Noncoding RNAs in Cardiovascular Disease

Authors, Journal, Affiliations, Type, DOI

• Authors: Saumya Das (Chair), Ravi Shah (Co-chair), Stefanie Dimmeler, Jane E. Freedman, Christopher Holley, Jin-Moo Lee, Kathryn Moore, Kiran Musunuru, Da-Zhi Wang, Junjie Xiao, Ke-Jie Yin; on behalf of AHA Council on Genomic and Precision Medicine
• Journal: Circulation: Genomic and Precision Medicine
• Date: August 2020
• Type: AHA Scientific Statement (systematic review + expert consensus)
• DOI: 10.1161/HCG.0000000000000062

Overview

This AHA Scientific Statement provides a comprehensive synthesis of all major noncoding RNA (ncRNA) classes and their roles in cardiovascular disease as of 2018. It covers miRNAs, snoRNAs, Y-RNAs, tRNA-derived fragments, long noncoding RNAs (lncRNAs), circular RNAs (circRNAs), and extracellular RNAs (exRNAs), alongside computational tools and experimental methodologies for studying each class. Key translational highlights include the clinical-stage siRNA inclisiran (PCSK9-targeting), anti-miR approaches to cardiac remodeling, and exRNAs as circulating biomarkers. The authors emphasise that methodological heterogeneity and annotation gaps remain major barriers to clinical translation, and that network-level approaches (rather than single-RNA studies) are needed to understand complex RNA–RNA interactions in disease.

Keywords

AHA Scientific Statements, cardiovascular diseases, microRNAs, RNA long noncoding, RNA small nucleolar, RNA untranslated

Key Takeaways

Computational Tools and Databases

• RNA-seq provides unprecedented transcriptome depth but creates annotation challenges; short reads (≤200 bp) map to sncRNA; longer reads to mRNA and lncRNAs
• Key databases: miRbase (miRNAs), gtRNAdb (tRNAs), piRNABank, GENCODE, UCSC for snoRNA/snRNA; frequent version updates complicate reproducibility
• RNA-binding protein databases (ATtRACT, RBPDB) enable search for protein–RNA interactions
• Multi-omics integration (genomics + transcriptomics + proteomics) is possible but challenged by heterogeneous experimental approaches; longitudinal personal omics pioneered in single individuals

Small Noncoding RNAs: MicroRNAs (miRNA)

• miRNAs are 17–22 nt ncRNAs that decrease target gene expression by altering mRNA stability or translation via the RISC complex
• Biogenesis: pri-miRNA (RNA Pol II) → pre-miRNA (~70 nt, Microprocessor) → cytoplasm export → Dicer cleavage → ~22 nt mature miRNA incorporated into RISC
• One miRNA can target hundreds of genes; one gene can be targeted by multiple miRNAs — supports key regulatory roles in cardiac development, physiology, and disease
• Therapeutic modulation: anti-miRs, antagomiRs (cholesterol-conjugated), LNA-anti-miRs; catheter-based intracoronary delivery enhances cardiac targeting; major challenges are organ-specific delivery and off-target effects

Small Noncoding RNAs: snoRNAs

• snoRNAs (60–200 nt) reside in the nucleolus and primarily guide rRNA/snRNA modifications; >200 well-defined snoRNAs with hundreds more predicted
• Box C/D snoRNAs (SNORDs) guide 2'-O-methylation; Box H/ACA snoRNAs (SNORAs) guide pseudouridylation
• Cardiometabolic: Rpl13a locus snoRNAs regulate oxidative stress and diabetes; Snord32A targets peroxidasin mRNA to alter cardiac reactive oxygen species; Snora73 regulates cholesterol homeostasis via Hummr
• Stroke/CHD: 6 miRNAs and 1 snoRNA (SNO1402) associated with prevalent stroke in FHS (n=2763); no clear functional connection established
• Congenital HD: lower snoRNA expression in tetralogy of Fallot; zebrafish knockdown of Snord94/Scarna1 → abnormal cardiac development

Small Noncoding RNAs: Y-RNAs

• Y-RNAs (83–112 nt; 4 types in humans) form stem-loop structures; stem highly conserved; loop harbours cleavage sites generating s-RNY fragments (~20–40 nt)
• s-RNY fragments measurable in blood as cell-free ribonucleoprotein complexes and in extracellular vesicles
• Atherosclerosis: atherogenic stimuli (ox-LDL, saturated fatty acids) strongly upregulate macrophage s-RNY; mouse atherosclerosis models show elevated plasma s-RNY; patients with CAD have higher serum s-RNY than controls (confounded by multiple co-morbidities)
• MI: in CADUCEUS trial cardiosphere-derived cells (CDCs), EVs contain 18% Y-RNA fragments; EV-YF1 (5'-half of Y-RNA 4) correlates with cardiac functional improvement in mouse MI model; direct injection reduces infarct mass and cardiomyocyte apoptosis
• Hypertrophy/fibrosis: EV-YF1 attenuates hypertrophy and reduces inflammation/fibrosis markers in angiotensin II mouse model; associated with reduced IL-10 expression

Small Noncoding RNAs: tRNA-Derived sRNAs (tsRNAs)

• tiRNAs (tRNA halves): generated by angiogenin cleavage at anticodon loop; accumulate under stress (hypoxia, heat shock, nutritional deprivation); inhibit translation, regulate apoptosis and stress granules
• tiRNAs upregulated in rat stroke model (tRNA[Val], tRNA[Gly]) and mouse hindlimb ischemia; synthetic tiRNAs inhibit endothelial proliferation/migration/tube formation — mechanism unclear
• tRNA-derived fragments (tRFs): generated by Dicer at T/D loops; some tRF-5' fragments interact with Argonaute proteins in RISC-like manner; tRFs enriched in isoproterenol-induced hypertrophic hearts; overexpression increases ANP, BNP, α-MHC and cardiomyocyte hypertrophy; tRFs can target Timp3 3'-UTR
• Heavy tRNA modification interferes with reverse transcription — bacterial AlkB enzyme pre-treatment recommended for tRNA/tRF sequencing

Long Noncoding RNAs (lncRNAs)

• lncRNAs are >200 nt transcripts sharing mRNA features (RNA Pol II, 5'-capped, polyadenylated, spliced); >30,000 predicted in human genome; lower abundance and poor conservation vs. mRNAs
• Classified by genomic location (lincRNA, intronic, eRNA, AS-lncRNA) and function (guides, scaffolds, decoys, ceRNAs)
• ceRNA function: lncRNAs sponge miRNAs, preventing their repression of mRNA targets — validation requires coprecipitation, miRNA target derepression upon lncRNA deletion, and loss of sponging after miRNA binding site mutation; ceRNA stoichiometry remains controversial
• CVD risk loci lncRNAs: ANRIL (9p21) — scaffold for CDKN2a/B regulation; circularised ANRIL isoforms take on atherogenic function; MIAT — variant associated with MI risk; H19 — variants increase IGF2 expression and CAD risk
• Metabolism: MALAT1 and H19 stabilise SREBP-1c to promote hepatic lipogenesis; LeXis (LXR-induced) reduces cholesterol synthesis; MeXis (macrophage LXR-induced) promotes ABCA1/cholesterol efflux; APOA1-AS and APOA4-AS regulate apolipoprotein expression
• Vascular biology: PUNISHER, MEG3, GATA6-AS regulate vessel formation; LincRNA-p21 restricts smooth muscle proliferation; MALAT1 regulates angiogenesis and contributes to diabetic retinopathy and atherosclerosis
• Cardiac hypertrophy/HF: Mhrt — cardiac-specific; repressed in pressure-overload hypertrophy; restoration is cardioprotective; potential HF biomarker; Chast — impedes cardiomyocyte autophagy by negatively regulating Plekhm1; Chaer — interacts with PRC2 catalytic subunit to alter epigenetic reprogramming; silencing either attenuated pathological remodeling in mice
• Stroke: Multiple lncRNAs differentially expressed in ischemic stroke whole blood; MIAT expression correlates with poor prognosis; ANRIL variants correlate with stroke risk at 9p21.3; Malat1 may protect against ischemic brain injury via proapoptotic/proinflammatory inhibition
• Tools: RNA-seq (with sufficient depth); CRISPRi/CRISPRa for lncRNA manipulation; ASOs/Gapmers for knockdown; chromatin isolation by RNA purification (ChIRP) + mass spectrometry for interactome mapping; digital droplet PCR + RNA-FISH for copy number and localisation
• Up to 20% of lncRNAs in human heart may encode micropeptides — blurring noncoding/coding boundary

Circular RNAs (circRNAs)

• circRNAs generated by back-splicing (3' exon end joined to upstream 5' exon end); lack poly(A) tails; detected by rRNA-depleted or ribonuclease R–enriched RNA-seq; validated by poly(A) depletion + RNase R resistance + Northern blotting
• Tools: circtools, CIRCexplorer2, CIRI2 for detection; StarBase3, CircNet, circInteractome for interaction prediction
• Biological examples: cZNF292 (endothelial angiogenic sprouting); circ_lrp4 (smooth muscle proliferation); circANRIL (atheroprotection); HRCR and circFoxo3 (cardiac hypertrophy and senescence)
• CDR1as has 63 binding sites for miR-7 — paradigmatic miRNA sponge; genetic deletion confirms in vivo function
• Detected in human blood (high stability); copy numbers are low — clinical utility as biomarkers requires larger validation studies

Extracellular RNAs (exRNAs)

• exRNAs encompass all circulating ncRNAs (miRNA, piRNA, snoRNA, lncRNA, circRNA, tRNA fragments) carried in EVs, bound to lipoproteins, or RNA-binding proteins
• FHS exRNA atlas (n=2763): miRNA, piRNA, snoRNA, circRNA, tRNA fragments all detectable in human circulation
• Major quantification challenges: low abundance, sample handling instability, heparin interference with RNA assays, biofluid selection (whole blood vs. plasma vs. serum vs. vesicle), technical biases in RNA isolation and library construction
• Three quantification platforms: RNA-seq (unbiased but library bias), RT-PCR (a priori target selection), hybridisation-based direct quantification (limited breadth)
• CVD biomarker associations reported for: AF (miR-21, miR-150), MI (miR-208b, miR-1, miR-133), cardiometabolic disorders, cardiac arrest
• Key unmet need: tissue-specific markers for exRNA carrier subtypes — the Extracellular RNA Communication Consortium (ERCC) is addressing this

ncRNAs in Exercise

• miRNAs dynamically regulated by exercise: upregulated — miR-21, miR-150, miR-222, miR-17-3p, miR-223; downregulated — miR-1, miR-133a, miR-208
• Proposed mechanisms: promotion of physiological cardiac hypertrophy, cardiomyocyte proliferation, angiogenesis, reduced fibrosis
• Circulating miR-222 and miR-17-3p increased after exercise training in both healthy athletes and HF patients
• lncRNAs in exercise: largely unstudied; one microarray study showed lncRNA FR030200 downregulation in aortic endothelium after exercise, cis-regulating col3a1

RNA Therapeutics in CVD

• siRNA (inclisiran): Targets PCSK9 mRNA in hepatocytes; N-acetylgalactosamine conjugate for liver targeting; phase II trial showed sustained LDL-C reduction at 240 days after 2 doses; phase III trials ongoing at time of publication
• siRNA (patisiran): Targets transthyretin mRNA; phase III trial showed improvement in multiple clinical endpoints for ATTR amyloidosis
• ASOs: First-generation ASOs against PCSK9 insufficient affinity; second-generation LNA-based ASOs show promise in primates and early human studies; RNase H-mediated degradation of RNA:DNA hybrid
• Anti-miR strategies: Systemic administration of LNA-anti-miRs or antagomiRs achieves target miRNA knockdown in vivo; intracoronary delivery enhances cardiac targeting; phase I trials ongoing for anti-miR-92a (NCT03603431, NCT03494712); phase I trial for CDR132L (anti-miR-132) for cardiac remodeling (NCT04045405)
• AAV-delivered miRNA overexpression: Promising results with AAV-miR-199 in porcine MI model, but dose-dependent toxicity observed — careful dosing essential before human translation
• Key hurdles: Cardiac-specific delivery, RNA chemistry stability, off-target effects, pharmacokinetics/pharmacodynamics

Limitations of the Document

• Literature review through 2018; field has evolved substantially
• Most functional data from in vitro or mouse/rat models — human translation unproven for most ncRNA classes
• exRNA biomarker studies show inconsistent associations across cohorts, biofluids, and quantification methods — lack of standardised protocols
• lncRNA annotation is species-dependent and lagging; >20% of putative lncRNAs may encode micropeptides (complicating classification)
• ceRNA sponging stoichiometry often not rigorously validated
• Single-RNA mechanistic studies dominate — complex RNA–RNA interaction networks understudied

Key Concepts Mentioned

• concepts/Noncoding-RNA-in-CVD — umbrella concept covering all ncRNA classes
• concepts/Long-Noncoding-RNA — lncRNA classification, function, CVD roles, tools
• concepts/Gene-Silencing-Therapy — siRNA, ASO, anti-miR modalities; inclisiran and patisiran clinical data added
• concepts/AAV-Gene-Delivery — AAV-delivered miRNA overexpression in porcine MI model

Key Entities Mentioned

• entities/ATTR-Amyloidosis — siRNA (patisiran) phase III benefit

Wiki Pages Updated

• wiki/sources/noncoding-rna-aha-2020.md (created)
• wiki/sourceindex.md (updated)
• wiki/wikiindex.md (updated)
• wiki/concepts/Noncoding-RNA-in-CVD.md (created)
• wiki/concepts/Long-Noncoding-RNA.md (created)
• wiki/concepts/Gene-Silencing-Therapy.md (updated — inclisiran/patisiran/anti-miR clinical data)