Abstract

   More than 10% of adults in the United States have type 2 diabetes
   mellitus (DM) with a 2–4 times higher prevalence of ischemic heart
   disease than the non-diabetics. Despite extensive research approaches
   to limit this life-threatening condition have proven unsuccessful,
   highlighting the need for understanding underlying molecular
   mechanisms. Long noncoding RNAs (lncRNAs), which regulate gene
   expression by acting as signals, decoys, guides, or scaffolds have been
   implicated in diverse cardiovascular conditions. However, their role in
   ischemic heart disease in DM remains poorly understood. We provide new
   insights into the lncRNA expression profile after ischemic heart
   disease in DM mice. We performed unbiased RNA sequencing of
   well-characterized type 2 DM model db/db mice or its control db/+
   subjected to sham or MI surgery. Computational analysis of the RNA
   sequencing of these LV tissues identified several differentially
   expressed lncRNAs between (db/db sham vs. db/db MI) including Gm19522
   and Gm8075. lncRNA Gm-19522 may regulate DNA replication via DNA
   protein kinases, while lncRNA Gm-8075 is associated with cancer gene
   dysregulation and PI3K/Akt pathways. Thus, the downregulation of
   lncRNAs Gm19522 and Gm8075 post-MI may serve as potential biomarkers or
   novel therapeutic targets to improve cardiac repair/recovery in
   diabetic ischemic heart disease.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12933-024-02441-6.

   Keywords: Transcriptome, Long noncoding RNA, Diabetes, Ischemic heart
   disease, Db/db mice

Introduction

   Diabetes mellitus (DM) is one of the leading chronic diseases globally,
   accounting for approximately 30 million cases, with an economic burden
   of $ 412 billion in the United States of America alone [[52]1, [53]2].
   Patients with diabetes mellitus are at a significantly higher risk of
   mortality and development of cardiovascular events compared to
   non-diabetic individuals. In fact, the prevalence of ischemic heart
   disease in diabetic patients is more than four times higher than that
   of non-diabetic individuals [[54]3, [55]4]. Current approaches to
   restrict heart failure in DM patients with ischemic events are limited.
   Therefore, identifying novel mechanisms and therapeutic targets that
   can modulate ischemic heart failure in DM patients will have enormous
   implications for the current methods of treating patients.

   Several noncoding RNAs have been implicated in diabetes and associated
   with heart failure [[56]5–[57]7]. LncRNAs are a new class of noncoding
   regulatory RNA, which are > 200 nt in size [[58]8]. They regulate gene
   expression by acting as signals, decoys, guides, or scaffolds in
   modifying epigenetic, transcriptional, and post-transcriptional
   mechanisms [[59]8]. Emerging evidence establish the role of lncRNAs in
   diabetic cardiomyopathy, such as inhibition of lncRNA zinc finger
   antisense 1 (ZFAS1) prevents cardiomyocyte ferroptosis and improves the
   progression of diabetic cardiomyopathy using db/db mice [[60]9].
   Additionally, elevated expression of lncRNA Kcnq1ot1 was observed in
   high-glucose treated cardiac fibroblast and diabetic myocardial tissues
   of streptozotocin (STZ) treated C57BL/6 mice [[61]6]. Furthermore,
   suppression of Kcnq1ot1 improved cardiac fibrosis in vivo and in vitro.
   However a systemic analysis of lncRNA in the context of diabetic
   ischemic heart disease has not been well-studied.

   To address this gap, we performed unbiased RNA sequencing of
   well-characterized DM model B6.BKS(D)-Lepr^db/J (db/db) mice subjected
   to sham or induced myocardial infarction (MI) surgery. We investigated
   transcriptome-wide changes in lncRNAs with or without MI. Computational
   analysis of the RNA sequencing of the left ventricular (LV) tissue
   identified differentially expressed lncRNAs between (db/db sham vs.
   db/db MI) which were association with cellular and molecular pathways
   that regulate various diseased conditions, including diabetic ischemic
   heart disease.

Materials and methods

Animal model

   The use of animals in this study followed the Guide for the Care and
   Use of Laboratory Animals published by the US National Institutes of
   Health. In addition, all experimental protocols have been approved by
   the Institutional Animal Care and Use Committee of Temple University
   School of Medicine. Eight-weeks-old leptin receptor-deficient db/+ and
   db/db male and female mice were purchased from Jackson Research
   Laboratory (Bar Harbor, ME).

Induction of myocardial infarction

   Mice were anesthetized with 2–3% isoflurane inhalation with an
   isoflurane delivery system. Following anesthesia, MI was induced in
   db/+ (5 male, 5 female) and db/db (5 male, 4 female) mice by ligation
   of the left anterior descending coronary artery (LAD) as described
   [[62]10–[63]12]. Mice were followed up for changes in LV cardiac
   function for 2- and 4-weeks post-MI, using transthoracic
   echocardiography as described [[64]10–[65]12].

RNA isolation

   Total RNA was extracted from LV tissue of db/+ and db/db mice post-sham
   or MI surgery using RNeasy Mini Kit (Qiagen), as per manufacturer’s
   protocol. All samples first underwent quality control assessment to
   ensure successful library preparation and sequencing. RNA sample
   quality was assessed by NanoDrop and Agilent 2100 BioAnalyzer. Samples
   with RNA integrity number (RIN) above 7, OD260/280: 2, and
   OD260/230 ≥ 2 were used. In addition, RNA degradation and contamination
   were monitored on 1% agarose gels before RNA-Seq [[66]13].

Library preparation and lncRNA sequencing

   Library preparation for non-coding RNAs was performed based on the
   manufacturer’s protocol (Illumina, San Diego, USA). In brief, a total
   300 ng of RNA was used to prepare the library. Quantity of prepared
   cDNA (complementary DNA) library was assessed with KAPA library
   quantification kit (KAPA Biosystem, Wilmington, USA). Further,
   sequencing was performed based on the manufacturer’s protocol
   (Illumina, San Diego, USA) by using pair-end 100 base sequencing.

Sequencing data analysis

   FASTQ files were preprocessed with fastp [[67]14] (version 0.21.0)
   using the default setting. After preprocessing of sequencing reads,
   STAR [[68]15] (version 020201) was used to map the reads to the
   reference genome (GRCm39.103). To calculate counts per million (CPM)
   values and derive differentially expressed genes, the R package, edgeR
   [[69]16] (version 3.30.3), was used. False discovery rate
   (FDR)-adjusted p-values were used for further analysis.

Data visualization and gene enrichment analyses

   RNA-Seq datasets were analyzed using iDEP [[70]17] (integrated
   Differential Expression and Pathway analysis), a comprehensive
   web-based resource that integrates 63 R/Bioconductor packages, 2 web
   services, alongside extensive annotation and pathway databases,
   accessible at [71]http://ge-lab.org/idep/. We assessed distribution of
   transformed data, batch effects, and generated Principal Component
   Analysis (PCA) and Heatmaps to ensure that our data were appropriately
   normalized and visually interpretable for pathway enrichment analysis.
   For pathway enrichment analysis, both cis- and trans-regulatory
   elements were analyzed using four different databases: (1) Kyoto
   Encyclopedia of Genes and Genomes (KEGG) Human database (2021 version),
   (2) Elsevier Pathway Collection, (3) BioPlanet 2019, and (4) The
   Molecular Signatures Database (MsigDB) 2020. Next, X2K network was
   generated to map predictive interactions between proteins, kinases, and
   transcription factors. To further analyze the functional predictions of
   lncRNAs, we used lncHUB2 [[72]18], a web server database available at
   [73]https://maayanlab.cloud/lncHUB2 that takes into account
   co-expression correlations based on ARCHS4’s processed RNA-seq samples,
   expression statistics in various tissues and cell lines, predictions on
   biological functions, molecular pathway, and subcellular localizations.

Real-time quantitative reverse transcription PCR

   We isolated total RNA from the LV tissue of db/+ and db/db mice
   post-sham or MI surgery and cDNA synthesis was performed using the cDNA
   synthesis kit (Cat # 4368813, Applied BioSystems, Waltham, MA, United
   States). Real-time polymerase chain reaction (RT-qPCR) was performed
   using the SYBR (Power Up SYBR Green Master Mix, Cat #A25742, Applied
   BioSystems, Waltham, MA, United States) and QuantStudio™ three
   Real-Time PCR. Primer sequences used are presented in supplementary
   Table [74]1.

Statistical analysis

   Results are presented as mean ± standard error of the mean (SEM). Data
   were analyzed using one way ANOVA followed by Tukey test for comparison
   of groups. Student t-test (two-tailed) was performed while comparing
   two groups. Statistical analysis was performed using GraphPad Prism 9,
   version 9.2.0 (GraphPad Software, Inc., La Jolla, CA, United States).
   Differences were considered statistically significant at p < 0.05.

Results

Establishment of MI in db/db mice

   We confirmed MI induction by echocardiography analysis in both cohorts
   of mice, including db/+ and db/db at 2 weeks and 4 weeks of post MI
   surgery. We observed MI-associated mortality rate of 40% in db + group
   and, as expected, a higher mortality rate of 56% in db/db mice at four
   weeks post-MI. Cardiac dysfunction was observed in these mice with
   reduced ejection fraction as well as increased end diastolic and
   systolic volume (Fig. [75]1A–E). The mRNA expression of atrial
   natriuretic peptide (Anp) and brain natriuretic peptides (Bnp),
   hallmarks of a failing heart [[76]19], were evaluated by qRT-PCR at 5
   weeks post-MI surgery, which revealed significant increases in these
   markers in both animal groups (Fig. [77]1F–I). This data shows we
   successfully induced MI in both groups of mice, and increased heart
   failure markers. Db/db MI mice serves as a model of diabetic ischemic
   heart disease study in this paper (Supplementary Table [78]2).

Fig. 1.

   [79]Fig. 1
   [80]Open in a new tab

   Establishment of heart failure model. M-mode representative images (A)
   depicting cardiac performance followed by the Fractional shortening (%)
   (B), ejection fraction (%) (C), end-diastolic volume (µL) (D),
   end-systolic volume (µL) (E) measurements in db/+ db/db mice before and
   after MI. mRNA expression of Anp (F–G) and Bnp (H–I) gene in db/+ and
   db/db mice hearts in sham and 5 weeks post-MI. Data is represented as
   Mean ± SEM, with 18s rRNA used as a housekeeping control for
   normalization. Two-tailed unpaired student t-test was used to compare
   the statistical significance among groups. *p < 0.05

Identification of differentially regulated lncRNA in LV tissue of db + or
db/db hearts with or without MI

   RNA isolated from the LV tissue of diabetic mice db/db and db/+ that
   underwent myocardial infarction surgery followed by deep RNA sequencing
   for lncRNA (Fig. [81]2A–D). Upon analysis, we identified several lncRNA
   exhibited differential expression. Bioinformatic analysis of sequencing
   data of db/db mice post MI revealed differentially expressed lncRNAs in
   LV tissue of mice (Fig. [82]2E). We chose to analyze the top 5
   upregulated and 5 downregulated lncRNA in the heart of db/db mice post
   MI (Fig. [83]2F–H).

Fig. 2.

   [84]Fig. 2
   [85]Open in a new tab

   LncRNA signature in LV tissue of db + or db/db hearts with or without
   MI. A Schematic outline of experimental design and downstream
   bioinformatic analysis. B Distribution of transformed RNA sequencing
   data. C Principal component analysis of RNA sequencing data. D A heat
   map of lncRNAs differentially expressed at 3 days following left
   anterior descending coronary artery (LAD) artery ligation or sham
   surgery in db + or db/db mice showing distinct profiles. E Volcano
   plots showing Log[2]-fold changes for the differentially regulated
   lncRNA in the LV tissue of db + or db/db hearts at 3 days post sham or
   MI surgery. Red dots indicate significantly up-regulated lncRNAs; blue
   dots indicate down-regulated lncRNAs. F A table showing top five up or
   down-regulated lncRNAs in the LV tissue of db + or db/db hearts at 3
   days post sham or MI surgery. G–H Normalized counts of the top 5 up or
   downregulated lncRNA in the LV tissue of db + or db/db hearts at 3 days
   post sham or MI surgery. Data are shown as Log[2] fold change > 2,
   p < 0.001, and FDR < 0.05

Validation of lncRNAs GM19522 and GM8075 in diabetic mouse hearts post-MI

   To the top two downregulated markers from the bioinformatic analysis
   Gm19522 and Gm8075, were selected for expression validation using
   real-time PCR. Gm19522-201 A and Gm19522-201B were significantly
   downregulated in post MI on day 3, week 3 and week 5 compared to sham
   mice (Fig. [86]3A, B). Similarly, the expression of Gm8075-201 A,
   Gm8075-201B and Gm8075-202B was significantly downregulated in post MI
   mice on day 3, week 3 and week 5 when compared to sham control mice
   (Fig. [87]3C–E). These results conform with bioinformatic analysis
   (Fig. [88]2E, H).

Fig. 3.

   [89]Fig. 3
   [90]Open in a new tab

   Validation of LncRNAs GM19522 and GM8075 in MI hearts of diabetic mice.
   Quantitative real-time PCR data showing downregulation in the
   expression of lncRNA GM19522-201 A (A), GM19522-201B (B), GM8075-201 A
   (C), GM8075-201B (D), and GM8075-202B (E) in sham or MI mice(db/+ or
   db/db) at 3 weeks and 5 weeks compared to sham surgery. Gapdh was used
   as housekeeping control. Data is represented as Mean ± SEM, calculated
   using. One-way ANOVA was used to compare the statistical significance
   among groups. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Pathway enrichment analysis

   To better understand the molecular pathways and biological functions
   associated with these lncRNAs, we performed pathway enrichment analysis
   using several databases, including the KEGG 2021 human database,
   Elsevier pathway collection, BioPlanet 2019, and MSigDB 2020
   (Fig. [91]4A) [[92]13]. Our results suggest that pathways related to
   RNA transport, protein synthesis, protein transport, and degradation
   may be dysregulated (Fig. [93]4A). Furthermore, we evaluated the
   downstream association of lncRNAs to transcription factors and
   associated kinases by performing Transcription Factor Enrichment
   Analysis (TFEA) (Fig. [94]4B) and Kinase Enrichment Analysis (KEA)
   (Fig. [95]4C). TEFA analysis uncovered important transcription factors,
   including TAF1, MYC, CREB1, MAX, and USF2 (Top 5 listed here)
   (Fig. [96]4B), while KEA analysis revealed changes in MAPK14, CDK2,
   CSNK2A1, CDK1 and ATM (Top 5 listed here) (Fig. [97]4C).

Fig. 4.

   [98]Fig. 4
   [99]Open in a new tab

   Pathway enrichment analysis. A A comprehensive analysis to identify
   downstream target genes of (cis- and trans-regulatory) of the
   differentially expressed lncRNAs using multiple databases 2021 Kyoto
   Encyclopedia of Genes and Genomes (KEGG) Human database, Elsevier
   Pathway Collection, BioPlanet 2019, and the Molecular Signatures
   Database 2020 (MsigDB). B lncRNAs potential downstream transcription
   factors and kinases analysis performed using Transcription factor
   enrichment analysis (left panel), and C identified key kinases involved
   via kinase enrichment analysis (right panel)

The eXpression2Kinases (X2K) network displays predictive interactions of
proteins, kinases and transcription factors

   Analysis of TFEA and KEA facilitated the construction of a network
   illustrating the interactions between transcription factors and kinases
   influenced by the dysregulated lncRNAs.The analysis of
   eXpression2Kinases (X2K) network revealed that SP1, MYC, MAX, ATF2, and
   CREB1 are the most affected transcription factors (Fig. [100]5, denoted
   in pink color) are associated with CDC2 and DNAPK kinases (Fig. [101]5,
   denoted in blue color).

Fig. 5.

   [102]Fig. 5
   [103]Open in a new tab

   The eXpression2Kinases (X2K) network analysis and predictive
   interactions. The X2K network illustrates the predictive interactions
   of proteins, kinases, and transcription factors. Pink nodes represent
   the top transcription factors predicted to regulate the expression of
   the input gene list; blue nodes represent the top predicted protein
   kinases known to phosphorylate the proteins within the expanded
   subnetwork. Green network edges/links represent kinase-substrate
   phosphorylation interactions between kinases and their substrates,
   while grey network edges represent physical protein-protein
   interactions

Discussion

   Current investigation aims to establish the link between the expression
   of lncRNAs and their potential role in diabetic ischemic heart
   diseases. In recent years, lncRNAs have emerged as potential biomarkers
   in diabetic heart diseases [[104]7, [105]20–[106]22]. One of the recent
   findings by Pant et al., reported that five Lnc RNAs including,
   XLOC015617, [107]AK035192, Gm10435, TCR-α chain, and MouselincRNA0135
   were found deregulated in mouse model of diabetic cardiomyopathy
   [[108]22]. In this study, we provide first evidence showing significant
   downregulation in the expression of lncRNAs (Gm 8075 and Gm 19522) with
   the progression of MI in diabetic mice. Although there is currently no
   literature specifically addressing these lncRNAs in the context of
   diabetic ischemic heart disease. Our results suggest they may have an
   involvement in key molecular pathways in the disease process. Future
   studies are warranted to determine functional signifance of these
   lncRNAs in diabetic ischemic heart disease. Our study provides a novel
   basis for further investigation into their potential role as
   biomarkers, therapeutic targets or in novel mechanisms. By exploring
   novel lncRNAs we may identify new insights into pathophysiology of MI
   in diabetes which would ultimately aid developing innovative approaches
   to diagnose or treat diabetic heart disease.

   Interestingly, our computational data analysis showed that these
   lncRNAs are involved in the regulation of various molecular pathways,
   including RNA transport, protein synthesis, and degradation. Our TEFA
   and KEA analysis establishes a link between transcription factors and
   kinases associated with lncRNAs. LncRNAs are known to modulate
   biological functions at cellular and molecular levels by regulating
   various genes [[109]8, [110]20]. LncRNAs can also affect the expression
   pattern of kinases and affect the outcome of the biological processes
   [[111]23]. In the current study, we investigated the interaction
   between transcription factors and kinases post-analysis of TFEA and KEA
   and found that CDC2 and DNAPK were the most affected kinases that
   showed the relationship with various transcription factors, including
   SP1, MYC, ATF2, YY1, and CREB1 (Fig. [112]5; transcription factors are
   highlighted in pink whereas kinases are highlighted in blue). These
   factors regulate the cell cycle, DNA damage response, and
   transcriptional control. For example, CDC2 is crucial for cell cycle
   progression and proliferation in vascular smooth muscle cells and
   cardiomyocytes. Overexpression of CDC2 promotes cardiomyocyte
   proliferation [[113]24, [114]25], an important process in cardiac
   repair mechanisms post-injury. Additionally, CDC2 is implicated in the
   hyperplastic proliferation of vascular smooth muscle cells in pulmonary
   arterial hypertension [[115]26], which could lead to vascular
   obstruction and elevate blood pressure in the pulmonary circulation,
   and lead to right heart failure. In diabetic ischemic heart disease,
   abnormal activation of CDC2 may cause maladaptive vascular and cardiac
   remodeling by promoting excessive proliferation or disrupting normal
   cell cycle control in cardiomyocytes. DNA-PK helps maintain genomic
   stability by repairing DNA double-strand breaks during cell division,
   thus preventing mutations and chromosomal abnormalities. In
   cardiomyocytes, activation of DNA-PK during ischemic conditions
   exacerbates mitochondrial dysfunction and apoptosis, contributing to
   pathological remodeling and cardiac dysfunction [[116]27]. Other
   studies showed that DNA-PK contributes to vascular proliferative
   disease processes such as neointimal formation in atherosclerosis-prone
   mice [[117]28]. Similarly, the transcription factor SP1, involved in
   cell growth, differentiation, and DNA damage response, has been
   associated with hypertrophic cardiomyopathy by regulating the
   expression of genes critical for structural integrity and mitochondrial
   function, essential for cardiomyocyte function under stress [[118]29].
   Finally, MYC has been extensively described for its role in cellular
   metabolism and growth by enhancing glucose metabolism and mitochondrial
   biogenesis in cardiomyocytes, which could initially be a beneficial
   adaptive mechanism in response to acute stress [[119]30–[120]32].
   However, persistent myc activation in the context of chronic stress
   could lead to hypertrophic cardiomyopathy and heart failure. In cardiac
   cells, CREB1 regulates genes essential for cell survival, metabolism,
   and mitochondrial function. Cardiac-specific overexpression of
   dominant-negative CREB leads to increased mortality and mitochondrial
   dysfunction in female mice [[121]33]. Its dysregulation in diabetes can
   lead to mitochondrial dysfunction and cell death, processes central to
   various cardiovascular diseases, including myocardial infarction and
   heart failure. The metabolic disturbances and hyperglycemia in diabetic
   ischemic heart disease might alter the activity of kinases and
   transcription factors like DNA-PK, CDC2, MYC, SP1, and CREB1,
   exacerbating cardiac dysfunction. These interactions highlight the
   complexity of diabetic ischemic heart disease pathogenesis and provide
   new opportunities to explore how these therapeutic targets are
   regulated to improve cardiac function in diabetes.

   Importantly, our KEGG analysis suggested the role of lncRNA Gm-19522 in
   various molecular processes, including DNA replication and repair,
   indicating a possible involvement of DNAPK (Supplementary Fig. [122]1).
   Interestingly, DNAPKs are known to the regulate cardiomyocyte-specific
   mitochondrial apoptosis in post ischemia [[123]27]. However, lncRNA
   Gm-8075 showed involvement in the dysregulation of genes of various
   cancers, including gastric cancer, breast cancer, and melanoma, as well
   as other metabolic pathways like PI3K/Akt. However, future studies are
   warranted to tease out the missing links of molecular mechanism that
   govern the relationship between lncRNAs, kinases and transcription
   factors. Future studies are warranted to determine lncRNAs interactome
   in diabetic ischemic heart disease.

   Db/db mice are a widely used model for studying diabetes induced
   cardiovascular disease [[124]34]. They develop diastolic dysfunction
   without additional stressors [[125]35]. In our study, 10-12-week-old
   db/db sham injury mice exhibited significant changes in lncRNA
   expression, including Gm43332 (upregulated) and Gm5420 (downregulated)
   compared to db + sham mice (Supplementary Fig. [126]2). Further KEGG
   pathway analysis revealed their potential roles in fatty acid
   biosynthesis, oxidative phosphorylation, protein digestion/absorption,
   mucin type O-glycan biosynthesis, and N-glycan biosynthesis. Further
   research is warranted to understand the specific roles of these lncRNAs
   in diabetes induced diastolic dysfunction.

Conclusions

   In summary, our current study highlights downregulation of lncRNAs
   Gm19522 and Gm 8075 expression post-MI in diabetic mice. These findings
   suggest that Gm19522 and Gm8075 may serve as potential biomarkers or as
   potential therapeutic targets for improving cardiac repair/recovery in
   diabetic ischemic heart disease.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [127]12933_2024_2441_MOESM1_ESM.tif^ (1.9MB, tif)

   Supplementary Figure-1: KEGG and ChEA analysis of our top target
   lncRNAs Gm19522 and Gm8075 in db/db MI vs db/db sham mice.
   [128]12933_2024_2441_MOESM2_ESM.tif^ (722.2KB, tif)

   Supplementary Figure-2: Volcano plots showing Log2-fold changes for the
   differentially regulated lncRNAs in the LV tissue of db/+or db/db
   hearts without MI.
   [129]12933_2024_2441_MOESM3_ESM.xlsx^ (10KB, xlsx)

   Supplementary Table 1: Primers sequences used in this study.
   [130]12933_2024_2441_MOESM4_ESM.xlsx^ (9.9KB, xlsx)

   Supplementary Table 2: Echocardiographic measurements.

Abbreviations

   Anp
          Atrial natriuretic peptide

   Bnp
          Brain natriuretic peptide

   cDNA
          Complementary DNA

   DM
          Diabetes mellitus

   FDR
          False discovery rate

   IDEP
          Integrated Differential Expression and Pathway Analysis

   KEA
          Kinase Enrichment Analysis

   KEGG
          Kyoto Encyclopedia of Genes and Genomes

   LAD
          Left anterior descending

   lncRNA
          Long non-coding RNA

   LV
          Left ventricle

   MI
          Myocardial infarctiom

   MsigDB
          Molecular Signatures Database

   PCA
          Principal Component Analysis

   RIN
          RNA integrity number

   RT-qPCR
          Real-time polymerase chain reaction

   SEM
          Standard error of the mean

   TFEA
          Transcription Factor Enrichment Analysis

   X2K
          eXpression2Kinases

   ZFAS1
          Zinc finger antisense 1

Author contributions

   VNSG conceived the idea, and planned experiments. AKR, NSM, IDK, CDL,
   MP, RR, SU, and MB, executed and performed the experiments. VNSG, AKR,
   MB, NN, RK, DAG, and RK wrote the manuscript. TL and MP helped design
   the graphical abstract and RK, and KD critically evaluated the
   manuscript and provided comments. AKR performed and analyzed (Fig. 1 F,
   G, H, I and 3 A, B, C, D and E). NSM helped in preparing in Fig. 3. RR
   helped acquire the data in Fig. 1 A-E and IDK, MP, and CDL analyzed the
   data(Figure 1 A-E). MB helped in acquiring the Figs. 4 and 5 and
   supplementary Figs. 1 and 2. SU helped in analyzing the bioinformatics
   data (Fig. 2). All authors reviewed the manuscript.

Funding

   This work was partially supported by American Heart Association
   Transformational Project Award 23TPA1140823, Innovative Project Award
   23IPA105444, and National Institute of Health R01 grant HL164794 and
   startup funds from the Temple University to V.N.S.G and [131]HL143892
   to R. K. AKR was supported with the American Heart Association-Post
   Doctoral Fellowship grant 2021AHA000POST000217114. MB was supported by
   the NIH/NHLBI 1K01HL159038-01A1, 5R25HL146166, American Heart
   Association- Career Development Award 24CDA1269532, and Unrestricted
   Research Grant. Authors declare no competing interests.

Data availability

   “Data is provided within the manuscript or supplementary information
   files”.

Declarations

Competing interests

   The authors declare no competing interests.

Footnotes

   Publisher’s note

   Springer Nature remains neutral with regard to jurisdictional claims in
   published maps and institutional affiliations.

References