Science - USA (2022-05-27)

lower 2C gene activation but greater down-
regulation of ESC-high genes (fig. S5, A to E).
Fto KO also caused several phenotypic changes
similar to those which occurred after LINE1
ASO treatment in mESCs, including cell cycle
dysregulation, self-renewal impairment, and
induction of capacity to form embryoid bodies
(EBs) (fig. S5, F to L). Moreover,Fto−/−mESCs
exhibited a reduced ability to integrate to
chimeric mice compared with WT mESCs (fig.
S5, M and N).
Evolutionarily young LINE1s display higher
RNA m^6 A levels and greater m^6 A increases
uponFtoKO (fig. S6, A and B). Longer and
less divergent LINE1 RNA also tends to ex-
hibit a higher m^6 A level (fig. S6, C and D). We
observed elevated m^6 Adensitynearthe5′end
withFtodepletion (Fig. 2A and fig. S6E),
which also responded toMettl3KO (fig. S6F).
Our analysis suggested that several young
LINE1 subfamilies were down-regulated and
hypermethylated uponFto KO (Fig. 2B and fig.
S6,GtoJ).WeusedaCRISPRdCas13bsystem
fused with WT FTO or a catalytically inactive
mutant (fig. S6K) ( 22 ),andobservedthatde-
livery of dCas13b-wtFTO by guide RNA target-
ing LINE1 RNA reversed its m^6 A level and
expression changes (fig. S6, L to N).
Consistent with the reported YTHDC1-
mediated destabilization of m^6 A-marked
carRNAs ( 22 ), we detected accelerated decay
of LINE1 RNA uponFtoKO (Fig. 2C), ac-
companiedbyelevatedbindingofLINE1RNA
by YTHDC1 (Fig. 2D). Both changes were re-
covered by targeting LINE1 RNA with dCas13b-
wtFTO inFto−/−mESCs (fig. S7, A and B).
LINE1 transcription was markedly reduced
withFto depletion (Fig. 2E and fig. S7, C to E).
Moreover,FtoKO led to greater decreases in
the nascent synthesis of m^6 A-marked com-
pared with unmarked LINE1 RNAs (Fig. 2F),
but not ERVK or Alu transcripts (fig. S7F). We
also observed reduced DNA association of
LINE1 RNA and decreased R-loop formation
around LINE1 loci withFto depletion (fig. S7, G
and H). These effects caused byFtoKO could
all be reversed by targeting dCas13b-wtFTO to
LINE1 RNA (fig. S7, I to K). Therefore, FTO
appears to mediate m^6 A demethylation of a
subset of LINE1 RNAs, maintaining their
levels on chromatin.

FtoKO leads to closed chromatin in mESCs

LINE1 RNA and m^6 A on carRNAs have been
shown to regulate chromatin state and tran-
scription ( 22 – 27 ). Indeed, we observed decreased
nascent RNA synthesis (fig. S8, A and B) ac-
companied by more closed chromatin (Fig. 3A)
uponFtoKO. Similar effects were observed
when treating WT mESCs with an FTO in-
hibitor (fig. S8, C and D). Additionally, LINE1
ASO treatment in WT mESCs led to more
closed chromatin (fig. S8E), whereas deliver-
ing dCas13b-wtFTO to LINE1 RNA largely re-

scued chromatin closure observed inFto−/−

mESCs (Fig. 3B and fig. S8F). We next moni-

tored LINE1 RNA m^6 A demethylation and

chromatin accessibility in a time-course of

FTO inhibition in WT mESCs (fig. S8, G to J).

Assay for transposase-accessible chromatin se-

quencing (ATAC-seq) results also validated re-

duced chromatin accessibility uponFto KO (fig.

S9, A to C), with gained-closed regions enriching

gene ontology (GO) terms relevant to develop-

ment (fig. S9D). Slightly decreased H3K4me3 and

H3K27ac and increased H3K9me3 and H3K27me3

were observed withFto depletion (fig. S9E).

YY1andEP300canberecruitedbycaRNA

to promote transcription ( 22 , 29 , 30 ). We found

notable enrichment at gained-closed regions

caused byFtoKO for H3K4me1, H3K4me3,

and H3K27ac, as well as YY1, EP300, and Pol II

binding, but not repressive histone marks (fig.

S10A). Consistently, chromatin immunopreci-

pitation sequencing experiments confirmed

reduced chromatin accessibility of these re-

gions uponFto depletion (fig. S10B).Fto KO–

induced gained-closed regions were also

affected byMettl3KO (fig. S10C) ( 22 ). In turn,

chromatin closure uponFtoKO could reduce

access of METTL3 to certain genomic regions,

potentially explaining m^6 A hypomethylation

within them.

We observed distinct profiles of H3K4me3

and H3K9me3 between loci with m^6 A-marked

and -unmarked LINE1 RNA (Fig. 3C). At

m^6 A-marked LINE1 RNA loci, we observed

decreased H3K4me3 and H3K27ac levels and

Pol II binding uponFtoKO, accompanied by

increased H3K9me3 levels (Fig. 3C and fig.

S11, A to C). Similar patterns were observed

for FTO-targeted LINE1 subfamilies but not

L1M2b,anunmarkedLINE1subfamily,or

IAPEz-int, an ERVK subfamily regulated by

METTL3 ( 23 – 25 )(fig.S11D).TargetingLINE1

RNA with dCas13b-wtFTO reversed the dys-

regulated histone marks (fig. S11E). Moreover,

FtoKO led to decreased chromatin associa-

tion and LINE1 RNA binding of YY1 and EP300

(fig. S11, F to H).

Altogether, our data reveal an interplay be-

tween LINE1 RNA m^6 A demethylation by FTO

and chromatin state. AfterFtodepletion, in-

creased m^6 A on LINE1 RNA could promote

Weiet al., Science 376 , 968–973 (2022) 27 May 2022 2of6

D

E

-0.75

-0.50

-0.25

0.00

0.25

-2 -1 0 1 2

r = -0.9153

p = 1.8e-40

Log 2 FC m^6 A Fto KO/WT

Log

FC exp 2

Fto

KO/WT

mESCs

NES = 1.75

p = 0.00019

NES = -1.62

p = 0.0011

0.00

0.25

0.50

Enrichment

score

2C Gene

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

high 0 5000 10000 15000low

ESC−high Gene

Ranked Log 2 FC exp Fto KO/WT

3.2e-112.5e-5< 2e-16

-2

0

2

4

eRNA

paRNA(+)paRNA(-)Repe

ats

Log

FC m 2

6 A

Fto

KO/WT

A

Log

FC exp 2

Fto

KO/WT

-2

-1

0

1

2

eRNA

paRNA(+)paRNA(-)Repeats

The rest

Hyper-m^6 A

0.0003 < 2e-16 < 2e-16 < 2e-16

B

C

Log 2 average m^6 A level (WT)

2.0 2.4 2.8

Log 2 FC m^6 A Fto KO/WT

Log 2 FC exp Fto KO/WT

m^6 A-marked

2e-16 4e-16 2e-16 2e-16 2e-16 2e-16

LINE1

Alu

ERVK

B4B2

ERVL−MaLR

Log 2 FC exp Fto KO/WT

-1^0 hyper vs the rest

The rest^1

Hyper-m^6 A

-0.2

-0.1

0.0

0.1

0.2

Log

FC 2

-Log

p 4

8

12

Number of hyper peaks

100 300 500

Log 2 average expression (WT)

2000 4000

Fig. 1. m^6 A on LINE1 RNA is a major substrate of FTO in mESCs.(A) Violin plots showing m^6 A-level

fold changes of hypermethylated m^6 A peaks on carRNAs uponFtoKO.Pvalues were determined using

Wilcoxon rank sum tests. (B) Boxplots showing expression fold changes of hypermethylated carRNAs versus

other m^6 A-marked carRNAs uponFtoKO.Pvalues were determined using Wilcoxon signed-rank tests.

(C) Summary of repeat RNAs on chromatin uponFtoKO. Top: number of hypermethylated peaks, average

m^6 A level, and expression. Middle: m^6 A-level and expression fold changes.Pvalues were determined

using Wilcoxon signed-rank tests. Bottom: expression fold changes of hypermethylated versus non-

hypermethylated repeat RNAs.Pvalues were determined using Wilcoxon rank sum tests. (D) Scatter plot

showing the negative correlation of fold changes between m^6 A level and expression of LINE1 RNA upon

FtoKO. LINE1 RNAs were categorized into 100 bins on the basis of their ranked m^6 A-level fold changes

uponFtoKO.r refers to Pearson’s correlation coefficient.P value was determined usingt distribution. (E)Geneset

enrichment analysis showing up-regulated 2C genes (top) and down-regulated ESC-high genes (bottom) from

mRNA-seq uponFtoKO.NES,normalizedenrichmentscore.

RESEARCH | RESEARCH ARTICLE