DNA Damage Response/TP53 Pathway Is Activated and Contributes to the Pathogenesis of Dilated Cardiomyopathy Associated with Lamin A/C Mutations

Abstract

Rationale:

Mutations in the LMNA gene, encoding lamin A/C (LMNA), are responsible for laminopathies. Dilated cardiomyopathy (DCM) is a major cause of mortality and morbidity in laminopathies.

Objective:

To gain insights into the molecular pathogenesis of DCM in laminopathies.

Methods and Results:

We generated a tet-off bigenic mice expressing either a wild type (WT) or a mutant LMNA (D300N) protein in cardiac myocytes. LMNA^D300N mutation is associated with DCM in progeroid syndromes. Expression of LMNA^D300N led to severe myocardial fibrosis, apoptosis, cardiac dysfunction, and premature death. Administration of doxycycline suppressed LMNA^D300N expression and prevented the phenotype.

Whole heart RNA-sequencing in 2-week old WT and LMNA^D300N mice led to identification of ~6,000 differentially expressed genes (DEGs). Gene set enrichment and Hallmark pathway analyses predicted activation of E2F, DNA Damage Response (DDR), TP53, NFκB and TGFβ pathways, which were validated by western blotting, qPCR of selected targets, and/or immunofluorescence staining. DEGs involved cell death, cell cycle regulation, inflammation, and epithelial-mesenchymal differentiation. RNA-sequencing of human hearts with DCM associated with defined LMNA pathogenic variants corroborated activation of the DDR/TP53 pathway in the heart. Increased expression of CDKN2A, a downstream target of E2F pathway and an activator of TP53, provided a plausible mechanism for activation of the TP53 pathway.

To determine pathogenic role of TP53 pathway in DCM, Tp53 gene was conditionally deleted in cardiac myocytes in mice expressing the LMNA^D300N protein. Deletion of Tp53 partially rescued myocardial fibrosis, apoptosis, proliferation of non-myocyte cells, left ventricular dilatation and dysfunction, and slightly improved survival.

Conclusions:

Cardiac myocyte-specific expression of LMNA^D300N, associated with DCM, led to pathogenic activation of the E2F/DDR/TP53 pathway in the heart and induction of myocardial fibrosis, apoptosis, cardiac dysfunction, and premature death. The findings denote the E2F/DDR/TP53 axis as a responsible mechanism for DCM in laminopathies and as a potential intervention target.

INTRODUCTION

Lamin A/C (LMNA) is a ubiquitously expressed protein constituent of nuclear inner membrane that interacts with genome to regulate gene expression. ^1–4 Mutations in the LMNA gene cause a diverse array of phenotypes, including dilated cardiomyopathy (DCM), which are collectively referred to as laminopathies. ^{3, 5} Cardiac involvement typically manifests with cardiac dilatation and dysfunction, conduction defects, arrhythmias, and sudden cardiac death, often necessitating implantation of a defibrillator/pacemaker. ^6–9 LMNA is among the common causal genes for hereditary DCM, accounting for up to 10% of familial DCM and a small fraction of arrhythmogenic cardiomyopathy cases. ^{6, 10–13}.

A notable phenotypic effect of LMNA mutations is progeria, which spans a broad spectrum ranging from the classic Hutchinson-Gilford Progeria Syndrome to atypical progeroid Werner syndrome. ^14–17 Cardiac involvement in progeroid syndromes includes arrhythmias, conduction defects, heart failure, atherosclerosis, and vascular calcification, among others. ^{16, 17} Cardiovascular involvement commonly manifests as DCM and leads to refractory heart failure and premature death. ^{6, 16, 17}

To gain insights into the molecular pathogenesis of myocardial involvement in DCM caused by LMNA mutation, a tet-Off gene expression system was used to express either a wild type (WT) or a mutant LMNA, namely LMNA^D300N, in cardiac myocytes. The LMNA variant p.Asp300Asn (LMNA^D300N) has been associated with DCM in patients with atypical progeroid/Werner syndrome and non-syndromic cardiac progeria. ^{17, 18} The latter is a unique progeroid phenotype that is primarily restricted to the heart, as opposed to Werner syndrome, which involves multiple organs, including the heart, as the predominant organ responsible for premature death. ^{17, 18} The main findings in the mouse model were corroborated in human heart samples from patients with DCM associated with defined pathogenic variants in the LMNA gene.

METHODS

Large data sets are already available to other investigators through GEO (GSE123916). Detailed information about material and methods are available in Online Supplementary Material. All other data and material are available from the corresponding author upon request.

Regulatory approvals.

The Institutional Review Board approved the use of human tissue samples. All subjected consented to the use of their cardiac tissues in research. Animal studies were in accord with the NIH Guide for the Care and Use of Laboratory Animals published and approved by the Animal Care and Use Committee.

Tet-off bigenic mice.

A tet-off binary transgene system was used to express either a FLAG-tagged wild type (WT) LMNA (LMNA^WT) or the mutant LMNA^D300N protein in cardiac myocytes. ¹⁹ The p.Asp300Asn point mutation was introduced by site-directed mutagenesis. The Lmna^WT and Lmna^D300N cDNAs were cloned downstream to the minimal Myh6 promoter in the tetO responder plasmid. ¹⁹ The constructs were microinjected into fertilized zygotes and the responder founders (tetO:Lmna^WT and tetO:Lmna^D300N) were generated and crossed to the activator mice (Myh6-tTA) to generate inducible cardiac-specific Myh6-tTA:tetO-Lmna^WT and Myh6-tTA:tetO-Lmna^D300N mice. The transgenes are constitutively expressed in cardiac myocytes in the bigenic mice and are turned off upon administration of Doxycycline. Mouse genotyping was performed by polymerase chain reaction (PCR) of tail genomic DNA. Sequence of the oligonucleotide primers used for genotyping is provided in Online Table I.

Survival analysis.

Kaplan-Meier survival curves were constructed using GraphPad Prism 7 software (https://www.graphpad.com/).

Gross morphology.

Heart weight was measured after explantation and flushing of blood from the cavities. Heart weight/body weight ratio was calculated and compared among groups.

Quantification of myocardial fibrosis.

Collagen volume fraction was calculated in picrosirius red stained myocardial thin sections as a percentage of the total myocardial area using ImageJ software (https://imagej.nih.gov/ij/index.html), as published. ^{4, 20–22}

Quantification of active TGFβ1 level.

TGFβ1 concentration in total heart lysate was measured using a sandwich enzyme immunoassay kit (R&D Systems, cat. # MB100B).

Echocardiography.

Cardiac size and function were assessed in age- and sex-matched littermates by 2D and M mode echocardiography using a Vevo 1100 ultrasound imaging system equipped with a 22–55 MHz MicroScan transducer (MS550D) (FUJIFILM VisualSonics Inc., Toronto, ON, Canada), as published. ^{4, 21–24} The leading-edge method was used and each measurement was made for at least 5 cycles and the mean values were used.

Immunoblotting.

Whole heart and nuclear proteins were extracted as published and protein concentrations were measured by a colorimetric assay. ^{4, 25} Aliquots of 30 μg of denatured protein extracts were loaded onto and separated on SDS-PAGE gels, and transferred to nitrocellulose membranes. Target proteins were detected using specific primary antibodies and the respective horseradish linked secondary antibodies, as listed in Online supplementary Table I.

Immunofluorescence.

Expression and localization of the proteins of interest were detected in 5 micrometer thin myocardial frozen sections upon probing with the primary antibodies of interest and incubation with the corresponding secondary antibody conjugates, as published. ^{4, 20} An antibody against anti pericentriolar material 1 (PCM1) protein was used to mark myocyte nuclei.^26–28 A detailed list of antibodies used in immunofluorescence staining of thin myocardial sections is provided in Online supplemental Table I.

Apoptosis.

Apoptosis was detected by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay, as published. ^4,29 The total number of TUNEL-positive cells and nuclei were counted in about 10 fields per each thin section and in at least 5 sections per heart using ImageJ cell counter software (https://imagej.net/Particle_Analysis).

Isolation of adult mouse cardiac myocytes.

Adult mouse cardiac myocytes were isolated per a published protocol and as reported previously, with minor modifications. ^{4, 30} In brief, hearts were rapidly explanted following euthanasia, washed with a perfusion buffer, and perfused with a myocyte digestion buffer containing collagenase type II in a retrograde manner. After completion of the enzymatic digestion, the heart was dissected free of vessels and atria, and minced in a stop buffer. The cell suspension was filtered through a 100 μm cell strainer and the myocytes were pelleted by centrifugation. Calcium was reintroduced in a step wise manner and the final cell pellet was resuspended in a RIPA buffer containing protease and phosphatase inhibitors for protein work or in Trizol reagent (Thermo Fisher Scientific, cat# 15596026) for RNA studies.

Isolation of adult mouse cardiac fibroblasts.

Adult mouse cardiac fibroblasts were isolated as published. ²⁴ In brief, mice were euthanized, hearts were explanted, ventricles were excised and transferred to a digestion buffer containing 0.1% collagenase II in DMEM. After repeated rounds of enzymatic digestion, the final cell suspension was centrifuged briefly at 20 g to remove myocytes and cell debris. Fibroblasts in the resulting supernatant were pelleted by centrifugation, washed with PBS, and re-suspended in a RIPA buffer containing protease and phosphatase inhibitors or the Trizol reagent, for protein and RNA experiments, respectively.

Quantitative RT-PCR (qPCR).

RNA was extracted from fresh frozen tissues or cells and treated with DNase 1 to remove the genomic DNA. Reverse transcription was performed and the transcript levels of genes of interest were determined by qPCR using specific TaqMan Gene expression assays or SYBR Green assays. Target gene expression levels were normalized to Gapdh mRNA levels. The ΔCT method was used to calculate the expression levels and presented as relative (to wild type control) and normalized (to Gapdh) values. The list of TaqMan assays and oligonucleotide primers used in the qPCR reactions is provided in Online supplementary Table I.

RNA Sequencing.

Bulk RNA-sequencing (RNA-Seq) was performed, as published. ^{4, 24} In brief, total RNA was extracted from the hearts of 2-week old Myh6-tTA:tetO-Lmna^D300N and WT littermate mice. RNA concentration was determined by spectrometry using a NanoDrop Spectrophotometer. RNA samples with an RNA Integrity Number (RIN) of > 8 were used to generate sequencing libraries after depletion of the rRNA. Sequencing was performed on the Illumina HiSeq 4000 instrument using the paired-end sample preparation chemistry.

Raw RNA sequencing reads were mapped to the Mouse reference genome build 10 (UCSCmm10/GRCm38) by Tophat2. ³¹ Gene expression was analyzed using Cufflinks where counts were presented as fragments per kilobase of transcript per million mapped fragments (FPKM) for each gene and transcript. ³² Transcripts with FPKM ≥ 1 in at least one sample were included in the analyses. Differentially expressed genes (DEGs) were identified using the edgeR analysis package in R statistical program with the significance level set at q<0.05. ³³

Rstudio was used to generate the heatmaps and volcano plots from the inferred FPKM values. (www.rstudio.com) Enriched upstream regulators were inferred using Gene Set Enrichment (GSEA, version 2.2.3, http://software.broadinstitute.org/gsea/). Circos plots were generated in RStudio using the GO-Chord option.

Conventional statistical methods.

Statistical analyses were performed as described. ^{4, 20, 21} Data were presented as mean ± SD. Data that followed a normal Gaussian distribution were analyzed by parametric and those departing from normality by non-parametric tests. Differences between the two groups were compared by t-test for normally distributed values and Kruskal-Wallis for those departing from normality. Differences among multiple groups were analyzed by one-way ANOVA, followed by Bonferroni pairwise comparison. Statistical analyses were performed either using GraphPad Prism 7 (www.graphpad.com) or STATA 10.1 (www.stata.com).

RESULTS

Expression of WT and mutant LMNA in cardiac myocytes.

Tet-off bigenic mice were generated to express either a LMNA^WT or the mutant LMNA (LMNA^D300N) protein specifically in cardiac myocytes (Online Figure I). The transgene proteins were tagged with a FLAG epitope at the N-terminal domain in order to detect their expressions without the confounding effect of the endogenous LMNA protein. The transgene proteins were stably expressed in the heart, as detected by immunoblotting of myocardial protein extracts using an anti-FLAG antibody (Figure 1A). Expression levels of the transgene proteins were higher than those of the endogenous LMNA protein, albeit accurate quantification of the LMNA protein levels was compounded by low solubility of LMNA in the extraction buffers. Immunofluorescence staining of thin myocardial sections with an anti-FLAG antibody showed incorporation of the transgene proteins into nuclear membrane (Figure 1B). Approximately a third of the nuclei in myocardial thin sections were stained positive for the expression of transgene LMNA, consistent with the known percentage of myocytes in the myocardium (Figure 1C). ^{26, 34} These data also indicate that tagging of LMNA at the N-terminal domain did not affect its localization to nuclear membrane.

Figure 1. Expression of LMNA^D300N in the heart and its effects on cardiac size and survival in mice:

A. Expression of the wild type (WT) and mutant LMNA (LMNA^D300N) in doxycycline-off bigenic mice. Immunoblots showing expression of FLAG-tagged LMNA as well as the endogenous LMNA are shown along with panels representing expression of nuclear envelope protein emerin and α-tubulin, subunit 1 (TUBA1A), as loading controls.

B. Incorporation of the transgene LMNA^WT and LMNA^D300N into nuclear membrane. Myocardial thin sections are co-stained with anti- FLAG and anti- emerin antibodies and counter-stained with DAPI.

C. Quantitative data showing approximately a third of nuclei in the myocardium were stained positive for expression of FLAG-tagged LMNA (N=3 per group, LMNA^WT = 31.8±2.9%, LMNA^D300N = 26.2±0.4%).

D. Gross cardiac morphology. The heart was grossly enlarged in the Myh6-tTA:tetO-Lmna^D300N mice, which was prevented upon administration of doxycycline.

E. Immunoblots showing effective shut down of transgene expression in the the Myh6-tTA:tetO-Lmna^D300N mice upon administration of doxycycline. A blot representing expression of emerin is shown as a control.

F. Kaplan-Meier survival plots showing premature death in the Myh6-tTA:tetO-Lmna^D300N mice (N=43), whereas survival was normal in bigenic mice expressing FLAG-tagged LMNA^WT (N=29, p<0.0001; Γ²=243.8).

G. Prevention of premature death in the Myh6-tTA:tetO-Lmna^D300N mice upon administration of doxycycline (Γ²=69.1; p<0.0001).

Gross cardiac morphology.

Expression of LMNA^D300N protein in cardiac myocyte was associated with gross cardiac enlargement and increased heart to body weight ratio (Figure 1D, and Online Figure II). Cardiac morphology and heart/body weight ratios were normal in all control groups, including Myh6-tTA, Myh6-tTA:tetO-Lmna^WT and tetO-Lmna^D300N mice. To determine whether cardiac enlargement resulted from expression of the mutant LMNA^D300N protein, doxycycline was added to the drinking water of pregnant mice and their newborn offspring throughout their lives. As shown in Figure 1E, treatment with doxycycline effectively suppressed expression of the mutant protein and prevented cardiac enlargement (Figure 1D and Online Figure II).

Cardiac dimensions and function.

Echocardiographic evaluation of cardiac function in 3 to 4-week old mice showed right and left ventricular dilatation, increased left ventricular mass index, and reduced ventricular fractional shortening in the Myh6-tTA:tetO-Lmna^D300N bigenic mice, as compared to WT or bigenic mice expressing the LMNA^WT protein (Table 1). Cardiac size and function in the control mice, including Myh6-tTA, Myh6-tTA:tetO- Lmna^WT, and tetO-Lmna^D300N were normal (Table 1). Suppressing expression of the transgene protein LMNA^D300N in the heart upon treatment of the Myh6-tTA:tetO-Lmna^D300N bigenic mice with doxycycline prevented cardiac dilatation and dysfunction.(Table 1).

TABLE 1.

Echocardiographic Phenotype Upon Cardiac Myocyte-Specific Expression of LMNA^D300N and After Turning Off Expression of the Transgene Protein with Doxycycline

	WT	Myh6-tTA	Myh6-tTA:TetO: LmnaWT	TetO: LmnaD300N	Myh6-tTA:TetO: LmnaD300N	p (ANOVA)	Myh6-tTA:TetO: LmnaD300N - Dox	p vs. Myh6-tTA: TetO:LmnaD300N
N	12	6	8	9	16	N/A	13	N/A
M/F	7/5	4/2	5/3	4/5	7/9	0.819	7/6	0.715
Age (days)	28.17±2.52	28.00±3.10	31.50±2.07	27.89±1.05&	26.25±2.96 &	0.001	66.46±3.84	<0.0001
BW (g)	16.3±2.23	19.4±1.63	17.9±2.33	16.7±1.8	14.4±2.83 #&	<0.0001	25.1±3.83	<0.0001
HR (bpm)	496.7±64.4	481.7±81.6	536.3±68.9	486.7±78.6	458.6±81.8	0.231	538.5±90.7	0.021
IVST(mm)	0.71±0.05	0.75±0.02	0.76±0.07	0.71±0.09	0.63±0.09 #&	0.001	0.83±0.08	<0.0001
PWT (mm)	0.68±0.17	0.78±0.06	0.75±0.14	0.72±0.14	0.66±0.13	0.288	0.75±0.07	0.027
LVEDD (mm)	3.01±0.27	2.93±0.41	2.98±0.20	2.87±0.19	3.57±0.48 *#$&	<0.0001	3.24±0.39	0.052
LVEDDi (mm/g)	0.19±0.08	0.15±0.01	0.17±0.02-	0.17±0.02+	0.25±0.03*#$&	<0.0001	0.13±0.02	<0.0001
LVESD (mm)	1.16±0.11	1.00±0.12	1.08±0.13	1.07±0.13	2.08±0.45*#$&	<0.0001	1.03±0.20	<0.0001
FS (%)	61.42±4.66	65.50±3.39	63.88±4.19	62.67±3.64	42.0±7.92 *#$&	<0.0001	68.31±4.23	<0.0001
LVM (mg)	58.68±10.16	64.39±11.99	64.22±10.37	56.50±12.70	71.77±25.53	0.211	80.14±21.43	0.354
LVMi (mg/g)	3.68±0.63	3.31±0.40	3.60±0.54	3.43±0.87	4.92±1.29 *#$&	<0.0001	3.20±0.82	<0.0001

Abbreviations: WT: Wild type; Dox: Doxycycline; M/F: Male/Female; BW: Body weight; g; grams; HR, heart rate; bpm, beats per minute; IVST, interventricular septum thickness; PWT, posterior wall thickness; LVEDD, left ventricular end diastolic diameter; LVEDDi, LVEDD divided by the body weight; LVESD, left ventricular end systolic diameter; FS, fractional shortening; LVM, left ventricular mass; LVMi, LVM divided by the body weight.

^*denotes p<0.05 v.s. WT

^#denotes p<0.05 v.s Myh6-tTA

^$denotes p<0.05 v.s. TetO:Lmna^D300N

^&denotes p<0.05 v.s. Myh6-tTA:TetO:Lmna^WT

Survival.

Expression of the LMNA^D300N protein in cardiac myocytes was associated with premature death leading to a near total mortality within 30 post-natal days, whereas expression of FLAG-tagged LMNA^WT had no effect on survival (Figure 1F). Likewise, the regulator Myh6-tTA and the target tetO-Lmna^D300N single transgenic mice had normal survival. To determine whether premature death resulted from expression of the LMNA^D300N protein in cardiac myocyte, expression of the transgene protein was suppressed upon administration of doxycycline to Myh6-tTA:tetO-Lmna^D300N bigenic mice during pregnancy and to the offspring for 3 months after birth. Administration of doxycycline was associated with normal survival in the Myh6-tTA:tetO-Lmna^D300N bigenic mice (Figure 1G). Collectively, the findings supported the causal role of expression of LMNA^D300N in cardiac dilatation and dysfunction and premature death.

Myocardial fibrosis.

To detect myocardial fibrosis, myocardial thin sections were stained with Masson trichrome or picrosirius red. Collagen volume fraction (CVF) was quantified in picrosirius red stained myocardial sections. Myh6-tTA:tetO-Lmna^D300N bigenic mice showed severe myocardial fibrosis, comprising approximately 15% of the myocardium (Figure 2, panels A and B). Administration of doxycycline, i.e., suppression of expression of the LMNA^D300N transgene protein, prevented myocardial fibrosis in the Myh6-tTA:tetO-Lmna^D300N mice (Figure 2, panels A and B). Corresponding to increased myocardial fibrosis, levels of phosphoSMAD2 (pSMAD2), an effector of the TGFβ1 pathway, were also increased in the myocardium of Myh6-tTA:tetO-Lmna^D300N mice, whereas pSMAD2 levels in the heart were normal in mice treated with doxycycline (Figure 2C and Online Figure III, panels A and B). Increased pSMAD2 levels were also corroborated in cardiac fibroblasts isolated from the hearts of Myh6-tTA:tetO-Lmna^D300N mice (Online Figure III, Panels C and D). Moreover, levels of active TGFβ1 protein, measured by a sandwich enzyme immunoassay technique, were also increased in the hearts of Myh6-tTA:tetO-Lmna^D300N mice, as compared to WT mice. (Figure 2D). Further, transcript levels of Tgfb1, Ctgf, and Col1a1 were increased in the heart of Myh6-tTA:tetO-Lmna^D300N mice (Figure 2E). The findings suggest that expression of the mutant LMNA^D300N in cardiac myocytes induces expression of paracrine factors, such as TGFB1, which activate the pro-fibrotic programs in cardiac fibroblasts, as reported previously for other forms of cardiomyopathies. ^35–37

Figure 2. Myocardial fibrosis and apoptosis upon expression of LMNA^D300N in the heart:

A. Panels represent micrographs of picrosirius red (upper two rows) and Masson trichrome (lower row)– stained myocardial thin sections, showing increased myocardial fibrosis in the Myh6-tTA:tetO-Lmna^D300N mouse hearts and its absence upon administration of doxycycline (Myh6-tTA:Lmna^D300N :14.27±4.235, Myh6-tTA;Lmna^WT: 2.363±0.616, and Myh6-tTA:Lmna^D300N treated with Dox: 2.363±0.616, p<0.0001).

B. Quantitative data on collagen volume fraction (CVF) in the experimental groups.

C. Immunoblots showing increased levels of pSMAD2 in the Myh6-tTA:tetO-Lmna^D300N as compared to other groups and its normalization upon treatment with doxycycline (p=0.0014).

D. Levels of active TGFβ1 in the heart of WT and the Myh6-tTA:tetO-Lmna^D300N mice, as determined by ELISA (33.30±2.71 vs. 41.97±4.64 pg/ml, respectively, p=0.0018).

E. Transcript levels of three fibrogenic genes in the heart, showing increased levels in the Myh6-tTA:tetO-Lmna^D300N mice.

F. TUNEL assay showing increased apoptosis in the heart of the Myh6-tTA:tetO-Lmna^D300N mice and its absence upon treatment with doxycycline (N=5, p<0.0001).

G. Quantitative data showing increased number of TUNEL positive cells in the heart in the Myh6-tTA:tetO-Lmna^D300N mice.

Myocardial apoptosis.

Apoptosis was detected by staining myocardial thin sections by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay (Figure 2F). The percentage of labeled nuclei was increased by more than 5-fold in the myocardium of Myh6-tTA:tetO-Lmna^D300N mice, as compared to WT mice (Figure 2, panels F and G). There was no difference in the percentage of TUNEL positive cells between the WT and Myh6-tTA or Myh6-tTA:tetO-Lmna^WT mice (Figure 2, panels F and G). Administration of doxycycline abrogated apoptosis in the myocardium of Myh6-tTA:tetO-Lmna^D300N mice (Figure 2, panels F and G). Identity of cardiac cells undergoing apoptosis was not determined.

Proliferation of non-myocytes cells in the heart.

Given that the LMNA^D300N mutant protein was expressed during proliferative phase of cardiac myocytes, potential effects of the mutant LMNA on proliferation of cardiac cells were examined. Accordingly, thin myocardial sections were stained for the expression of cell cycle markers PCNA and Ki67 (MKI67) along with PCM1, the latter to mark cardiac myocytes. ^26–28 The approach enabled identification of myocyte (PCM1-positive) and non-myocyte (PCM1-negative) cell types that had entered cell cycle. Approximately, 25% of nuclei in thin myocardial sections stained positive for PCM1 in the WT mice, whereas only about 12% of nuclei were positive for PCM1 staining in the Myh6-tTA:tetO-Lmna^D300N mouse hearts (Online Figure IV, p<0.0001). The latter is likely reflective of the increased number of non-myocytes cells and increased apoptosis in the heart of Myh6-tTA:tetO-Lmna^D300N mice. Expression of cell proliferation markers were detected only in non-myocyte cells, identified as PCM1-negative cells, in the heart (Online Figure IV). The number of non-myocyte cells expressing Ki67 or PCNA was increased significantly in the Myh6-tTA:tetO-Lmna^D300N mouse hearts (Online Figure IV).

Collectively, the morphological and histological data indicate that expression of LMNA^D300N leads to premature death, cardiac dilatation and dysfunction, myocardial fibrosis, apoptosis, and increased proliferation of non-myocyte cells in the heart.

Phenotypic effects of reducing endogenous LMNA levels in the Myh6-tTA:tetO-Lmna^D300N mice.

To address the possibility that the phenotype was the consequence of over-expression of the mutant LMNA^D300N, the Myh6-tTA:tetO-Lmna^D300N mice was crossed to Lmna^+/− heterozygous mice to express the mutant protein in the background of LMNA haploinsufficiency. The approach has been reported to reduce the dose of LMNA and eliminate progeria-like phenotype in a mouse model of laminopathies. ³⁸ The Myh6-tTA:tetO-Lmna^D300N:Lmna^+/− mice were generated and characterized for survival, cardiac size and function, and myocardial fibrosis. Expression of the LMNA^D300N protein in the heart and its incorporation into myocyte nuclear membrane in the background of Lmna^+/− were confirmed by immunoblotting and immunofluorescence staining of myocardial thin sections, respectively (Online Figure V, panels A and B, respectively). Phenotypic findings were notable for reduced survival, comparable to that in the Myh6-tTA:tetO-Lmna^D300N mice with a median survival of about 26 days and a near total mortality by about 1 month of age (Online Figure V, panel C). Echocardiographic data showed no discernible differences in indices of cardiac size and function between Myh6-tTA:tetO-Lmna^D300N and Myh6-tTA:tetO-Lmna^D300N:Lmna^+/− mice, as the heart remained dilated with reduced systolic function despite reducing the endogenous LMNA levels (Online Table II). Likewise, heart to body weight ratio remained increased and similar to that in the Myh6-tTA:tetO-Lmna^D300N mice (Online Figure V, panel D). Moreover, myocardial fibrosis, quantified on picrosirius red stained myocardial thin sections was not reduced in mice expressing the LMNA^D300N protein in the background of Lmna^+/−, as compared to that in Myh6-tTA:tetO-Lmna^D300N mice (Online Figure V, Panels E and F).

Differentially expressed genes (DEGs).

To gain insights into the mechanisms responsible for the observed phenotype in the Myh6-tTA:tetO-Lmna^D300N mice, dysregulated genes were analyzed by RNA-Seq and biological pathways were identified in the hearts of 2-week old Myh6-tTA:tetO-Lmna^D300N mice, as compared to WT littermates. The 2-week time point was selected to identify early changes and reduce potential confounding effects of the ensuing cardiac dysfunction and mortality at a later age. Accordingly, myocardial total RNA was extracted from the hearts of 2-week old WT and Myh6-tTA:tetO-Lmna^D300N mice, depleted of ribosomal RNA, and analyzed by RNA-Seq (N=4 per group). Principle component analysis distinctly segregated the two genotypes (Online Figure VI, Panel A). Pearson correlation matrix showed strong correlation in the transcript levels among mice per each genotype (Online Figure VI, panel B). A total of ~ 6,600 transcripts were dysregulated (FDR<0.05) with a relatively equal distribution of up- and down-regulated transcripts (Figure 3, panels A-C). Spp1 (228-fold), Ereg (134-fold), Il6 (64-fold), Crlf1 (45-fold), Gdf15 (35-fold), encoding secreted phosphoprotein 1, epiregulin, interleukin 6, cytokine receptor like factor 1, and growth differentiation factor 15, respectively, were among the most upregulated genes (Figure 3, panel D). In contrast, Apa2, encoding apolipoprotein a 2 (75-fold) and Adipoq, coding for adiponectin (40-fold) were the most down-regulated genes (Figure 3, panel D), albeit transcript levels of the latter two genes in the heart were relatively low. A complete list of differentially expressed genes (FDR<0.05) has been submitted to GEO (GSE123916).

Figure 3. Dysregulated gene expression and activation of E2F transcription factor pathway in the hearts of Myh6-tTA:tetO-Lmna^D300N mice:

A. Volcano plots showing differentially expressed genes (q<0.05) in the hearts of the Myh6-tTA:tetO-Lmna^D300N mice. Green reflects down-regulated and red upregulated genes. The dash line denotes significance level at an FDR of 0.05.

B. Pie chart demonstrating the number of dysregulated genes at FDR of < 0.05.

C. Heat map of differentially expressed genes (DEGs), red indicating upregulated and blue down-regulated genes

D. List of top upregulated and down-regulated genes in the hearts of the Myh6-tTA:tetO-Lmna^D300N mice.

E. Hallmark signature depicting the top dysregulated pathways, along with the number of genes in each pathway and normalized enrichment score (NES).

F. Western blots showing expression of E2F (Myh6-tTA:tetO-Lmna^D300N vs Myh6-tTA:tetO-Lmna^WT, 1.2-fold change, p=0.150) and retinoblastoma 1 (RB1) showing a marked reduction in the Myh6-tTA:tetO-Lmna^D300N vs Myh6-tTA:tetO-Lmna^WT hearts (fold change=16, p=0.0369). Blots representing LMNA and TUBA1A proteins in the heart of WT and the Myh6-tTA:tetO-Lmna^D300N mice are shown as controls.

G. Quantitative data showing reduced levels of RB1 protein in the hearts of Myh6-tTA:tetO-Lmna^D300N mice (N=3 per group).

H. Gene Set Enrichment Analysis (GSEA) plot of genes regulated by E2F transcription factors, showing activation of the pathway.

I. Heat map of the dysregulated genes that are known targets of E2F transcription factors

J. Quantitative PCR showing increased transcript levels of selected E2F transcription factor target genes (N=4 per group).

Transcriptional regulators of DEGs.

Analysis of DEGs identified Hallmark signatures for cell cycle/death/differentiation, fibrosis, and inflammation, as the three categories of biological pathways dysregulated in the hearts of Myh6-tTA:tetO-Lmna^D300N mice (Figure 3E and Online Figure VII).

E2F transcription factor pathway was among the top dysregulated pathway along with TNFA (via NFκB), MTORC1, and TP53 pathways. Immunoblotting showed no discernible change in the levels of E2F protein, whereas levels of RB1, which is stabilized by LMNA and is a binding partner and suppressor of E2F transcription factors ^39–41, was markedly reduced in the hearts of Myh6-tTA:tetO-Lmna^D300N mice (Figure 3, panels F and G). ⁴⁰ GSEA plot and heat map of transcript levels of selected E2F target genes are presented in Figure 3, panels H and I, which predict increased E2F transcriptional activities in the hearts of Myh6-tTA:tetO-Lmna^D300N mice, likely resulting from the removal of inhibitory effects of RB1. ⁴⁰ Quantitative PCR analysis of transcript levels of selected E2F target genes confirmed the results of RNA-Seq data, indicating increased E2F transcriptional activity (Figure 3, panel J).

Given the prominence of myocardial fibrosis and the transcriptome signature identifying epithelial mesenchymal transition and TGFB1 pathway, as the major dysregulated pathways (Figure 3, panel E), DEGs were further analyzed to identify genes involved in fibrosis. GSEA analysis of the DEGs showed marked enrichment of profibrotic gene set in the heart (normalized enrichment score =0.66, and q <0.0001, Online Figure VIII, Panel A). Heat map of DEGs involved in fibrosis is shown in Online Figure VIII, panel B). DEGs were also analyzed to identify genes whose proteins are secreted proteins, which might serve as paracrine factors. Analysis identified a large number of DEGs as putative paracrine factors that could mediate the profibrotic programs in cardiac fibroblasts (Online Figure VIII, panel C).

Given that the LMNA^D300N transgene protein was expressed during cardiac development and metabolic switch, effects of its expression on transcript levels of DEG involved in cardiac metabolism were analyzed by GSEA. Transcript levels of genes involved in oxidative phosphorylation was suppressed as were those involved in fatty acid metabolism, whereas those involved in glycolysis were increased (Online Figure IX).

Activation of TP53 pathway.

TP53 pathway was among the top differentially dysregulated transcriptional pathways in the hearts of Myh6-tTA:tetO-Lmna^D300N mice (Figure 3, panel E). Heat map of differentially expressed gene in the TP53 pathway and the corresponding GSEA plot are shown in Figure 4, panels A and B, respectively. Analysis of myocardial protein extracts by immunoblotting showed increased expression levels of TP53 and its downstream signature target CDKN1A as well as MYC proteins (Figure 4, panels C and D). Transcript levels of several TP53 target genes, such as Cdkn1a, Cdkn2a, and Myc, which are also shared by the E2F pathways and shown in Figure 3, panel J, were increased in the heart of Myh6-tTA:tetO-Lmna^D300N mice. Moreover, transcript levels of Tp53 and its selected target genes were increased in cardiac myocytes isolated from the heart the Myh6-tTA:tetO-Lmna^D300N mice (Figure 4-E).

Figure 4. Activation of the TP53 pathway in the hearts of Myh6-tTA:tetO-Lmna^D300N mice:

A. Heat map of DEGs in the hearts of the Myh6-tTA:tetO-Lmna^D300N mice, which are the targets of TP53 pathway

B. GSEA plot of TP53 targets genes, showing enrichment of the targets in the hearts of the Myh6-tTA:tetO-Lmna^D300N mice. Red and blue colors correspond to genes whose expression is high and low, respectively, in the Myh6-tTA:tetO-Lmna^D300N mouse hearts.

C. Immunoblots showing increased levels of TP53, CDKN1A, and MYC in the hearts of the Myh6-tTA:tetO-Lmna^D300N mice, as compared to WT mice.

D. Quantitative data of protein levels analyzed in panel C (N=3 per group).

E. qPCR data showing increased transcript levels of selective genes, known targets of the TP53 pathway in cardiac myocytes isolated from the Myh6-tTA:tetO-Lmna^D300N mice (N=3 per group).

Dysregulated biological pathways.

Pathway analysis of DEGs showed prominent representation of genes involved in the apoptotic processes (45 genes, q=4.1*10⁻²¹), cell cycle G1 check point (13 genes, q=9*10⁻¹⁶), DNA damage response (DDR, 14 genes, q=3.7*10⁻¹⁵), and cellular metabolic process (46 genes, q=3.7*10⁻¹⁵, Figure 5-A). The top dysregulated pathways were further analyzed and validated by an alternative method. A GSEA plot, and a heat map of DEGs along with the transcript levels of selected genes, as determined by qPCR, involved in apoptosis are shown in Online Figure X. Preponderance of the genes involved in all forms of cell death among the DEGs (apoptosis: 45 genes, q=4.1*10⁻²¹ and other forms of cell death: 39 genes, q=1.7*10⁻¹⁹), was in accord with the phenotypic data showing increased number of TUNEL positive cells in the hearts of Myh6-tTA:tetO-Lmna^D300N mice (Figure 2-F). Likewise, increased apoptosis was in accord with activation of the TP53 pathway, which is known to activate the apoptotic programs. ⁴² Moreover, in accord with activation of the TNFA-NFκB pathway, expression levels of a number of pro-inflammatory cytokines, including Il6, Cxcl1, Cxcl2, Cxcl10, and Lif, were increased in the hearts of Myh6-tTA:tetO-Lmna^D300N mice (Online Figure XI). Finally, TP53 targets involved in cellular metabolism encompassed genes involved in ATP biosynthesis, protein phosphorylation, phospholipid formation and gene expression (Online Figure XII). Similar GO analysis for MYC targets (120 genes) showed enrichment of genes mostly involved in RNA metabolism and processing (45/120) while subset of genes was also shown to be involved in carbohydrate and protein metabolism (Online Figure XII).

Figure 5. Biological pathways dysregulated by TP53 activation in the hearts of Myh6-tTA:tetO-Lmna^D300N mice:

A. Circos plot showing DEGs in the TP53 pathway and their corresponding biological pathways affected, depicting prominence of cell death, cell cycle, metabolism and DNA damage response (DDR).

B. Heat map of dysregulated genes involved in DDR, showing increased levels in the hearts of the Myh6-tTA:tetO-Lmna^D300N mice.

C. Immunoblots showing increased levels of DDR markers; TP53, pH2AFX, and ATM, in the heart of the Myh6-tTA:tetO-Lmna^D300N mice.

D. Quantitative data on expression levels of proteins shown in panel C (N=3 per group).

E. Immunofluorescence stained thin myocardial sections (N=5 per group) along with a magnified panel (inserts) showing increased number of cells stained positive for the expression of pH2AFX, a DDR marker in mice expressing the mutant LMNA^D300N protein (N=3 mice per group; p=0.0068).

F. Quantitative data on the number of myocardial cells stained positive for pH2AFX in WT and Myh6-tTA:tetO-Lmna^D300N mice.

G. qPCR data showing increased transcript levels of selected genes involved in DDR in the hearts of the Myh6-tTA:tetO-Lmna^D300N mice.

Genes involved in DDR pathway were among the most differentially expressed in the heart of Myh6-tTA:tetO-Lmna^D300N mice (Figure 5-A). A heat map of DEGs involved in DDR is shown in Figure 5-B. Immunoblotting further substantiated increased expression of selected proteins involved in the DDR, including phospho-H2AFX (pH2AFX) and ATM in addition to TP53 (Figure 5, panels C and D). In conjunction with increased levels of pH2AFX on immunoblotting, immunofluorescence staining also showed an approximately 10-fold increase in the number of nuclei stained positive for the expression of pH2AFX (Figure 5, panels E and F). Given the restrictive expression of the transgene protein LMNA^D300N in cardiac myocytes, regulated by the Myh6 promoter, activation of the DDR/TP53 pathway was further analyzed in cardiac myocytes isolated from the hearts of Myh6-tTA:tetO-Lmna^D300N mice. The findings are notable for increased transcript levels of selective genes also indicated activation of the DDR/TP53 pathway in isolated cardiac myocytes (Figure 5, panel G).

Activation of DDR/TP53 pathway in human DCM caused by pathogenic variants in the LMNA gene.

To determine whether findings in the Myh6-tTA:tetO-Lmna^D300N mice were extendable to human DCM, associated with defined pathogenic variants in LMNA gene, RNA-Seq was performed in ribosome-depleted RNA extracts, isolated from ventricular tissue samples of patients with DCM and controls, as described. ⁴ Analysis of transcriptional regulators of DEGs also identified activation of TP53 among the most dysregulated pathways in the heart of DCM patients (Figure 6, panels A and B). Immunoblot analysis of ventricular protein extracts showed increased expression levels of TP53, pH2AFX, ATM, and POLH, markers of DDR, in DCM hearts as compared to controls (Figure 6, panels C and D). Likewise, immunofluorescence staining of myocardial thin sections showed an approximately 5-fold increase in the number of nuclei stained positive for pH2AFX protein in the DCM heart samples as compared to controls (Figure 6, panels E and F). Furthermore, qPCR analysis of several DDR/TP53 pathway genes in myocardial RNA extracts showed increased levels of CDKN1A, GADD45A, GADD45B, and POLH in DCM hearts (Figure 6, Panel G).

Figure 6. Activation of TP53 pathway in human hearts with DCM associated with defined pathogenic variants in LMNA gene:

A. Dysregulated pathways in human hearts with DCM associated with defined pathogenic variants in the LMNA gene, identifying TP53 as a major altered transcriptional regulator.

B. Circos plot showing DEGs and their corresponding biological pathways in human hearts with DCM.

C. Immunoblot of selected proteins involved in the DDR/TP53 pathway.

D. Quantitative data showing individual data points, mean and 95% confidence interval of protein levels of the DDR/TP53 pathways depicted in panel C (N=5 per group).

E. Immunofluorescence myocardial micrographs showing staining for pH2AFX in control and human hearts with DCM associated with pathogenic variants in the LMNA gene.

F. Quantitative data showing increased number of cells expressing pH2AFX in human hearts with DCM (N=3 to 5 per group)

G. QPCR showing increased transcript levels of four genes involved in the DDR pathways in human hearts with DCM associated with pathogenic variants in the LMNA gene

Partial rescue of cardiac phenotype induced by expression of LMNA^D300N upon deletion of Tp53 gene.

In view of the above data indicating activation of the TP53 pathway in human DCM as well as in the Myh6-tTA:tetO-Lmna^D300N mouse hearts, and given the well-established role of the TP53 pathway in apoptosis, inflammation, and DDR, major phenotypes in the Myh6-tTA:tetO-Lmna^D300N mice, Tp53 gene was deleted in cardiac myocytes by crossing Tp53^F/F, Myh6-Cre and Myh6-tTA:tetO-Lmna^D300N mice. ⁴³ Efficiency of deletion of the Tp53 gene in cardiac myocytes was documented in isolated cardiac myocytes at RNA and protein levels. Levels of Tp53 mRNA and proteins were significantly reduced, but not totally absent, in cardiac myocytes isolated from the Myh6-Cre:Tp53^F/F mouse hearts as compared to WT myocytes (Online Figure XIII, panels A, C, and E). Levels of Tp53 mRNA and protein levels in the whole heart extracts from WT and Myh6-Cre:Tp53^F/F were unchanged, reflecting ubiquitous expression of the TP53 protein in non-myocyte cells in the heart (Figure 7, panel A and B, and Online Figure XIII, panels B, D, and F). Partial deletion of Tp53 in cardiac myocytes in the background of expression of LMNA^D300N (Myh6-tTA:tetO-Lmna^D300N:Myh6-Cre:Tp53^F/F, simply referred to as Lmna^D300N:Tp53^−/− mice) was associated with reduced levels of ATM, pH2AFX, and POLH proteins, as compared to the corresponding levels in the Myh6-tTA:tetO-Lmna^D300N hearts (Figure 7, panels A and B). Consistent with the partial deletion of the Tp53 gene in cardiac myocytes and ubiquitous expression of this protein in other cells in the heart, the mean level of the TP53 protein in Lmna^D300N:Tp53^−/− mice was higher than that in the WT mice (Figure 7, panels A and B). Immunofluorescence staining for pH2AFX showed a lower number of cells expressing this classic DDR marker (Figure 7, panels C and D). Moreover, transcript levels of selected TP53 target genes implicated in DDR, which were increased in the hearts of Myh6-tTA:tetO-Lmna^D300N mice, were attenuated and/or normalized in the hearts of Lmna^D300N:Tp53^−/− mice (Figure 7-E). However, the molecular rescue was incomplete, likely reflective of cellular heterogeneity of the heart, partial deletion of the Tp53 gene, and multiplicity of the involved mechanisms, as transcript levels of several other DDR genes, including Myc, Rad52, and Topbp1 were not rescued (Online Figure XIV). ⁴³

Figure 7. Effects of deletion of Tp53 gene in the Myh6-tTA:tetO-Lmna^D300N mice on activation of the DDR/TP53 pathway:

A. Expression levels of selected proteins in the DDR/TP53 pathway in the main experimental groups (N=3 per group)

B. Quantitative data showing increased levels of TP53, ATM, pH2AFX and POLH in the hearts of Myh6-tTA:tetO-Lmna^D300N mice and their partial normalization upon deletion of the Tp53 gene

C. Immunofluorescence micrographs showing high magnification of myocardial sections stained for pH2AFX in the main experimental groups.

D. Quantitative data showing increased number of cells stained positive for the expression pH2AFX in the hearts of Myh6-tTA:tetO-Lmna^D300N mice and its partial rescue upon deletion of the Tp53 gene (N=5 per group)

E. qPCR data showing partial rescue of a number of DDR/TP53 genes upon deletion of Tp53 gene in cardiac myocytes in the Myh6-tTA:tetO-Lmna^D300N mice (N=5 per group)

Given that myocardial fibrosis was a prominent phenotype in the Myh6-tTA:tetO-Lmna^D300N hearts (Figure 2) and a pro-fibrotic transcriptome signature was markedly activated (Online Figure VIII), myocardial sections were stained for picrosirius red or Masson trichrome to assess effects of deletion of Tp53 gene on fibrosis in the Lmna^D300N:Tp53^−/− mice. The findings were notable for attenuated myocardial fibrosis in the Lmna^D300N:Tp53^−/− mice, as indicated by a 3-fold reduction in CVF and reduced transcript levels of Col1a1, Col2a1, and Col3a1 in the myocardium (Figure 8, panels A-C and Online Figure XV). Similarly, given the prominence of apoptosis in the myocardium of Myh6-tTA:tetO-Lmna^D300N and in view of well-established role of TP53 in apoptosis, rescue of apoptosis was analyzed in the Lmna^D300N:Tp53^−/− mice. TUNEL staining of thin myocardial sections showed attenuation of apoptosis in the hearts of Lmna^D300N:Tp53^-/ mice, as compared to Myh6-tTA:tetO- Lmna^D300N (Figure 8, panels D and F). Finally, myocardial thin sections were also analyzed for the expression of the proliferation markers in the experimental groups. Consistent with the data on increased proliferation of non-myocyte cells in the heart in the Myh6-tTA:tetO-Lmna^D300N mice, as compared to controls, PCM1-positive cells, i.e., myocytes, did not show expression of the proliferation markers. In contrast, percent nuclei stained negative for PCM1, i.e., non-myocyte cells, and positive for expression of the proliferation markers PCNA and Ki67 were significantly increased in the Myh6-tTA:tetO-Lmna^D300N mouse hearts (Figure 8, panels E, G, and H). Deletion of Tp53 gene in cardiac myocytes attenuated proliferation of the non-myocyte cells in Myh6-tTA:tetO-Lmna^D300N mouse hearts (Figure 8, panels E, G and H). Consistent with the histological findings, echocardiographic evaluation of cardiac size and function showed partial rescue of cardiac dysfunction, as indicated by improved indices of left ventricular size and fractional shortening in the Lmna^D300N:Tp53^−/− mice as compared to Myh6-tTA:tetO-Lmna^D300N mice (Figure 9, panels A-C and Online Table III). Finally, survival analysis showed a modest improvement in overall survival upon partial deletion of Tp53 in the Myh6-tTA:tetO-Lmna^D300N mice (Figure 8, panel D). Collectively, the data indicate partial rescue of cardiac phenotype caused by mutant LMNA^D300N protein upon deletion of the Tp53 gene.

Figure 8. Partial rescue of interstitial fibrosis, apoptosis, and cell proliferation upon deletion of Tp53 gene in the Myh6-tTA:tetO-Lmna^D300N mice:

A. Micrographs representing low and high magnification picrosirius red stained thin myocardial sections showing interstitial fibrosis in the main experimental groups in the rescue experiments

B. Quantitative data on CVF showing increased levels in the Myh6-tTA:tetO-Lmna^D300N mice and its partial rescue upon deletion of Tp53 gene in cardiac myocytes. Fibrosis was attenuated in the hearts of Myh6-tTA:tetO-Lmna^D300N mice upon deletion of Tp53 gene in cardiac myocytes (Myh6-tTA:tetO-Lmna^D300N:14.27±4.24 vs Myh6-tTA:tetO-Lmna^D300N;Myh6-Cre:Tp53^F/F:2.38±0.59, p<0.0001).

C. qPCR data showing partial rescue of increased transcript levels of three procollagen genes upon deletion of Tp53 gene in cardiac myocytes in the Myh6-tTA:tetO-Lmna^D300N mice.

D. TUNEL-stained thin myocardial sections in the main experimental groups showing the apoptotic cells.

E. Immunofluorescence myocardial thin panels stained for two markers of cell proliferation Ki67 and PCNA as well as PCM1, the latter to mark cardiac myocytes, in the experimental groups. PCM1 positive cells, i.e., cardiac myocytes, did not express markers of cell proliferation, whereas, cardiac cells negative for PCM1 expression, i.e., non-myocyte cells stained positive for expression of proliferation markers in Myh6-tTA:tetO-Lmna^D300N mouse hearts. Cell proliferation was partial rescued upon deletion of the Tp53 gene (Myh6-tTA:tetO-Lmna^D300N:Myh6-Cre:Trp53^F/F group).

F. Quantitative data showing percent nuclei stained positive for markers of cell proliferations.

G and H. Quantitative data showing percent of nuclei stained for Ki67 and PCNA in thin myocardial sections. Only PCM1-negative cells (non-myocyte cells) stained positive for the expression of cell proliferation markers.

Figure 9. Partial rescue of cardiac dysfunction and survival upon deletion of Tp53 gene in the Myh6-tTA:tetO-Lmna^D300N mice:

A-C. Partial rescue of echocardiographic indices of cardiac size and function upon deletion of Tp53 gene in cardiac myocytes in the Myh6-tTA:tetO-Lmna^D300N mice.; A. left ventricular end diastolic diameter (LVEDD: 3.701±0.426 in Myh6-tTA:tetO-Lmna^D300N and 3.303±0.386 in Myh6-tTA:tetO-Lmna^D300N:Myh6-Cre-Tp53^F/F, Bonferroni pairwise comparison p=0.0335), B. left ventricular end systolic diameter (LVESD: 2.877±0.539 in Myh6-tTA:tetO-Lmna^D300N and 2.076±0.510 in Myh6-tTA:tetO-Lmna^D300N:Myh6-Cre-Tp53 ^F/F, Bonferroni pairwise p<0.0001), and C. left ventricular fractional shortening (LVEF: 22.694±7.568 in Myh6-tTA:tetO-Lmna^D300N and 37.636±9.062 in Myh6-tTA:tetO-Lmna^D300N:Myh6-Cre-Tp53 ^F/F, Bonferroni pairwise p<0.0001).

D. Kaplan-Meier survival plots showing premature death in the Myh6-tTA:tetO-Lmna^D300N group and a modest improvement in survival upon myocyte-specific deletion of the Tp53 gene (X²=126.6; p<0.0001). Sex-matched littermates are used as controls.

DISCUSSION

Exploiting the large effect size of a recently described rare pathogenic variant in the LMNA gene responsible for DCM in atypical progeroid syndromes, namely the LMNA^D300N mutation, we generated a tet-off mouse model to turn on or suppress expression of the mutant LMNA^D300N protein specifically in cardiac myocytes and to determine the underpinning mechanisms. ^{17, 18} The phenotype was remarkable for severe myocardial fibrosis, proliferation of non-myocyte cells, apoptosis, cardiac dysfunction, and increased mortality, starting within 3 weeks after birth and leading to near total mortality within two months. The causal role of the LMNA^D300N in the induced cardiac pathology was established upon suppressing expression of the transgene, which abrogated the phenotype. Cardiac transcriptome analysis from 2-week old mice, representing the early changes in gene expression occurring prior to the onset of cardiac dysfunction, indicated activation of the E2F/DDR/TP53 pathway in cardiac myocytes. Activation of this complex pathway was verified at multiple levels, including data showing increased TP53 and CDKN1A protein levels, the later the canonical downstream target of the TP53 pathway. These findings were also corroborated in human hearts with DCM carrying defined pathogenic variants in the LMNA gene. Upon establishing the relevance to human DCM and in order to substantiate the pathogenic role of TP53 in DCM, the Tp53 gene was deleted specifically in cardiac myocyte in the LMNA^D300N mice. Deletion of the Tp53 gene partially rescued molecular, histological, and functional phenotypes and modestly improved survival. Collectively, the findings provide insights into the molecular pathogenesis of DCM, the major cause of morbidity and mortality in laminopathies, and denote E2F/DDR/TP53 as a potential therapeutic target in laminopathies involving the heart. ^6–9

The findings on the pathogenic activation of E2F/DDR/TP53 pathway in DCM, caused by defined LMNA mutations, although novel, are in principle in accord with the data showing increased TP53 protein levels in human heart failure as well as in experimental models of cardiac hypertrophy and failure. ^44–48 Likewise, the findings are also in agreement with the known biological functions of TP53 in inducing apoptosis, inflammation, cell senescence, expression of senescence-associated secretory phenotype, and cell cycle arrest.^{42, 43, 49–52} In accord with the TP53 functions, myocardial apoptosis, detected by TUNEL assay was increased, NFκB pathway was activated, and transcript levels of several pro-inflammatory cytokines were increased in the hearts in mice expressing the LMNA^D300N protein. Moreover, pathogenic activation of TP53 in DCM is aligned with the role of this multi-functional transcription factor in regulating expression of a large number of cardiac genes involved in mitochondrial function, oxidative stress, metabolism, signaling transduction, sarcomere and cytoskeletal structure, extracellular matrix proteins, and angiogenesis. ^44,46 Dysregulated gene expression and biological pathways observed in the present study, are also in accord with the known functions of TP53, providing further support to pathogenic activation of the TP53 pathway in DCM-associated with LMNA mutations.

The mechanisms responsible for activation of TP53 pathway, known to regulate expression of a large number of cardiac genes under pathological conditions, are partially known and also partly illustrated in the present study. ^{44, 53} Adult hearts typically express low levels of TP53 under physiological states, as its levels are tightly regulated by E3 ubiquitin ligase mouse double minute 2 (MDM2). ⁵⁴ A putative mechanism for activation of TP53 in our model and human DCM pertains to reduced expression levels of RB1, which is stabilized by the wild type LMNA. ^40,39 RB1 is known to inhibit gene expression through the E2F transcription factor pathways. ^55,41 Accordingly, expression of the mutant LMNA^D300N results in destabilization of RB1 and activation of the E2F transcription factor pathway, which induces expression of a number of cell cycle regulators, including CDKN2A. The latter is known to inhibit E3 ubiquitin ligase MDM2, which forms a complex with TP53 and targets it for degradation, and hence, resulting in suppressed TP53-mediated transcription. ⁵⁶ Consequently, inhibition of MDM2 activity by CDKN2A is expected to result in unchecked activation of TP53 in the heart of Myh6-tTA:tetO-Lmna^D300N mouse (Online Figure XVI). The putative mechanism is not expected to be the sole mechanism for activation of TP53 in laminopathies. Given the complexity of LMNA functions in regulating gene expression directly through lamin-associated domain and indirectly through secondary mechanisms, and in view of complexity of mechanisms involved in regulation of TP53 functions, several other alternative mechanisms might be also responsible for pathogenic activation of TP53 in the heart and pathogenesis of DCM caused by LMNA mutations. ^{43, 53} The partial rescue of the observed phenotypes might reflect partial deletion of the Tp53 gene in cardiac myocytes but also suggests the presence of alternative mechanisms involved in the pathogenesis of DCM in laminopathies. A notable alternative but an interactive pathway with the TP53 pathway is the MYC, which was significantly upregulated in the hearts of Myh6-tTA:tetO-Lmna^D300N mice as well as human hearts with DCM carrying defined pathogenic variants.

The study has notable shortcomings. First and foremost, it was an over-expression study using the tet-off bigenic approach in mice. ¹⁹ The approach enabled ascertainment of causality of the LMNA^D300N protein in the pathogenesis of cardiac phenotype in the Myh6-tTA:tetO-Lmna^D300N mice. However, it does not abrogate potential fortuitous effects, in part or as a whole, resulting from over-expression of LMNA^D300N in the heart. Nevertheless, reducing the dose of the endogenous LMNA protein, upon expression of the LMNA^D300N protein in the background of LMNA haploinsufficiency (Lmna^+/−), did not affect survival, fibrosis, and other cardiac phenotypes. Although the approach has been reported to rescue progeria-like phenotype in a mouse model of laminopathies, may not be sufficient to overcome over-expression of the LMNA^D300N protein. ³⁸ It is also noteworthy that over-expression of the wild type LMNA using the same transgenic approach did not induce a phenotype in the heart, a finding that is consistent with the previous reports. ^{57, 58} Others, however, have reported osteoblastic differentiation upon over-expression of LMNA in non-cardiac tissues. ⁵⁹ Despite these precautions, the possibility of fortuitous effects resulting from over-expression of the LMNA^D300N cannot be excluded unambiguously. It should also be noted that the transgene LMNA was tagged at the N-terminal domain, which seemingly did not affect its stable expression and incorporation into the nuclear membrane. This is in contrast to tagging of LMNA at the C-terminal domain, which might affect its post-translational processing. Another shortcoming of the study is that gene expression (RNA-Seq) was performed in the whole heart transcriptome and not in isolated myocytes. The approach enabled comparison of the data and extension of the findings in the mouse models to human DCM. However, it does not allow discerning cell-type specific changes. We speculate that some of the phenotypic changes, such as fibrosis, likely reflect secondary changes in non-myocyte cells in response to paracrine factors expressed by the myocytes (Online Figures VIII and XVI). Activation of the DDR/TP53 pathway in cardiac myocytes might induce expression of senescence-associated secretary phenotype that affect the non-myocyte cells in the heart, with regards to fibrosis, proliferation, and apoptosis. Thus, the findings reflect changes in the whole heart but not cell-type specific changes resulting from expression of LMNA^D300N in cardiac myocytes. It is also notable that despite a considerable improvement in the echocardiographic indices of left ventricular function in the Myh6-tTA:TetO-Lmna^D300N mice upon partial deletion of the Tp53 gene, survival was improved only modestly. The disparity might reflect incomplete deletion of the Tp53 gene in cardiac myocytes, varying threshold effects on survival and cardiac function, as well as the complexity of the mechanisms involved in the pathogenesis of the clinical phenotypes in laminopathies. The findings also raise the possibility of premature death arising from cardiac arrhythmias in the Myh6-tTA:tetO-Lmna^D300N mice. The latter is a known characteristic of cardiac involvement in laminopathies, which was not explored in the present study, partly due to small size of the mice and premature death within a few weeks of life.⁹

In conclusion, the findings of the present study denote pathogenic activation of the DDR/TP53 pathway in DCM caused by defined LMNA mutations, which leads to induction of apoptosis, inflammation, fibrosis, dysregulated cell cycle progression, cardiac dysfunction and premature death. The findings suggest DDR/TP53 pathway might be a potential target of intervention to prevent, attenuate, and reverse heart failure in laminopathies.

NOVELTY AND SIGNIFICANCE.

What Is Known?

• Mutations in the LMNA gene cause a diverse array of phenotypes, including dilated cardiomyopathy (DCM), which are collectively referred to as laminopathies.

• DCM is a major cause of mortality and morbidity in laminopathies.

• The molecular pathogenesis of DCM due to LMNA mutations is largely unknown.

What New Information Does This Article Contribute?

• We generated doxycycline-off bigenic mice expressing the LMNA missense mutation p.Asp300Asn (LMNA^D300N), known to cause DCM and progeroid syndrome in cardiac myocytes.

• The LMNA^D300N mice showed severe cardiac fibrosis, myocardial apoptosis, cardiac dysfunction, and a near total mortality within 3–4 weeks of life. These changes were prevented upon switching off the expression of the transgene at birth.

• Cardiac transcriptome analysis, performed prior to the onset of cardiac dysfunction, showed increased levels of genes involved in the DNA damage response (DDR) and TP53 pathways..

• Activation of the DDR/TP53 pathways were also detected in human DCM associated with pathogenic variants in the LMNA gene.

• Targeted deletion of Tp53 in cardiac myocyte in mice partially rescued the phenotype.

Expression of a rare pathogenic LMNA variant in cardiac myocytes led to the activation of the DDR and TP 53 pathways and ensuing dysregulation of biological pathways downstream of TP53, including cell cycle, apoptosis, and senescence-associated secretory phenotype. The latter was reflected by the increased expression levels of transforming growth factor β and cytokines, resulting in myocardial fibrosis and increased proliferation of non-myocyte cells in the heart. The underlying mechanism involved destabilization of retinoblastoma by mutant LMNA, subsequent activation of E2F transcription factors and its downstream target - CDKN2A, which target E3 ligase MDM2 to activate the DDR/TP 53 pathways. These findings implicate the DDR/TP53 pathways as potential therapeutic targets in DCM caused by the LMNA mutation.

Nonstandard Abbreviations and Acronyms:

ATM

Ataxia telangiectasia mutated

CDKN2A

Cyclin-dependent kinase inhibitor 2A

CVF

Collagen volume fraction

DAPI

4′, 6 Diamidino-2-phenylindole dihydrochloride

DCM

Dilated cardiomyopathy

DDR

DNA damage response

DEGs

Differentially expressed genes

Dox

Doxycycline

E2F

E2F transcription factor

FPKM

Fragments per kilobase of transcript per million mapped fragments

GAPDH

Glyceraldehyde 3-phosphate dehydrogenase

GSEA

Gene Set Enrichment Analysis

H2AFX

H2A histone family member X

Ki67 (MKI67)

Marker of Proliferation Ki-67

LMNA

Lamin A/C

LMNA^D300N

Lamin A/C carrying an Asp to Asn missense mutation at amino acid position 300

NFκB

Nuclear factor kappa B

PCM1

Pericentriolar protein 1

PCNA

Proliferating cell nuclear antigen

PCR

Polymerase chain reaction

qPCR

Quantitative polymerase chain reaction

RNA-Seq

RNA-sequencing

TF

Transcription factor

TGFβ

Transforming growth factor beta

TP53

Tumor protein 53

TUBA1A

α-tubulin alpha 1 subunit

TUNEL

Terminal deoxynucleotidyl transferase dUTP nick end labeling

WT

Wild type