Long Noncoding RNAs

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY

VOL. 67, NO. 10, 2016

ª 2016 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER

http://dx.doi.org/10.1016/j.jacc.2015.12.051

THE PRESENT AND FUTURE STATE-OF-THE-ART REVIEW

Long Noncoding RNAs From Clinical Genetics to Therapeutic Targets? Reinier A. Boon, PHD,a,b Nicolas Jaé, PHD,a Lesca Holdt, MD, PHD,b,c Stefanie Dimmeler, PHDa,b

ABSTRACT Recent studies suggest that the majority of the human genome is transcribed, but only about 2% accounts for proteincoding exons. Long noncoding RNAs (lncRNAs) constitute a heterogenic class of RNAs that includes, for example, intergenic lncRNAs, antisense transcripts, and enhancer RNAs. Moreover, alternative splicing can lead to the formation of circular RNAs. In support of putative functions, GWAS for cardiovascular diseases have shown predictive singlenucleotide polymorphisms in lncRNAs, such as the 9p21 susceptibility locus that encodes the lncRNA antisense noncoding RNA in the INK4 locus (ANRIL). Many lncRNAs are regulated during disease. For example, metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) and myocardial infarction-associated transcript (MIAT) were shown to affect endothelial cell functions and diabetic retinopathy, whereas lincRNA-p21 controls neointima formation. In the heart, several lncRNAs were shown to act as microRNA sponges and to control ischemia-reperfusion injury or act as epigenetic regulators. In this review, the authors summarize the current understanding of lncRNA functions and their role as biomarkers in cardiovascular diseases. (J Am Coll Cardiol 2016;67:1214–26) © 2016 by the American College of Cardiology Foundation.

I

Listen to this manuscript’s

n the early days of molecular biology, the catego-

was solved stepwise by the discovery of intronic

rization of ribonucleic acids (RNAs) was rela-

sequences (4,5) and various functional noncoding

tively

RNAs, such as the RNA component of RNaseP (6),

straightforward,

dividing

RNAs

into

protein-coding messenger RNAs (mRNAs) and a

small nuclear RNAs (7), longer noncoding transcripts

reasonable number of functional noncoding RNAs,

(e.g., Xist and H19 [8,9]), and microRNAs (miRNAs)

such as transfer RNAs or ribosomal RNAs. However,

(10) (Figure 1A). Meanwhile, the biogenesis and func-

by comparing the sizes of haploid genomes (C values)

tion of many of these noncoding RNAs is well

from different organisms, it was soon discovered that

described. MiRNAs, for example, are small cyto-

these values do not necessarily correlate with devel-

plasmic RNAs of 20 to 23 nt in length exhibiting an

opmental complexity (1). The assumption was that

important gene-regulatory potential. They are tran-

the majority of the genome is noncoding DNA or

scribed as primary transcripts, processed to precur-

“junk” DNA (2). Nevertheless, this supposed non-

sors in the nucleus, and subsequently exported to

functionality was already questioned at this time (1),

the cytoplasm to be diced into mature double-

and subsequent RNA-driven hybridization reactions

stranded miRNAs. Of this duplex, 1 strand is incorpo-

with nonrepetitive sea urchin DNA revealed that so-

rated into RNA-induced silencing complexes to

called heterogeneous nuclear RNA covers a larger

silence gene expression at the post-transcriptional

portion of the genome than mRNA isolated from pol-

level (see He et al. [11] for a detailed review). In

ysomal fractions (3). This obvious imbalance between

contrast, small nuclear RNAs are part of small nuclear

RNA being transcribed and known RNA species

ribonucleoproteins, which participate in the dynamic

audio summary by JACC Editor-in-Chief Dr. Valentin Fuster. From the aInstitute for Cardiovascular Regeneration, University of Frankfurt, Frankfurt, Germany; bGerman Center of Cardiovascular Research, Berlin, Germany; and the cInstitute of Laboratory Medicine, Ludwig-Maximilians-University Munich, Munich, Germany. Drs. Boon, Jaé, and Dimmeler have filed for a patent regarding long noncoding RNAs in cardiovascular disease. Dr. Holdt has reported that she has no relationships relevant to the contents of this paper to disclose. Manuscript received October 13, 2015; revised manuscript received November 26, 2015, accepted December 14, 2015.

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

1215

Long Noncoding RNAs in Cardiovascular Disease

assembly of the spliceosome on nascent pre-mRNAs

methyl-CpG-binding domain protein 1 to re-

ABBREVIATIONS

and the process of splicing (12). Finally, with the

cruit H3K9 methyltransferases to its own

AND ACRONYMS

advent of next-generation sequencing technologies,

imprinted gene network, thus establishing

it became evident that the human genome is nearly

repressive chromatin marks, which, in turn,

pervasively transcribed, which means that the major-

control gene expression of the given network

ity of its bases are associated with at least 1 primary

members (29). An additional function of

transcript (13,14); however, protein-coding exons ac-

lncRNAs is their involvement in develop-

count for only about 2% of the genome (15,16).

mental differentiation. The lncRNA Neat1, for

Today,

long

noncoding

RNAs

(lncRNAs)

are

example, is expressed in a broad range of

ANRIL = antisense noncoding RNA in the INK4 locus

APF = autophagy-promoting factor

Bvht = Braveheart CAD = coronary artery disease CARL = cardiac-apoptosis

frequently defined as non-protein-coding transcripts

tissues

larger than 200 nt to distinguish them from small

cell

noncoding RNAs (17); however, their classification is

Neat1 is implicated in the assembly of para-

not standardized. Accordingly, lncRNA nomenclature

speckles (nuclear subdomains of unclear

CDKN2B = cyclin-dependent

relies on diverse empirical features, for example,

function) by recruiting paraspeckle proteins,

kinase inhibitor 2B

origin of transcription, tissue specificity, molecular

as well as in the maintenance of these dy-

CHRF = cardiac hypertrophy

function, or mechanism of action (see Ma et al. [18]

namic nuclear structures (32).

and

is

differentiation

up-regulated (30,31).

upon

Functionally,

for a detailed review). Regarding their genomic

The detailed mechanism of action is

localization, lncRNAs can be transcribed from inter-

closely linked with lncRNA function. Those

related long noncoding RNA

Cas9 = CRISPR-associated protein 9

related factor

circRNA = circular RNA CRISPR = clustered regularly interspaced short palindromic

genic regions (long intervening noncoding RNAs),

lncRNAs implicated in the regulation of

from within introns of protein-coding genes (intronic

chromatin states often recruit chromatin-

lncRNAs), or from the antisense strand of a given

modifying complexes to genes in the vicin-

association studies

lncRNA = long noncoding RNA

repeats

GWAS = genome-wide

gene (natural antisense transcripts) (19). Interest-

ity of their own transcription (so-called cis

ingly, the primary sequence of lncRNAs is only poorly

regulation). A prominent example is the

conserved; however, this circumstance is partially

lncRNA antisense noncoding RNA in the INK4

associated lung

compensated by a higher degree of structural con-

locus (ANRIL), which is transcribed close to

adenocarcinoma transcript 1

servation

the genomic locus of INK4A and interacts

Frequently,

(20)

to

lncRNAs

maintain

from

with components of the polycomb repressive complex 1 (PRC1) to induce silencing of

genes in different species), indicating a putative

INK4A (33). Beyond being restricted to their

functional relation, despite the previously mentioned

sites of transcription, lncRNAs can also re-

sequence divergence (22). Back-splicing of exons,

cruit epigenetic complexes to distant chro-

thereby forming circular RNAs (circRNAs), can also

mosomal sites in order to regulate gene

generate lncRNAs (23,24). In contrast to the previ-

expression (so-called trans regulation). For

ously

better

example, the lncRNA HOTAIR, which is

conserved and are more stable because of their

transcribed from the HOXC locus, represses

resistance to exonucleases (24,25).

transcription in trans across 40 kb of the

lncRNAs,

also

transcribed

(21).

conserved genomic regions (same order of flanking

described

are

functionality

circRNAs

are

LncRNA molecular functions are very diverse

HOXD locus (34). Besides their well-known

MALAT1 = metastasis-

MDRL = mitochondrial dynamic related long noncoding RNA

Mhrt = myosin heavy chainassociated RNA transcript

MIAT = myocardial RNA transcript

miRNA = micro–RNA mRNA = messenger RNA NFIA = nuclear factor IA PRC1 = polycomb repressive complex 1

PRC2 = polycomb repressive complex 2

(Central Illustration). One of the best-studied aspects

influence on transcription and chromatin

is their role in the epigenetic regulation of allelic

states, lncRNAs can also act as molecular

endothelial cell-enriched

expression. A generic model for this is the well-

scaffolds to organize protein complexes or,

migration/differentiation-

known process of X chromosome dosage compensa-

conversely, as decoys or sinks for specific

tion in female mammals. X chromosome inactivation

target molecules to limit their availability.

is achieved by expression of the lncRNA Xist, which

For instance, HOTAIR is, in addition to its

coats 1 X chromosome and acts as a scaffold for the

epigenetic function, known to act at the post-

recruitment of silencing factors (e.g., polycomb

translational level by serving as assembly

repressive complex 2 [PRC2]) (26,27). Related to X

platform for protein ubiquitination (35), whereas the

chromosome inactivation is the developmental pro-

lncRNA PANDA inhibits the expression of apoptotic

cess of genomic imprinting, which results in the

genes by sequestering the transcription factor NF-YA,

exclusive expression of specific genes from only 1

thereby acting as molecular sink (36). Another

parent chromosome (28). In this context, the highly

example of an lncRNA acting as a molecular decoy is

expressed long intervening noncoding RNA H19,

lncRNA-ATB, which competitively binds miR-200 to

now known to be part of an imprinted gene network,

sequester this miRNA from its mRNA targets ZEB1/2,

was among the first imprinted genes identified (9).

thereby promoting prometastatic functions of trans-

Recent studies suggest that H19 interacts with

forming growth factor– b (37). Finally, lncRNAs are

SENCR = smooth muscle and

associated long noncoding RNA

shRNA = small hairpin RNA siRNA = small interfering RNA SNP = single-nucleotide polymorphism

1216

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

F I G U R E 1 Noncoding RNAs of the Human Transcriptome

Human genome

mRNAs

Translation

Other noncoding RNAs*

Noncoding transcriptome

Long noncoding RNAs

miRNAs piRNAs snoRNAs snRNAs tRNAs H1 RNA RMRP 7SL RNA TERC scaRNAs rRNAs Y RNAs

Long noncoding RNAs (lncRNAs) Long intervening noncoding RNAs (lincRNAs) Intronic long noncoding RNAs (lncRNAs) Natural antisense transcripts (NATs) Enhancer RNAs (eRNAs) Circular RNAs (circRNAs)

Post-transcriptional gene regulation Germ line transposon silencing RNA modification pre-mRNA splicing Translation pre-tRNA processing mtRNA processing, rRNA processing Direction of protein traffic Maintenance of telomere ends RNA modification Translation DNA replication, small RNA maturation

The human genome is nearly pervasively transcribed, resulting in an almost noncoding transcriptome with various RNA species. *Depicted RNA classification does not rely on a strict 200 nt cutoff. H1 RNA ¼ RNA component of ribonuclease P; miRNAs ¼ microRNAs; mRNA ¼ messenger RNA; mtRNA ¼ mitochondrial RNA; piRNAs ¼ piwi-interacting RNAs; RMRP ¼ RNA component of RNase MRP; rRNA ¼ ribosomal RNA; scaRNAs ¼ small Cajal body-specific RNAs; 7SL RNA ¼ signal recognition particle RNA; snoRNAs ¼ small nucleolar RNAs; snRNAs ¼ small nuclear RNAs; TERC ¼ telomerase RNA component; tRNAs ¼ transfer RNAs; Y RNAs ¼ part of the RoRNP.

involved in several steps of post-transcriptional

IDENTIFICATION OF NONCODING RNAs AS HOT SPOTS

gene regulation, including mRNA processing and

FOR SINGLE-NUCLEOTIDE POLYMORPHISMS: LESSONS

mRNA stability, as well as translation. In this context,

TO

mRNA processing can be altered through 2 different

STUDIES. Genome-wide association studies (GWAS)

mechanisms. First, natural antisense transcripts can

have

form RNA-RNA duplexes with their corresponding

morphisms (SNPs) in noncoding regions may have

sense transcripts, thereby altering splice-site recog-

functional consequences. SNPs in 3 0 untranslated

nition and spliceosome recruitment, a mechanism

regions of mRNAs may interfere with miRNA binding

observed, for example, during processing of the first

or other regulatory processes that control mRNA

intron of neuroblastoma MYC mRNA (38). Second,

stability or translation (44). For example, it has

lncRNAs can indirectly regulate splicing, as seen for

been

metastasis-associated lung

tran-

disease–associated variation in the 3 0 untranslated

script 1 (MALAT1), which regulates splicing factor

region of the basic-helix-loop-helix transcription

phosphorylation and can interact with U1 small nu-

factor TCF21, which critically controls coronary artery

clear RNA, a component of the spliceosome (39,40).

smooth muscle cell and cardiac fibroblast maturation,

adenocarcinoma

LEARN

FROM

suggested

recently

GENOME-WIDE

that

reported

ASSOCIATION

single-nucleotide

that

a

coronary

poly-

heart

With respect to mRNA stability, lncRNAs are impli-

disrupts a miR-224 binding site (45). Variations in

cated in processes leading to stabilization, as well as

miRNAs can themselves alter mRNA targeting and

destabilization, of a given transcript. An interesting

may thereby dysregulate the gene expression pat-

example is the lncRNA BACE1AS, which forms an

terns controlled by the miRNA. SNPs may also affect

RNA-RNA duplex with BACE1 mRNA, thus preventing

the expression of lncRNAs. In addition, variations

miR-485-5p-mediated

(41).

may affect the structure of lncRNAs by interfering

Reciprocally, the long intervening noncoding RNA H19

with RNA folding and the formation of RNA structures

was shown to coordinate the interaction of protein K

or by modulating protein-RNA interactions. Although

homology-type splicing regulatory proteins and labile

little is known regarding SNPs in noncoding regions,

repression

of

BACE1

transcripts in the cytoplasm, which favors protein K

the first hints that SNPs in noncoding regions may

homology-type splicing regulatory protein–mediated

indeed be relevant were provided in 2007 (46–48).

destabilization (42). Finally, lncRNAs can also exert

Among the 58 genomic loci (p < 107) of athero-

their effects at the level of protein synthesis,

sclerosis susceptibility identified by GWAS (49),

as seen for the mouse lncRNA Uchl1AS, which is

the most significant genetic locus was found at

required for the association of Uchl mRNA with active

chromosome 9p21. SNPs in this region have been

polysomes (43).

associated with different endpoints of atherosclerotic

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

CENTRAL I LLU ST RAT ION

Long Noncoding RNAs in Cardiovascular Disease

The RNA World and Cardiovascular Disease: Long Noncoding RNAs

Boon, R.A. et al. J Am Coll Cardiol. 2016; 67(10):1214–26.

Depending on the subcellular localization, long noncoding RNAs (lncRNAs) can act via various mechanisms. By mimicking transcription factor binding sites, doublestranded regions of lncRNAs can bind to transcription factors, thereby functioning as a decoy. Via binding to chromatin modifiers, lncRNAs contribute to epigenetic silencing or activation of gene expression. LncRNAs that bind to exon/intron junctions of pre–messenger RNA (mRNA) can influence (alternative) splicing. Several lncRNAs act as a scaffold in ribonucleoprotein (RNP) complexes. In the cytoplasm, numerous lncRNAs have been shown to bind microRNAs (miRNAs), thereby preventing the miRNAs from binding to and regulating their mRNA targets. By masking binding sites for proteins or miRNAs, cytoplasmic lncRNAs can alter the stability of mRNAs.

cardiovascular disease (46–48,50). Chromosome 9p21

exon of ANRIL, and mechanistic studies have shown

spans a region of approximately 50 kb but does not

that the 9p21 risk allele disrupts an inhibitory STAT1

comprise protein-coding genes (51) and has therefore

binding site, leading to up-regulation of ANRIL

been considered a “gene desert.” The closest protein-

expression (52,55). ANRIL is expressed in cells and

coding genes are cyclin-dependent kinase inhibitor

tissues that are relevant in atherogenesis, and mul-

2A CDKN2A and cyclin-dependent kinase inhibitor 2B

tiple isoforms have been identified that can be clas-

(CDKN2B), which are located >100 kb from associated

sified according to their exon structures in 4 major

SNPs. Chromosome 9p21 encodes different transcripts

linear and circular spliced forms (56,57). The molec-

of the lncRNA ANRIL (also known as CDKN2B anti-

ular function of linear ANRIL has been investigated

sense RNA 1). A potential function of this lncRNA in

using RNA interference and overexpression studies

vascular disease is supported by several studies

(56,58). Targeting exon 1 (all linear ANRIL isoforms)

showing its expression to be associated with risk for

or exon 19 (long ANRIL isoforms only) by small

coronary atherosclerosis (52), carotid arteriosclerosis

interfering RNAs (siRNAs) in smooth muscle cells

(53), peripheral artery disease (54), and other vascular

reduced cell viability (58). Results from this study are

disease (see Holdt and Teupser [51] for a review). The

in line with overexpression studies in which ANRIL

top associated SNPs are found adjacent to the last

increased cellular proliferation (56). In addition,

1217

1218

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

ANRIL increased adhesion and decreased apoptosis,

through multiple let-7-binding sites (68,69), thereby

which are proatherosclerotic cellular mechanisms

preventing

(56). At the molecular level, ANRIL binds to epige-

target genes associated with various cardiovascular

netic effector proteins of PRC1 and PRC2 (33,56,59)

diseases (70).

let-7

from

inhibiting

expression

of

and recruits them to promoter regions, thereby

In summary, there is increasing evidence that

epigenetically regulating gene expression patterns

lncRNAs may control atherosclerosis risk at non-

(56,60,61). The recruitment is mediated though re-

coding regions identified by GWAS of cardiovascular

petitive Alu elements in ANRIL, which are mirrored in

disease. Apart from ANRIL at 9p21, no other lncRNAs

the promoters of its target genes (56).

at GWAS loci identified in European populations

Two other studies have investigated lncRNAs at

have thus far been functionally characterized. Results

genomic loci associated with myocardial infarction

from the first functional studies of lncRNAs from

and coronary artery disease (CAD) in Japanese and

Asian association studies hold great promise for

Chinese cohorts (62,63). Because these loci were not

novel mechanistic insights in the pathology of

replicated in GWAS performed in populations of Eu-

atherosclerosis.

ropean descent, it remains to be determined whether these loci and lncRNAs also contribute to athero-

REGULATION OF lncRNAs IN

sclerosis risk in non-Asian populations. The lncRNA

CARDIOVASCULAR DISEASE

myocardial infarction-associated transcript (MIAT) is transcribed from chromosome 22q12.1 and was iden-

Because lncRNAs can be linked to cardiovascular

tified as a risk allele for myocardial infarction in a

disease, it is highly likely that lncRNAs are also

large-scale case-control association study in Japanese

regulated in certain cardiovascular disease states.

subjects (62). The risk allele of a SNP in exon 5 was

The first evidence that lncRNAs are dynamically

shown to increase MIAT expression, likely through

regulated in the heart came from studies of cardiac

binding

development during embryogenesis in mice (71,72).

of

a

hitherto

uncharacterized

protein.

Although the function of MIAT in the heart is still

In recent years, several lncRNAs have been

unknown, MIAT expression was reduced in periph-

described

eral blood mononuclear cells isolated from patients

required for cardiac development. The first of these

as

highly

induced

during

and

even

with ST-segment elevation myocardial infarction.

lncRNAs was termed Braveheart (Bvht) (71). Bvht was

Recent studies suggest that MIAT controls endothe-

identified by screening gene expression databases

lial cell functions and also plays a role in diabetic

for lncRNAs highly expressed in the heart. Bvht does

retinopathy (discussed in later text).

not code for any protein, but Bvht depletion dimin-

In a Chinese population, variations of lncRNA H19

ished the capacity of embryonic stem cells to

at chromosome 11p15.5 have been associated with the

differentiate into cardiomyocytes. Furthermore, Bvht

risk for CAD (63). Here it was shown that subjects

was found to induce a transcriptional program

carrying 3 or 4 risk alleles at the H19 locus had a

downstream of the transcription factor MESP1, a key

significantly increased risk for CAD and higher Gen-

cardiac differentiation transcription factor. At the

sini scores than those with 0 to 2 risk alleles (63). The

same time that Bvht was described, another lncRNA,

rs2067051 polymorphism site is located in the 5 0 flank

Fendrr, was found to be required for heart develop-

of H19 and is near the imprinting control regions,

ment in mice (72). The investigators found that ge-

which are critical for regulation of H19 expression.

netic deletion of Fendrr caused ventricular defects

The second SNP in the H19 locus is located in the last

during development. Like Bvht, Fendrr regulates a

exon of H19, but how this contributes to H19 expres-

transcriptional network required for cardiac devel-

sion levels or function is unclear at the moment. H19

opment via interaction with histone modifying

is an imprinted lncRNA, which is induced during

complexes (PRC2 and TrxG/MLL). In contrast to

embryogenesis

birth,

Bvht, which does not seem to have a human homo-

except in adult skeletal muscle and heart. H19

log, Fendrr is also present in the human genome, but

expression is increased in various tumors (64).

its function during human development or cardio-

Moreover, H19 is highly expressed in the neointima of

vascular disease is not known.

and

down-regulated

after

rats 7 to 14 days after vascular injury (65) and is

Recently, 3 novel lncRNAs were identified as

reexpressed in human atherosclerotic plaques (66).

induced during differentiation of human embryonic

Genetic studies showed that polymorphisms in H19

stem cells to endothelial cells (73). These lncRNAs,

are correlated with high blood pressure (67), a known

termed TERMINATOR, ALIEN, and PUNISHER, not

risk factor for CAD. Mechanistic studies have shown

only increase in expression during differentiation to

that H19 can bind and sequester the miRNA let-7

endothelial cells, they also contribute to endothelial

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

commitment and identity. Knockdown of any of the 3

valve interstitial cells. Furthermore, HOTAIR was also

lncRNAs results in cardiovascular defects in devel-

reduced in bicuspid aortic valves in comparison with

oping zebrafish embryos.

tricuspid aortic valves, suggesting a role in valve

Studies that assessed enhancer regions that func-

calcification.

tion as transcriptional activators in the human heart

lncRNAs AS BIOMARKERS. A potential application

found that many of these regions localize to non-

for lncRNAs lies in the use of circulating lncRNA

coding parts of the genome (74,75), raising the sug-

levels as biomarkers for cardiovascular disease.

gestion that enhancer RNAs could arise from these

Several pioneering studies have shown this to be

loci that regulate cardiac gene expression. Several

feasible. In 2014, the group of Thum (84) described

other lncRNAs are also regulated during development

the mitochondrial lncRNA uc022bqs.1, which the in-

or in various cardiovascular disease models (76,77).

vestigators termed LIPCAR, as increased in the

Finally, recent studies report that circRNAs are

plasma of patients with cardiac remodeling after

expressed in endothelial cells and are regulated by

acute myocardial infarction and in patients with

hypoxia in vitro (78). Among the many circRNAs

chronic heart failure. Furthermore, plasma LIPCAR

identified, the circular form of ZNF292 (cZNF292)

levels can be used as a prognostic indicator for car-

was shown to regulate angiogenesis, and its inhibi-

diovascular mortality, although it is not clear whether

tion reduces sprouting angiogenesis and endothelial

circulating LIPCAR arises from the mitochondria or

cell proliferation (78). Although the mechanism un-

nuclei of cardiac myocytes or even from other cell

derlying the biological effects of cZNF292 are not

types (85). Another study from 2014 (86) described

fully clear, a function as a miR sponge has been

the association of whole blood cell expression

excluded (78).

levels of the lncRNAs aHIF, ANRIL, KCNQ1OT1, MIAT,

REGULATION OF lncRNAs DURING PATHOLOGY. The

and MALAT1 with acute myocardial infarction. The

first evidence that lncRNAs are differentially regu-

investigators

lated in human cardiac pathologies came from RNA

KCNQ1OT1 could be used as prognostic markers for

deep sequencing experiments by Yang et al. (79). The

left ventricular function. The studies mentioned thus

investigators were able to detect 18,480 lncRNAs in

far focused on markers for cardiac dysfunction, but

human left ventricles. This study showed that the

because (coronary) atherosclerosis is the underlying

lncRNA expression pattern distinguished between

cause of an acute ischemic insult, biomarkers that

ischemic and nonischemic failing hearts. The func-

reflect coronary atherosclerosis are also highly clini-

tion of the specifically regulated lncRNAs was not

cally relevant. Yang et al. (87) found that the lncRNA

assessed. By studying expression of lncRNAs that are

they termed Coromarker is increased in plasma from 2

regulated after myocardial infarction in mice, the

cohorts of patients with CAD compared with healthy

group of Pedrazzini (80) identified several human

control subjects. Whether Coromarker is released

homologs that are differentially expressed in heart

directly by cells in atherosclerotic plaques or rather

samples of patients with dilated cardiomyopathy

reflects the proinflammatory state of circulating cells

and aortic stenosis. The investigators hypothesized

remains to be seen. Finally, circulating levels of the

that

lncRNA Tapsaki predicted mortality of patients with

1

particular

lncRNA

(Novlnc6),

which

is

down-regulated in patients with dilated cardiomy-

further

showed

that

ANRIL

and

acute kidney injury (88).

opathy, controls cardiomyocyte identity. A descrip-

Very recent studies additionally suggest that

tive study of lncRNA expression in right heart failure

circRNAs, which are regulated in endothelial cells

identified several lncRNAs that are differentially

in vitro (78), might be potential biomarkers. Memczak

expressed in patient heart samples compared with

et al. (89) reported that circRNAs can be detected at

unused donor hearts (81). LncRNAs are likely also

higher levels in peripheral blood compared with the

regulated during progression of cardiomyopathy, and

linear host genes and might be suitable as biomarkers

expression differences among pathologies, such as

for disease. However, their regulation in cardiovas-

hypertrophic and dilated cardiomyopathy, undoubt-

cular pathologies is unclear.

edly also exist. However, to date, no studies

As with any circulating biomarker for cardiovas-

describing lncRNA expression in these human dis-

cular disease, the cellular origin of the circulating

eases have been published. An interesting study by

lncRNAs is often unclear, and it is not known

Carrion et al. (82) described the repression of a

whether these lncRNAs are causally involved in the

lncRNA, HOTAIR, previously described as controlling

pathophysiology of the underlying disease. Further-

expression of the HOXD locus (34) and enhancing

more, quantification of RNA relies, at least with

metastasis (83), by cyclic stretch in human aortic

the majority of techniques, on polymerase chain

1219

1220

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

se-

occurring during diabetes, are regulated by lncRNAs

quences, which is inherently troublesome with blood

reaction–mediated

amplification

of

specific

(Figure 2A). Particularly diabetic retinopathy, which

samples (90). Detection of linear lncRNAs, which

is characterized by pathological proliferation of

might be even more susceptible to degradation (if

retinal vessels and increased vasopermeability, can

not adequately protected) and are often expressed at

be ameliorated by interfering with lncRNA ex-

very low levels, may be more challenging compared

pression in vivo. Inhibition of the hypoxia- and

with miRNAs. Moreover, to perform an unbiased

diabetes-induced lncRNA MALAT1 (93–95) prevented

lncRNA transcriptome analysis with small volumes

diabetes-induced vascular complications (95). Specif-

of plasma or whole blood from large patient cohorts

ically,

by RNA deep sequencing could prove technically

improved retinal function and reduced retinal cell

challenging and is likely very costly. However,

death (95). MALAT1 inhibition reduced endothelial cell

because there are potentially more noncoding genes

hyperproliferation and vascular inflammation without

intraocular

injection

of

MALAT1

shRNA

than coding genes, analyzing signatures or sets

directly affecting endothelial barrier function (95,96).

of circulating lncRNAs could prove very powerful

These findings are consistent with the previously

as biomarkers for specific (cardiovascular) disease

described inhibition of endothelial cell proliferation by

states.

pharmacological gapmer-mediated MALAT1 silencing or genetic deletion of MALAT1 (93). Gapmers are

FIRST INSIGHTS INTO THE PATHOPHYSIOLOGICAL ROLE OF NONCODING RNAs IN

single-stranded oligonucleotides that consist of a DNA stretch flanked by locked nucleic acid nucleotides, inducing ribonuclease H-dependent cleavage of the

CARDIOVASCULAR DISEASE

target RNA, and showed an efficient inhibition of nuclear localized MALAT1 in vitro (93). Intraperitoneal

Whereas transcriptomic analysis showed that many

injection of gapmers furthermore efficiently sup-

lncRNAs are regulated in cardiovascular disease,

pressed MALAT1 in various organs, particularly in the

relatively little is known regarding the causal func-

intima in vivo (93). Whereas inhibition of pathological

tional contribution of lncRNAs in disease models

angiogenesis by MALAT1 silencing using shRNAs or

in vivo. To understand their biological functions,

gapmers may be useful under conditions of diabetic

lncRNAs can be genetically deleted by conventional

retinopathy, endothelial proliferation is required for

gene-targeting strategies, for example, by intro-

ischemia-induced neovascularization and inhibition

ducing a transcriptional stop signal to block RNA

of MALAT1 in this context may be detrimental (93).

polymerase (reviewed by Bassett et al. [91]). Short

Diabetic retinopathy is also controlled by the

RNAs, such as siRNAs, modified antisense oligonu-

lncRNA MIAT (97,98). MIAT knockdown by shRNA

cleotides, or gapmers can also achieve inhibition.

in rats reduced endothelial cell apoptosis and

Whereas

oligonucleotides

vascular leakage and counteracted diabetes-induced

preferentially target cytoplasmic RNAs, gapmers

up-regulation of proinflammatory proteins, thereby

introduce ribonuclease H-dependent cleavage of nu-

alleviating retinal vessel impairment (98). The study

clear RNAs. Alternatively, viral vectors can deliver

additionally suggests that MIAT is acting as a miR-150

small hairpin RNAs (shRNAs) targeting the respective

sponge to control endothelial cell functions and dia-

lncRNA. Overexpression of lncRNAs may be more

betic retinopathy.

siRNAs

and

antisense

challenging because some lncRNAs affect gene

A very recent study described the identification of

expression in their local genomic environments; as

2 lncRNAs that are induced by hypoxia in endothelial

such, conventional overexpression with plasmids or

cells and contribute to angiogenesis (99). In vitro

viral vectors may not result in appropriate localiza-

silencing of either of these 2 lncRNAs, called

tion of these lncRNAs. Alternatively, a recently

linc00323-003 and MIR503HG, results in reduction of

described strategy using clustered regularly inter-

proliferation of endothelial cells and subsequent in-

spaced short palindromic repeats (CRISPR)/CRISPR-

hibition of angiogenic sprouting. Moreover, embed-

associated protein 9 (Cas9) technology, which allows

ding endothelial cells in which linc00323-003 and

the activation of a specific promoter by using a fusion

MIR503HG are silenced in human induced pluripo-

protein of catalytically inactive Cas9 (dCas9) with the

tent

transcriptional activator domain VP64, may be used

ex vivo results in a diminished contribution of endo-

(92) to activate expression of the endogenous

thelial cells to these engineered heart tissues. Inter-

lncRNA.

estingly, both lncRNAs inhibit the expression of the

VASCULAR DISEASE. D i a b e t i c

stem

cell–derived

engineered

heart

tissue

retinopathy and

transcription factor GATA2, thereby potentially con-

a n g i o g e n e s i s . Vascular complications, such as those

trolling endothelial cell proliferation (99). Because

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

F I G U R E 2 Long Noncoding RNAs Regulate Vascular Function

Normal

Diabetic retinopathy

MIAT

VEGF

° Vascular leakage ° Cell death (in vivo)

Other mechanisms

° Pro-angiogenic signaling °

miR-150 Cell cycle genes (e.g. cyclins) MALAT 1 Normal

° Proliferation ° Cell death

p38-P

Hypoxia

° Endothelial cell migration tube formation

Other mechanisms

linc00323-003

° Pericyte loss ° Proliferation ° Migration

GATA2

° Endothelial tube formation

MIR503HG

Lipid homeostasis HDL

LDL

NFIA miR-382-5p RP5-833A20.1

ANRIL

SENCR

linc-p21 MDM2

PRC1/2 Cell viability Proliferation

MDK MYOCD PTN ACTA2 Contractile phenotype

P53 Proliferation Apoptosis

(A) The long noncoding RNAs (lncRNAs) myocardial infarction-associated transcript (MIAT) and metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) have been implicated in angiogenesis and inflammation in diabetic retinopathy. Both lncRNAs are up-regulated in diabetic retinopathy and affect endothelial cell function. MALAT1 is also induced by hypoxia. Linc00323-003 and MIR503HG are both induced by hypoxia and regulate endothelial proliferation and migration via GATA2. (B) The lncRNA RP5-833A20.1 was reported to control cholesterol homeostasis, with potential implications for atherosclerosis. Antisense noncoding RNA in the INK4 locus (ANRIL), smooth muscle and endothelial cell-enriched migration/differentiation-associated long noncoding RNA (SENCR), and linc-p21 were shown to be involved in phenotypic control of cells in the vascular wall. HDL ¼ high-density lipoprotein; LDL ¼ low-density lipoprotein; NFIA ¼ nuclear factor IA; PRC1/2 ¼ PRC1, polycomb repressive complex 1/2.

1221

1222

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

of the lack of evolutionary conservation of these

cells and endothelial cells. When the investigators

lncRNAs, their contribution to angiogenesis in vivo in

inhibited SENCR in smooth muscle cells, the cells

small animal models could not be assessed; hence, the

exhibited a more promigratory and less contractile

in vivo function of these lncRNAs is unclear.

phenotype, suggesting a potential role in vascular

Neointima formation and atherosclerosis. Neointima

pathologies characterized by smooth muscle dedif-

formation is influenced by linc-p21, which regulates

ferentiation and migration, such as restenosis and

p53-dependent gene expression and thereby re-

atherosclerosis.

presses smooth muscle cell proliferation and stimu-

LncRNAs may also regulate risk factors for CAD,

lates apoptosis of vascular smooth muscle cells (100)

such as metabolism or hypertension. For example,

(Figure 2B). In vivo silencing of linc-p21 by lentiviral

liver-enriched lncLSTR was shown to regulate sys-

vector-delivered siRNAs reduced neointima forma-

temic lipid metabolism in mice (103), which may

tion after endothelial injury (100).

secondarily affect atherosclerosis.

In vitro studies suggested that the lncRNA RP5-

CARDIAC DISEASE. Several studies addressed the

833A20.1 influences cholesterol metabolism and

role of lncRNAs in cardiac hypertrophy and remodel-

atherosclerosis. RP5-833A20.1 was shown to decrease

ing (Figure 3). Wang et al. (104) identified AK017121,

nuclear factor IA (NFIA) by inducing hsa-miR-382-5p

which they named cardiac-apoptosis related lncRNA

expression, thereby reducing high-density lipopro-

(CARL), as anoxia-regulated lncRNA, which sup-

tein cholesterol but increasing low-density lipopro-

pressed apoptosis and mitochondrial fission in vitro

tein and proinflammatory cytokines. Overexpression

and reduced ischemia-reperfusion injury in vivo.

of the putative lncRNA target NFIA induced an athe-

Mechanistically, the investigators showed that CARL

roprotective lipid status and reduced atherosclerotic

binds to miR-539 and acts as a sponge to block miR-539

lesion formation in ApoE/ mice, however, the

functions in mitochondrial fission. CARL thereby

causal involvement of the lncRNA itself has not been

derepressed the miR-539 target PHB2 and prevented

demonstrated (101).

anoxia-induced mitochondrial fission. Mitochondrial

Using RNA sequencing of human coronary artery

fission was also regulated by the lncRNA mitochon-

smooth muscle cells, Bell et al. (102) identified a

drial dynamic related lncRNA (MDRL). MDRL inhibits

lncRNA, smooth muscle and endothelial cell-enriched

mitochondrial fission and apoptosis in vitro by

migration/differentiation-associated long noncoding

down-regulating miR-361, which controls processing

RNA (SENCR), enriched in vascular smooth muscle

of another miRNA, miR-484. In vivo intracoronary

F I G U R E 3 Long Noncoding RNAs Regulate Cardiac Function

Hypertrophy

Cardiomyocyte cell death miR-539 Mitochondrial fission

miR484

miR361

CARL

Mhrt

MDRL

CHRF

Brg1 Myd88

miR-489 Autophagy

ATG7

miR188-3p

APF

Novlnc6 Nkx2.5 Bmp10

Several long noncoding RNAs (lncRNAs) were shown to regulate cardiac physiology. Cardiac-apoptosis related long noncoding ribonucleic acid (CARL), mitochondrial dynamic related long noncoding ribonucleic acid (MDRL), and autophagy-promoting factor (APF) regulate cardiomyocyte cell death by inhibiting microRNAs. Cardiac hypertrophy related factor (CHRF) also functions as a microRNA sponge in the context of cardiac hypertrophy and inhibits miR-489. The lncRNA myosin heavy chain-associated ribonucleic acid transcript (Mhrt) regulates chromatin remodeling and cardiac hypertrophy. Novlnc6 is involved in cardiac development.

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

delivery of adenoviral-expressed MDRL reduced

viral gene delivery of siRNAs or overexpression of

myocardial

ischemia-reperfusion

transcripts, antisense oligonucleotides might be

injury by inhibition of mitochondrial fission and

used to block lncRNA functions. Although the ef-

apoptosis (104).

ficiency of antisense oligonucleotide delivery is

infarction

after

AK048451, which was subsequently named cardiac hypertrophy

related

factor

(CHRF),

was

lower in the heart compared with the liver, recent

highly

studies suggest that antisense miRNA inhibitors,

induced by angiotensin II in vitro and in murine

so-called antimiRs or antagomirs, can be used to

transverse aortic constriction models, as well as in

improve cardiovascular diseases in small animals

human heart failure in vivo. CHRF is widely expressed

(63,64). Some of these findings have also been

in cardiovascular cells and other tissues but has a

confirmed in more clinically relevant large animal

specific function in cardiomyocytes (105). CHRF

models (108). Anti-miRNAs have been successfully

induced cardiomyocyte hypertrophy in vitro and

and safely used in clinical phase 2 studies,

apoptosis in vivo. The mechanism of action was

although for the treatment of liver disease (109),

attributed to binding of miR-489, which derepresses

which is easier to target by therapeutic RNAs

the miR-489 target Myd88 to regulate cardiomyocyte

compared with the heart. Whether lncRNAs can be

hypertrophy

lncRNA

targeted by a similar strategy remains to be deter-

autophagy-promoting factor (APF) was reported to

mined. Gapmers might be useful, particularly for

regulate autophagic cell death in vitro, and inhibition

nuclear-expressed lncRNAs. Locked nucleic acid

of APF by siRNA reduced ischemia-reperfusion injury

gapmers have, for example, been used for targeting

in vivo in mice. APF thereby targeted miR-188-3p,

liver PCSK9 in nonhuman primates, but a phase 1

which represses the autophagy-related protein 7 (106).

clinical study was halted (110). Gapmers against

A cluster of antisense transcripts from the myosin

survivin and hypoxia-inducing factor 1a have been

heavy chain 7 locus (named myosin heavy chain-

applied weekly over as long as 1 year without

associated RNA transcripts [Mhrt]) were shown to be

safety concerns (111,112). Future approaches to

highly expressed in the murine heart (107). Mhrt is

therapeutically interfere with lncRNA functions

cardiac specific and prevents stress-induced cardiac

may also include small molecules specifically

(105).

More

recently,

the

remodeling by directly interfering with the Brg1

designed to interfere with RNA-protein complexes.

helicase, thereby inhibiting a stress-induced gene

 LncRNAs exhibit diverse functions, and under-

expression program. The investigators also identified

standing the molecular mechanism mediated by a

a human ortholog of Mhrt, which is also induced and

given lncRNA may be complicated. Particularly if

could perhaps be used therapeutically to prevent

the lncRNA interferes with chromatin structure or

pathological cardiac remodeling in humans (107).

epigenetic control mechanisms, deciphering the

PROMISE AND CHALLENGES IN USING lncRNAs AS THERAPEUTIC TARGETS OR BIOMARKERS Increasing evidence suggests that lncRNAs may play crucial roles in the regulation of pathophysiological processes in the cardiovascular system and might be therapeutically be targeted. However, the field of lncRNAs is still in its infancy and faces many challenges:

exact mechanism by which the lncRNA functions is not trivial. Furthermore, some lncRNAs may act through

more

than

1

mechanism

of

action.

Although several studies carefully document that lncRNAs act as sponges for miRNAs to control cardiac functions, it is unclear whether such a mechanism can be generally considered. Because most lncRNAs are expressed at low levels, it is hard to reconcile a sponge function for highly expressed miRNAs with only 1 miRNA binding site per lncRNA. The Stoffel group recently reported that

 Many lncRNAs are not conserved at the sequence

changes in the abundance of miRNA target sites are

level and are expressed only in primates. Unless

unlikely to cause significant effects (113). Although

structural conservation is established in the future,

this study only used the 3 0 untranslated region of

it will be challenging to understand the biological

mRNA, the stoichiometry of target RNA and miRNA

properties of these regulatory RNAs. Humanized

has to be carefully considered. In addition, it is

models or organoid cultures will be necessary to

thus far unclear whether bound miRNA may

screen for biological properties of primate-specific lncRNAs before embarking on costly and ethically disputed nonhuman primate studies.  Therapeutically targeting lncRNAs may potentially be used to combat cardiovascular disease. Besides

instead affect lncRNA expression or functions.  Mechanistic studies are complicated by the fact that lncRNAs are often expressed in many transcript variants, which may or may not have common

functions.

Like

proteins, RNAs

can

be

1223

1224

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

modified and such modifications (e.g., methyl-

Forschungsgemeinschaft (SFB834 and SFB902), the

ation, editing) may affect lncRNA functions.

European Research Council (Drs. Boon and Dimmeler),

Finally, some of the transcripts annotated as non-

and the Leducq Network “Mirvad” for support.

coding RNAs may code for proteins. For example, lncRNAs may encode short peptides that have been overlooked in the past, as recently reported for a cardiac lncRNA (114).

REPRINT REQUESTS AND CORRESPONDENCE: Prof.

Stefanie

Dimmeler,

Institute

of

Cardiovascular

Regeneration, Centre for Molecular Medicine, J. W.

ACKNOWLEDGMENTS The authors thank Rie Manavski

Goethe University Frankfurt am Main, Theodor-

for help with the graphic design and the German

Stern-Kai 7, D-60590 Frankfurt, Germany. E-mail:

Center for Cardiovascular Research, the Deutsche

[email protected].

REFERENCES 1. Thomas CA Jr. The genetic organization of chromosomes. Annu Rev Genet 1971;5:237–56.

Consortium. Initial sequencing and analysis of the human genome. Nature 2001;409:860–921.

2. Ohno S. So much “junk” DNA in our genome. Brookhaven Symp Biol 1972;23:366–70.

16. Venter JC, Adams MD, Myers EW, et al. The sequence of the human genome. Science 2001;

3. Hough BR, Smith MJ, Britten RJ, et al. Sequence complexity of heterogeneous nuclear RNA in sea urchin embryos. Cell 1975;5:291–9. 4. Chow LT, Gelinas RE, Broker TR, et al. An amazing sequence arrangement at the 50 ends of adenovirus 2 messenger RNA. Cell 1977;12:1–8. 5. Berget SM, Moore C, Sharp PA. Spliced segments at the 50 terminus of adenovirus 2 late mRNA. Proc Natl Acad Sci U S A 1977;74:3171–5. 6. Stark BC, Kole R, Bowman EJ, et al. Ribonuclease P: an enzyme with an essential RNA component. Proc Natl Acad Sci U S A 1978;75: 3717–21. 7. Lerner MR, Steitz JA. Antibodies to small nuclear RNAs complexed with proteins are produced by patients with systemic lupus erythematosus. Proc Natl Acad Sci U S A 1979;76:5495–9. 8. Brown CJ, Ballabio A, Rupert JL, et al. A gene from the region of the human X inactivation centre is expressed exclusively from the inactive X chromosome. Nature 1991;349:38–44. 9. Bartolomei MS, Zemel S, Tilghman SM. Parental imprinting of the mouse H19 gene. Nature 1991; 351:153–5. 10. Lee RC, Feinbaum RL, Ambros V. The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell 1993;75:843–54. 11. He L, Hannon GJ. MicroRNAs: small RNAs with a big role in gene regulation. Nat Rev Genet 2004; 5:522–31. 12. Patel SB, Bellini M. The assembly of a spliceosomal small nuclear ribonucleoprotein particle. Nucleic Acids Res 2008;36:6482–93. 13. ENCODE Project Consortium. Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature 2007;447:799–816. 14. Djebali S, Davis CA, Merkel A, et al. Landscape of transcription in human cells. Nature 2012;489: 101–8. 15. Lander ES, Linton LM, Birren B, et al., for the International Human Genome Sequencing

29. Monnier P, Martinet C, Pontis J, et al. H19 lncRNA controls gene expression of the imprinted gene network by recruiting MBD1. Proc Natl Acad Sci U S A 2013;110:20693–8.

291:1304–51.

30. Chen LL, Carmichael GG. Altered nuclear

17. Mercer TR, Dinger ME, Mattick JS. Long noncoding RNAs: insights into functions. Nat Rev Genet 2009;10:155–9.

retention of mRNAs containing inverted repeats in human embryonic stem cells: functional role of a nuclear noncoding RNA. Mol Cell 2009;35: 467–78.

18. Ma L, Bajic VB, Zhang Z. On the classification of long non-coding RNAs. RNA Biol 2013;10: 925–33. 19. Moran VA, Perera RJ, Khalil AM. Emerging functional and mechanistic paradigms of mammalian long non-coding RNAs. Nucleic Acids Res 2012;40:6391–400. 20. Johnsson P, Lipovich L, Grandér D, et al. Evolutionary conservation of long non-coding RNAs; sequence, structure, function. Biochim Biophys Acta 2014;1840:1063–71. 21. Pang KC, Frith MC, Mattick JS. Rapid evolution of noncoding RNAs: lack of conservation does not mean lack of function. Trends Genet 2006;22:1–5. 22. Ulitsky I, Shkumatava A, Jan CH, et al. Conserved function of lincRNAs in vertebrate embryonic development despite rapid sequence evolution. Cell 2011;147:1537–50. 23. Jeck WR, Sorrentino JA, Wang K, et al. Circular RNAs are abundant, conserved, and associated with ALU repeats. RNA 2012;19:141–57. 24. Memczak S, Jens M, Elefsinioti A, et al. Circular RNAs are a large class of animal RNAs with regulatory potency. Nature 2013;495:333–8.

31. Sunwoo H, Dinger ME, Wilusz JE, et ε/b nuclear-retained non-coding RNAs regulated upon muscle differentiation essential components of paraspeckles.

al. MEN are upand are Genome

Res 2009;19:347–59. 32. Mao YS, Sunwoo H, Zhang B, et al. Direct visualization of the co-transcriptional assembly of a nuclear body by noncoding RNAs. Nat Cell Biol 2011;13:95–101. 33. Yap KL, Li S, Muñoz-Cabello AM, et al. Molecular interplay of the noncoding RNA ANRIL and methylated histone H3 lysine 27 by polycomb CBX7 in transcriptional silencing of INK4a. Mol Cell 2010;38:662–74. 34. Rinn JL, Kertesz M, Wang JK, et al. Functional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell 2007;129:1311–23. 35. Yoon JH, Abdelmohsen K, Kim J, et al. Scaffold function of long non-coding RNA HOTAIR in protein ubiquitination. Nat Commun 2013;4:2939. 36. Hung T, Wang Y, Lin MF, et al. Extensive and coordinated transcription of noncoding RNAs within cell-cycle promoters. Nat Genet 2011;43: 621–9.

25. Cocquerelle C, Mascrez B, Hétuin D, et al. Missplicing yields circular RNA molecules. FASEB J 1993;7:155–60.

37. Li W, Kang Y. A new Lnc in metastasis: long

26. Brown CJ, Hendrich BD, Rupert JL, et al. The human XIST gene: analysis of a 17 kb inactive X-

38. Krystal GW, Armstrong BC, Battey JF. N-myc mRNA forms an RNA-RNA duplex with endogenous antisense transcripts. Mol Cell Biol 1990;10: 4180–91.

specific RNA that contains conserved repeats and is highly localized within the nucleus. Cell 1992;71: 527–42.

noncoding RNA mediates the prometastatic functions of TGF-b. Cancer Cell 2014;25:557–9.

27. Zhao J, Sun BK, Erwin JA, et al. Polycomb proteins targeted by a short repeat RNA to the mouse X chromosome. Science 2008;322:750–6.

39. Tripathi V, Ellis JD, Shen Z, et al. The nuclearretained noncoding RNA MALAT1 regulates alternative splicing by modulating SR splicing factor phosphorylation. Mol Cell 2010;39:925–38.

28. Surani MA, Barton SC, Norris ML. Development of reconstituted mouse eggs suggests imprinting

40. Engreitz JM, Sirokman K, McDonel P, et al. RNA-RNA interactions enable specific targeting of

of the genome during gametogenesis. Nature 1984;308:548–50.

noncoding RNAs to nascent pre-mRNAs and chromatin sites. Cell 2014;159:188–99.

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

41. Faghihi MA, Zhang M, Huang J, et al. Evidence for natural antisense transcript-mediated inhibition of microRNA function. Genome Biol 2010;11: R56. 42. Giovarelli M, Bucci G, Ramos A, et al. H19 long noncoding RNA controls the mRNA decay promoting function of KSRP. Proc Natl Acad Sci U S A 2014;111:E5023–8. 43. Carrieri C, Cimatti L, Biagioli M, et al. Long non-coding antisense RNA controls Uchl1 translation through an embedded SINEB2 repeat. Nature 2012;491:454–7. 44. Meurs KM, Mauceli E, Lahmers S, et al. Genome-wide association identifies a deletion in the 30 untranslated region of striatin in a canine model of arrhythmogenic right ventricular cardiomyopathy. Hum Genet 2010;128:315–24. 45. Miller CL, Haas U, Diaz R, et al. Coronary heart disease-associated variation in TCF21 disrupts a miR-224 binding site and miRNA-mediated regulation. PLoS Genet 2014;10:e1004263. 46. Helgadottir A, Thorleifsson G, Manolescu A, et al. A common variant on chromosome 9p21 affects the risk of myocardial infarction. Science 2007;316:1491–3. 47. McPherson R, Pertsemlidis A, Kavaslar N, et al. A common allele on chromosome 9 associated with coronary heart disease. Science 2007;316: 1488–91. 48. Samani NJ, Erdmann J, Hall AS, et al., for WTCCC and the Cardiogenics Consortium. Genomewide association analysis of coronary artery disease. N Engl J Med 2007;357:443–53. 49. Holdt LM, Teupser D. From genotype to phenotype in human atherosclerosis—recent findings. Curr Opin Lipidol 2013;24:410–8. 50. Erridge C, Gracey J, Braund PS, et al. The 9p21 locus does not affect risk of coronary artery disease through induction of type 1 interferons. J Am Coll Cardiol 2013;62:1376–81. 51. Holdt LM, Teupser D. Recent studies of the human chromosome 9p21 locus, which is associated with atherosclerosis in human populations. Arterioscler Thromb Vasc Biol 2012;32:196–206. 52. Holdt LM, Beutner F, Scholz M, et al. ANRIL expression is associated with atherosclerosis risk at chromosome 9p21. Arterioscler Thromb Vasc Biol 2010;30:620–7. 53. Congrains A, Kamide K, Oguro R, et al. Genetic variants at the 9p21 locus contribute to atherosclerosis through modulation of ANRIL and CDKN2A/B. Atherosclerosis 2012;220:449–55. 54. Tsai PC, Liao YC, Lin TH, et al. Additive effect of ANRIL and BRAP polymorphisms on anklebrachial index in a Taiwanese population. Circ J 2012;76:446–52. 55. Harismendy O, Notani D, Song X, et al. 9p21 DNA variants associated with coronary artery disease impair interferon-g signalling response. Nature 2011;470:264–8. 56. Holdt LM, Hoffmann S, Sass K, et al. Alu elements in ANRIL non-coding RNA at chromosome 9p21 modulate atherogenic cell functions through trans-regulation of gene networks. PLoS Genet 2013;9:e1003588.

Long Noncoding RNAs in Cardiovascular Disease

57. Burd CE, Jeck WR, Liu Y, et al. Expression of linear and novel circular forms of an INK4/ARFassociated non-coding RNA correlates with atherosclerosis risk. PLoS Genet 2010;6: e1001233. 58. Congrains A, Kamide K, Katsuya T, et al. CVDassociated non-coding RNA, ANRIL, modulates expression of atherogenic pathways in VSMC. Biochem Biophys Res Commun 2012;419:612–6. 59. Kotake Y, Nakagawa T, Kitagawa K, et al. Long non-coding RNA ANRIL is required for the PRC2 recruitment to and silencing of p15(INK4B) tumor suppressor gene. Oncogene 2011;30:1956–62. 60. Bochenek G, Häsler R, El Mokhtari NE, et al. The large non-coding RNA ANRIL, which is associated with atherosclerosis, periodontitis and several forms of cancer, regulates ADIPOR1, VAMP3 and C11ORF10. Hum Mol Genet 2013;22: 4516–27. 61. Zhao W, Smith JA, Mao G, et al. The cis and trans effects of the risk variants of coronary artery disease in the Chr9p21 region. BMC Med Genomics 2015;8:21. 62. Ishii N, Ozaki K, Sato H, et al. Identification of a novel non-coding RNA, MIAT, that confers risk of myocardial infarction. J Hum Genet 2006;51: 1087–99. 63. Gao W, Zhu M, Wang H, et al. Association of polymorphisms in long non-coding RNA H19 with coronary artery disease risk in a Chinese population. Mutat Res 2015;772:15–22. 64. Matouk IJ, Halle D, Gilon M, et al. The noncoding RNAs of the H19-IGF2 imprinted loci: a focus on biological roles and therapeutic potential in lung cancer. J Transl Med 2015;13:113. 65. Kim DK, Zhang L, Dzau VJ, et al. H19, a developmentally regulated gene, is reexpressed in rat vascular smooth muscle cells after injury. J Clin Invest 1994;93:355–60. 66. Han DK, Khaing ZZ, Pollock RA, et al. H19, a marker of developmental transition, is reexpressed in human atherosclerotic plaques and is regulated by the insulin family of growth factors in cultured rabbit smooth muscle cells. J Clin Invest 1996;97: 1276–85. 67. Tragante V, Barnes MR, Ganesh SK, et al. Gene-centric meta-analysis in 87,736 individuals of European ancestry identifies multiple bloodpressure-related loci. Am J Hum Genet 2014;94: 349–60. 68. Kallen AN, Zhou XB, Xu J, et al. The imprinted H19 lncRNA antagonizes let-7 microRNAs. Mol Cell 2013;52:101–12. 69. Yan L, Zhou J, Gao Y, et al. Regulation of tumor cell migration and invasion by the H19/let-7 axis is antagonized by metformin-induced DNA methylation. Oncogene 2015;34:3076–84. 70. Bao MH, Feng X, Zhang YW, et al. Let-7 in

72. Grote P, Wittler L, Hendrix D, et al. The tissuespecific lncRNA Fendrr is an essential regulator of heart and body wall development in the mouse. Dev Cell 2013;24:206–14. 73. Kurian L, Aguirre A, Sancho-Martinez I, et al. Identification of novel long noncoding RNAs underlying vertebrate cardiovascular development. Circulation 2015;131:1278–90. 74. May D, Blow MJ, Kaplan T, et al. Large-scale discovery of enhancers from human heart tissue. Nat Genet 2011;44:89–93. 75. Blow MJ, McCulley DJ, Li Z, et al. ChIP-Seq identification of weakly conserved heart enhancers. Nat Genet 2010;42:806–10. 76. Matkovich SJ, Edwards JR, Grossenheider TC, et al. Epigenetic coordination of embryonic heart transcription by dynamically regulated long noncoding RNAs. Proc Natl Acad Sci U S A 2014;111: 12264–9. 77. Uchida S, Dimmeler S. Long non-coding RNAs in cardiovascular diseases. Circ Res 2015;116: 737–50. 78. Boeckel JN, Jaé N, Heumüller AW, et al. Identification and characterization of hypoxiaregulated endothelial circular RNA. Circ Res 2015;117:884–90. 79. Yang KC, Yamada KA, Patel AY, et al. Deep RNA sequencing reveals dynamic regulation of myocardial noncoding RNAs in failing human heart and remodeling with mechanical circulatory support. Circulation 2014;129:1009–21. 80. Ounzain S, Micheletti R, Beckmann T, et al. Genome-wide profiling of the cardiac transcriptome after myocardial infarction identifies novel heart-specific long non-coding RNAs. Eur Heart J 2015;36:353–68a. 81. Di Salvo TG, Guo Y, Su YR, et al. Right ventricular long noncoding RNA expression in human heart failure. Pulm Circ 2015;5:135–61. 82. Carrion K, Dyo J, Patel V, et al. The long noncoding HOTAIR is modulated by cyclic stretch and WNT/b-CATENIN in human aortic valve cells and is a novel repressor of calcification genes. PLoS One 2014;9:e96577. 83. Gupta RA, Shah N, Wang KC, et al. Long noncoding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature 2010;464: 1071–6. 84. Kumarswamy R, Bauters C, Volkmann I, et al. Circulating long noncoding RNA, LIPCAR, predicts survival in patients with heart failure. Circ Res 2014;114:1569–75. 85. Dorn GW II LIPCAR: a mitochondrial lnc in the noncoding RNA chain? Circ Res 2014;114:1548–50. 86. Vausort M, Wagner DR, Devaux Y. Long noncoding RNAs in patients with acute myocardial infarction. Circ Res 2014;115:668–77.

cardiovascular diseases, heart development and cardiovascular differentiation from stem cells. Int J Mol Sci 2013;14:23086–102.

87. Yang Y, Cai Y, Wu G, et al. Plasma long noncoding RNA, CoroMarker, a novel biomarker for diagnosis of coronary artery disease. Clin Sci (Lond) 2015;129:675–85.

71. Klattenhoff CA, Scheuermann JC, Surface LE,

88. Lorenzen JM, Schauerte C, Kielstein JT, et al.

et al. Braveheart, a long noncoding RNA required for cardiovascular lineage commitment. Cell 2013; 152:570–83.

Circulating long noncoding RNATapSaki is a predictor of mortality in critically ill patients with acute kidney injury. Clin Chem 2015;61:191–201.

1225

1226

Boon et al.

JACC VOL. 67, NO. 10, 2016 MARCH 15, 2016:1214–26

Long Noncoding RNAs in Cardiovascular Disease

89. Memczak S, Papavasileiou P, Peters O, et al. Identification and characterization of circular RNAs as a new class of putative biomarkers in human blood. PLoS One 2015;10:e0141214.

99. Fiedler J, Breckwoldt K, Remmele CW, et al. Development of long noncoding RNA-based strategies to modulate tissue vascularization. J Am Coll Cardiol 2015;66:2005–15.

90. Boeckel JN, Thomé CE, Leistner D, et al.

100. Wu G, Cai J, Han Y, et al. LincRNA-p21 reg-

Heparin selectively affects the quantification of microRNAs in human blood samples. Clin Chem 2013;59:1125–7.

ulates neointima formation, vascular smooth muscle cell proliferation, apoptosis, and atherosclerosis by enhancing p53 activity. Circulation 2014;130:1452–65.

91. Bassett AR, Akhtar A, Barlow DP, et al. Considerations when investigating lncRNA function in vivo. Elife 2014;3:e03058. 92. Maeder ML, Linder SJ, Cascio VM, et al. CRISPR RNA-guided activation of endogenous human genes. Nat Methods 2013;10:977–9. 93. Michalik KM, You X, Manavski Y, et al. Long noncoding RNA MALAT1 regulates endothelial cell function and vessel growth. Circ Res 2014;114:1389–97. 94. Yan B, Tao ZF, Li XM, et al. Aberrant expression of long noncoding RNAs in early diabetic retinopathy. Invest Ophthalmol Vis Sci 2014;55:941–51. 95. Liu JY, Yao J, Li XM, et al. Pathogenic role of lncRNA-MALAT1 in endothelial cell dysfunction in diabetes mellitus. Cell Death Dis 2014;5:e1506. 96. Puthanveetil P, Chen S, Feng B, et al. Long non-coding RNA MALAT1 regulates hyperglycaemia induced inflammatory process in the endothelial cells. J Cell Mol Med 2015;19:1418–25. 97. Jaé N, Dimmeler S. Long noncoding RNAs in diabetic retinopathy. Circ Res 2015;116:1104–6. 98. Yan B, Yao J, Liu JY, et al. lncRNA-MIAT regulates microvascular dysfunction by functioning as a competing endogenous RNA. Circ Res 2015;116: 1143–56.

101. Hu YW, Zhao JY, Li SF, et al. RP5–833A20.1/ miR-382–5p/NFIA-dependent signal transduction pathway contributes to the regulation of cholesterol homeostasis and inflammatory reaction. Arterioscler Thromb Vasc Biol 2015;35:87–101. 102. Bell RD, Long X, Lin M, et al. Identification and initial functional characterization of a human vascular cell-enriched long noncoding RNA. Arterioscler Thromb Vasc Biol 2014;34:1249–59. 103. Li P, Ruan X, Yang L, et al. A liver-enriched long non-coding RNA, lncLSTR, regulates systemic lipid metabolism in mice. Cell Metab 2015;21:

107. Han P, Li W, Lin CH, et al. A long noncoding RNA protects the heart from pathological hypertrophy. Nature 2014;514:102–6. 108. Hinkel R, Penzkofer D, Zühlke S, et al. Inhibition of microRNA-92a protects against ischemia/ reperfusion injury in a large-animal model. Circulation 2013;128:1066–75. 109. Janssen HLA, Reesink HW, Lawitz EJ, et al. Treatment of HCV infection by targeting microRNA. N Engl J Med 2013;368:1685–94. 110. Krieg AM. Targeting LDL cholesterol with LNA. Mol Ther Nucleic Acids 2012;1:e6. 111. Jeong W, Rapisarda A, Park SR, et al. Pilot trial of EZN-2968, an antisense oligonucleotide inhibitor of hypoxia-inducible factor-1 alpha (HIF-1a), in patients with refractory solid tumors. Cancer Chemother Pharmacol 2014;73:343–8. 112. Tanioka M, Nokihara H, Yamamoto N, et al. Phase I study of LY2181308, an antisense oligo-

455–67.

nucleotide against survivin, in patients with advanced solid tumors. Cancer Chemother Pharmacol 2011;68:505–11.

104. Wang K, Sun T, Li N, et al. MDRL lncRNA regulates the processing of miR-484 primary transcript by targeting miR-361. PLoS Genet 2014; 10:e1004467.

113. Denzler R, Agarwal V, Stefano J, et al. Assessing the ceRNA hypothesis with quantitative measurements of miRNA and target abundance. Mol Cell 2014;54:766–76.

105. Wang K, Liu F, Zhou LY, et al. The long noncoding RNA CHRF regulates cardiac hypertrophy by targeting miR-489. Circ Res 2014;114: 1377–88. 106. Wang K, Liu CY, Zhou LY, et al. APF lncRNA regulates autophagy and myocardial infarction by targeting miR-188–3p. Nat Commun 2015;6:6779.

114. Anderson DM, Anderson KM, Chang CL, et al. A micropeptide encoded by a putative long noncoding RNA regulates muscle performance. Cell 2015;160:595–606.

KEY WORDS cardiac, long noncoding RNA, microRNA, vascular