Ctnna3 Deciency Promotes Heart Regeneration by Enhancing Cardiomyocyte Proliferation in Neonatal Mice

Heart regeneration requires renewal of lost cardiomyocytes. However, the mammalian heart loses its proliferative capacity soon after birth, and the molecular signaling underlying the loss of cardiac proliferation postnatally is not fully understood. Here we report that ablation of Ctnna3, coding for an αT-catenin protein and highly expressed in hearts, accelerated heart regeneration following heart apex resection in neonatal mice. Our results show that Ctnna3 deciency enhances cardiomyocyte proliferation in hearts from P7 mice by upregulating Yap expression. Our study demonstrates that Ctnna3 deciency is sucient to promote heart regeneration and cardiomyocyte proliferation in neonatal mice and indicates that functional interference of α-catenins might help to stimulate myocardial regeneration after injury.


Introduction
Heart failure resulting from the loss of cardiomyocytes during injury and disease is a primary cause of human morbidity and mortality [1]. The adult mammalian heart only has limited cardiomyocyte turnover, and this is not su cient to restore contractile function after heart injury. In contrast, the neonatal mammalian heart possesses a remarkable capacity for cardiac regeneration, but this heart regeneration ability after injury is lost by 7 days of age, a time point that coincides with the onset of cardiomyocyte proliferative arrest [2][3][4]. Multiple experimental strategies have been explored to restore functional myocardium for repairing the injured heart. These are: (a) cell therapy using embryonic stem cells, induced pluripotent stem cells (iPS) and cardiac progenitor cells; (b) reprogramming of nonmyocytes, e.g. cardiac broblasts, to a cardiac cell fate (cardiomyocytes) using cardiogenic genes and small molecules; (c) re-activation of cardiomyocyte mitosis in the adult heart [5][6][7][8]. Genetic fate-mapping experiments in the neonatal mice [3] and adult zebra sh [9,10] indicate that the regenerated cardiomyocytes are mainly derived from reactivation of preexisting cardiomyocytes, rather than activation of undifferentiated stem or progenitor cells. Thus, the identi cation of genes and a thorough understanding of the mechanisms underlying the regulation of cardiomyocyte proliferation and regeneration may provide critical information for design of new therapeutic strategies to heart failure.
Here we present the experimental evidence that Ctnna3 de ciency solely is su cient to promote heart regeneration following heart apex resection in neonatal mice. Our study revealed that cardiomyocyte proliferation was increased in neonatal Ctnna3 de cient mice, probably by up-regulating Yap expression. These results suggest that αT-catenin does contribute to regulation of heart regeneration and cardiomyocyte proliferation in neonatal mice.

Isolation And Culture Of Neonatal Mouse Cardiomyocytes
Cardiomyocytes from neonatal mice were isolated as previously described [19]. Brie y, hearts from neonatal mice at postnatal day 1 (P1) were disassociated with collagenase II. The disassociated cells were plated in a 10cm plate for 2 hours with DMEM/F12 medium, and the supernatant was collected and cells were re-plated on laminin-coated glass coverslips in 12-well plates at 3x10 5 cells per well. On the following day, the proliferating cells were labeled by EdU in fresh medium for 24 hours and detected using Cell-light EdU Apollo 643 In Vitro Kit (#C10310-2, RiboBio, Guangzhou, China) following the manufacturers' instructions.

Neonatal Mouse Apical Resection
Neonatal mouse apical resection of hearts from P1 pups was performed as described previously [21].

Histological And Immuno uorescent Analyses
Tissue processing, frozen sections and immuno uorescent microscopic analysis were performed as previously described [22]. Brie y, for frozen sections, mouse hearts were dissected out, xed with 4% PFA in PBS overnight, dehydrated with 30% sucrose at 4℃ for 3 days and then embedded in OCT (Richard-Allan Scienti c). Sections were collected at 10µm.
For histological analysis, mouse hearts were para nized and were sectioned at 4 µm followed by hematoxylin/eosin (H&E) staining or Masson's trichrome (MT) staining (G1340, Solarbio, Beijing, China) as previously described [23].For quanti cation of cardiac brosis, images of the MT stained sections were captured with Leica Aperio VERSA microscope (Leica Biosystems, Germany) and the area of brosis in heart apex was determined using Visiopharm software (Visiopharm, Horsholm, Denmark).
For in vivo EdU (5-ethynyl-2′-deoxyuridine) assay, WT or Ctnna3 −/− mice at P2 were injected intraperitoneally with 50mg/kg EdU. Twenty four hours after injection, the mouse hearts were dissected out and processed for frozen section. EdU assay was performed using the Kit (#C10310-2, RiboBio, Guangzhou, China) following the manufacturers' instructions.

Western Blotting
Standard Western blot protocol was followed. Protein extracts were prepared with RIPA buffer and then subjected to SDS-PAGE. The antibodies against the following proteins were used in our study: αΤ-catenin (#13974-1-AP, Proteintech, USA), β-catenin(#51067-2-AP, Proteintech, USA), and Yap (14074T, Cell Signaling, USA). The protein bands were detected with an ECL Western Blotting Analysis System. The images were obtained by Tanon-5200, and the density of bands was determined with Image J.

Statistical analysis
Unpaired Student's t-test by GraphPad Prism was conducted for statistical analysis.

Results
Loss of Ctnna3 accelerates heart regeneration in neonatal mice after heart apex resection.
Although the mammalian adult heart is generally considered nonregenerative, neonatal mouse hearts have a genuine capacity to regenerate following apex resection [1,3]. To evaluate whether Ctnna3 affects heart regeneration, we performed surgical apical resection of hearts (5%~10% of the ventricular myocardium) of WT and Ctnna3 −/− neonatal mice at P1 and harvested hearts at 7 and 14 day(s) post-resection (dpr) (Fig. 1A) for histological analysis and immuno uorescence staining with antibody against PCNA, separately. The results revealed that, as reported by Porrello [24], the resection plane was characterized by progressive regeneration of the apex with some restoration of the resected myocardium within 14 days (Fig. 1B). The Masson's trichrome staining showed that the accumulation of brotic tissue (blue staining in Fig. 1C) in hearts from Ctnna3 −/− mice at 14dpr dramatically decreased compared with WT mice (Fig. 1C and D). The number of PCNA-positive cells in the border zone of regenerated hearts of Ctnna3 −/− neonatal mice at 7 dpr was signi cantly higher than the counterpart in hearts of WT neonatal mice ( Fig. 1E and F). As heart regeneration is thought to occur primarily through cardiomyocyte proliferation [25, 10], our results suggest that loss of Ctnna3 may enhance heart regeneration by promoting cardiomyocyte proliferation in neonatal mice.

Ctnna3 de ciency promotes cell proliferation in neonatal mouse hearts
Since cardiomyocytes in neonatal mouse heart retain active proliferation before postnatal day 7 [26], we tested whether Ctnna3 de ciency promoted cardiomyocyte proliferation in neonatal mice. The ventricles of WT and Ctnna3 −/− mice at P1, P3, P7 were sectioned and stained separately with antibodies against phospho-histone H3 (PH3) (a marker of mitosis)[27], and Sarcomeric α-actinin (a speci c marker for αskeletal and α-cardiac muscle actinins) [28]. Quanti cation of PH3-positive cardiomyocytes revealed that cardiomyocyte proliferation was strikingly increased in the ventricles of Ctnna3 −/− neonatal mice at P3 and P7 ( Fig. 2A, B), but not at P14 compared to WT counterparts.
To further verify the enhanced cell proliferation in Ctnna3 −/− neonatal mice, the proliferating cells in Ctnna3 −/− and WT neonatal mice at P7 were in vivo EdU-pulse labeled and detected by EdU staining (see Materials and Methods). The results showed that the number of EdU-positive cells was signi cantly increased in ventricles and atriums of Ctnna3 −/− neonatal mice compared to those in control littermates ( Fig. 2C-F). These results demonstrate that Ctnna3 de ciency promotes cell proliferation in neonatal mouse hearts.

Ctnna3 de ciency promotes proliferation of primary cardiomyocytes
Since heart is mainly composed of cardiomyocytes, in addition to several other types of cells, such as cardiac broblasts, endothelial cells, and smooth muscle cells [29], we hypothesized that Ctnna3 de ciency promoted the proliferation of cardiomyocytes. To test this hypothesis, primary cardiomyocytes were isolated from P5 WT and Ctnna3 −/− mouse hearts and pulse-labelled with EdU in vitro followed by immunostaining with antibody against sarcomeric α-actinin and EdU stainng. Our study revealed that a signi cant increase of the proportion of the EdU-positive cardiomyocytes (proliferating cardiomyocytes) from Ctnna3 −/− neonatal mice compared to that from WT mice at P5 (Fig. 3A and B). These results demonstrate that the loss of Ctnna3 promotes proliferation of cardiomyocytes in neonatal mice.

Loss of Ctnna3 enhances Yap expression
Alpha-catenins directly bind to both β-catenin and actin laments, thereby coupling stable actin laments to the cadherin adhesion molecules [11,16,13]. β-catenin is also a well-documented positive regulator for cell proliferation. To study the mechanism(s) by which αT-catenin de ciency promotes cell proliferation, the expression level of β-catenin in hearts from Ctnna3 −/− and WT mice at P7 was evaluated by Western blotting. The result showed that the protein level of β-catenin was not signi cantly changed between WT and Ctnna3 −/− hearts (Fig. 3C and D).
The Hippo pathway is a key regulatory signaling pathway for heart development and organ size [30]. Thus, we inspected whether loss of Ctnna3 only had any effect on Yap expression in the neonatal heart.
Our study revealed that the protein level of Yap signi cantly increased in hearts from P7 Ctnna3 −/− mice compared to that in control mouse heats (Fig. 3C and D). This result suggests that Ctnna3 de ciency may promote neonatal cardiomyocyte proliferation by enhancing Yap expression.

Discussion
Previous study showed that cardiac-speci c Ctnna1 (αE-catenin) de cient adult mice displayed progressive dilated cardiomyopathy [17] and Ctnna3 (αT-catenin) de cient adult mice exhibited dilated cardiomyopathy only after acute ischemia[18], but without affecting the number of cardiomyocytes. Li, et al., also demonstrated that simultaneously cardiac-speci c deleting Ctnna1 and Ctnna3 in mice at perinatal stage resulted in an increased cardiomyocyte number and cardiomyocyte proliferation in the postnatal heart [19]. Although these results strongly suggest that Ctnna1 and Ctnna3 jointly play an important but redundant role in the inhibition of cardiomyocyte proliferation, whether Ctnna3 contributes to heart regeneration or Ctnna3 is adequate to suppress cardiomyocyte proliferation at neonatal stage needs to be addressed. Here we present in vivo evidences that Ctnna3 de cient only is su cient to enhance heart regeneration and cardiomyocyte proliferation in neonatal mice. Hence, functional interference of α-catenins may provide a potential strategy to promote myocardial regeneration after injury.

Declarations Declaration of competing interest
The authors declare that they have no con icts of interest with the contents of this article.

Figure 3
Ctnna3 de ciency promotes proliferation of primary cardiomyocytes and up regulates YAP expression. A.
Representative uorescent images of cultured primary mouse cardiomyocytes from neonatal WT and Ctnna3-/-mice. The cultured cells were pulse-labeled with EdU for 24 hrs followed by uorescent staining.