed in Table 3.

Table 3. Additional Neonatal Cardiac Regeneration Studies.

. .

Reference	Authors	Year	Journal	Model	Cardiac Regeneration?
Hippo signaling impedes adult heart regeneration. 29	Heallen, T., Morikawa, Y., Leach, J., Tao, G., Willerson, J. T., Johnson, R. L. & Martin, J. F.	2013	Development	Apical Resection	Yes
Extending the time window of mammalian heart regeneration by thymosin beta 4. 33	Rui, L., Yu, N., Hong, L., Feng, H., Chunyong, H., Jian, M., Zhe, Z. & Shengshou, H.	2014	Journal of Cellular and Molecular Medicine	Apical Resection	Yes
Transcriptional Reversion of Cardiac Myocyte Fate During Mammalian Cardiac Regeneration. 30	O’meara, C. C., Wamstad, J. A., Gladstone, R. A., Fomovsky, G. M., Butty, V. L., Shrikumar, A., Gannon, J. B., Boyer, L. A. & Lee, R. T.	2015	Circulation Research	Apical Resection	Yes
Actin cytoskeletal remodeling with protrusion formation is essential for heart regeneration in Hippo-deficient mice. 31	Morikawa, Y., Zhang, M., Heallen, T., Leach, J., Tao, G., Xiao, Y., Bai, Y., Li, W., Willerson, J. T. & Martin, J. F.	2015	Science Signaling	Apical Resection	Yes
Sympathetic Reinnervation Is Required for Mammalian Cardiac Regeneration. 32	White, I. A., Gordon, J., Balkan, W. & Hare, J. M.	2015	Circulation Research	Apical Resection	Yes
REST regulates the cell cycle for cardiac development and regeneration.43	Zhang, D., Wang, Y., Lu, P., Wang, P., Yuan, X., Yan, J., Cai, C., Chang, C.- P., Zheng, D., Wu, B. & Zhou, B.	2017	Nature Communicati ons	Apical Resection	Yes
Early postnatal rat ventricle resection leads to long‐term preserved cardiac function despite tissue hypoperfusion. 44	Zogbi, C., Saturi De Carvalho, A. E. T., Nakamuta, J. S., Caceres, V. D. M., Prando, S., Giorgi, M. C. P., Rochitte, C. E., Meneghetti, J. C. & Krieger, J. E.	2014	Physiological Reports	Apical resection in rats	Yes
Dystrophin glycoprotein complex sequesters Yap to inhibit cardiomyocyte proliferation. 42	Morikawa, Y., Heallen, T., Leach, J., Xiao, Y. & Martin, J. F.	2017	Nature	Apical Resection	Yes
Hippo pathway effector Yap promotes cardiac regeneration. 45	Xin, M., Kim, Y., Sutherland, L. B., Murakami, M., Qi, X., Mcanally, J., Porrello, E. R., Mahmoud, A. I., Tan, W., Shelton, J. M., Richardson, J. A., Sadek, H. A., Bassel-Duby, R. & Olson, E. N.	2013	Proceedings of the National Academy of Sciences	LAD coronary artery ligation	Yes
Hippo signaling impedes adult heart regeneration. 29	Heallen, T., Morikawa, Y., Leach, J., Tao, G., Willerson, J. T., Johnson, R. L. & Martin, J. F.	2013	Development	LAD coronary artery ligation	Yes
The Oxygen- Rich Postnatal Environment Induces Cardiomyocyte Cell-Cycle Arrest through DNA Damage Response. 46	Puente, Bao N., Kimura, W., Muralidhar, Shalini A., Moon, J., Amatruda, James F., Phelps, Kate L., Grinsfelder, D., Rothermel, Beverly A., Chen, R., Garcia, Joseph A., Santos, Celio X., Thet, S., Mori, E., Kinter, Michael T., Rindler, Paul M., Zacchigna, S., Mukherjee, S., Chen, David J., Mahmoud, Ahmed I., Giacca, M., Rabinovitch, Peter S., Asaithamby, A., Shah, Ajay M., Szweda, Luke I. & Sadek, Hesham A.	2014	Cell	Ischemia reperfusion injury	No at P21. Yes if treated with NAC from birth.
A Proliferative Burst during Preadolescence Establishes the Final Cardiomyocyte Number. 47	Naqvi, N., Li, M., Calvert, John W., Tejada, T., Lambert, Jonathan P., Wu, J., Kesteven, Scott H., Holman, Sara R., Matsuda, T., Lovelock, Joshua D., Howard, Wesley W., Iismaa, Siiri E., Chan, Andrea Y., Crawford, Brian H., Wagner, Mary B., Martin, David I. K., Lefer, David J., Graham, Robert M. & Husain, A.	2014	Cell	LAD coronary artery ligation	Yes
MicroRNA-34a Plays a Key Role in Cardiac Repair and Regeneration Following Myocardial Infarction. 51	Yang, Y., Cheng, H.-W., Qiu, Y., Dupee, D., Noonan, M., Lin, Y.-D., Fisch, S., Unno, K., Sereti, K.-I. & Liao, R.	2015	Circulation Research	Ligation of main branch coronary artery.	Yes
Ablation of periostin inhibits post- infarction myocardial regeneration in neonatal mice mediated by the phosphatidylinositol 3 kinase/glyocogen synthase kinase 3β/cyclin D1 signalling pathway. 52	Chen, Z., Xie,J., Hao, H., Lin, H., Wang, L., Zhang, Y., Chen, L., Cao, S. Huang, X., Liao, W., Bin, J., Liao, Y.	2017	Cardiovascular Research	Ligation of main branch coronary artery.	Yes
c-kit+ precursors support postinfarction myogenesis in the neonatal, but not adult, heart. 53	Jesty, S. A., Steffey, M. A., Lee, F. K., Breitbach, M., Hesse, M., Reining, S., Lee, J. C., Doran, R. M., Nikitin, A. Y., Fleischmann, B. K. & Kotlikoff, M. I.	2012	Proceedings of the National Academy of Sciences	Cryoinjury	Yes
FGF10 Signaling Enhances Epicardial Cell Expansion during Neonatal Mouse Heart Repair. 54	Rubin, N., Darehzereshki, A., Bellusci, S., Kaartinen, V. & Ling Lien, C.	2013	J Cardiovasc Dis Diagn	Cryoinjury	No
Differential regenerative capacity of neonatal mouse hearts after cryoinjury. 55	Darehzereshki, A., Rubin, N., Gamba, L., Kim, J., Fraser, J., Huang, Y., Billings, J., Mohammadzadeh, R., Wood, J., Warburton, D., Kaartinen, V. & Lien, C.-L.	2015	Developmental Biology	Cryoinjury	No
Postnatal telomere dysfunction induces cardiomyocyte cell-cycle arrest through p21 activation. 56	Aix, E., Gutiérrez- Gutiérrez, Ó., Sánchez- Ferrer, C., Aguado, T. & Flores, I.	2016	The Journal of Cell Biology	Cryoinjury	Yes
The transcription factor GATA4 promotes myocardial regeneration in neonatal mice. 57	Malek Mohammadi, M., Kattih, B., Grund, A., Froese, N., Korf- Klingebiel, M., Gigina, A., Schrameck, U., Rudat, C., Liang, Q., Kispert, A., Wollert, K. C., Bauersachs, J. & Heineke, J.	2017	EMBO Molecular Medicine	Cryoinjury	Yes (but small scar present by 60 days post cryoinjury)
A p53-based genetic tracing system to follow postnatal cardiomyocyte expansion in heart regeneration.58	Xiao, Q., Zhang, G., Wang, H., Chen, L., Lu, S., Pan, D., Liu, G. & Yang, Z.	2017	Development	Cryoinjury, Apical Resection, and Left coronary artery ligation.	Yes

Discussion

In the current review, we provide a comprehensive overview of neonatal heart regeneration studies since it was first reported in 2011. Overall, the published reports indicate that there is overwhelming consensus that both apical resection and ischemic myocardial infarction in neonatal mice result in induction of cardiomyocyte proliferation and heart regeneration. In addition, cryoinjury, which results in delayed regeneration in zebrafish, is generally associated with incomplete regeneration in neonatal mice, although nontransmural cryoinjury appears to result in complete regeneration. A number of clear conclusions, and unanswered questions, have emerged from the wide array of studies conducted on the neonatal heart to date; First, neonatal heart regeneration in mice is mediated by proliferation of pre-existing cardiomyocytes, and is lost when cardiomyocytes exit cell cycle shortly after birth⁶⁰. Whether there is a specific population of cardiomyocytes that mediate this regenerative response is not clear. In other words, do all neonatal cardiomyocytes contribute to the regenerative response?

Second, both the size and the type of injury dictate the mode of repair. It is clear from the inadvertent results by Andersen et al, as well as by careful studies by the Lee group, that larger apical resection injuries are not associated with complete regeneration, while smaller injuries result in a robust regenerative response, and complete regeneration. However, whether this is merely a function of inability of the neonatal heart to repair the large injury in time before cell cycle arrest ensues, or whether larger injury fails to induce adequate cardiomyocyte mitosis, is unclear. Intriguingly, a similar phenomenon is seen in zebrafish. For example, although complete regeneration is seen following resection of 20% of the ventricle, zebrafish cannot sufficiently regenerate when the apical resection is greater than 25%.¹

Third, the postnatal regenerative window in both mice and rats appears to be shorter than 1 week after birth, and it is more likely that robust regeneration is mostly seen after injury in the first 2–3 days of life. However, it is unclear whether this phenomenon is also seen in large mammals, and in humans. Anecdotal reports over the past several decades suggest that newborn humans can recover left ventricular function following various degrees of myocardial infarction.^61–63 However, because cardiac injury is often associated with transient contractile dysfunction (stunning or hibernation), and in the absence of the privilege of serial histological analysis that is often used in animal studies, it is impossible to make valid conclusions regarding the regenerative capacity of the human heart without prospective viability studies, coupled with functional and anatomic recovery.

Finally, it is peculiar that the actual definition of cardiac regeneration remains open to interpretation, which creates confusion, particularly for investigators outside the regeneration field. Ventricular regeneration from a histological perspective involves recovery of myocardial wall thickness, with generation of cardiomyocytes of normal size, with normal sarcomeric structure and content, alignment of cardiomyocytes in the correct orientation, normal capillary density and size, generation of coronary vessels with normal size and density, as well as regeneration of the epicardium and endocardium. From anatomic and functional perspectives, regeneration of the ventricle involves recovery of normal ventricular size, architecture and chamber dimensions, accompanied by recovery of systolic function and myocardial contractility. In our view, satisfying the majority of these parameters, when assessing ventricular regeneration, should amount to successful regeneration. There are however a number of other open questions that should be addressed; for example, if ECM deposition is an integral part of the wound healing response seen in the early stages of myocardial regeneration, what does the persistence of epicardial or interstitial ECM mean if the aforementioned parameters are achieved? And what are the clinical implications of this phenomenon? Would this contribute to the development of impaired ventricular relaxation following recovery of systolic function? Or create a nidus for ventricular arrhythmias?

Another important issue is assessment of true myogenesis. From a technical standpoint, future studies should consider adding more rigorous methods for assessment of cardiomyogenesis to the existing repertoire of assays. For example, standardization of the use of z-stacking confocal microscopy to localize mitotic events to cardiomyocytes, immunostaining isolated cardiomyocytes for pH3, Ki67, Aurkb, BrdU, EdU and anillin for FACS analysis, and co-staining with cardiomyocyte-specific nuclear markers, in addition to using conditional cardiomyocyte-specific genetic lineage tracing models such as Myh6CreERT2;MADM-11^GT/GTmice ⁶⁴ and αMHC^CreER;R26^VT2/GK (Rainbow) mice.⁴¹

The cardiac injury models outlined in the current review have been applied to various knockout and transgenic mice and are now being utilized as a standard assay to assess the endogenous regenerative potential of the mammalian heart, and to examine factors that promote or impair cardiac regeneration. For example, studies have used this approach to determine whether proregenerative factors prolong the postnatal window of heart regeneration, which in our experience appears to be a relatively easier task than reactivation of cardiomyocyte proliferation in the adult heart. Another important insight that was gained from neonatal heart regeneration studies is the fact that endogenous myocardial regeneration in mammals is mediated by proliferation of pre-existing cardiomyocytes, rather than an extracardiac or resident cardiac stem or progenitor cells. These insights have important translational implications if we aim to utilize the innate mechanisms of neonatal heart regeneration to induce regeneration in adult mammals.

Scientists are sceptics by nature, which is the basis of successful peer review, and the single most important driver of scientific progress. As such, the neonatal heart regeneration phenomenon has been critiqued from several angles; A “so what” argument has been made given the known proliferative property of neonatal cardiomyocytes. In other words, it should not be surprising that the neonatal heart can regenerate when cardiomyocytes can still divide. However, this phenomenon is not simply haphazard cardiomyocyte proliferation, but rather a full regenerative response as outline earlier. A “what does this really mean?” argument was also made, questioning whether this phenomenon really amounts to regeneration since it occurs during an organ growth phase. This is more of a philosophical question in our view, because other regenerative organisms, such as zebrafish, display a phenomenon known as indeterminate growth⁶⁵, which means that the organism can continue to grow, albeit slowly, throughout it’s lifespan depending on nutrient and space availability. If regeneration in zebrafish occurs in the setting of continual growth capacity, does this then mean that regeneration in zebrafish is not “true” regeneration? Finally, there is of course the “no way” argument such as that made by Andersen and Sheikh and their colleagues, questioning whether this neonatal regeneration or even cardiomyocyte proliferation actually happens in the first place. Although this question is now settled, it is essential that conceptual, technical and mechanistic questions continue to be pursued if the cardiac regeneration field is to ever reach a stage of therapeutic applications. As such, the neonatal regenerative response appears to be modifiable by the degree and type of injury, as well as any number of factors including technical, genetic and environmental factors etc. In a recent editorial, Gerald Dorn commented on apical resection studies by the Lee group, which reproduced the regenerative response, and outlined the pitfalls of injury size discrepancy and other technical consideration by reminding us that “Biology is not Binary”.⁶⁶

Clinical Perspective.

What is new?

• Neonatal mice in the first week of life can regenerate their hearts after injury by apical resection or myocardial infarction, and this process is mediated by proliferation of pre-existing cardiomyocytes.

• Neonatal mouse cardiac injury models have been replicated by many groups and continue to be useful models for identifying factors that impair or promote cardiac regeneration.

• Factors that promote neonatal cardiomyocyte proliferation can prolong the postnatal regenerative window, and may reverse cell cycle arrest in adult cardiomyocytes.

What are the clinical implications?

• Promoting adult cardiomyocyte proliferation can regenerate the myocardium and restore cardiac function in heart failure patients.

• Anecdotal reports suggest that the neonatal human heart may be able to regenerate after injury, although these observations have relied on surrogate measures rather than on true viability assessment.