YTHDC1 binding, which could recruit nuclear
decay machinery ( 22 ) and histone modifiers
that add repressive histone marks ( 23 , 24 ),
resulting in reduced LINE1 RNA abundance
and more closed chromatin (Fig. 3D).
FTO regulates gene expression mostly through
intragenic LINE1 RNA
We observed thatFtoand LINE1 RNA abun-
dance increased after EB differentiation induc-
tion, but LINE1 RNA increase was diminished
uponFtoKO (fig. S12A). Targeting dCas13b-
wtFTO to LINE1 RNA inFto−/−mESCs partial-
ly rescued aberrantly expressed differentiation
markers associated with induced EB differentia-
tionandimpairedself-renewal(fig.S12,BtoG).
Therefore, we investigated the mechanism
underlying the impact ofFto KO on mESC dif-
ferentiation. We found that genes contain-
ing LINE1 RNA were down-regulated uponFto
KO compared with genes beyond 10 kb from
LINE1 RNA (Fig. 4A and fig. S13A) ( 26 ). Theexpression of LINE1-containing genes and
intragenic LINE1 RNA tended to decrease
withFtodepletion, accompanied by increased
intragenic LINE RNA m^6 A level (Fig. 4B and
fig. S13B). LINE1-containing genes also dis-
played greater transcription rate reductions
compared with other genes uponFto KO (fig.
S13, C to E).
Both intragenic and intergenic LINE1 RNAs
showed suppressed expression uponFtoKO,
but intragenic RNAs were generally moreWeiet al., Science 376 , 968–973 (2022) 27 May 2022 3of6
Fig. 2. FTO regulates LINE1 RNA level through
m^6 A demethylation.(A)m^6 A peak profiles on
m^6 A-marked LINE1 RNA in WT andFto−/−mESCs.
(B) Volcano plot showing differentially expressed
subfamilies of m^6 A-marked repeat RNAs upon
FtoKO. (C) Cumulative distribution and boxplots
(inset) showing nuclear LINE1 RNA lifetime in
WT andFto−/−mESCs.Pvalue was determined
using Wilcoxon signed-rank test. (D) Relative
enrichment of LINE1 RNA by YTHDC1 from cross-
linking and immunoprecipitation quantitative
polymerase chain reaction (qPCR) in WT andFto−/−
mESCs.Pvalues were determined using unpaired
two-tailedt tests. Error bars and means ± SD
are shown forn = 3 experiments. (E) Cumulative
distribution and boxplots (inset) showing LINE1 RNA
transcription rate in WT andFto−/−mESCs.Pvalue
was determined using Wilcoxon signed-rank test.
(F) Cumulative distribution and boxplots (inset)
showing the difference in transcription rate changes
between m^6 A-marked and -unmarked LINE1 RNA
uponFtoKO.Pvalue was determined using
Wilcoxon rank sum test.
E
WT0.000.250.500.751.00-0.5 0.0 0. 5 1.0 1.5
Transcription rateno
it
ca
rf
ev
it
al
u
m
uC0 .51.00.0FtoKO< 2e-16F
no
it
ca
rf
ev
it
al
u
m
u
C0.50 .30.000.250.500.751.00 non-m (^6) A
- 0 -0.50.0
Δtranscription Fto KO/WT
- 0 -0.50.0
-0.6-0.30 .0m^6 A< 2e-16C
WT0.000.250.500.751.000 5 10 15 20 25
Nuclear half lifetimen
oit
ca
rf
ev
it
al
u
mu
C1050< 2e-16FtoKO0 2000400060008000
Location along LINE1 RNAm
fo
yti
sn
e
D6(
A0(^1) ×
(^4) - )
0
1
3
2
WT
FtoKO
A
L1MdTf_I & II
L1MdA_III & V
L1MdF_II
0 L1MdGf_II IAPEz−int
5
10
15
-2 -1 0 12
ERVK
LINE1
go
L-
p (^01)
eu
la
v
Log 2 FC exp Fto KO/WT
B
D
Enrichment by YTHDC1
LINE1 RNAL1Md_Tf
L1Md_A
WT
Fto-/-
0
50
100
150
0.004
0.0007
0.002
Fig. 3. FTO affects chromatin state through
LINE1 RNA m^6 A demethylation.(A) Left: DNase
I–terminal deoxynucleotidyl transferase–mediated
deoxyuridine triphosphate nick end labeling (TUNEL)
assay showing more closed chromatin upon
FtoKO. Scale bars, 20mm. The nucleus was
counterstained with 4′,6-diamidino-2-phenylindole
(DAPI). Representative images were selected from
three independent experiments. Right: relative
TUNEL intensity in WT andFto−/−mESCs.Pvalue
was determined using an unpaired two-tailed
t test. TUNEL intensity was quantified by ImageJ.
(B) TUNEL signals of WT (gray) andFto−/−mESCs
rescued by dCas13b-wtFTO with guide RNA
(gRNA) targeting LINE1 RNA (red) or control gRNA
(blue). (C) Profiles of H3K4me3 and H3K9me3 levels
on LINE1 RNA loci from 3.0 kb upstream of the
transcription start site (TSS) to 3.0 kb downstream
of the transcription end site (TES) in WT and
Fto−/−mESCs. (D) Schematic model showing how
FtoKO affects LINE1 RNA abundance and local
chromatin state.
C
(^1) E
NI
L
n
o)
M
K
P
R(
eg
ar
ev
o
C
4
K 3
Hm3
e
9
K 3
H
m
(^3) e
0
2
4
6
8
10
0
2
4
6
8
10
Non-m^6 A m^6 A-marked
Non-m^6 A
0
4
8
12
0
4
8
12 m
(^6) A-marked
-3.0 TSS TES 3.0-3. 0 TSS TES 3.0
WT
Fto KO
A DAPI B
T
W
O
K
TUNEL Merge
0.0
0.5
1.0
1.5 <0.0001
Nrom
ez
il
adTE
N
U
is
Lga
nl
WTFto
-/- -0^103103104105
TUNEL_561nm
80
60
40
20
0
100
dCas13b-wtFTO rescue
WT mESCs
Fto-/- mESCs
Control
gRNA
LINE1
gRNA
Nrom
ez
il
adtome
do
SETDB1
NEXT
complex
LIN
E1 RN
A
D
Histone
Pol II
FTO (depleted)
YTHDC 1
H3K4me3/H3K27ac
H3K9me3
m^6 A LINE1
RN
A
RESEARCH | RESEARCH ARTICLE