Transcriptional regulatory functions of nuclear long noncoding RNAs

Abstract

Several nuclear localised intergenic long noncoding RNAs (lncRNAs) have been ascribed regulatory roles in transcriptional control and their number is growing rapidly. Initially, these transcripts were shown to function locally, near their sites of synthesis, by regulating the expression of neighbouring genes. More recently, lncRNAs have been demonstrated to interact with chromatin at several thousand different locations across multiple chromosomes and to modulate large-scale gene expression programs. Although the molecular mechanisms involved in targeting lncRNAs to distal binding sites remain poorly understood, the spatial organisation of the genome may have a role in specifying lncRNA function. Recent advances indicate that intergenic lncRNAs may exert more widespread effects on gene regulation than previously anticipated.

Full text

Transcriptional
regulatory
functions
of
nuclear
long
noncoding
RNAs
Keith
W.
Vance
and
Chris
P.
Ponting
MRC
Functional
Genomics
Unit,
Department
of
Physiology,
Anatomy
and
Genetics,
University
of
Oxford,
South
Parks
Road,
Oxford,
OX1
3PT,
UK
Several
nuclear
localised
intergenic
long
noncoding
RNAs
(lncRNAs)
have
been
ascribed
regulatory
roles
in
transcriptional
control
and
their
number
is
growing
rapidly.
Initially,
these
transcripts
were
shown
to
func-
tion
locally,
near
their
sites
of
synthesis,
by
regulating
the
expression
of
neighbouring
genes.
More
recently,
lncRNAs
have
been
demonstrated
to
interact
with
chro-
matin
at
several
thousand
different
locations
across
multiple
chromosomes
and
to
modulate
large-scale
gene
expression
programs.
Although
the
molecular
mechanisms
involved
in
targeting
lncRNAs
to
distal
binding
sites
remain
poorly
understood,
the
spatial
or-
ganisation
of
the
genome
may
have
a
role
in
specifying
lncRNA
function.
Recent
advances
indicate
that
inter-
genic
lncRNAs
may
exert
more
widespread
effects
on
gene
regulation
than
previously
anticipated.
Emerging
roles
for
nuclear
lncRNAs
The
mammalian
genome
contains
large
numbers
of
non-
coding
RNA
(ncRNA)
loci
that
interdigitate
between,
with-
in,
and
among
protein-coding
genes
on
either
strand.
To
date,
more
than
10
000
mammalian
intergenic
lncRNAs
[>200
nucleotides
(nt)]
(see
Glossary)
have
been
catalo-
gued;
the
majority
of
these
are
expressed
at
lower
levels
compared
with
protein-coding
transcripts,
and
are
more
tissue
specific
[1–3].
A
small
number
of
intergenic
lncRNAs
have
been
implicated
in
a
variety
of
biological
processes
[4].
The
functions,
if
any,
of
the
remaining
transcripts
remain
unknown
and,
in
contrast
to
protein-coding
sequence,
can-
not
yet
be
predicted
from
sequence
alone
[5].
Some
intergenic
lncRNAs
function
as
transcriptional
regulators
that
can
act
locally,
near
their
sites
of
synthesis,
to
regulate
the
expression
of
nearby
genes,
or
distally
to
regulate
gene
expression
across
multiple
chromosomes
(Figure
1).
Here,
we
draw
upon
recent
studies
to
review
the
functions
of
nuclear
localised
intergenic
lncRNAs
in
regulating
gene
transcription
and
chromatin
organisation,
their
local
and
distal
modes
of
action,
their
mechanisms
of
genomic
targeting,
and
the
nature
of
their
interactions
with
chromatin.
lncRNAs
function
at
their
sites
of
synthesis
to
regulate
local
gene
expression
Intergenic
lncRNAs
have
been
divided,
on
the
basis
of
chromatin
marks
at
their
promoters,
into
two
broad
cate-
gories:
those
emanating
from
enhancer
regions
or
those
transcribed
from
promoter-like
lncRNA
loci
[6].
Most,
if
not
all,
transcriptional
enhancer
elements
are
transcribed
to
produce
often
exosome
sensitive,
unspliced
transcripts
termed
‘enhancer
RNAs’
(eRNAs).
The
level
of
these
tran-
scripts
tends
to
correlate
positively
with
expression
levels
of
neighbouring
protein
coding
genes
[7,8].
A
subset
of
enhancers
also
appears
to
be
associated
with
polyadeny-
lated,
more
stable,
and
often
spliced
lncRNAs
variously
called
elncRNAs,
1d-eRNAs,
or
ncRNA-activating
lncRNAs
(ncRNA-a)
[6,9–11].
All
of
these
transcripts
are
likely
generated
bidirectionally
with
RNAs
transcribed
from
either
or
both
strands
being
rapidly
degraded,
as
seen
for
unstable
antisense
promoter
upstream
transcripts
(PROMPTS)
[12]
and
for
intragenic
enhancer
produced
transcripts
[13].
Thus,
further
experiments
will
be
needed
to
determine
the
relative
proportions
and
functions
of
enhancer-associated
lncRNA
loci
that
are
uni-
or
bi-direc-
tional,
capped,
and
polyadenylated
or
unpolyadenylated,
and
multi-
or
mono-exonic.
It
is
currently
unknown
whether
eRNAs
or
elncRNAs
are
commonly
simply
a
by-product
of
or
an
actual
cause
of
enhancer
action
on
neighbouring
protein-coding
genes.
However,
a
small
but
growing
number
of
eRNAs
and
elncRNAs
have
been
shown
to
function
at
their
site
of
synthesis
in
a
RNA-dependent
manner
to
regulate
posi-
tively
the
expression
of
neighbouring
protein
coding
genes
on
the
same
chromosome
[14–18].
In
one
study,
multiple
Review
0168-9525/
ß
2014
The
Authors.
Published
by
Elsevier
Ltd.
This
is
an
open
access
article
under
the
CC
BY
license
(http://creativecommons.org/licenses/by/3.0/).
http://dx.doi.org/
10.1016/j.tig.2014.06.001
Corresponding
authors:
Vance,
K.W.
(Keith.Vance@dpag.ox.ac.uk);
Ponting,
C.P.
(Chris.Ponting@dpag.ox.ac.uk).
Keywords:
long
noncoding
RNA;
transcription;
chromatin
conformation;
RNA–protein
interactions.
Glossary
Cis-acting
lncRNA:
a
lncRNA
that
functions
close
to
its
site
of
synthesis
to
regulate
the
expression
of
nearby
genes
on
the
same
chromosome
in
an
allele-
specific
manner.
Enhancer-associated
lncRNA:
a
lncRNA
whose
genomic
locus
is
marked
by
high
levels
of
histone
H3
lysine
4
mono-
compared
to
tri-methylation.
Intergenic
lncRNA:
a
lncRNA
whose
genomic
locus
does
not
overlap
transcribed
protein
coding
gene
sequence.
lncRNA:
an
RNA
molecule,
greater
than
200
nt
in
length,
which
is
not
predicted
to
encode
protein.
Promoter-associated
lncRNA:
a
lncRNA
whose
genomic
locus
is
marked
by
high
levels
of
histone
H3
lysine
4
tri-methylation
relative
to
monomethylation.
Proximity
transfer:
translocation
of
an
RNA
molecule
from
its
site
of
transcription
to
distal
binding
sites,
located
in
close
spatial
proximity.
Trans-acting
lncRNA:
a
lncRNA
that
regulates
the
expression
of
genes
on
a
different
chromosome
and/or
on
the
homologous
chromosome
from
where
it
is
transcribed.
TIGS-1128;
No.
of
Pages
8
Trends
in
Genetics
xx
(2014)
1–8
1

17B-oestradiol
(E2)-induced
eRNA
transcripts
were
found
to
interact
with
cohesin
in
vitro
and
to
induce
looping
interactions
between
their
enhancer
elements
and
the
promoters
of
nearby
target
genes
[15].
In
another,
two
elncRNAs
(ncRNA-a3
and
ncRNA-a7)
bound
to
compo-
nents
of
the
Mediator
complex
and
also
promoted
enhanc-
er–promoter
looping
interactions
to
regulate
local
gene
expression
[19].
In
a
third
study,
upon
depletion
of
an
eRNA
transcribed
from
the
MyoD1
core
enhancer
region,
both
MyoD1
chromatin
accessibility
and
RNA
polymerase
II
(PolII)
occupancy
were
reduced
and
MyoD1
expression
was
decreased
[17].
Therefore,
enhancer-associated
tran-
scripts
can
modulate
enhancer
activity
by
altering
local
chromatin
accessibility
and/or
structure.
Nevertheless,
other
studies
showed
that
eRNAs
generated
from
p53-bound
enhancers
acted
on
pre-existing
chromatin
conformations
to
increase
enhancer
activity
by
unknown
mechanisms
[16]
and
that
inhibition
of
eRNA
production
at
estrogen
receptor
(ER)
bound
enhancers,
for
example
by
blocking
transcriptional
elongation,
had
no
effect
on
chro-
matin
looping
yet
still
inhibited
target
gene
activation
[18].
Thus,
enhancer-associated
lncRNAs
may
have
multiple
RNA-dependent
mechanisms
of
transcriptional
control.
lncRNA
loci
have
also
been
ascribed
RNA-independent
functions
in
gene
activation,
for
example
that
arise
from
transcription
through
loci
affecting
local
chromatin
acces-
sibility,
as
described
during
fbp1+
gene
activation
in
yeast
[20].
In
another
example,
the
activity
of
the
human
growth
hormone
(hGH)
HS1
enhancer
was
shown
to
be
stimulated
by
lncRNA
transcription
that
initiates
immediately
down-
stream
of
HS1
and
is
noncontiguous
with
the
hGH
target
promoter
[21].
To
investigate
the
molecular
mechanism
of
(A)
(B)
(C)
chr A
chr B
chr A PCG
lncRNA
lncRNA
PCG
lncRNA PCG
lncRNA
PCG
PCG
lncRNA
chr A chr B
chr A
chr B
Local effects tethered to site of synthesis
Distal effects tethered to site of synthesis
Distal effects away from site of synthesis
TRENDS in Genetics
Figure
1.
Local
and
distal
modes
of
long
noncoding
RNA
(lncRNA)-mediated
transcriptional
regulation.
(A)
DNA
looping
interactions
bring
a
lncRNA
locus
into
close
physical
proximity
with
a
genomically
adjacent
protein
coding
gene
(PCG).
Such
lncRNAs
function
close
to
their
sites
of
synthesis
to
regulate
the
expression
of
nearby
genes
on
the
same
chromosome.
(B)
Chromatin
conformation
changes
bring
two
distantly
located
loci
into
close
spatial
proximity.
lncRNAs
in
this
category
function
close
to
their
site
of
synthesis,
but
their
genomic
PCG
targets
are
located
on
different
or
homologous
chromosomes
(chr).
(C)
lncRNAs
translocate
from
their
sites
of
synthesis
to
regulate
transcription
of
distantly
located
target
genes
on
the
same
or
different
chromosomes.
Review Trends
in
Genetics
xxx
xxxx,
Vol.
xxx,
No.
x
TIGS-1128;
No.
of
Pages
8
2

this
stimulation,
a
transcriptional
terminator
was
inserted
into
the
locus,
which
led
to
reduced
lncRNA
transcription,
and
to
a
concomitant
decrease
in
hGH
expression.
When
the
sequence
of
this
lncRNA
was
replaced
by
that
of
an
unrelated
bacteriophage
RNA,
the
enhancing
effect
of
the
natural
transcript
was
recapitulated.
Taken
together,
these
studies
show
that
enhancer-associated
lncRNAs
can
also
act
locally,
near
their
site
of
synthesis,
using
either
RNA-dependent
or
-independent
mechanisms
to
increase
the
transcriptional
activity
of
chromosomally
proximal
protein
coding
genes.
Local
changes
in
gene
expression
are
presumed
to
be
mediated
by
cis-acting
lncRNA
modes
of
action
on
the
same
chromosome,
and
in
an
allele-specific
manner.
However,
trans-acting
lncRNA
mechanisms
could
also
operate
to
control
nearby
gene
expression.
The
lncRNA
Jpx,
for
ex-
ample,
is
transcribed
from
the
active
X
chromosome
and
can
upregulate
expression
of
the
adjacent
Xist
gene
on
the
other
future
inactive
X
allele
in
trans
during
the
process
of
X
chromosome
inactivation
[22].
Evf-2,
in
addition,
mod-
ulates
the
activity
of
an
enhancer
element
present
within
its
locus
by
binding
distal-less
homeobox
2
(DLX2)
and
methyl
CpG
binding
protein
2
(MECP2)
and
inhibiting
DNA
methylation
of
this
enhancer
to
control
expression
of
the
genomically
adjacent
Dlx6
gene
[23,24].
These
effects
occur
in
trans
because
Evf-2
and
DLX2
cooperate
to
increase
the
activity
of
the
Dlx6
enhancer
when
they
are
co-expressed
from
transiently
transfected
plasmids
in
a
reporter
assay,
and
Evf-2
inhibits
DNA
methylation
when
expressed
ectopically
from
a
transgene
in
a
mouse
model.
In
general,
further
mechanistic
studies
will
be
needed
to
assess
the
relative
contributions
of
cis-
and
trans-acting
lncRNA
mechanisms
controlling
local
gene
expression.
lncRNAS
with
distal
regulatory
functions
Many
experiments
thus
far
have
focussed
on
the
possible
mechanisms
by
which
an
enhancer-associated
lncRNA
or
transcription
of
its
locus
regulates
the
expression
level
of
an
adjacent
protein-coding
gene.
However,
long-range
intra-chromosomal
interactions
between
eRNA
expressing
loci
and
distantly
located
loci
have
also
been
documented
[15,16].
The
TFF1
and
NRIP1
eRNA
containing
loci,
locat-
ed
27
Mb
apart
on
chromosome
21,
are
brought
into
close
spatial
proximity
by
long-range
DNA
looping
interactions.
This
looping
is
induced
by
E2
and
appears
to
be
dependent
on
the
NRIP1
eRNA.
Therefore,
a
subset
of
eRNAs
may
have
so
far
uncharacterised
distal
regulatory
roles.
What
has
been
studied
less
often
is
whether
lncRNA
transcripts
can
function
in
trans
at
distal
genomic
loca-
tions.
To
address
this
and
other
issues,
several
techniques
have
been
recently
developed
[25–28]
to
map
the
occupan-
cy
of
lncRNAs
genome-wide
(Box
1).
Although
these
approaches
are
providing
insights
into
the
direct
or
indi-
rect
binding
of
RNAs
to
genomic
locations,
it
is
important
also
to
understand
their
limitations.
One
of
these
is
that
false
positive
inferences
can
arise
from
the
direct
DNA
binding
of
the
antisense
oligonucleotides
used
to
capture
the
RNA
and
also
from
the
cross-linking
of
spatially
adja-
cent
genomic
regions
in
the
nucleus.
Interpretation
of
results
is
further
complicated
by
the
observation
that
the
binding
of
transcription
factors
to
DNA
is
often
insuf-
ficient
to
alter
transcription
[29,30].
Thus,
we
expect
that
a
large
number
of
lncRNA
binding
events
are
also
inconse-
quential.
For
example,
one
eRNA,
transcribed
from
an
E2-
regulated
FoxC1
enhancer,
has
been
shown
to
occupy
15
binding
locations
on
multiple
chromosomes,
all
well
away
from
its
endogenous
locus;
however,
none
of
these
were
located
within
regulatory
regions
of
E2-responsive
genes
and,
thus,
probably
represent
nonfunctional
genomic
inter-
actions
[15].
Therefore,
to
prioritise
binding
events
that
are
functional,
it
is
necessary
to
identify
genes
that
are
both
directly
bound
and
regulated
by
lncRNA
transcripts,
for
example
by
intersecting
lncRNA
genomic
binding
profiles
with
lncRNA-induced
gene
expression
changes.
Determination
of
the
genomic
binding
profiles
for
sev-
eral
other
types
of
lncRNA
shows
that
a
single
lncRNA
transcript
can
interact
with
multiple
binding
sites
on
different
chromosomes
away
from
its
site
of
transcription.
Hotair,
a
lncRNA
transcribed
from
the
homeobox
(Hox)
C
locus,
was
shown
initially
to
function
in
trans
to
repress
the
transcription
of
genes
in
the
HoxD
gene
cluster
on
another
chromosome
[31].
Subsequently,
it
was
found
to
associate
Box
1.
Mapping
genomic
binding
sites
of
lncRNAs
Four
techniques
have
been
developed
recently
that
map
the
interaction
of
RNA
with
chromatin:
chromatin
oligoaffinity
precipita-
tion
(ChOP);
chromatin
isolation
by
RNA
purification
(ChIRP);
capture
hybridisation
analysis
of
RNA
targets
(CHART);
and
RNA
antisense
purification
(RAP)
[25–28,50].
Each
uses
biotinylated
antisense
oligonucleotides
to
capture
RNA
from
cross-linked
chromatin
ex-
tracts,
in
combination
with
quantitative
(q)PCR
and/or
high-through-
put
sequencing,
to
identify
their
associated
DNA
binding
sites.
Although
these
techniques
are
based
on
an
analogous
concept,
they
differ
in
their
chromatin
cross-linking,
shearing,
and
hybridisation
conditions,
the
size
and
number
of
oligonucleotide
probes
used
to
capture
target
RNAs,
and
the
method
of
elution
of
the
associated
DNA
fragments.
Whereas
the
ChOP
method
has
only
been
used
to
examine
RNA
occupancy
at
discrete
loci
[28,65],
ChIRP,
CHART,
and
RAP
have
been
applied
to
map
RNA
chromatin
occupancy
genome-wide.
A
pool
of
approximately
24
short
20-nt
probes
are
used
in
ChIRP
to
enrich
for
RNA
targets
from
nonreversible
glutaraldehyde
cross-
linked
extract
[25],
whereas
in
RAP
a
large
pool
(1054
probes
for
the
17-kb
Xist
transcript)
of
long
120-nt
capture
oligonucleotides
are
used
to
pull
down
target
transcripts
from
a
combination
of
glutarate
and
formaldehyde
cross-linked
cells
[26].
The
use
of
capture
oligonucleo-
tides
that
span
the
whole
length
of
the
target
molecule
in
both
ChIRP
and
RAP
has
been
reported
to
improve
the
effectiveness
of
this
approach
to
uniformly
capture
long
RNAs
[25].
Moreover,
when
RAP
was
used
to
purify
the
highly
abundant
Xist
transcript
from
female
lung
fibroblasts,
approximately
70%
of
the
total
RNA-Seq
reads
in
the
pull
down
corresponded
to
Xist
transcript,
representing
a
massive
enrichment
[26].
CHART,
by
comparison,
uses
RNaseH
sensitivity
mapping
to
design
a
small
number
of
18–28-nt
antisense
oligonu-
cleotides
against
regions
of
the
target
RNA
that
are
accessible
for
hybridisation
[48,50].
These
are
then
used
to
capture
RNA
targets
from
formaldehyde
cross-linked
chromatin.
In
the
CHART
protocol,
hybridisation
is
performed
at
room
temperature
and
RNaseH
is
used
to
elute
RNA–chromatin
complexes
that
specifically
interact
with
antisense
capture
DNA
oligonucleotides,
which
is
effective
in
reducing
the
number
of
false
positive
interactions
that
are
generated
due
to
direct
DNA
binding
of
antisense
oligonucleotides.
The
size,
abundance,
and
subcellular
localisation
of
the
target
RNA
are
likely
to
influence
the
efficacy
of
these
approaches
and,
therefore,
the
optimal
method
should
be
determined
for
each
transcript.
Review Trends
in
Genetics
xxx
xxxx,
Vol.
xxx,
No.
x
TIGS-1128;
No.
of
Pages
8
3

with
approximately
800
binding
locations
of
up
to
1
kb
in
length
across
multiple
chromosomes.
These
focal
binding
sites
are
reported
to
be
embedded
within
larger
polycomb
domains
and
are
enriched
within
genes
that
become
dere-
pressed
upon
Hotair
depletion
[25].
Another
lncRNA,
pros-
tate-specific
transcript
1
(non-protein
coding)
(PCGEM1),
which
binds
to
the
androgen
receptor
(AR),
associates
with
2142
binding
locations
on
the
genome,
the
majority
(ap-
proximately
70%)
of
which
correspond
to
AR-bound
H3K4me1-modified
enhancers.
PCGEM1
association
with
AR-bound
enhancers
appears
to
increase
AR-mediated
gene
activation
without
affecting
AR
levels
[32].
Paupar
is
an
intergenic
lncRNA
that
interacts
with
chromatin
at
over
2800
sites
located
on
multiple
chromo-
somes
and
controls
large-scale
gene
expression
programs
in
a
transcript-dependent
manner
[33].
It
is
transcribed
from
a
conserved
enhancer
upstream
of
the
paired
box
6
(Pax6)
gene
and
its
depletion
significantly
alters
the
expression
of
Pax6
and
942
other
genes
distributed
across
the
genome.
Paupar
binding
sites,
defined
by
CHART-Seq
(Box
1),
overlap
func-
tional
elements,
such
as
DNase
I
hypersensitive
sites
(HSs),
and
are
enriched
at
gene
promoters.
Control
CHART-Seq
experiments
using
lacZ
probes
showed
that
binding
to
such
sites
by
RNAs
can
be
nonspecific.
Consequently,
candidate
lncRNA genomic binding profile
Binding site not within
regulatory region
of target gene
Binding site within
regulatory region
of target gene
Putave funconal interacon
Nonfunconal binding?
Binding site overlaps
DNase HS or chroman
defined regulatory region
Binding site does
not overlap
DNase HS
Binds transcriponal regulatory element
Transcript less likely to have
transcriponal regulatory role
Transcriptome
profiling
Chromosome
conformaon
capture
Chroman
mapping
Binding site within
close spaal proximity
to lncRNA locus
Binding site
distant to
lncRNA locus
Proximity transferTranscript translocaon
Binding site enriched for
lncRNA complementary
sequences
Binding site enriched
for transcripon
factor mofs
Indirect DNA binding
Direct RNA–DNA–DNA interacon
Funconal and biochemical analysis
Computaonal
analysis
Funconal bindingRegulatory element
associaon
Guiding mechanism
Chroman interacon
TRENDS in Genetics
Figure
2.
Workflow
diagram
detailing
a
combined
experimental
and
computational
pipeline
for
investigating
trans-acting
long
noncoding
RNA
(lncRNA)
transcriptional
regulatory
functions.
Likely
functional
lncRNA
genomic
binding
sites
are
identified
within
the
regulatory
regions
of
target
genes
that
are
differentially
expressed
upon
lncRNA
depletion
or
overexpression.
Gene
Ontology
analyses
of
lncRNA
bound
and
regulated
genes
provide
clues
regarding
lncRNA
function.
Binding
sites
that
associate
with
transcriptional
regulatory
regions
are
further
selected
for
based
on
DNase
I
hypersensitive
site
mapping
and
chromatin
status.
Putative
functional
binding
locations
are
integrated
with
chromosome
conformation
capture-based
experiments
to
provide
insights
into
the
mechanism
of
genomic
targeting.
Computational
sequence
analyses
generate
predictions
regarding
how
lncRNAs
interact
with
DNA.
These
criteria
are
used
to
inform
the
design
of
reporter
assays
and
biochemical
experiments
aimed
at
understanding
trans-acting
lncRNA
function
and
mode
of
action
at
candidate
loci.
Review Trends
in
Genetics
xxx
xxxx,
Vol.
xxx,
No.
x
TIGS-1128;
No.
of
Pages
8
4

functional
Paupar
binding
sites
were
predicted
to
be
only
those
within
the
regulatory
regions
of
genes
that
were
differentially
expressed
upon
Paupar
depletion.
Paupar
was
then
shown
using
reporter
assays
to
modulate
the
transcriptional
activity,
in
trans
and
in
a
dose-dependent
manner,
of
three
out
of
five
such
candidate
regions
tested.
These
experiments
demonstrate
that
a
lncRNA
can
have
dual
functions
both
locally,
to
regulate
the
expression
of
its
neighbouring
protein
coding
gene,
and
distally
at
regulatory
elements
genome-wide.
In
this
case,
the
distal
functions
of
Paupar
rely,
in
part,
on
it
being
guided
to
its
genomic
binding
sites
by
formation
of
a
complex
with
PAX6,
a
DNA-binding
protein.
The
Firre
lncRNA
also
appears,
from
its
genome-wide
binding
profile,
to
act
locally
as
well
as
distally.
It
occupies
a
large
5-Mb
domain
surrounding
its
site
of
synthesis
on
the
X
chromosome
and
interacts
with
five
additional
domains
on
four
different
autosomes
[34].
Only
one
of
these
binding
events
was
shown
to
alter
the
expression
of
a
gene
within
the
bound
region.
The
ncRNA
Ctbp1-as
has
also
been
shown
to
function
both
locally,
to
repress
Ctbp1
expression
through
a
sense-antisense
mediated
mecha-
nism,
and
distally
to
increase
AR
transcriptional
activity
in
prostate
cancer
cells
[35].
Although
such
studies
are
as
yet
limited
in
number,
they
suggest
that
the
ability
of
lncRNAs
to
function
both
locally,
as
well
as
distally,
to
regulate
large-scale
gene
expression
programs
may
be
more
widespread
than
origi-
nally
anticipated.
As
genome-wide
binding
profiles
for
more
lncRNAs
are
mapped
and
their
direct
transcriptional
targets
are
identified,
there
will
be
increasing
opportu-
nities
to
elucidate
their
presumed
heterogeneous
molecu-
lar
mechanisms
(Figure
2).
lncRNA
genome
targeting
The
mechanisms
by
which
lncRNAs
target
specific
genomic
sequences
are
not
understood.
It
is
easy
to
envisage
that
lncRNA
transcripts
could
participate
in
regulating
local
gene
expression
by
accumulating
to
comparatively
high
concentrations
at
their
sites
of
synthesis.
However,
it
is
more
difficult
to
explain
how
lowly
expressed,
and
often
unstable,
nuclear
lncRNAs
act
by
binding
many
different
chromosomal
regions
that
lie
distant
to
their
site
of
tran-
scription.
Recent
reports
suggest
that
the
3D
conformation
of
the
genome
guides
lncRNAs
to
distal
binding
sites.
This
process
of
‘proximity
transfer’
was
first
proposed
for
Xist
on
the
basis
of
its
transferral
from
its
site
of
synthesis
to
distal,
yet
spatially
close,
binding
sites
along
the
X
chro-
mosome;
however,
confirmation
of
this
model
will
require
data
at
higher
resolution
than
the
1-Mbp
intervals
used
in
the
initial
study
[26].
The
mechanism
of
proximity
transfer
is
further
sup-
ported
by
observations
concerning
the
Hottip
lncRNA
lo-
cus.
Chromosomal
looping
interactions
were
found
that
brought
this
locus
into
close
spatial
proximity
to
its
target
genes
in
the
HOXA
cluster
[36].
Furthermore,
transcrip-
tion
was
activated
when
the
Hottip
transcript
was
recruited
to
its
target
promoters
in
reporter
assays,
where-
as
induction
of
Hottip
expression
from
an
ectopic
site
had
no
effect
[36].
The
spatial
organisation
of
the
genome
might
also
permit
lncRNAs
to
span
multiple
binding
locations
across
different
chromosomes,
including
their
sites
of
syn-
thesis.
Consistent
with
this,
the
binding
domains
of
Firre
on
different
chromosomes
appear
to
be
located
in
close
spatial
proximity
within
the
nucleus
[34].
Several
nuclear
lncRNAs
are
able
to
regulate
transcrip-
tion
when
expressed
in
trans
from
ectopic
loci
[22,24,33,37].
This
suggests
an
alternative
model
in
which
lncRNAs
are
translocated
from
their
site
of
synthesis
as
components
of
ribonucleoprotein
complexes
to
bind
specif-
ically
and
regulate
the
expression
of
distantly
located
target
genes.
One
such
lncRNA
could
be
NeST,
which
is
involved
in
controlling
the
immune
response
to
microbial
infection.
NeST
can
activate
interferon
gamma
(Ifn)g
tran-
scription
in
trans,
both
when
expressed
from
a
transgene
and
also
from
its
genomic
locus,
by
interacting
with
WD
repeat
domain
5
(WDR5)
and
by
altering
Ifng
histone
H3K4
tri-methylation
[37].
Additionally,
and
in
contrast
to
the
proximity
transfer
model,
Xist
might
be
a
diffusible
factor;
when
Xist
is
expressed
from
a
transgene
in
female
mouse
embryonic
fibroblasts,
it
diffused
from
the
ectopic
site
of
synthesis
and
acted
on
the
endogenous
Xist
locus
in
trans
[38].
Similarly,
Evf-2,
when
expressed
from
a
tran-
siently
transfected
plasmid,
cooperated
with
the
DLX2
protein
to
activate
the
Dlx-5/6
enhancer
in
a
luciferase
reporter
in
trans
[24],
and
depletion
of
the
endogenous
Paupar
transcript
modulated,
in
a
dose-dependent
man-
ner,
the
transcriptional
activity
of
a
number
of
its
genomic
binding
sites
when
inserted
into
transiently
transfected
reporters
[33].
Therefore,
coordination
among
sites
of
lncRNA
synthe-
sis,
the
spatial
organisation
of
the
genome
in
the
nucleus,
and
specific
lncRNA
interactions
with
transcription
and
chromatin
regulatory
proteins
are
all
likely
to
have
roles
in
facilitating
binding
of
lncRNA
transcripts
to
their
genomic
targets
(Figure
2).
lncRNA–chromatin
interactions
The
genomic
associations
observed
between
lncRNAs
and
chromatin
could
be
accomplished
through
direct
base
pair-
ing
between
RNA
and
DNA
sequences
[39]
(Figure
3A).
This
is
exemplified
by
promoter
associated
RNA
(pRNA),
a
low-abundance
RNA
transcribed
from
upstream
of
the
pre-
rRNA
transcription
start
site
that
can
interact
with
com-
plementary
sequences
within
the
rDNA
promoter
forming
a
RNA–DNA–DNA
triplex,
possibly
through
Hoogsteen
base-pairing
[40].
Furthermore,
because
RNA
base
pairs
with
itself,
RNA–RNA
interactions
between
complemen-
tary
sequences
at
transcribed
loci
could
also
guide
lncRNAs
to
their
genomic
targets
(Figure
3B).
It
remains
unclear
how
widespread
direct
RNA–DNA
or
RNA–RNA
targeting
may
be.
lncRNAs
that
associate
with
sequence-specific
DNA
binding
transcription
factors
could
be
targeted
to
the
genome
indirectly
through
RNA–protein–DNA
interac-
tions
(Figure
3C).
YY1,
for
example,
is
a
zinc
finger-con-
taining
transcription
factor
that
may
recruit
Xist
to
chromatin
by
binding
DNA
and
Xist
RNA
through
different
sequence
domains
[38].
Other
candidates
for
RNA–protein
DNA-binding
complexes
include
Gas5
and
glucocorticoid
receptor
[41],
Panda
and
nuclear
transcription
factor
Y
(NFYA)
[42],
Lethe
and
nuclear
factor
kB
(NF-kB)
[43],
Jpx
Review Trends
in
Genetics
xxx
xxxx,
Vol.
xxx,
No.
x
TIGS-1128;
No.
of
Pages
8
5

and
CCCTC-binding
factor
(CTCF)
[44],
Paupar
and
PAX6
[33],
Rmst
and
SRY
(sex
determining
region
Y)-box
(SOX2)
[45],
and
Prncr1
and
AR
[32].
Each
of
the
Gas5,
Panda,
Lethe,
and
Jpx
lncRNAs
appears
to
inhibit
DNA
binding
of
their
associated
transcription
factors
at
several
target
sites,
whereas
knockdown
of
Paupar
and
Prncr1
levels
had
no
effect
on
PAX6
or
AR
occupancy
where
tested.
By
contrast,
Rmst
appears
to
be
required
for
the
correct
association
of
SOX2
with
promoter
regions
of
neurogenic
target
genes
[45].
Therefore,
lncRNAs
can
actively
modu-
late
the
DNA
binding
activity
of
their
associated
transcrip-
tion
factors
as
well
as
acting
as
non-DNA
binding
cofactors,
as
has
been
described
for
Six3OS
and
SRA
[46,47],
whose
precise
regulatory
roles
need
to
be
investigated.
Chromatin
modification
and
structure
may
also
modu-
late
how
lncRNAs
are
recruited
to
the
genome.
Xist
appears
to
target
active
chromatin
because
domains
that
are
initially
occupied
by
Xist
are
unusually
enriched
in
actively
transcribed
genes
and
open
chromatin
[48],
where-
as
AR-associated
lncRNAs
Pcgem1
and
Prncr1
preferen-
tially
interact
with
enhancer-associated
histone
modifications
in
vitro
[32].
Computational
approaches
are
beginning
to
predict
how
lncRNAs
interact
with
chro-
matin
or
DNA:
firstly,
by
suggesting
the
candidature
of
lncRNA-associated
transcription
factors
from
enrichments
of
their
binding
motifs;
secondly
by
proposing
the
involve-
ment
of
the
lncRNA
in
transcriptional
enhancement
or
repression
from
enrichments
of
relevant
chromatin
marks;
and
thirdly
by
identifying
near
complementary
DNA
sequence
within
lncRNA-associated
regions
that
might
indicate
direct
RNA–DNA–DNA
triplex
formation
[25,33,49,50].
Mechanisms
of
action
Several
lncRNAs
associate
with
chromatin-modifying
com-
plexes
and
transcriptional
regulatory
proteins
in
the
nu-
cleus.
High
throughput
RNA-immunoprecipitation
(RNA-
IP)
experiments
have
indicated
that
individual
chromatin-
modifying
complexes,
such
as
polycomb
repressive
complex
2
(PRC2)
and
mixed-lineage
leukemia
(MLL),
interact
with
thousands
of
RNA
transcripts,
including
lncRNAs
[51,52].
However,
substantial
numbers
of
transcripts
are
known
to
bind
nonspecifically
and
reproducibly
with
various
RNA-
binding
proteins
in
RNA-IP
based
assays
[53].
Further-
more,
purified
PRC2
complex,
for
example,
binds
RNA
nonspecifically
in
vitro
in
a
size-dependent
manner
[54].
Thus,
studies
will
need
to
distinguish
specific
from
non-
specific
RNA–protein
interactions.
lncRNAs
that
bind
proteins
specifically
might
act
as
guides
to
target
chromatin-modifying
complexes
to
the
genome.
The
lncRNA
Mistral,
for
example,
which
is
tran-
scribed
from
the
intergenic
region
between
the
Hoxa6
and
Hoxa7
genes,
forms
a
RNA–DNA
hybrid
structure
at
its
site
of
synthesis
which
recruits
MLL1
complex
proteins
[55].
By
contrast,
the
lateral
mesoderm-specific
lncRNA
Fendrr
can
associate
with
PRC2
and
regulate
Pitx2
expres-
sion
in
trans.
Fendrr
is
predicted
to
interact
with
a
short
stretch
of
complementary
sequence,
of
fewer
than
40
nt,
in
the
Pitx2
promoter,
which
can
form
a
RNA–DNA–DNA
triplex
in
vitro.
Given
that
Pitx2
promoter
PRC2
occupancy
and
histone
H3K27
tri-methylation
are
decreased
in
Fendrr
knockout
embryos,
Fendrr
may
have
a
role
in
recruiting
PRC2
to
the
Pitx2
promoter
[56].
Thus,
such
nuclear
lncRNAs
have
the
potential
to
target
chromatin
remodelling
complexes
either
to
their
sites
of
synthesis
or
to
distally
located
loci
in
trans.
Several
recent
studies
suggest
that
lncRNAs
modulate
the
structure
and
function
of
their
associated
protein
com-
plexes.
The
Drosophila
roX2
lncRNA
appears
to
function
as
lncRNA
DNA
helix
(A)
(B)
(C)
Direct RNA–DNA interacons
RNA–RNA interacons at transcribed loci
Indirect RNA–protein–DNA associaons
DNA
helix
lncRNA
DNA
binding
protein
Transcripon unit
Polymerase
lncRNA
DNA
helix
Transcript
TRENDS in Genetics
Figure
3.
Different
modes
of
long
noncoding
RNA
(lncRNA)–chromatin
association.
(A)
Single-stranded
lncRNAs
directly
interact
with
complementary
double-stranded
DNA
target
sequences
through
hydrogen
bonding
to
form
a
RNA-
DNA-DNA
triplex
structure.
lncRNAs
are
predicted
to
bind
in
the
major
groove
of
the
DNA
through
either
Hoogsteen
or
reverse
Hoogsteen
base
pairing.
(B)
lncRNAs
base
pair
with
RNA
sequences
at
transcribed
loci.
This
may
involve
Watson–Crick
base
pairing
(G–C,
A–U)
between
complementary
nucleotides
as
well
as
non-
Watson–Crick
base
pairing
(G–U,
A–A)
which
does
not
require
exact
sequence
complementarity.
(C)
Indirect
recruitment
of
lncRNAs
to
the
genome
through
RNA–protein–DNA
interactions.
This
includes
lncRNA
associations
with
sequence
specific
DNA
binding
transcription
factors,
non-DNA
binding
transcriptional
cofactors
and
histone
proteins.
Post-translational
histone
modifications,
such
as
acetylation,
methylation,
and
ubiquitination,
may
influence
lncRNA–histone
binding.
Review Trends
in
Genetics
xxx
xxxx,
Vol.
xxx,
No.
x
TIGS-1128;
No.
of
Pages
8
6

a
critical
complex
subunit
that
is
necessary
for
the
correct
assembly
of
a
functional
male-specific
lethal
(MSL)
dosage
compensation
complex.
Its
stem
loop-containing
structured
domains
bind
the
MLE
RNA
helicase
and
MSL2
ubiquitin
ligase
components
of
the
MSL
dosage
compensation
com-
plex
in
a
sequential
manner
[57,58].
Interaction
of
roX2
with
the
MLE
RNA
helicase
results
in
an
ATP-dependent
con-
formational
change
in
a
roX2
stem-loop
structure
and
a
subsequent
increase
in
its
association
with
MSL2.
Other
lncRNAs
may
also
promote
the
ordered
recruitment
of
functional
ribonucleoprotein
complexes.
For
example,
Pcgem1
and
Prncr1
bind
AR
sequentially,
thereby
stimulat-
ing
both
ligand-dependent
and
ligand-independent
AR
con-
trolled
gene
expression
programs
[15].
Prncr1
first
associates
with
DOT1-like,
histone
H3
methyltransferase
(DOT1L)
and
binds
to
acetylated
enhancer-bound
AR
caus-
ing
DOT1L
to
methylate
the
AR.
AR
methylation
subse-
quently
induces
recruitment
of
Pcgem1,
in
complex
with
pygopus
homolog
2
(PYGO2),
which
because
of
its
H3K4me3-binding
ability,
may
stimulate
DNA
looping
interactions
between
AR-bound
transcriptional
enhancers
and
target
promoters.
These
studies
raise
the
possibility
that
a
single
lncRNA
molecule
contains
multiple
structural
motifs,
upon
which
multiple
different
proteins
bind,
which
enhance
the
effi-
ciency
of
genomic
targeting
and
transcriptional
regulation
[59].
Hotair,
for
example,
interacts
with
PRC2
through
its
50end,
whereas
its
30region
associates
with
(co)repressor
for
element-1-silencing
transcription
factor
(CoREST)
in
vitro
[60].
A
scaffolding
role
for
lncRNAs
may
also
translo-
cate
gene
loci
between
different
nuclear
compartments
to
allow
transcriptional
activation
or
repression
in
response
to
various
stimuli.
This
is
exemplified
by
the
differential
interactions
of
taurine
upregulated
1
(Tug1)
and
metasta-
sis
associated
lung
adenocarcinoma
transcript
1(Malat1)
with
methylated
and
unmethylated
Pc2
protein,
respec-
tively,
and
the
subsequent
relocation
of
growth
control
genes
from
Tug1-containing
polycomb
bodies,
where
they
are
repressed,
to
interchromatin
granules
for
assembly
of
activator
complexes,
where
they
are
activated
[61].
Concluding
remarks
Roles
for
lncRNAs
as
regulators
of
chromatin
organisation
and
gene
expression
were
initially
described
for
H19
and
Xist
lncRNAs
in
genomic
imprinting
and
X
chromosome
inactivation,
respectively.
It
has
since
become
apparent
that
the
genome
encodes
large
numbers
of
nuclear
local-
ised
intergenic
transcripts.
The
number
of
these
tran-
scripts
that
derive
from
serendipitous
transcription
or
that
fail
to
have
functions
(as
opposed
to
mere
effects
[62])
remains
unknown.
Evolutionary
evidence
for
selected
effect
functionality
[62]
of
lncRNAs,
in
general,
is
meagre
[3,63]
and
the
proportion
of
lncRNA
sequence
that
is
under
purifying
selection
appears
to
be
small,
approximately
5%
[64].
Therefore,
further
detailed
studies
on
a
larger
sample
of
lncRNAs
are
needed
to
estimate
the
proportion
of
lncRNAs
that
are
functional,
as
well
as
to
define
their
structure–function
relations,
and
to
better
understand
the
mechanisms
of
these
transcripts
in
regulating
genome
organisation
and
gene
transcription.
We
also
await
the
results
of
lncRNA
loss-of-function
studies
in
animal
model
systems
that
discriminate
lncRNA
from
DNA
sequence-
mediated
effects
that
might
identify
nuclear
lncRNAs
that
are
essential
for
embryonic
development
and
adult
tissue
homeostasis
in
vivo.
Acknowledgements
The
authors’
work
is
funded
by
an
European
Research
Council
Advanced
Grant
and
by
the
Medical
Research
Council.
References
1
Cabili,
M.N.
et
al.
(2011)
Integrative
annotation
of
human
large
intergenic
noncoding
RNAs
reveals
global
properties
and
specific
subclasses.
Genes
Dev.
25,
1915–1927
2
Derrien,
T.
et
al.
(2012)
The
GENCODE
v7
catalog
of
human
long
noncoding
RNAs:
analysis
of
their
gene
structure,
evolution,
and
expression.
Genome
Res.
22,
1775–1789
3
Marques,
A.C.
and
Ponting,
C.P.
(2009)
Catalogues
of
mammalian
long
noncoding
RNAs:
modest
conservation
and
incompleteness.
Genome
Biol.
10,
R124
4
Ulitsky,
I.
and
Bartel,
D.P.
(2013)
lincRNAs:
genomics,
evolution,
and
mechanisms.
Cell
154,
26–46
5
Ponjavic,
J.
et
al.
(2009)
Genomic
and
transcriptional
co-localization
of
protein-coding
and
long
non–coding
RNA
pairs
in
the
developing
brain.
PLoS
Genet.
5,
e1000617
6
Marques,
A.C.
et
al.
(2013)
Chromatin
signatures
at
transcriptional
start
sites
separate
two
equally
populated
yet
distinct
classes
of
intergenic
long
noncoding
RNAs.
Genome
Biol.
14,
R131
7
Andersson,
R.
et
al.
(2014)
An
atlas
of
active
enhancers
across
human
cell
types
and
tissues.
Nature
507,
455–461
8
Kim,
T.K.
et
al.
(2010)
Widespread
transcription
at
neuronal
activity-
regulated
enhancers.
Nature
465,
182–187
9
De
Santa,
F.
et
al.
(2010)
A
large
fraction
of
extragenic
RNA
pol
II
transcription
sites
overlap
enhancers.
PLoS
Biol.
8,
e1000384
10
Natoli,
G.
and
Andrau,
J.C.
(2012)
Noncoding
transcription
at
enhancers:
general
principles
and
functional
models.
Annu.
Rev.
Genet.
46,
1–19
11
Orom,
U.A.
et
al.
(2010)
Long
noncoding
RNAs
with
enhancer-like
function
in
human
cells.
Cell
143,
46–58
12
Ntini,
E.
et
al.
(2013)
Polyadenylation
site-induced
decay
of
upstream
transcripts
enforces
promoter
directionality.
Nat.
Struct.
Mol.
Biol.
20,
923–928
13
Kowalczyk,
M.S.
et
al.
(2012)
Intragenic
enhancers
act
as
alternative
promoters.
Mol.
Cell
45,
447–458
14
Lam,
M.T.
et
al.
(2013)
Rev-Erbs
repress
macrophage
gene
expression
by
inhibiting
enhancer-directed
transcription.
Nature
498,
511–515
15
Li,
W.
et
al.
(2013)
Functional
roles
of
enhancer
RNAs
for
oestrogen-
dependent
transcriptional
activation.
Nature
498,
516–520
16
Melo,
C.A.
et
al.
(2013)
eRNAs
are
required
for
p53-dependent
enhancer
activity
and
gene
transcription.
Mol.
Cell
49,
524–535
17
Mousavi,
K.
et
al.
(2013)
eRNAs
promote
transcription
by
establishing
chromatin
accessibility
at
defined
genomic
loci.
Mol.
Cell
51,
606–617
18
Hah,
N.
et
al.
(2013)
Enhancer
transcripts
mark
active
estrogen
receptor
binding
sites.
Genome
Res.
23,
1210–1223
19
Lai,
F.
et
al.
(2013)
Activating
RNAs
associate
with
Mediator
to
enhance
chromatin
architecture
and
transcription.
Nature
494,
497–501
20
Hirota,
K.
et
al.
(2008)
Stepwise
chromatin
remodelling
by
a
cascade
of
transcription
initiation
of
non-coding
RNAs.
Nature
456,
130–134
21
Yoo,
E.J.
et
al.
(2012)
An
RNA-independent
linkage
of
noncoding
transcription
to
long-range
enhancer
function.
Mol.
Cell.
Biol.
32,
2020–2029
22
Tian,
D.
et
al.
(2010)
The
long
noncoding
RNA,
Jpx,
is
a
molecular
switch
for
X
chromosome
inactivation.
Cell
143,
390–403
23
Berghoff,
E.G.
et
al.
(2013)
Evf2
(Dlx6as)
lncRNA
regulates
ultraconserved
enhancer
methylation
and
the
differential
transcriptional
control
of
adjacent
genes.
Development
140,
4407–4416
24
Feng,
J.
et
al.
(2006)
The
Evf-2
noncoding
RNA
is
transcribed
from
the
Dlx-5/6
ultraconserved
region
and
functions
as
a
Dlx-2
transcriptional
coactivator.
Genes
Dev.
20,
1470–1484
25
Chu,
C.
et
al.
(2011)
Genomic
maps
of
long
noncoding
RNA
occupancy
reveal
principles
of
RNA-chromatin
interactions.
Mol.
Cell
44,
667–678
26
Engreitz,
J.M.
et
al.
(2013)
The
Xist
lncRNA
exploits
three-dimensional
genome
architecture
to
spread
across
the
X
chromosome.
Science
341,
1237973
Review Trends
in
Genetics
xxx
xxxx,
Vol.
xxx,
No.
x
TIGS-1128;
No.
of
Pages
8
7

27
Simon,
M.D.
(2013)
Capture
hybridization
analysis
of
RNA
targets
(CHART).
Curr.
Protoc.
Mol.
Biol.
http://dx.doi.org/10.1002/
0471142727.mb2125s101
101.21.25.1-21.25.16
28
Mariner,
P.D.
et
al.
(2008)
Human
Alu
RNA
is
a
modular
transacting
repressor
of
mRNA
transcription
during
heat
shock.
Mol.
Cell
29,
499–509
29
Cusanovich,
D.A.
et
al.
(2014)
The
functional
consequences
of
variation
in
transcription
factor
binding.
PLoS
Genet.
10,
e1004226
30
Zhou,
X.
and
O’Shea,
E.K.
(2011)
Integrated
approaches
reveal
determinants
of
genome-wide
binding
and
function
of
the
transcription
factor
Pho4.
Mol.
Cell
42,
826–836
31
Rinn,
J.L.
et
al.
(2007)
Functional
demarcation
of
active
and
silent
chromatin
domains
in
human
HOX
loci
by
noncoding
RNAs.
Cell
129,
1311–1323
32
Yang,
L.
et
al.
(2013)
lncRNA-dependent
mechanisms
of
androgen-
receptor-regulated
gene
activation
programs.
Nature
500,
598–602
33
Vance,
K.W.
et
al.
(2014)
The
long
non-coding
RNA
Paupar
regulates
the
expression
of
both
local
and
distal
genes.
EMBO
J.
33,
296–311
34
Hacisuleyman,
E.
et
al.
(2014)
Topological
organization
of
multichromosomal
regions
by
the
long
intergenic
noncoding
RNA
Firre.
Nat.
Struct.
Mol.
Biol.
21,
198–206
35
Takayama,
K.I.
et
al.
(2013)
Androgen-responsive
long
noncoding
RNA
CTBP1-AS
promotes
prostate
cancer.
EMBO
J.
32,
1665–1680
36
Wang,
K.C.
et
al.
(2011)
A
long
noncoding
RNA
maintains
active
chromatin
to
coordinate
homeotic
gene
expression.
Nature
472,
120–124
37
Gomez,
J.A.
et
al.
(2013)
The
NeST
long
ncRNA
controls
microbial
susceptibility
and
epigenetic
activation
of
the
interferon-gamma
locus.
Cell
152,
743–754
38
Jeon,
Y.
and
Lee,
J.T.
(2011)
YY1
tethers
Xist
RNA
to
the
inactive
X
nucleation
center.
Cell
146,
119–133
39
Buske,
F.A.
et
al.
(2011)
Potential
in
vivo
roles
of
nucleic
acid
triple-
helices.
RNA
Biol.
8,
427–439
40
Schmitz,
K.M.
et
al.
(2010)
Interaction
of
noncoding
RNA
with
the
rDNA
promoter
mediates
recruitment
of
DNMT3b
and
silencing
of
rRNA
genes.
Genes
Dev.
24,
2264–2269
41
Kino,
T.
et
al.
(2010)
Noncoding
RNA
gas5
is
a
growth
arrest-
and
starvation-associated
repressor
of
the
glucocorticoid
receptor.
Sci.
Signal.
3,
ra8
42
Hung,
T.
et
al.
(2011)
Extensive
and
coordinated
transcription
of
noncoding
RNAs
within
cell-cycle
promoters.
Nat.
Genet.
43,
621–629
43
Rapicavoli,
N.A.
et
al.
(2013)
A
mammalian
pseudogene
lncRNA
at
the
interface
of
inflammation
and
anti-inflammatory
therapeutics.
Elife
2,
e00762
44
Sun,
S.
et
al.
(2013)
Jpx
RNA
activates
Xist
by
evicting
CTCF.
Cell
153,
1537–1551
45
Ng,
S.Y.
et
al.
(2013)
The
long
noncoding
RNA
RMST
interacts
with
SOX2
to
regulate
neurogenesis.
Mol.
Cell
51,
349–359
46
Rapicavoli,
N.A.
et
al.
(2011)
The
long
noncoding
RNA
Six3OS
acts
in
trans
to
regulate
retinal
development
by
modulating
Six3
activity.
Neural
Dev.
6,
32
47
Yao,
H.
et
al.
(2010)
Mediation
of
CTCF
transcriptional
insulation
by
DEAD-box
RNA-binding
protein
p68
and
steroid
receptor
RNA
activator
SRA.
Genes
Dev.
24,
2543–2555
48
Simon,
M.D.
et
al.
(2013)
High-resolution
Xist
binding
maps
reveal
two-
step
spreading
during
X-chromosome
inactivation.
Nature
504,
465–469
49
Buske,
F.A.
et
al.
(2012)
Triplexator:
detecting
nucleic
acid
triple
helices
in
genomic
and
transcriptomic
data.
Genome
Res.
22,
1372–1381
50
Simon,
M.D.
et
al.
(2011)
The
genomic
binding
sites
of
a
noncoding
RNA.
Proc.
Natl.
Acad.
Sci.
U.S.A.
108,
20497–20502
51
Yang,
Y.W.
et
al.
(2014)
Essential
role
of
lncRNA
binding
for
WDR5
maintenance
of
active
chromatin
and
embryonic
stem
cell
pluripotency.
Elife
3,
e02046
52
Zhao,
J.
et
al.
(2010)
Genome-wide
identification
of
polycomb-
associated
RNAs
by
RIP-seq.
Mol.
Cell
40,
939–953
53
Friedersdorf,
M.B.
and
Keene,
J.D.
(2014)
Advancing
the
functional
utility
of
PAR-CLIP
by
quantifying
background
binding
to
mRNAs
and
lncRNAs.
Genome
Biol.
15,
R2
54
Davidovich,
C.
et
al.
(2013)
Promiscuous
RNA
binding
by
Polycomb
repressive
complex
2.
Nat.
Struct.
Mol.
Biol.
20,
1250–1257
55
Bertani,
S.
et
al.
(2011)
The
noncoding
RNA
Mistral
activates
Hoxa6
and
Hoxa7
expression
and
stem
cell
differentiation
by
recruiting
MLL1
to
chromatin.
Mol.
Cell
43,
1040–1046
56
Grote,
P.
et
al.
(2013)
The
tissue-specific
lncRNA
Fendrr
is
an
essential
regulator
of
heart
and
body
wall
development
in
the
mouse.
Dev.
Cell
24,
206–214
57
Ilik,
I.A.
et
al.
(2013)
Tandem
stem-loops
in
roX
RNAs
act
together
to
mediate
X
chromosome
dosage
compensation
in
Drosophila.
Mol.
Cell
51,
156–173
58
Maenner,
S.
et
al.
(2013)
ATP-dependent
roX
RNA
remodeling
by
the
helicase
maleless
enables
specific
association
of
MSL
proteins.
Mol.
Cell
51,
174–184
59
Guttman,
M.
and
Rinn,
J.L.
(2012)
Modular
regulatory
principles
of
large
non-coding
RNAs.
Nature
482,
339–346
60
Tsai,
M.C.
et
al.
(2010)
Long
noncoding
RNA
as
modular
scaffold
of
histone
modification
complexes.
Science
329,
689–693
61
Yang,
L.
et
al.
(2011)
ncRNA-
and
Pc2
methylation-dependent
gene
relocation
between
nuclear
structures
mediates
gene
activation
programs.
Cell
147,
773–788
62
Doolittle,
W.F.
(2014)
Distinguishing
between
‘function’
and
‘effect’
in
genome
biology.
Genome
Biol.
Evol.
6,
1234–1237
63
Haerty,
W.
and
Ponting,
C.P.
(2013)
Mutations
within
lncRNAs
are
effectively
selected
against
in
fruitfly
but
not
in
human.
Genome
Biol.
14,
R49
64
Ponjavic,
J.
et
al.
(2007)
Functionality
or
transcriptional
noise?
Evidence
for
selection
within
long
noncoding
RNAs.
Genome
Res.
17,
556–565
65
Pandey,
R.R.
et
al.
(2008)
Kcnq1ot1
antisense
noncoding
RNA
mediates
lineage-specific
transcriptional
silencing
through
chromatin-level
regulation.
Mol.
Cell
32,
232–246
Review Trends
in
Genetics
xxx
xxxx,
Vol.
xxx,
No.
x
TIGS-1128;
No.
of
Pages
8
8