ARTICLE Corrected: Author correction DOI: 10.1038/s41467-017-02210-y OPEN REST regulates the cell cycle for cardiac development and regeneration Donghong Zhang1, Yidong Wang1, Pengfei Lu1, Ping Wang1, Xinchun Yuan2, Jianyun Yan3, Chenleng Cai3, Ching-Pin Chang4, Deyou Zheng 5, Bingruo Wu1 & Bin Zhou 6,7 Despite the importance of cardiomyocyte proliferation in cardiac development and regen- eration, the mechanisms that promote cardiomyocyte cell cycle remain incompletely 1234567890 understood. RE1 silencing transcription factor (REST) is a transcriptional repressor of neu- ronal genes. Here we show that REST also regulates the cardiomyocyte cell cycle. REST binds and represses the cell cycle inhibitor gene p21 and is required for mouse cardiac development and regeneration. Rest deletion de-represses p21 and inhibits the cardiomyocyte cell cycle and proliferation in embryonic or regenerating hearts. By contrast, REST overexpression in cul- tured cardiomyocytes represses p21 and increases proliferation. We further show that p21 knockout rescues cardiomyocyte cell cycle and proliferation defects resulting from Rest deletion. Our study reveals a REST-p21 regulatory axis as a mechanism for cell cycle pro- gression in cardiomyocytes, which might be exploited therapeutically to enhance cardiac regeneration. 1 Department of Genetics, Albert Einstein College of Medicine, Bronx, NY 10461, USA. 2 Department of Medical Ultrasound, The First Affiliated Hospital of Nanchang University, Nanchang 330006, China. 3 Department of Developmental and Regenerative Biology, The Black Family Stem Cell Institute, and The Mindich Child Health and Development Institute, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA. 4 Department of Medicine, Indian University School of Medicine, Indianapolis, IN 46202, USA. 5 Departments of Genetics, Neurology and Neuroscience, Albert Einstein College of Medicine, Bronx, NY 10461, USA. 6 Departments of Genetics, Pediatrics, and Medicine (Cardiology), The Wilf Cardiovascular Research Institute, The Institute for Aging Research, Albert Einstein College of Medicine, Bronx, NY 10461, USA. 7 Department of Cardiology of First Affiliated Hospital and State Key Laboratory of Reproductive Medicine, Nanjing Medical University, Nanjing, Jiangsu 210029, China. Correspondence and requests for materials should be addressedto B.Z. (email: [email protected]) NATURE COMMUNICATIONS | 8: 1979 | DOI: 10.1038/s41467-017-02210-y | www.nature.com/naturecommunications 1 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02210-y ardiomyocyte proliferation is required for generating cardiomyocytes to proliferate prevents the replenishment of lost myocardial mass and building a functional four-chamber or dysfunctional cells in a diseased heart11. Because heart diseases C 1–3 11 heart during embryonic development . After birth, car- are the number one cause of death worldwide , it is important to diomyocytes continue to proliferate in a short neonatal period, identify the regulatory factors of the cardiomyocyte cell cycle, which is crucial for the final cardiac growth surge as well as for which may be used as therapeutic targets for these devastating – regeneration of injured mouse neonatal hearts4 6. The vast conditions. majority of cardiomyocytes then exits the cell cycle and stops Several transcription factors, such as GATA412, TBX2013, – proliferating after preadolescence7 10. The inability of BRG114, YAP15,16, ERBB217, PITX218, and MEIS119 have been a b c 100 3 Cre GT/GT mKO Tnt ;Rest Control Control Rest 18 33 12 12 19 Tnt Cre;Rest GT/+ 1.00 ± 0.18 0.20 ± 0.12* 75 44 17 GT/GT DAPI 18 Rest / E12.5 121 KDa 50 11 Rest GT/+ REST 18 REST 17 43 18 / Rest mKO 25 TnT 12 36 KDa Embryos/stage (%) 18 43 17 18 GAPDH 0 E11.5 E12.5 E14.5 E16.5 E18.5 d Control Rest mKO Control Rest mKO e Control E10.5 E12.5 Rest mKO 50 * m) 40 µ 30 * 20 10 Wall thickness ( thickness Wall 0 E10.5 E12.5 f Ki67/TnT/DAPI EdU/TnT/DAPI PH3/TnT/DAPI Aurora B/TnT/DAPI Control E12.5 mKO Rest 60 60 * 4 2.5 * * 50 50 * 3 2.0 40 40 1.5 30 30 2 1.0 20 20 EdU/TnT (%) 1 PH3/TnT (%) KI67/TnT (%) 10 10 0.5 Aurora B/TnT (%) Aurora 0 0 0 0 Control Rest mKO Control Rest mKO Control Rest mKO Control Rest mKO Fig. 1 Rest is required for myocardial development and cardiomyocyte proliferation. a, b Immunostaining and western blot analyses showing the efficient deletion of Rest in cardiomyocytes in TnTCre;RestGT/GT (RestmKO) hearts. Controls were RestGT/GT mice. c A summarizing graph of the distribution and number of embryos at different stages indicates death of Restmko embryos between E12.5 and E16.5. d, e H&E-stained sections of E10.5 and E12.5 hearts indicate thin ventricular wall resulting from Rest deletion. f Quantitative immunostaining of cell cycle markers showing reduced proliferation of RestmKO cardiomyocytes. n = 4/group, mean ± s.d., *p < 0.05 by unpaired two-tailed Student’s t test. Scale bars = 40 µm 2 NATURE COMMUNICATIONS | 8: 1979 | DOI: 10.1038/s41467-017-02210-y | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02210-y ARTICLE a TAM Rest imKO b P3(iP1) P5(iP3) P9(iP7) c 120 100 iP1 Control (n=30) REST imKO imKO imKO 80 Rest imKO (iP1) (n=22) 60 imKO Control Rest Control Rest Control Rest Rest (iP3) (n=19) iP3 REST 40 Rest imKO (iP7) (n=41) REST 121 KDa Survival (%) 20 p<0.001 iP7 REST (P) 2 4681012 14 21 28 35 P1 P3 P7 GAPDH 36 KDa d e fgControl Control Rest imKO (iP1) Rest imKO (iP1) 12 0.6 120 * * * * 9 90 0.4 * P3 6 60 8 * 6 0.2 3 30 4 (mm) thickness LV 2 mass corrected (mg) LV Heart/body 0 0.0 0 weight (mg/g) weight 0 LVPW LVAW LVEF (%) LVFS (%) h ijControl Rest imKO (iP1) 80 * 100 * 200 200 90 60 80 150 G1/G0 S G2/M 150 G1/G0 S G2/M 70 40 20 * 100 100 15 * 20 50 50 10 5 0 0 (%) Cardiomyocyte/stage 0 Cardiomyocyte/total cell (%) Cardiomyocyte/total 0 0 50 100 150 200 (K) 0 50 100 150 200 (K) G1/G0 S G2/M k Control Rest imKO (iP1) Control Rest imKO (iP1) 15 * 2.5 * 2.0 10 DAPI DAPI 1.5 / / 1.0 TnT TnT / 5 * / PH3/TnT (%) 0.5 Ki67 PH3 0 0.0 20 * 1.5 * DAPI 15 / DAPI / 1.0 TnT 10 / TnT / 0.5 5 EdU/TnT (%) KI67/TnT (%) * EdU Aurora B/TnT (%) Aurora 0 B Aurora 0.0 Fig. 2 Rest is required for neonatal cardiomyocyte proliferation and cardiac function. a A graph of experimental design for generating myocardial Rest knockout (RestimKO) at different postnatal (P) stages using the inducible TnTMerCreMer deleter mice. b Western blots showing effective deletion of Rest. c Survival curve of control and RestimKO mice, p < 0.001, log-rank (Mantel–Cox) testing between mice with Rest deletion at different stages and control (Tam- treated Rest+/+;TnTMerCreMer/+ mice). d Representative H&E staining and heart/body weight ratio revealing underdeveloped RestimKO hearts at P3 after Rest deletion at P1. n = 5/group. Scale bar = 100 µm. e–g Echocardiography showing structure defects and decreased left ventricle function of P3 RestimKO. n = 6 for RestimKO; n = 8 for Control. h–j FACS shows reduced percentage of cardiomyocytes isolated from P3 RestimKO hearts (h) and arrested cell cycle of the RestimKO cardiomyocytes at G0/G1 phase (i, j). k Immunostaining for cell cycle markers indicating reduced proliferation of Rest null cardiomyocytes. n = 4/ group for FACS and immunostaining. Scale bar = 40 µm. Mean ± s.d., *p < 0.05, unpaired two-tailed Student’s t test shown to be essential for cardiomyocyte proliferation during binds a cis-element of 21 nucleotide base pairs, named RE1 development and for regeneration following neonatal heart motif22, to silence the expression of neuronal genes in the non- injury. Transcriptional repressor element-1 silencing transcrip- neuronal cells required for neurogenesis23,24. We have recently tion factor (REST), also known as neuron-restrictive silencer reported REST acting as a transcription repressor in mouse factor (NRSF), is widely expressed in the embryonic tissues20,21.It embryonic hearts20. In this study, we reveal that suppression of NATURE COMMUNICATIONS | 8: 1979 | DOI: 10.1038/s41467-017-02210-y | www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02210-y abE12.5 P3 c p21 50 E12.5 P3 Inter gene Control Control Repeat * * RE-1 motif 40 Coding UTR Mm9 30 Intron Hg18 /DAPI 20 0 3 kb 6 kb 9 kb mko imko p21-RE1 TnT p21/TnT (%) / Rest Rest (iP1) 10 WT p21 Mut1 0 Mut2 Del mKo imKo Control Control Rest Rest 690 bp d efE12.5 Control P1 Rest imKO (iP1) 2.5 P21 0.6 Del * * * 2.0 0.5 Mut2 * 0.4 * 1.5 -RE-1 motif Mut1 * 0.3 * p21 * 1.0 * * 0.2 WT NS NS 0.5 NS pGL3- * 0.1 enhancer Enrichment of RE-1 motif 0.0 Relative change to 5% input 0.0 0 0.5 1.0 1.5 lgG lgG lgG lgG Luciferase activity REST REST REST REST 5% input 5% input 5% input H3K4me3H3K9me3 H3K9/14ac H3K27me3 Fig. 3 REST binds p21 to repress its expression. a, b Immunostaining (n = 4/group) showing upregulated p21 protein level in the Rest deleted E12.5 and P3 hearts. Scale bar = 40 µm. c Top, Sequence alignment of p21 between mouse and human indicating a conserved RE-1 motif (red vertical line) located at the 3′ UTR. Bottom, The p21-RE1 motif and its mutated (Mut) or deleted (Del) counterpart. d Luciferase reporter assays with a 690-bp mouse DNA fragment containing the wild type (WT), mutated, or deleted RE1 motif reveals that the intact p21-RE1 motif is required for repressing the reporter gene transcription in primary cultured P1 mouse cardiomyocytes.