Article

FTO Plays an Oncogenic R

ole in Acute MyeloidLeukemia as a N6-Methyladenosine RNADemethylase

Highlights

d FTO is highly expressed in certain subtypes of AMLs such as

MLL-rearranged AML

d FTO promotes leukemic oncogene-mediated cell

transformation and leukemogenesis

d FTO targets a set of genes (e.g., ASB2 and RARA) in AML as

an m6A RNA demethylase

d The FTOxASB2/RARA axis mediates AML cell growth and

(ATRA-induced) differentiation

Li et al., 2017, Cancer Cell 31, 127–141January 9, 2017 ª 2017 Elsevier Inc.http://dx.doi.org/10.1016/j.ccell.2016.11.017

Authors

Zejuan Li, Hengyou Weng, Rui Su, ...,

Jie Jin, Chuan He, Jianjun Chen

Correspondencechuanhe@uchicago.edu (C.H.),chen3jj@ucmail.uc.edu (J.C.)

In Brief

Li et al. show that FTO, an

N6-methyladenosine (m6A) demethylase,

is highly expressed in subtypes of AML,

promotes leukemogenesis, and inhibits

all-trans-retinoic acid-induced leukemia

cell differentiation. FTO exerts its

oncogenic role by regulating mRNA

targets such as ASB2 and RARA by

reducing their m6A levels.

Accession Numbers

GSE34184

GSE30285

GSE76414

GSE84944

GSE85008

Cancer Cell

Article

FTO Plays an Oncogenic Role in AcuteMyeloid Leukemia as a N6-MethyladenosineRNA DemethylaseZejuan Li,1,7,10 Hengyou Weng,1,2,10 Rui Su,2,10 Xiaocheng Weng,3,8,10 Zhixiang Zuo,1,2,9,10 Chenying Li,2,4 Huilin Huang,2

Sigrid Nachtergaele,3 Lei Dong,2 Chao Hu,1,2,4 Xi Qin,2 Lichun Tang,5 Yungui Wang,1,2,4 Gia-Ming Hong,1 Hao Huang,1

Xiao Wang,3 Ping Chen,1 Sandeep Gurbuxani,6 Stephen Arnovitz,1 Yuanyuan Li,1 Shenglai Li,1 Jennifer Strong,2

Mary Beth Neilly,1 Richard A. Larson,1 Xi Jiang,1,2 Pumin Zhang,5 Jie Jin,4 Chuan He,3,* and Jianjun Chen1,2,11,*1Section of Hematology/Oncology, Department of Medicine, University of Chicago, Chicago, IL 60637, USA2Department of Cancer Biology, University of Cincinnati College of Medicine, Cincinnati, OH 45219, USA3Departments of Chemistry, Biochemistry and Molecular Biology, Institute for Biophysical Dynamics, Howard Hughes Medical Institute,

University of Chicago, Chicago, IL 60637, USA4Key Laboratory of Hematopoietic Malignancies, Department of Hematology, The First Affiliated Hospital of Zhejiang University, Hangzhou,

Zhejiang 310003, China5Department of Molecular Physiology and Biophysics, Baylor College of Medicine, Houston, TX 77030, USA6Department of Pathology7Department of Human Genetics

University of Chicago, Chicago, IL 60637, USA8College of Chemistry and Molecular Sciences, Key Laboratory of Biomedical Polymers of Ministry of Education, Wuhan University, Hubei,

Wuhan 430072, PR China9Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer

Medicine, Guangzhou 510060, China10Co-first author11Lead Contact

*Correspondence: chuanhe@uchicago.edu (C.H.), chen3jj@ucmail.uc.edu (J.C.)

http://dx.doi.org/10.1016/j.ccell.2016.11.017

SUMMARY

N6-Methyladenosine (m6A) represents the most prevalent internal modification in mammalian mRNAs.Despite its functional importance in various fundamental bioprocesses, the studies of m6A in cancer havebeen limited. Here we show that FTO, as anm6A demethylase, plays a critical oncogenic role in acutemyeloidleukemia (AML). FTO is highly expressed in AMLs with t(11q23)/MLL rearrangements, t(15;17)/PML-RARA,FLT3-ITD, and/or NPM1 mutations. FTO enhances leukemic oncogene-mediated cell transformation andleukemogenesis, and inhibits all-trans-retinoic acid (ATRA)-induced AML cell differentiation, through regu-lating expression of targets such as ASB2 and RARA by reducing m6A levels in these mRNA transcripts.Collectively, our study demonstrates the functional importance of the m6A methylation and the correspond-ing proteins in cancer, and provides profound insights into leukemogenesis and drug response.

INTRODUCTION

N6-Methyladenosine (m6A) is the most abundant internal modi-

fication in mRNA mainly occurring at the consensus motif of

Significance

The identification of FTO as the first N6-methyladenosine RNAdemonstrated in mammalian mRNA transcriptomes have spuin post-transcriptional regulation. However, little is known aHere we show that FTO functions as an oncogene that promleukemogenesis, and inhibits all-trans-retinoic acid (ATRA)-meerts its oncogenic role as an m6A demethylase by targeting a sstudy reveals a previously unrecognized mechanism of genetance of FTO and m6A modification in cancer.

G[G > A]m6AC[U > A > C] (Fu et al., 2014; Meyer and Jaffrey,

2014; Niu et al., 2013; Yue et al., 2015). Although m6A was first

discovered in the 1970s (Desrosiers et al., 1974; Perry and Kel-

ley, 1974), the lack of technologies to study RNA modifications

demethylase and the high prevalence of m6A methylationrred immense interest in the function of m6A modificationsbout the function of FTO or m6A modifications in cancer.otes leukemic oncogene-mediated cell transformation anddiated leukemia cell differentiation. Mechanistically, FTO ex-et of critical transcripts such as ASB2 and RARA. Thus, ourregulation in tumorigenesis and highlights functional impor-

Cancer Cell 31, 127–141, January 9, 2017 ª 2017 Elsevier Inc. 127

limited research on m6A and the field has not advanced for

several decades. The identification of the fat mass- and

obesity-associated protein (FTO) as the first RNA demethylase

(Jia et al., 2011) revived RNA methylation research, suggesting

that m6A is a reversible and dynamic RNA modification that

may impact biological regulation analogous to the well-studied

reversible DNA and histone modifications (Jia et al., 2013). In

2012, two groups reported transcriptome-wide approaches for

m6A RNA immunoprecipitation followed by next-generation

sequencing (termed as m6A-seq or MeRIP-seq) and detected

m6A peaks in more than 7,000 mRNA transcripts and hundreds

of long non-coding RNAs (lncRNAs) in both human and mouse

cells, with many of the m6A peaks conserved between humans

and mice (Dominissini et al., 2012; Meyer et al., 2012). Thus,

these studies suggest that m6A methylation in mRNAs is a prev-

alent modification that likely possesses functional importance.

Recent studies have shown that m6A modification in mRNAs

or non-coding RNAs plays critical roles in tissue development,

stem cell self-renewal and differentiation, control of heat shock

response, and circadian clock controlling, as well as in RNA

fate and functions such as mRNA stability, splicing, transport,

localization and translation, primary microRNA processing,

and RNA-protein interactions (Alarcon et al., 2015; Chen

et al., 2015; Dominissini et al., 2012; Geula et al., 2015; Liu

et al., 2015; Meyer et al., 2012, 2015; Wang et al., 2014a,

2014b, 2015; Zhao et al., 2014; Zheng et al., 2013; Zhou et al.,

2015). FTO and ALKBH5, the second RNA demethylase identi-

fied in 2013 (Zheng et al., 2013), both belong to the AlkB family

and catalyze m6A demethylation in a Fe(II)- and a-ketogluta-

rate-dependent manner, and are referred to as m6A ‘‘erasers’’

(Fu et al., 2014; Yue et al., 2015). METTL3 and METTL14 were

identified as m6A ‘‘writers’’ that form a heterodimer with the sup-

port of WTAP to catalyze m6Amethylation (Bokar et al., 1997; Liu

et al., 2014; Ping et al., 2014; Wang et al., 2014b). YTHDF1/2/3

were identified as m6A ‘‘readers’’ that preferentially bind to

RNA that contains m6A at the G[G > A] m6ACU consensus

sequence and lead to different biological consequence (Domi-

nissini et al., 2012; Wang et al., 2014a, 2015), such as inducing

RNA decay (Wang et al., 2014a) and promoting mRNA transla-

tion (Wang et al., 2015).

FTOwas known to be robustly associatedwith increased body

mass and obesity in humans (Dina et al., 2007; Frayling et al.,

2007; Scuteri et al., 2007). Animal model studies showed that

Fto deficiency protected from obesity and caused growth retar-

dation (Fischer et al., 2009; Gao et al., 2010), while overexpres-

sion of Fto led to increased food intake and obesity (Church

et al., 2010). In humans, loss-of-function mutations in FTO also

caused severe growth retardation and multiple malformations

that resulted in premature death (Boissel et al., 2009). As the first

identified RNA demethylase that regulates the demethylation of

target mRNAs, FTO has been reported to regulate dopaminergic

signaling in brain (Hess et al., 2013), and also regulate mRNA

splicing of adipogenetic regulatory factors and thus play a critical

role in adipogenesis (Ben-Haim et al., 2015; Zhao et al., 2014).

However, the impact of FTO, especially as an RNA demethy-

lase, in cancer development and progression has yet to be

investigated.

Acutemyeloid leukemia (AML) is one of themost common and

fatal forms of hematopoietic malignancies with distinct genetic

128 Cancer Cell 31, 127–141, January 9, 2017

(e.g., t(11q23)/MLL-rearranged, inv(16), t(8;21), and t(15;17))

and molecular (e.g., FLT3-ITD and NPM1 mutations) abnormal-

ities and variable response to treatment (Chen et al., 2010;

Dohner et al., 2015;Marcucci et al., 2005).With standard chemo-

therapies, only 35%–40% of younger (aged <60 years) and

5%–15% of older (aged R60 years) patients with AML survive

more than 5 years (Dohner et al., 2015). Thus, to develop effec-

tive targeted therapies to treat AML is an urgent and significant

unmet medical need, which relies on better understanding of

the molecular mechanisms underlying the pathogenesis and

drug response of AML.

In the present study, we sought to determine the biological

function of FTO in the pathogenesis and drug response of

AMLs and also investigate the underlying molecular mechanism

through identification of critical mRNA targets of FTO.

RESULTS

FTO Is Highly Expressed in Certain Subtypes of AMLIn analysis of our in-house microarray dataset of 100 human

AML with t(11q23)/MLL-rearranged, t(8;21), inv(16), or t(15;17)

and nine normal control samples, we found that FTO was ex-

pressed at a significantly higher level in MLL-rearranged AML

than in both normal controls (p = 0.04) and non-MLL-rearranged

AML samples (p = 0.002); among non-MLL-rearranged AML,

FTO is expressed at a significantly higher level in t(15;17)

AML, also called acute promyelocytic leukemia (APL), than in

t(8;21) and inv(16) AMLs (Figure 1A). Our qPCR assay also

showed that FTO is expressed at a significantly higher level in

CD34+ bone marrow (BM) cells isolated from primary MLL-rear-

ranged AML patients relative to normal CD34+ BM cells isolated

from healthy donors (Figure 1B). In contrast, ALKBH5, another

m6A demethylase (Zheng et al., 2013), was not significantly up-

regulated in MLL-rearranged AML relative to normal controls

(Figures S1A and S1B), although a very recent study suggests

involvement of ALKBH5 in breast cancer cell proliferation

in vitro (Zhang et al., 2016).

Consistently, in analysis of two large-cohort AML datasets,

including GSE37642 set (n = 562) and GSE14468 set (n = 518),

we observed that FTO is expressed at a significantly higher level

in t(11q23) and t(15;17) AMLs compared with other subtypes of

cytogenetically abnormal AMLs including inv(16) and t(8;21)

AMLs (Figures S1C and S1D). The expression level of FTO

in normal karyotype (NK) AMLs is comparable with that in

t(11q23) and t(15;17) AMLs (Figures S1C and S1D). Within NK-

AMLs, FTO is expressed at a significantly higher level in AML

with FLT3-ITD and/or NPM1 mutations (i.e., NPM1c) (especially

that with both) comparedwith thosewithout (Figures 1C and 1D).

We also observed that FTO was aberrantly upregulated at the

protein level in human primary AML specimens of the above

subtypes (Figure 1E).

Expression of FTO Can Be Upregulated by the RelevantLeukemic OncogenesWe then conducted both western blot and qPCR assays of

mouse BM progenitor cell samples transformed by MLL-AF9

(the most common form of MLL fusions; Chen et al., 2010),

PML-RARA (fusion gene of t(15;17)), and the FLT3-ITD/NPM1

mutant, showing that FTO can be upregulated by the oncogenes

Figure 1. FTO Is Highly Expressed in Certain

AML Subtypes

(A) Comparison of FTO expression between hu-

man primary AML cases with MLL rearrange-

ments/t(11q23) (MLL) and those without MLL re-

arrangements (non-MLL), or AML cases with

inv(16), t(8;21) or t(15;17), or normal controls (NC).

The expression values were detected by Affyme-

trix exon arrays (Huang et al., 2013). The expres-

sion values were log2-transformed and mean

centered.

(B) qPCR analysis of FTO expression in human

CD34+ AML BM cells isolated from ten primary

MLL-rearranged AML patients (CD34_MLL) and

normal CD34+ BM cells isolated from six healthy

donors (CD34_NC). The average expression level

of FTO in the CD34_NC samples was set as 1.

(C and D) The expression patterns of FTO across

cytogenetically normal (or normal karyotype [NK])

AMLs within GSE37642 set (n = 562) and

GSE14468 set (n = 518) AML datasets.

(E) Western blot assay of FTO expression in human

primary AML specimens with different fusion

genes or mutant oncogenes and a healthy donor

sample (NC). Mononuclear cells isolated from pri-

mary AML patients and the healthy donor were

used for the assay.

(F) The protein level of Fto (upper panel) and m6A

level (lower panel; by QQQ-MS) in the represen-

tative samples of the control group (from a primary

BMT recipient) or MA9 leukemic group (one each

from primary and secondary BMT recipients).

(G) ChIP-qPCR assays of the enrichment of MLL-N

(i.e., MLL N-terminal, representing both wild-type

MLL and MLL-fusion proteins), MLL-C (i.e., MLL

C-terminal, representing wild-type MLL only), and

H3K79me2/3 at the promoter region of FTO (CpG

site) and a distal upstream region (control site) in

MONOMAC-6, KOCL-48, and K562 cells. IgG was

used as a negative control.

(H) qPCR analysis of Fto and Alkbh5 expression in

mouse MLL-ENL-ERtm cells after withdrawal of

4-hydroxytamoxifen (4-OHT).

*p < 0.05, **p < 0.01; t test. See also Figure S1.

at both protein and RNA levels (Figure S1E). Moreover, Fto, but

not Alkbh5, was also significantly upregulated by MLL-fusion

in mouse leukemic BM cells collected from primary BM trans-

plantation recipient mice, relative to normal control cells;

notably, the upregulation of Fto was further enhanced after

transplantation of the primary MLL-AF9 leukemic BM cells into

secondary recipients (Figures 1F, S1F, and S1G). Fto overex-

pression is accompanied with mRNA m6A level decrease in the

samples (Figure 1F).

Can

We also used MLL-rearranged leuke-

mia as a model to further investigate

whether FTO is directly upregulated by

the oncogenic proteins. Through chro-

matin immunoprecipitation (ChIP) assays,

we found that MLL (see MLL-C binding)

and particularly MLL-fusion proteins (see

the portion of MLL-N binding exceeding

that of MLL-C) are enriched at the CpG

area (CpG sites), but not the distal upstream site (control site),

of the FTO locus in human MONOMAC-6 (a MLL-AF9 AML

line) and KOCL-48 (a MLL-AF4 AML line) cells; the locus

also shows a significant enrichment of H3K79 methylation

(H3K79me2/3), a mark of active transcription (Bernt et al.,

2011; Okada et al., 2005) (Figure 1G). No such significant enrich-

ments were observed in a control cell line, K562 (a human eryth-

roleukemic cell line) (Figure 1G). Thus, our data suggest that FTO

is likely a direct target of MLL fusions. In addition, we observed

cer Cell 31, 127–141, January 9, 2017 129

that Fto (but not Alkbh5) expression was significantly (p < 0.01)

downregulated in MLL-ENL-ERtm mouse myeloid cells car-

rying tamoxifen-inducible MLL-ENL (Zeisig et al., 2004) when

expression of MLL-ENL was depleted after withdrawal of 4-hy-

droxytamoxifen (Figure 1H), indicating that Fto expression in

MLL-ENL-ERtm cells relies on the presence of MLL-ENL.

FTO Promotes Cell Proliferation/Transformation andSuppresses Apoptosis In VitroBoth gain- and loss-of-function studies were performed to

investigate the pathological role of FTO in AML. As shown in Fig-

ures 2A–2F, lentivirally transduced wild-type FTO, but not the

FTO mutant (carrying two point mutations, H231A and D233A,

which disrupt the enzymatic activity of FTO; Jia et al., 2011;

Lin et al., 2014), promoted cell growth/proliferation and viability,

while decreasing apoptosis and the global mRNA m6A level, in

two MLL-rearranged AML cell lines (MONOMAC-6 and MV4-

11); the opposite is true when endogenous expression of FTO

was knocked down by FTO small hairpin RNAs (shRNAs).

Similar phenomena were observed when FTO was overex-

pressed by a retroviral construct or knocked down by small

interfering RNAs (Figures S2A–S2D). In contrast, while forced

expression of FTO exhibited moderate (though significant) ef-

fects, FTO knockdown exhibited no significant effects on cell

growth/proliferation, growth, or apoptosis in K562 AML cells (a

control line) (Figures S2E–S2I); thus, FTO is less functionally

essential in K562 cells than in MLL-rearranged AML cells, likely

due to its much lower endogenous expression in K562 cells

(Figure S2J). We next performed mouse BM colony-forming/re-

plating assays to investigate the function of FTO in MLL-AF9-

mediated cell transformation. As expected, co-overexpression

of FTO, but not FTO-Mut, significantly increased colony

numbers after replating, compared with the MLL-AF9 alone

group; knockdown expression of Fto led to the opposite (Fig-

ures 2G and S2K).

Similar patterns were observed in the PML-RARA/t(15;17) and

FTL3-ITD/NPM1 leukemic cell models when both gain- and loss-

of-function studies of FTO/Fto were conducted (see Figures 3

and S3), demonstrating the oncogenic role of FTO in these

AML subtypes.

Fto Significantly Enhances Leukemogenesis In VivoTo investigate the role of Fto in vivo, we conducted mouse BM

transplantation (BMT) assays. We found that forced expres-

sion of Fto significantly (p < 0.05; log-rank test) accelerated

MLL-AF9-induced leukemogenesis in recipient mice (Figures

4A and S4), and this pattern is repeatable (Figure 4B). Notably,

forced expression of Fto significantly increased c-Kit+ imma-

ture blast cell proportion (Figure 4C) and decreased global

m6A level in leukemic BM cells (Figure 4D). Conversely, the

knockdown of Fto by shFto-1 or shFto-2 significantly delayed

MLL-AF9-mediated leukemogenesis in mice (Figure 4E). Simi-

larly, MLL-AF9-tranduced BM progenitor cells from Fto+/�

mice (Gao et al., 2010) caused leukemia in recipient mice

significantly slower than did MLL-AF9-tranduced BM progeni-

tor cells from wild-type (Fto+/+) mice (Figure 4F), and were

associated with a decrease in c-Kit+ cell proportion (Figure 4G)

and an increase in m6A abundance in leukemic BM cells

(Figure 4H).

130 Cancer Cell 31, 127–141, January 9, 2017

Transcriptome-wide m6A-Seq and RNA-Seq Assays toIdentify Potential Targets of FTOTo identify potential mRNA targets of FTO the m6A levels

of which are reduced by FTO in AML cells, we retrovirally

transduced MSCV-PIG-FTO (i.e., FTO) or MSCV-PIG (i.e., Ctrl/

Control) into human MONOMAC-6 AML cells and then selected

individual stable clones under selection of puromycin (0.5 mg/ml).

Two FTO-overexpressing (namely FTO_1 and _2) and two

control (namely Ctrl_1 and _2) stable cell lines were selected

for transcriptome-wide m6A-sequencing (m6A-seq) and RNA-

sequencing (RNA-seq) assays. The FTO-overexpressing stable

lines exhibit a 4- to 5-fold increase in FTOprotein level (Figure 5A)

and a noticeable decrease in m6A level (Figures 5B and 5C),

compared with the control lines. As shown in Table S1, 17.7 to

83.1 million reads were generated from each m6A-seq or RNA-

seq (also serving as the input control of the corresponding

m6A-seq) library. A total of 13,278 m6A peaks were identified

by both MACS (Zhang et al., 2008) and exomePeak (Meng

et al., 2013) methods from at least two of the four m6A-seq li-

braries (Figure S5A). Consistent with previous studies (Chen

et al., 2015; Dominissini et al., 2012; Meyer et al., 2012), the

most common m6A motif GGAC is significantly enriched in our

13,278 m6A peaks (Figure S5B); the m6A peaks are mostly

located in exons (>87%; Figure S5C) and are especially enriched

in the vicinity of the stop codon (Figures S5D and S5E).

Potential Target Genes of FTO Tend to Be NegativelyRegulated by FTOWe next compared the abundance (normalized to input) of

the 13,278 m6A peaks between FTO-overexpressing cells (i.e.,

FTO_1/2) and the control cells (i.e., Ctrl_1/2). A total of 2,785 and

3,180 m6A peaks showed a significant decrease and increase

(p < 0.005; fold-change R1.2), respectively, in abundance, in

FTO_1/2 cells relative to Ctrl_1/2 cells, and they were thus termed

hypo-andhyper-methylatedm6Apeaks, respectively (FigureS5F).

Through analysis of the RNA-seq data, we identified 322 hypo-

methylatedm6A peaks themRNA transcripts of whichwere signif-

icantly (p < 0.005; fold-changeR1.2) downregulated (275; Hypo-

down) or upregulated (47; Hypo-up) in FTO_1/2 cells relative to

Ctrl_1/2 cells (Figure 5D). Notably, 85% of the 322 hypo-methyl-

ated m6A peaks are associated with downregulated mRNA tran-

scripts in FTO-overexpressing cells, significantly more frequent

(p < 0.0001; c2 test) than the rate (54%) in the 308 hyper-methyl-

ated m6A peaks (Figure 5D). Thus, mRNA transcripts carrying

hypo-methylated m6A peaks, which are likely potential targets of

FTO since FTO is anm6A demethylase, tend to be downregulated

in FTO-overexpressing AML cells.

Through searching the Molecular Signature Database

(MSigDB) of GSEA (Subramanian et al., 2005), we found that

the Hypo-up transcripts were significantly enriched with target

genes of SOX2, NANOG, and LEF1, key transcription factors

for embryonic stem cell pluripotency or WNT signaling activation

(Figure S5G). Hypo-down transcripts were significantly enriched

with genes involving the interferon signaling and immune sys-

tem; interestingly, the most enriched genes are a set of potential

direct target genes of PML-RARA, a fusion protein resulting from

t(15;17) in APL (Figure S5G). The enriched gene sets in Hyper-up

or Hyper-down transcripts are shown in Figure S5H. We next

investigated the correlation between FTO and the above genes

Figure 2. Biological Effects of Forced Expression or Knockdown of FTO/Fto Expression in MLL-Rearranged AML

(A) Western blotting confirmation of forced expression and knockdown of FTO by lentiviral constructs in MONOMAC-6 and MV4-11 cells. FTO, pMIRNA1-FTO;

FTO-Mut, pMIRNA1-FTO-Mut; Ctrl, control vector (empty pMIRNA1); shFTO-1, pLKO.1-shFTO-1; shNS-1, pLKO.1-shNS-1; shFTO-2, pGFP-C-shLenti-

shFTO-2; shNS-2, pGFP-C-shLenti-shNS-2.

(B–E) Effects of forced expression or knockdown of FTO expression on cell growth/proliferation (B), viability (C) and (D), and apoptosis (E) in MONOMAC-6 and

MV4-11 cells. h, hour.

(F) m6A dot blot assays of MONOMAC-6 and MV4-11 cells with or without forced expression or knockdown of FTO. MB, methylene blue staining (as loading

control).

(G) Effects of forced expression of FTO or FTO-Mut (with the above pMIRNA1 constructs) or knockdown of Fto expression (with pGFP-V-RS-Fto shRNAs) on

colony-forming/replating capacity of mouse normal BM progenitor cells transduced by MSCVneo-MLL-AF9 (MA9). Colony cells were replated every 7 days.

*p < 0.05, **p < 0.01, ***p < 0.001; NS, not significant (p > 0.05); t test. See also Figure S2.

Cancer Cell 31, 127–141, January 9, 2017 131

Figure 3. Effects of Forced Expression or Knockdown of FTO/Fto in PML-RARA and FLT3-ITD/NPM1-Mutant AML

(A) Confirmation of overexpression and knockdown of FTO by western blotting in PL-21/t(15;17) AML cells.

(B) Effects of forced expression or knockdown of FTO expression on cell growth/proliferation in PL-21 (upper panels) and NB4/t(15;17) (lower panels) AML cells.

(C–E) Effects of forced expression or knockdown of FTO expression on cell viability in PL-21 (C) and NB4 (E), and apoptosis in PL-21 (D) cells.

(F and G) Effects of forced expression of FTO or FTO-Mut, and knockdown of Fto expression on colony-forming/replating capacity of mouse BM progenitor cells

carrying PML-RARA (F) or FLT3-ITD/NPM1-mutant (G). Colony cells were replated every 7 days.

*p < 0.05, **p < 0.01, ***p < 0.001. See also Figure S3.

in expression across four large cohorts of AML patient datasets

(see Table S2). We found that eight Hypo-down (i.e., ASB2,

KCNG1, PPARD, RAB17, RARA, SLC11A1, SLCO4A1, and

TBC1D9) and three Hypo-up (C21orf59, MZF1, and TXLNA)

genes exhibited a significantly negative and positive correlation,

respectively, with FTO in expression across all four datasets (Ta-

ble S2). We next conducted gene-specific m6A qPCR assays for

five such m6A-Hypo genes (ASB2, PPARD, RARA, SLC11A1,

and TXLNA) and confirmed the m6A-level decrease in four out

of the five genes (80%; Figure 5E), demonstrating the reliability

of our transcriptome-wide m6A-seq data.

We then conducted m6A-seq and RNA-seq of MA9/FLT3-

ITD leukemic cells (i.e., MLL-AF9 plus FLT3-ITD-transformed

human CD34+ cord blood cells; Wunderlich et al., 2013) with

or without FTO-shRNA knockdown. Remarkably, more than

132 Cancer Cell 31, 127–141, January 9, 2017

90% m6A-Hypo transcripts identified from FTO-overexpressing

MONOMAC-6 cells turned into m6A-Hyper in FTO-knockdown

MA9/FLT3-ITD cells (Figure 5F), with approximately 85% of the

Hypo-down and 47% of Hypo-up transcripts became Hyper-

up and Hyper-down, respectively (see Figures 5F and 5G), which

might be genuine targets of FTO. Overall, the vast majority of

potential targets of FTO are likely negatively regulated by FTO.

Eight out of the 11 m6A-Hypo genes listed in Table S2 showed

expected patterns in the m6A and RNA level changes in

FTO-knockdown cells (Table S3).

ASB2 and RARA Are Functionally Important TargetGenes of FTO in AMLInterestingly, among the genes listed in Tables S2 and S3, ASB2

and RARA have been reported to be upregulated during normal

Figure 4. The Role of Fto in Leukemogenesis Mediated by MLL-AF9

(A and B) Effect of forced expression of Fto on MLL-AF9 (MA9)-induced leukemogenesis. Kaplan-Meier curves are shown for two cohorts of transplanted mice

including MSCVneo-MA9+MSCV-PIG (MA9) and MSCVneo-MA9+MSCV-PIG-Fto (MA9+Fto) from two independent BMT assays. Six mice per group in (A) and

five mice per group in (B). The p values were calculated by log-rank test.

(C and D) Flow cytometry analysis of Mac-1+ and c-Kit+ cell populations (C) andm6A dot blot analysis (D) in BM cells of the representative leukemic mice from the

BMT assay shown in (A).

(E) Effect of depleted expression of Fto by shRNAs on MA9-induced leukemogenesis. Kaplan-Meier curves are shown for four cohorts of transplanted mice

including MSCVneo + MSCV-PIG (Control), MSCVneo-MA9+MSCV-PIG (MA9), MSCVneo-MA9+pGFP-V-RS-shFto-1 (MA9+shFto-1), and MSCVneo-MA9+

pGFP-V-RS-shFto-2 (MA9+shFto-2). Five mice were studied per group.

(F) Effect of depleted expression of Fto by genetic knockout (heterozygous) on MA9-induced leukemogenesis. Kaplan-Meier curves are shown for two cohorts of

recipient mice transplanted with MSCVneo-MA9-transduced wild-type donor cells (MA9_Fto+/+) and MSCVneo-MLL-AF9-transduced Fto+/� donor cells

(MA9_Fto+/�). Five mice were studied per group.

(G and H) Flow cytometry analysis of Mac-1+ and c-Kit+ cell populations (G) andm6A dot blot analysis (H) in BM cells of the representative leukemic mice from the

BMT assay shown in (F).

*p < 0.05, **p < 0.01; t test. See also Figure S4.

hematopoiesis and in all-trans-retinoic acid (ATRA)-induced

differentiation of leukemia cells and function as key regulators

during the processes (Glasow et al., 2005; Guibal et al., 2002;

Kohroki et al., 2001; Sakamoto et al., 2014; Wang et al., 2012;

Zhu et al., 2001). Notably, ASB2 can also degrade MLL during

hematopoietic differentiation via ubiquitination (Wang et al.,

2012). Consistent with their downregulation in FTO-overex-

pressing AML cells, ASB2 and RARA exhibit a significant inverse

correlation with FTO in expression across all four independent

AML cohorts (see Table S2 and Figure 5H). Our m6A-seq

data indicate that FTO targets the 30 UTR of ASB2 and both 30

UTR and 50 UTR of RARA transcripts; FTO overexpression

and knockdown causes a significant decrease and increase,

respectively, in the m6A level of the UTR(s) (Figures 5I and 5J).

We then focused on these two FTO potential targets for further

studies.

As expected, we showed that in both MONOMAC-6 and NB4/

t(15; 17) AML cells, forced expression of wild-type FTO, but

not mutant FTO, substantially reduced expression of ASB2

and RARA, while increasing expression of MLL, a negative

downstream target of ASB2 (Wang et al., 2012) (Figures 6A

and S6A). The opposite phenomena were observed when we

Cancer Cell 31, 127–141, January 9, 2017 133

Figure 5. Identification of Potential Targets

of FTO in AML via Transcriptome-wide m6A-

Seq and RNA-Seq Assays

(A) Western blot assay of FTO expression in human

MONOMAC-6 AML cell lines with or without forced

expression of FTO, including FTO_1/2 and Ctrl_1/2

cell lines. The upper panel shows the image and the

lower panel shows the relative quantitative infor-

mation of FTO expression at the protein level in

different AML cell lines.

(B and C) The m6A dot blot assay (B) and relative

quantitative information (C) of global m6A abun-

dance in transcriptomes of the above four cell lines.

(D) Distribution of genes with a significant change

in both m6A level and overall transcript (i.e.,

expression) level in FTO-overexpressing (FTO_1

and FTO_2) compared to control (Ctrl_1 and Ctrl_2)

MONOMAC-6 AML cells.

(E) Gene-specific m6A qPCR validation of m6A level

changes of five representative m6A-Hypo genes in

MONOMAC-6 cells.

(F) Distribution of the FTO-induced hypo-genes,

including the Hypo-down and Hypo-up groups

shown in (D), in FTO-knockdown MA9/FLT3-ITD

AML cells relative to the control AML cells.

(G) Up to 84.5% Hypo-down genes in FTO-over-

expressing (FTO OE) MONOMAC-6 AML cells

display a Hyper-up pattern in FTO-knockdown (FTO

KD) MA9/FLT3-ITD AML cells; 46.7% Hypo-up

genes in the FTO OE cells display Hyper-down

pattern in the FTO KD cells.

(H) Correlation of expression between FTO and

ASB2 or RARA across the 109 (100 AML and 9

normal control) samples shown in Figure 1A.

(I and J) The m6A abundances in ASB2 and RARA

mRNA transcripts in FTO-overexpressing (FTO_1

and FTO_2) and control (Ctrl_1 and Ctrl_2)

MONOMAC-6 AML cells (I), and in FTO-knockdown

(shFTO-1) and control (shNS) MA9/FLT3-ITD AML

cells (J), as detected by m6A-seq. The m6A peaks

shown in the green rectangles are those that have a

significant reduced abundance (p < 0.005; fold-

change >1.2) in FTO_1/2 than in Ctrl_1/2 cells.

*p < 0.05, **p < 0.01; t test. See also Figure S5

and Tables S1–S3.

knocked down endogenous expression of FTO (Figures 6B, 6C,

and S6B). Forced expression or depletion of FTO affected

expression of RARA, but not that of other RAR genes such as

RARG (Figure S6C). In addition, we observed similar changes

in RARA and ASB2 expression after overexpression or knock-

down of FTO in nuclear and cytoplasm (Figure S6D).

We then cloned ASB2- and RARA-coding regions or shRNAs

into lentiviral vectors and investigated their functions in AML

cells. As expected, forced expression of either ASB2 or RARA

largely recapitulated the phenotypes caused by FTO knockdown

in both MONOMAC-6 and NB4 cells (Figures 6D and 6E versus

Figures 2B and 3B). Moreover, the effects of FTO overexpression

can be largely rescued by forced expression of RARA or ASB2

(Figures S6E–S6J). Conversely, knockdown of ASB2 or RARA

significantly enhanced AML cell growth and viability, which

mimics the effect of FTO overexpression, and was sufficient

to rescue the inhibitory effect of FTO knockdown on AML

cell growth and viability (Figures 6F–6I). Collectively, our data

134 Cancer Cell 31, 127–141, January 9, 2017

demonstrate that ASB2 and RARA are functionally important

targets of FTO.

FTO-Mediated Regulation of ASB2 and RARA Dependson Its m6A Demethylase Activity and the m6AModifications in the Target mRNA TranscriptsTo elucidate the molecular mechanism underlying FTO-medi-

ated regulation of expression of its targets such as ASB2 and

RARA, we first validated effects of FTO expression changes on

the m6A levels in target mRNA transcripts. Using gene-specific

m6A qPCR assays, we demonstrated that forced expression

and knockdown of FTO reduced and increased, respectively,

the m6A levels on ASB2 and RARAmRNA transcripts (Figure 7A

and data not shown). More importantly, to assess the require-

ment of the target mRNA m6A modifications for FTO-mediated

gene regulation, we conducted luciferase reporter andmutagen-

esis assays (Figure 7B). As expected, compared with mutant

FTO or empty vector, ectopically expressed wild-type FTO

Figure 6. ASB2 and RARA Are Two Critical Target Genes of FTO in AML

(A–C) Western blot assays of FTO, ASB2, RARA, and MLL in MONOMAC-6 or NB4 AML cells with lentivirally transduced FTO (pmiRNA1-FTO), FTO mutant

(H231A and D233A; pmiRNA1-FTO-Mut), or control (Ctrl; pmiRNA1) construct (A), as well as shFTO-1/shNS-1 (B) and shFTO-2/shNS-2 (C). ACTIN was used as

the endogenous control protein for loading control.

(D and E) Effects of forced expression of ASB2 and RARA on cell growth/proliferation in MONOMAC-6 (D) and NB4 (E) AML cells.

(F) Western blot assays of ASB2 in MONOMAC-6 cells transduced with shFTO + shASB2 (pLKO.1-shFTO-1+ pTRIPZ-shASB2), shFTO (pLKO.1-shFTO-1 +

pTRIPZ), shNS (pLKO.1-shNS + pTRIPZ), or shASB2 (pLKO.1-shNS-1 + pTRIPZ-shASB2).

(G) Effects of FTO and/or ASB2 knockdown on cell growth/proliferation (left panel) and viability (right panel) in MONOMAC-6 cells.

(H) Western blot assays of RARA in MONOMAC-6 AML cells transduced with shFTO + shRARA (pLKO.1-shFTO-1+ shRARA), shFTO (pLKO.1-shFTO-1+ shNS),

shNS (pLKO.1-shNS + shNS), or shRARA (pLKO.1-shNS-1 + shRARA).

(I) Effects of FTO and/or RARA knockdown on cell growth/proliferation (left panel) and viability (right panel) in MONOMAC-6 cells.

*p < 0.05, **p < 0.01, ***p < 0.001; t test. See also Figure S6.

substantially reduced luciferase activity of the individual reporter

constructs bearing wild-type ASB2 30 UTR, RARA 30 UTR, orRARA 50 UTR that have intact m6A sites, while mutations in

the m6A sites abrogated the inhibition (Figure 7B). To validate

whether the effects observed above are related tom6Amodifica-

tions, we conducted gene-specific m6A qPCR of the cloned

wild-type and mutant 30 UTR fragments of ASB2 or RARA in

HEK293T cells that were used for the luciferase reporter assays.

As expected, while the wild-type 30 UTR fragments have a high

abundance of m6A modifications, the mutant 30 UTR fragments

contain no or minimal levels of m6A modifications; co-trans-

fected wild-type FTO, but not FTO mutant, significantly reduced

the m6A abundance on the wild-type 30 UTR fragments of ASB2

or RARA in HEK293T cells (see Figure 7C). Together, our data

Cancer Cell 31, 127–141, January 9, 2017 135

Figure 7. FTO-Mediated Regulation of Expression of ASB2 and RARA Relies on its m6A Demethylase Activity and the m6A Modifications in

Target mRNAs

(A) Gene-specific m6A qPCR analysis of m6A level in mRNA transcripts of each gene in MONOMAC-6 and NB4 cells transduced with FTO (FTO), FTO mutant

(FTO-Mut), or control vector (Ctrl).

(legend continued on next page)

136 Cancer Cell 31, 127–141, January 9, 2017

demonstrate that FTO-mediated gene regulation relies on its de-

methylase activity and m6A modifications in the target mRNA

transcripts.

A set of m6A-modified transcripts could be recognized by the

YTHDF2 reader, which exhibits a shorter half-life than non-meth-

ylated ones (Wang et al., 2014a). Thus, it is surprising that most

of the FTO potential targets we identified tend to be negatively

regulated by FTO in AML cells (Figures 5D, 5F, and 5G). To

analyze the effect of m6A level decrease on the stability of poten-

tial target transcripts of FTO, we conducted RNA stability assays

(Wang et al., 2014a) and showed that forced expression and

knockdown of FTO shortened and prolonged, respectively, the

half-life of ASB2 and RARA mRNA transcripts in AML cells (Fig-

ures 7D, S7A, and S7B). Thus, FTO-induced repression of ASB2

andRARA expression is at least in part due to the decreased sta-

bility of ASB2 and RARAmRNA transcripts upon FTO-mediated

decrease of m6A level in their mRNA transcripts. Consistent with

the effect of FTO overexpression, we found that knockdown

of METTL3 and especially of METTL14, two m6A writers, also

resulted in the downregulation of ASB2 and RARA levels

(Figure S7C).

Our results suggest that alternative m6A reading processes

may exist which recognize m6A sites in FTO target transcripts

and promote their stability. YTHDF1 and YTHDF2 are two well-

established m6A readers with each recognizing a few thousands

ofmethylated transcripts inmammalian cells (Wang et al., 2014a,

2015). Nonetheless, knockdown of either of them did not lead to

a significant reduction at the mRNA levels of ASB2 orRARA (Fig-

ures 7E and 7F), indicating that YTHDF1 and YTHDF2 are un-

likely the readers that can promote the stability of ASB2 and

RARA mRNAs. While YTHDF1 knockdown resulted in a minor

decrease in the protein levels of ASB2 and RARA, YTHDF2

knockdown showed no effect on their protein levels at all (Fig-

ure 7G). Furthermore, neither forced expression nor knockdown

of FTO affected expression of METTL3, METTL14, YTHDF1, or

YTHDF2 (Figure 7H). Thus, the effects of FTO overexpression

or knockdown on ASB2 and RARA mRNA stability are unlikely

due to the changes in the above m6A writers or readers. The

reader(s) that promotes the mRNA stability of FTO target tran-

scripts (e.g., ASB2 and RARA) has yet to be identified.

The FTOxASB2/RARA Axis Contributes to theResponse of APL Cells to ATRA TreatmentWhile ATRA-based differentiation therapy has transformed APL

from a highly fatal disease to a highly curable one (Huang

et al., 1988; Wang and Chen, 2008), it is still important to better

(B) Relative luciferase activity of pMIR-REPORT-ASB2-30 UTR (left panel), pMIR

panel) with either wild-type or mutant (A-to-T mutation) m6A sites after co-tran

HEK293T cell. Firefly luciferase activity was measured and normalized to Renilla

(C) Luciferase reporter assay-related gene-specific m6A qPCR analysis of m6A le

RARA 30 UTR in HEK293T cells. For each luciferase reporter construct, we desig

fragment and pMIR-Report vector fragment. Primer 1 covers the joint of firefly Luc

RARA-30 UTR and SV40 poly A.

(D) The mRNA half-life (t1/2) of ASB2 or RARA in NB4 cells transduced with FTO (F

of FTO (shFTO-1) or not (shNS-1).

(E and F) Relative expression of ASB2 and RARA after knockdown of m6A reade

(G) Western blot assay of ASB2 and RARA expression after depletion of m6A rea

(H) Relative expression of m6A writers and readers with forced or depleted expre

*p < 0.05, **p < 0.01; t test. See also Figure S7.

understand the underlying molecular mechanism(s). As shown in

Figure 8A, upon ATRA treatment, FTO is significantly downregu-

lated in NB4 APL cells, associated with a significant upregulation

ofRARA and especially ASB2. To investigate the potential role of

FTO in ATRA-induced APL cell differentiation, we compared

FTO-overexpressing NB4 cells and control NB4 cells on their

response to ATRA treatment through flow cytometry assays.

As ATRA can induce APL cells toward granulocytic and mono-

cytic differentiation (Arteaga et al., 2013; Carlesso et al., 1999;

Gocek et al., 2014), anti-CD11b (a granulocytic differentiation

marker) and anti-CD14 (a monocytic differentiation marker) anti-

bodies were used in the flow-cytometric assays. As shown in

Figure 8B, forced expression of FTO, but not FTO mutant,

noticeably increased the undifferentiated NB4 cell population

(i.e., CD11b�/CD14� cells) after 48 hr of 500 nMATRA treatment.

We then conducted a loss-of-function study, and found that

depletion of FTO expression by two individual different shRNAs

could substantially enhance ATRA-induced cell differentiation,

resulting in a striking decrease in undifferentiated population of

NB4 cells after 48 hr of treatment with a lower concentration of

ATRA (100 nM) (Figure 8C). Figure 8D shows a summary of the

results from three independent experiments. Consistent with

the phenotype caused by FTO knockdown, forced expression

of either RARA or ASB2 could also substantially enhance NB4

cell differentiation (Figures 8E and 8F). Collectively, FTO inhibits

ATRA-induced differentiation of APL cells likely through post-

transcriptionally repressing expression/function of RARA and

ASB2, and such function of FTO relies on its m6A demethylase

activity.

DISCUSSION

In the present study, we demonstrate that FTO, an obesity risk-

associated gene and the first m6A eraser identified, plays a crit-

ical oncogenic role in hematopoietic cell transformation and

leukemogenesis. In brief, FTO expression can be upregulated

by certain oncogenic proteins (e.g., MLL-fusion proteins, PML-

RARA, FLT3-ITD, and NPM1 mutant) and thereby FTO is aber-

rantly upregulated in certain subtypes of AMLs, e.g., t(11q23)/

MLL-rearranged, t(15;17), FLT3-ITD, and/or NPM1-mutated

AMLs. In vitro, we show that forced expression of FTO signifi-

cantly enhances the viability and proliferation/growth of human

AML cells, while inhibiting the apoptosis of the cells, and also

significantly enhances leukemic oncogenes-mediated transfor-

mation/immortalization of normal hematopoietic stem/progeni-

tor cells; the opposite is true when expression of FTO is knocked

-REPORT-RARA-30 UTR (middle panel), and pGL3-basic-RARA-50 UTR (right

sfection with FTO (FTO), FTO mutant (FTO-Mut), or control vector (Ctrl) into

luciferase activity.

vels in exogenous mRNA transcripts of firefly Luc-ASB2 30 UTR or firefly Luc-

ned two different pairs of primers crossing the inserted ASB2 or RARA 30 UTRandASB2-30 UTR orRARA-30 UTR. Primer 2 covers the joint ofASB2-30 UTR or

TO), FTOmutant (FTO-Mut), or control vector (Ctrl) or with depleted expression

rs, YTHDF1 or YTHDF2, in MONOMAC-6 (E) and NB4 (F) cells.

ders in MONOMAC-6 cells.

ssion of FTO in MONOMAC-6 cells.

Cancer Cell 31, 127–141, January 9, 2017 137

Figure 8. The Potential Role of the FTOxRARA/ASB2 Axis in ATRA-Induced NB4

Cell Differentiation and a Schematic Model

of FTO Signaling in AML

(A) Expressional changes of FTO, RARA, and

ASB2 in NB4 cells 48 hr post-treatment with

100 nM ATRA as detected by qPCR. *p < 0.05,

**p < 0.01, ***p < 0.001; t test.

(B) Flow-cytometric analyses of NB4 cells trans-

duced with FTO (FTO), FTO mutant (FTO-Mut), or

control vector (Ctrl) 48 hr post-treatment with

500 nM ATRA or DMSO.

(C) Flow-cytometric analyses of NB4 cells with

knockdown of FTO (shFTO-1 or shFTO-2) 48 hr

post-treatment with 100 nM ATRA or DMSO.

(D) The summary of results from three independent

experiments with those shown in (B) or (C) as

representatives.

(E and F) The representative (E) or summary (F;

from triplicates) results of flow-cytometric ana-

lyses of NB4 cells with forced expression of RARA

or ASB2 after exposed to 100 nM ATRA or DMSO

for 48 hr.

(G) The schematic model of the role and underlying

mechanism of FTO in leukemogenesis and ATRA-

induced differentiation of leukemic cells.

down. In vivo, we show that forced and depleted expression

of Fto significantly promotes and inhibits, respectively, MLL-

fusion-mediated leukemogenesis in mice. Our transcriptome-

wide m6A-seq assay and the subsequent validation and

functional studies suggest that ASB2 and RARA are two critical

target genes of FTO. As an m6A RNA demethylase, FTO reduces

the m6A levels of ASB2 and RARAmainly at UTRs, which in turn

leads to the downregulation of these two genes at the RNA level

and especially at the protein level. Mechanistically, we demon-

strate that the biological function of FTO relies on its m6A

demethylase activity as mutations in the FTO catalytic domain

sufficiently abrogate the function of FTO. Moreover, our lucif-

erase reporter/mutagenesis assays indicate that the m6A sites

in the UTRs of its critical target genes such as ASB2 and

RARA are essential for FTO to post-transcriptionally regulate

their expression. Our data and previous studies (Glasow et al.,

2005; Guibal et al., 2002; Kohroki et al., 2001; Zhu et al., 2001)

demonstrate the anti-leukemic effects of ASB2 and RARA.

Thus, the FTOxASB2/RARA axis likely plays a critical role in

the pathogenesis of AMLs. Furthermore, we provide evidence

that this axis likely also plays an essential role in mediating the

response of leukemic cells to ATRA treatment. A schematic

model summarizing our discoveries is shown in Figure 8G.

Epidemiology studies demonstrate a strong association

between FTO SNPs or overweight/obesity and the risk of

138 Cancer Cell 31, 127–141, January 9, 2017

various types of cancers, such as

breast, prostate, kidney, and pancreatic

cancers, as well as hematopoietic

malignancies including myeloma, lym-

phoma, and leukemia (Castillo et al.,

2012; Li et al., 2012a; Soderberg et al.,

2009). Therefore, it is possible that

increased expression of FTO, caused

by its obesity-associated SNPs (Berulava and Horsthemke,

2010; Church et al., 2010), may contribute (to some

extent) to the higher risk of individuals with overweight and

obesity in developing various types of cancers. If so, FTO

may play an oncogenic role also in other cancers, besides

leukemia.

Previous studies suggest that mRNA transcripts with m6A

modifications tend to be less stable (Schwartz et al., 2014;

Wang et al., 2014a), largely due to the relocation of such mRNAs

by YTHDF2 to RNA decay sites (Wang et al., 2014a). Surpris-

ingly, herein we show that over 80% of FTO potential targets

tend to be negatively regulated by FTO in AML cells. We show

that forced and depleted expression of FTO substantially

shortens and prolongs, respectively, the half-life of its critical

targets such as ASB2 and RARA, suggesting that FTO-medi-

ated repression of ASB2 and RARA expression is at least in

part due to the decreased stability of ASB2 and RARA mRNA

transcripts. Our data further suggest that an additional reading

process may exist which controls the stability of FTO target

transcripts. Neither YTHDF2 nor YTHDF1 target all m6A sites

in mammalian cells. Other reading processes have been sug-

gested (Liu et al., 2015). Although certain m6A sites on ASB2

or RARA could be affected by YTHDF2 or YTHDF1, our data

indicate that the FTO-targeted sites exhibit effects on mRNA

distinct from these known reading processes. It will be very

interesting to uncover such an alternative reading process in

the future.

Interestingly, all the subtypes of AMLs with high levels

of endogenous FTO expression, such as those carrying

t(11q23), t(15;17), NPM1 mutations, and/or FLT3-ITD, are

more sensitive to ATRA than the other AML subtypes (Dos

Santos et al., 2013; El Hajj et al., 2015; Hu et al., 2009; Huang

et al., 1988; Iijima et al., 2004; Jiang et al., 2016; Lo-

Coco et al., 2013; Martelli et al., 2015; Niitsu et al., 2001;

Schlenk et al., 2009). It is possible that the survival/prolifera-

tion of these subtypes of AML cells relies more on the FTO

signaling, and thus they are more responsive to ATRA treat-

ment, as ATRA can release the expression/function of ASB2

and RARA, two negative targets of FTO, and thereby trigger

cell differentiation.

In summary, we provide compelling in vitro and in vivo

evidence demonstrating that FTO, an m6A demethylase,

plays a critical oncogenic role in cell transformation and

leukemogenesis as well as in ATRA-mediated differentiation

of leukemic cells, through reducing m6A levels in mRNA

transcripts of its critical target genes such as ASB2 and RARA

and thereby triggering corresponding signaling cascades. Our

study highlights the functional importance of the m6A modifi-

cation machinery in cancer, and provides profound insights

into the molecular mechanisms underlying tumorigenesis by

revealing a previously unrecognized mechanism of gene regula-

tion in cancer. In addition, given the functional importance of

FTO in leukemogenesis and drug response, targeting FTO

signaling by selective inhibitors may represent a promising ther-

apeutic strategy to treat leukemia, especially in combination with

ATRA treatment. As FTO has also been implicated in other types

of cancers, our discoveries may have a broad impact in cancer

biology and cancer therapy.

EXPERIMENTAL PROCEDURES

Leukemic Patient Samples

The leukemia patient samples were obtained at the time of diagnosis or

relapse and with informed consent at the University of Chicago Hospital or

other collaborative hospitals, and were approved by the institutional review

board of the institutes/hospitals.

The Care and Maintenance of Animals

This was approved by the Institutional Animal Care and Use Committee of the

University of Chicago or the University of Cincinnati.

Cell Culture and Transfection, ChIP, qPCR, In Vitro Functional Study

Assays, and In Vivo Bone Marrow Transplantation Studies

These assays were performed as previously described (Huang et al., 2013;

Jiang et al., 2012; Li et al., 2012b) with some modifications.

Global m6A Quantitative Assays, m6A-Seq, RNA-Seq, Gene-Specific

m6A qPCR, and RNA Stability Assays

These assays were conducted as described previously (Dominissini et al.,

2013; Jia et al., 2011; Liu et al., 2014; Wang et al., 2014a) with some modifica-

tions (see Supplemental Information for details).

ACCESSION NUMBERS

The microarray, m6A-seq and RNA-seq data have been deposited in the

GEO repository under the accession numbers GEO: GSE34184, GSE30285,

GSE76414, GSE84944, and GSE85008.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures,

seven figures, and three tables and can be found with this article online at

http://dx.doi.org/10.1016/j.ccell.2016.11.017.

AUTHOR CONTRIBUTIONS

Z.L. and J.C. conceived the project. Z.L., R.S., C.H., and J.C. designed the

research and supervised the experiments conducted in the laboratories.

Z.L., H.W., R.S., X.W., Z.Z., C.L., H.H., S.N., L.D., C.H., X.Q., L.T., G.M.H.,

Y.W., H.H., X.W., P.C., S.G., S.A., Y.L., S.L., J.S., X.J., and J.C. performed

the experiments and/or data analyses; Z.L., H.H., M.B.N., R.A.L., X.J., P.Z.,

J.J., C.H., and J.C. contributed the reagents/analytic tools and/or grant sup-

port; Z.L., H.W., R.S., X.W., Z.Z., C.H., and J.C. wrote the paper. All authors

discussed the results and commented on the manuscript.

ACKNOWLEDGMENTS

We thank Dr. Michelle M. Le Beau for providing primary AML patient samples

and James Mulloy for providing the MA9/FLT3-ITD AML cell line. This work

was supported in part by NIH R01 grants CA178454 (J.C.), CA182528 (J.C.),

CA214965 (J.C.), and GM071440 (C.H.), Leukemia & Lymphoma Society

(LLS) Special Fellowship (Z.L.), LLS Translational Research grant (J.C.), Amer-

ican Cancer Society (ACS) Research Scholar grant (J.C.), ACS-IL Research

Scholar grant (Z.L.), Gabrielle’s Angel Foundation for Cancer Research (J.C.,

Z.L., X.J., and H.H.), China Scholarship Council (CSC) visiting scholar (X.W.),

and Foundation of Innovation Team for Basic and Clinical Research of Zhe-

jiang Province (grant 2011R50015) (J.J.). S.N. is an HHMI Fellow of the Damon

Runyon Cancer Research Foundation (DRG-2215-15). C.H. is an investigator

of the Howard Hughes Medical Institute (HHMI).

Received: January 12, 2016

Revised: June 2, 2016

Accepted: November 21, 2016

Published: December 22, 2016

REFERENCES

Alarcon, C.R., Lee, H., Goodarzi, H., Halberg, N., and Tavazoie, S.F. (2015).

N6-Methyladenosine marks primary microRNAs for processing. Nature 519,

482–485.

Arteaga, M.F., Mikesch, J.H., Qiu, J., Christensen, J., Helin, K., Kogan, S.C.,

Dong, S., and So, C.W. (2013). The histone demethylase PHF8 governs reti-

noic acid response in acute promyelocytic leukemia. Cancer Cell 23, 376–389.

Ben-Haim, M.S., Moshitch-Moshkovitz, S., and Rechavi, G. (2015). FTO: link-

ing m6A demethylation to adipogenesis. Cell Res. 25, 3–4.

Bernt, K.M., Zhu, N., Sinha, A.U., Vempati, S., Faber, J., Krivtsov, A.V., Feng,

Z., Punt, N., Daigle, A., Bullinger, L., et al. (2011). MLL-rearranged leukemia is

dependent on aberrant H3K79 methylation by DOT1L. Cancer Cell 20, 66–78.

Berulava, T., and Horsthemke, B. (2010). The obesity-associated SNPs in

intron 1 of the FTO gene affect primary transcript levels. Eur. J. Hum. Genet.

18, 1054–1056.

Boissel, S., Reish, O., Proulx, K., Kawagoe-Takaki, H., Sedgwick, B., Yeo,

G.S., Meyre, D., Golzio, C., Molinari, F., Kadhom, N., et al. (2009). Loss-of-

function mutation in the dioxygenase-encoding FTO gene causes severe

growth retardation and multiple malformations. Am. J. Hum. Genet. 85,

106–111.

Bokar, J.A., Shambaugh, M.E., Polayes, D., Matera, A.G., and Rottman, F.M.

(1997). Purification and cDNA cloning of the AdoMet-binding subunit of the

human mRNA (N6-adenosine)-methyltransferase. RNA 3, 1233–1247.

Carlesso, N., Aster, J.C., Sklar, J., and Scadden, D.T. (1999). Notch1-induced

delay of human hematopoietic progenitor cell differentiation is associated with

altered cell cycle kinetics. Blood 93, 838–848.

Castillo, J.J., Mull, N., Reagan, J.L., Nemr, S., and Mitri, J. (2012). Increased

incidence of non-Hodgkin lymphoma, leukemia, and myeloma in patients

Cancer Cell 31, 127–141, January 9, 2017 139

with diabetes mellitus type 2: a meta-analysis of observational studies. Blood

119, 4845–4850.

Chen, J., Odenike, O., and Rowley, J.D. (2010). Leukaemogenesis: more than

mutant genes. Nat. Rev. Cancer 10, 23–36.

Chen, T., Hao, Y.J., Zhang, Y., Li, M.M., Wang, M., Han, W., Wu, Y., Lv, Y.,

Hao, J., Wang, L., et al. (2015). m(6)A RNA methylation is regulated by

microRNAs and promotes reprogramming to pluripotency. Cell Stem Cell

16, 289–301.

Church, C., Moir, L., McMurray, F., Girard, C., Banks, G.T., Teboul, L., Wells,

S., Bruning, J.C., Nolan, P.M., Ashcroft, F.M., and Cox, R.D. (2010).

Overexpression of Fto leads to increased food intake and results in obesity.

Nat. Genet. 42, 1086–1092.

Desrosiers, R., Friderici, K., and Rottman, F. (1974). Identification of methyl-

ated nucleosides in messenger RNA from Novikoff hepatoma cells. Proc.

Natl. Acad. Sci. USA 71, 3971–3975.

Dina, C., Meyre, D., Gallina, S., Durand, E., Korner, A., Jacobson, P., Carlsson,

L.M., Kiess,W., Vatin, V., Lecoeur, C., et al. (2007). Variation in FTO contributes

to childhood obesity and severe adult obesity. Nat. Genet. 39, 724–726.

Dohner, H., Weisdorf, D.J., and Bloomfield, C.D. (2015). Acute myeloid leuke-

mia. N. Engl. J. Med. 373, 1136–1152.

Dominissini, D., Moshitch-Moshkovitz, S., Schwartz, S., Salmon-Divon, M.,

Ungar, L., Osenberg, S., Cesarkas, K., Jacob-Hirsch, J., Amariglio, N.,

Kupiec, M., et al. (2012). Topology of the human and mouse m6A RNA meth-

ylomes revealed by m6A-seq. Nature 485, 201–206.

Dominissini, D., Moshitch-Moshkovitz, S., Salmon-Divon, M., Amariglio, N.,

and Rechavi, G. (2013). Transcriptome-wide mapping of N(6)-methyladeno-

sine by m(6)A-seq based on immunocapturing and massively parallel

sequencing. Nat. Protoc. 8, 176–189.

Dos Santos, G.A., Kats, L., and Pandolfi, P.P. (2013). Synergy against PML-

RARa: targeting transcription, proteolysis, differentiation, and self-renewal in

acute promyelocytic leukemia. J. Exp. Med. 210, 2793–2802.

El Hajj, H., Dassouki, Z., Berthier, C., Raffoux, E., Ades, L., Legrand, O.,

Hleihel, R., Sahin, U., Tawil, N., Salameh, A., et al. (2015). Retinoic acid and

arsenic trioxide trigger degradation of mutated NPM1, resulting in apoptosis

of AML cells. Blood 125, 3447–3454.

Fischer, J., Koch, L., Emmerling, C., Vierkotten, J., Peters, T., Bruning, J.C.,

and Ruther, U. (2009). Inactivation of the Fto gene protects from obesity.

Nature 458, 894–898.

Frayling, T.M., Timpson, N.J., Weedon, M.N., Zeggini, E., Freathy, R.M.,

Lindgren, C.M., Perry, J.R., Elliott, K.S., Lango, H., Rayner, N.W., et al.

(2007). A common variant in the FTO gene is associated with body mass index

and predisposes to childhood and adult obesity. Science 316, 889–894.

Fu, Y., Dominissini, D., Rechavi, G., and He, C. (2014). Gene expression regu-

lation mediated through reversible m(6)A RNA methylation. Nat. Rev. Genet.

15, 293–306.

Gao, X., Shin, Y.H., Li, M., Wang, F., Tong, Q., and Zhang, P. (2010). The fat

mass and obesity associated gene FTO functions in the brain to regulate post-

natal growth in mice. PLoS One 5, e14005.

Geula, S., Moshitch-Moshkovitz, S., Dominissini, D., Mansour, A.A., Kol, N.,

Salmon-Divon, M., Hershkovitz, V., Peer, E., Mor, N., Manor, Y.S., et al.

(2015). Stem cells. m6A mRNA methylation facilitates resolution of naive plu-

ripotency toward differentiation. Science 347, 1002–1006.

Glasow, A., Prodromou, N., Xu, K., von Lindern, M., and Zelent, A. (2005).

Retinoids and myelomonocytic growth factors cooperatively activate RARA

and induce human myeloid leukemia cell differentiation via MAP kinase path-

ways. Blood 105, 341–349.

Gocek, E., Marchwicka, A., Bujko, K., and Marcinkowska, E. (2014). NADPH-

cytochrome P450 reductase is regulated by all-trans retinoic acid and by 1,25-

dihydroxyvitamin D3 in human acute myeloid leukemia cells. PLoS One 9,

e91752.

Guibal, F.C., Moog-Lutz, C., Smolewski, P., Di Gioia, Y., Darzynkiewicz, Z.,

Lutz, P.G., and Cayre, Y.E. (2002). ASB-2 inhibits growth and promotes

commitment in myeloid leukemia cells. J. Biol. Chem. 277, 218–224.

140 Cancer Cell 31, 127–141, January 9, 2017

Hess, M.E., Hess, S., Meyer, K.D., Verhagen, L.A., Koch, L., Bronneke, H.S.,

Dietrich, M.O., Jordan, S.D., Saletore, Y., Elemento, O., et al. (2013). The fat

mass and obesity associated gene (Fto) regulates activity of the dopaminergic

midbrain circuitry. Nat. Neurosci. 16, 1042–1048.

Hu, J., Liu, Y.F., Wu, C.F., Xu, F., Shen, Z.X., Zhu, Y.M., Li, J.M., Tang, W.,

Zhao, W.L., Wu, W., et al. (2009). Long-term efficacy and safety of all-trans ret-

inoic acid/arsenic trioxide-based therapy in newly diagnosed acute promyelo-

cytic leukemia. Proc. Natl. Acad. Sci. USA 106, 3342–3347.

Huang, M.E., Ye, Y.C., Chen, S.R., Chai, J.R., Lu, J.X., Zhoa, L., Gu, L.J., and

Wang, Z.Y. (1988). Use of all-trans retinoic acid in the treatment of acute pro-

myelocytic leukemia. Blood 72, 567–572.

Huang, H., Jiang, X., Li, Z., Li, Y., Song, C.X., He, C., Sun, M., Chen, P.,

Gurbuxani, S., Wang, J., et al. (2013). TET1 plays an essential oncogenic

role in MLL-rearranged leukemia. Proc. Natl. Acad. Sci. USA 110, 11994–

11999.

Iijima, K., Honma, Y., and Niitsu, N. (2004). Granulocytic differentiation of

leukemic cells with t(9;11)(p22;q23) induced by all-trans-retinoic acid. Leuk.

Lymphoma 45, 1017–1024.

Jia, G., Fu, Y., Zhao, X., Dai, Q., Zheng, G., Yang, Y., Yi, C., Lindahl, T., Pan, T.,

Yang, Y.G., and He, C. (2011). N6-Methyladenosine in nuclear RNA is a major

substrate of the obesity-associated FTO. Nat. Chem. Biol. 7, 885–887.

Jia, G., Fu, Y., andHe, C. (2013). Reversible RNA adenosinemethylation in bio-

logical regulation. Trends Genet. 29, 108–115.

Jiang, X., Huang, H., Li, Z., Li, Y., Wang, X., Gurbuxani, S., Chen, P., He, C.,

You, D., Zhang, S., et al. (2012). Blockade of miR-150 maturation by MLL-

fusion/MYC/LIN-28 is required for MLL-associated leukemia. Cancer Cell

22, 524–535.

Jiang, X., Hu, C., Arnovitz, S., Bugno, J., Yu, M., Zuo, Z., Chen, P., Huang, H.,

Ulrich, B., Gurbuxani, S., et al. (2016). miR-22 has a potent anti-tumor role with

therapeutic potential in acute myeloid leukemia. Nat. Commun. 7, 11452.

Kohroki, J., Fujita, S., Itoh, N., Yamada, Y., Imai, H., Yumoto, N., Nakanishi, T.,

and Tanaka, K. (2001). ATRA-regulated Asb-2 gene induced in differentiation

of HL-60 leukemia cells. FEBS Lett. 505, 223–228.

Li, G., Chen, Q., Wang, L., Ke, D., and Yuan, Z. (2012a). Association between

FTO gene polymorphism and cancer risk: evidence from 16,277 cases and

31,153 controls. Tumour Biol. 33, 1237–1243.

Li, Z., Huang, H., Chen, P., He, M., Li, Y., Arnovitz, S., Jiang, X., He, C., Hyjek,

E., Zhang, J., et al. (2012b). miR-196b directly targets both HOXA9/MEIS1 on-

cogenes and FAS tumour suppressor in MLL-rearranged leukaemia. Nat.

Commun. 2, 688.

Lin, L., Hales, C.M., Garber, K., and Jin, P. (2014). Fat mass and obesity-asso-

ciated (FTO) protein interacts with CaMKII and modulates the activity of CREB

signaling pathway. Hum. Mol. Genet. 23, 3299–3306.

Liu, J., Yue, Y., Han, D., Wang, X., Fu, Y., Zhang, L., Jia, G., Yu, M., Lu, Z.,

Deng, X., et al. (2014). AMETTL3-METTL14 complexmediatesmammalian nu-

clear RNA N6-adenosine methylation. Nat. Chem. Biol. 10, 93–95.

Liu, N., Dai, Q., Zheng, G., He, C., Parisien, M., and Pan, T. (2015). N(6)-

Methyladenosine-dependent RNA structural switches regulate RNA-protein

interactions. Nature 518, 560–564.

Lo-Coco, F., Avvisati, G., Vignetti, M., Thiede, C., Orlando, S.M., Iacobelli, S.,

Ferrara, F., Fazi, P., Cicconi, L., Di Bona, E., et al. (2013). Retinoic acid

and arsenic trioxide for acute promyelocytic leukemia. N. Engl. J. Med. 369,

111–121.

Marcucci, G., Mrozek, K., and Bloomfield, C.D. (2005). Molecular heterogene-

ity and prognostic biomarkers in adults with acute myeloid leukemia and

normal cytogenetics. Curr. Opin. Hematol. 12, 68–75.

Martelli, M.P., Gionfriddo, I., Mezzasoma, F., Milano, F., Pierangeli, S., Mulas,

F., Pacini, R., Tabarrini, A., Pettirossi, V., Rossi, R., et al. (2015). Arsenic

trioxide and all-trans retinoic acid target NPM1 mutant oncoprotein levels

and induce apoptosis in NPM1-mutated AML cells. Blood 125, 3455–3465.

Meng, J., Cui, X., Rao, M.K., Chen, Y., and Huang, Y. (2013). Exome-based

analysis for RNA epigenome sequencing data. Bioinformatics 29, 1565–1567.

Meyer, K.D., and Jaffrey, S.R. (2014). The dynamic epitranscriptome: N6-

methyladenosine and gene expression control. Nat. Rev. Mol. Cell Biol. 15,

313–326.

Meyer, K.D., Saletore, Y., Zumbo, P., Elemento, O., Mason, C.E., and Jaffrey,

S.R. (2012). Comprehensive analysis of mRNAmethylation reveals enrichment

in 3’ UTRs and near stop codons. Cell 149, 1635–1646.

Meyer, K.D., Patil, D.P., Zhou, J., Zinoviev, A., Skabkin, M.A., Elemento, O.,

Pestova, T.V., Qian, S.B., and Jaffrey, S.R. (2015). 5’ UTR m(6)A promotes

cap-independent translation. Cell 163, 999–1010.

Niitsu, N., Hayashi, Y., Sugita, K., and Honma, Y. (2001). Sensitization by

5-aza-2’-deoxycytidine of leukaemia cells with MLL abnormalities to induction

of differentiation by all-trans retinoic acid and 1alpha,25-dihydroxyvitamin D3.

Br. J. Haematol. 112, 315–326.

Niu, Y., Zhao, X., Wu, Y.S., Li, M.M., Wang, X.J., and Yang, Y.G. (2013).

N6-Methyl-adenosine (m6A) in RNA: an old modification with a novel epige-

netic function. Genomics Proteomics Bioinformatics 11, 8–17.

Okada, Y., Feng, Q., Lin, Y., Jiang, Q., Li, Y., Coffield, V.M., Su, L., Xu, G., and

Zhang, Y. (2005). hDOT1L links histone methylation to leukemogenesis. Cell

121, 167–178.

Perry, R.P., and Kelley, D.E. (1974). Existence of methylated messenger RNA

in mouse L cells. Cell 1, 37–42.

Ping, X.L., Sun, B.F., Wang, L., Xiao, W., Yang, X., Wang, W.J., Adhikari, S.,

Shi, Y., Lv, Y., Chen, Y.S., et al. (2014). Mammalian WTAP is a regulatory

subunit of the RNA N6-methyladenosine methyltransferase. Cell Res. 24,

177–189.

Sakamoto, K., Imamura, T., Yano, M., Yoshida, H., Fujiki, A., Hirashima, Y.,

and Hosoi, H. (2014). Sensitivity of MLL-rearranged AML cells to all-trans ret-

inoic acid is associated with the level of H3K4me2 in the RARalpha promoter

region. Blood Cancer J. 4, e205.

Schlenk, R.F., Dohner, K., Kneba, M., Gotze, K., Hartmann, F., Del Valle, F.,

Kirchen, H., Koller, E., Fischer, J.T., Bullinger, L., et al. (2009). Gene mutations

and response to treatment with all-trans retinoic acid in elderly patients with

acute myeloid leukemia. Results from the AMLSG Trial AML HD98B.

Haematologica 94, 54–60.

Schwartz, S., Mumbach, M.R., Jovanovic, M., Wang, T., Maciag, K., Bushkin,

G.G., Mertins, P., Ter-Ovanesyan, D., Habib, N., Cacchiarelli, D., et al. (2014).

Perturbation of m6A writers reveals two distinct classes of mRNA methylation

at internal and 5’ sites. Cell Rep. 8, 284–296.

Scuteri, A., Sanna, S., Chen, W.M., Uda, M., Albai, G., Strait, J., Najjar, S.,

Nagaraja, R., Orru, M., Usala, G., et al. (2007). Genome-wide association

scan shows genetic variants in the FTO gene are associated with obesity-

related traits. PLoS Genet. 3, e115.

Soderberg, K.C., Kaprio, J., Verkasalo, P.K., Pukkala, E., Koskenvuo, M.,

Lundqvist, E., and Feychting, M. (2009). Overweight, obesity and risk of hae-

matological malignancies: a cohort study of Swedish and Finnish twins. Eur.

J. Cancer 45, 1232–1238.

Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L.,

Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., and

Mesirov, J.P. (2005). Gene set enrichment analysis: a knowledge-based

approach for interpreting genome-wide expression profiles. Proc. Natl.

Acad. Sci. USA 102, 15545–15550.

Wang, Z.Y., and Chen, Z. (2008). Acute promyelocytic leukemia: from highly

fatal to highly curable. Blood 111, 2505–2515.

Wang, J., Muntean, A.G., and Hess, J.L. (2012). ECSASB2 mediates MLL

degradation during hematopoietic differentiation. Blood 119, 1151–1161.

Wang, X., Lu, Z., Gomez, A., Hon, G.C., Yue, Y., Han, D., Fu, Y., Parisien, M.,

Dai, Q., Jia, G., et al. (2014a). N6-Methyladenosine-dependent regulation of

messenger RNA stability. Nature 505, 117–120.

Wang, Y., Li, Y., Toth, J.I., Petroski, M.D., Zhang, Z., and Zhao, J.C. (2014b).

N6-Methyladenosine modification destabilizes developmental regulators in

embryonic stem cells. Nat. Cell Biol. 16, 191–198.

Wang, X., Zhao, B.S., Roundtree, I.A., Lu, Z., Han, D., Ma, H., Weng, X., Chen,

K., Shi, H., and He, C. (2015). N(6)-Methyladenosine modulates messenger

RNA translation efficiency. Cell 161, 1388–1399.

Wunderlich, M., Mizukawa, B., Chou, F.S., Sexton, C., Shrestha, M.,

Saunthararajah, Y., and Mulloy, J.C. (2013). AML cells are differentially sensi-

tive to chemotherapy treatment in a human xenograft model. Blood 121,

e90–e97.

Yue, Y., Liu, J., and He, C. (2015). RNA N6-methyladenosine methylation in

post-transcriptional gene expression regulation. Genes Dev. 29, 1343–1355.

Zeisig, B.B., Milne, T., Garcia-Cuellar, M.P., Schreiner, S., Martin, M.E., Fuchs,

U., Borkhardt, A., Chanda, S.K., Walker, J., Soden, R., et al. (2004). Hoxa9 and

Meis1 are key targets for MLL-ENL-mediated cellular immortalization. Mol.

Cell. Biol. 24, 617–628.

Zhang, Y., Liu, T., Meyer, C.A., Eeckhoute, J., Johnson, D.S., Bernstein, B.E.,

Nusbaum, C., Myers, R.M., Brown, M., Li, W., and Liu, X.S. (2008). Model-

based analysis of ChIP-seq (MACS). Genome Biol. 9, R137.

Zhang, C., Samanta, D., Lu, H., Bullen, J.W., Zhang, H., Chen, I., He, X., and

Semenza, G.L. (2016). Hypoxia induces the breast cancer stem cell phenotype

by HIF-dependent and ALKBH5-mediated m6A-demethylation of NANOG

mRNA. Proc. Natl. Acad. Sci. USA 113, E2047–E2056.

Zhao, X., Yang, Y., Sun, B.F., Shi, Y., Yang, X., Xiao, W., Hao, Y.J., Ping, X.L.,

Chen, Y.S., Wang, W.J., et al. (2014). FTO-dependent demethylation of N6-

methyladenosine regulates mRNA splicing and is required for adipogenesis.

Cell Res. 24, 1403–1419.

Zheng, G., Dahl, J.A., Niu, Y., Fedorcsak, P., Huang, C.M., Li, C.J., Vagbo,

C.B., Shi, Y., Wang, W.L., Song, S.H., et al. (2013). ALKBH5 is a mammalian

RNA demethylase that impacts RNA metabolism and mouse fertility. Mol.

Cell 49, 18–29.

Zhou, J., Wan, J., Gao, X., Zhang, X., Jaffrey, S.R., and Qian, S.B. (2015).

Dynamic m(6)A mRNA methylation directs translational control of heat shock

response. Nature 526, 591–594.

Zhu, J., Heyworth, C.M., Glasow, A., Huang, Q.H., Petrie, K., Lanotte, M.,

Benoit, G., Gallagher, R., Waxman, S., Enver, T., and Zelent, A. (2001).

Lineage restriction of the RARalpha gene expression in myeloid differentiation.

Blood 98, 2563–2567.

Cancer Cell 31, 127–141, January 9, 2017 141

Cancer Cell, Volume 31

Supplemental Information

FTO Plays an Oncogenic Role in Acute

Myeloid Leukemia as a N6-Methyladenosine

RNA Demethylase

Zejuan Li, Hengyou Weng, Rui Su, Xiaocheng Weng, Zhixiang Zuo, Chenying Li, HuilinHuang, Sigrid Nachtergaele, Lei Dong, Chao Hu, Xi Qin, Lichun Tang, YunguiWang, Gia-Ming Hong, Hao Huang, Xiao Wang, Ping Chen, SandeepGurbuxani, Stephen Arnovitz, Yuanyuan Li, Shenglai Li, Jennifer Strong, Mary BethNeilly, Richard A. Larson, Xi Jiang, Pumin Zhang, Jie Jin, Chuan He, and Jianjun Chen

1  

SUPPLEMENTAL DATA

Figure S1 (related to Figure 1). Relative expression of ALKBH5/Alkbh5 or FTO/Fto. (A)

Comparison of ALKBH5 expression between human primary AML cases with MLL

rearrangements/t(11q23) (MLL) and those without MLL rearrangements (non-MLL), or AML

2  

cases with inv(16), t(8;21) or t(15;17), or normal controls (NC) based Affymetrix exon array data.

The expression values were detected by microarrays (Huang et al., 2013). The expression values

of ALKBH5 of the individual samples were log2-transformed and then normalized. The mean

value of ALKBH5 expression across the entire set of 109 samples was set as 0. (B) qPCR

analysis of ALKBH5 expression in human CD34+ AML BM cells isolated from 10 primary

MLL-rearranged AML patients (CD34_MLL) and normal CD34+ BM cells isolated from 6

healthy donors (CD34_NC). The average expression level of ALKBH5 in the CD34_NC samples

was set as 1. (C,D) Relative expression of FTO in different AML subtypes in GSE37642 set

(n=562) (C) and GSE14468 set (n=518) (D). (E) The protein (left panel) and RNA (right panel)

levels of Fto in mouse BM progenitor cells transduced with different oncogenes or control

constructs were detected by Western blot and qPCR assays, respectively. NPM1c, NPM1

mutation, which leads to cytoplasmic delocalization of Nucleophosmin. MLL, t(11q23)/MLL-

rearranged; MA9, MLL-AF9/t(9;11) fusion gene. (F) Relative expression of Fto in mouse BM

cells of the control group or MLL-AF9 (MA9) leukemic group collected from primary (_P) or

secondary (_S) BMT recipient mice as detected by qPCR. (G) Comparison of Alkbh5 expression

between MA9 leukemic BM cells and normal control BM cells collected from the

aforementioned primary or secondary BMT recipients. The expression values were detected by a

microarray assay (Li et al., 2012a). Expression values of Alkbh5 were log2-transformed and then

normalized. The mean value of Alkbh5 across the 3 normal control samples from primary BMT

was set as 0. Error bar, mean±SEM.

3  

Figure S2 (related to Figure 2). Effects of forced expression and knockdown of FTO

expression on cell apoptosis and global m6A abundance in leukemia cells. (A-D) Effects of

forced expression (with MSCV-PIG-FTO) or knockdown (with siRNAs) of FTO expression (A)

on cell viability (B), apoptosis (C) and growth (D) of human MONOMAC-6/t(9;11) AML cells.

(E-I) Effects of forced expression or knockdown of FTO on cell growth/proliferation (E),

viability (F), apoptosis (G and H), and protein level of FTO (I) in K562 AML cells. FTO,

pMIRNA1-FTO; FTO-Mut, pMIRNA1-FTO-Mut; Ctrl, control vector (empty pMIRNA1);

shFTO-1, pLKO.1-shFTO-1; shNS-1, pLKO.1-shNS-1; shFTO-2, pGFP-C-shLenti-shFTO-2;

shNS-2, pGFP-C-shLenti-shNS-2. (J) Relative expression of endogenous FTO in MONOMAC-6,

4  

MV4-11 and K562 cells by qPCR. NC, human normal CD34+ BM cells, which was used as

normal control for expression level normalization and statistical analysis. (K) Relative

expression of FTO/Fto (the antibody recognizes both FTO and Fto) in secondary passage of

colony-forming cells shown in Figure 2G as detected by Western Blotting. *, p<0.05; **, p<0.01;

NS, not significant (p>0.05); t-test. Error bar, mean±SD.

Figure S3 (related to Figure 3). Forced expression of FTO, but not FTO mutant,

significantly decreased global m6A levels, while knockdown of FTO by two different

shRNAs increased global m6A levels in NB4 AML cells. FTO, pMIRNA1-FTO; FTO-Mut,

pMIRNA1-FTO-Mut; Ctrl, control vector (empty pMIRNA1); shFTO-1, pLKO.1-shFTO-1;

shNS-1, pLKO.1-shNS-1; shFTO-2, pGFP-C-shLenti-shFTO-2; shNS-2, pGFP-C-shLenti-shNS-

2; m6A, m6A dot blot staining; MB, methylene blue staining (as loading control).

5  

Figure S4 (related to Figure 4). Relative expression of Fto in mouse leukemic BM cells and histopathology analysis of representative bone marrow transplantation (BMT) mice shown in Figure 4A. (A) Relative expression of Fto in leukemic BM cells from representative BMT recipient mice as detected by qPCR. **, p<0.01; t-test. Error bar, mean±SD. (B) Wright-Giemsa stained peripheral blood (PB) and BM, and hematoxylin and eosin (H&E) stained spleen and liver of the BMT recipient mice at the end point. The length of bars (showing under the right

bottom of each individual image) represents 25 m for PB and BM, and 100 m for spleen and liver.

6  

Figure S5 (Related to Figure 5). Transcriptome-wide m6A-seq and analysis of m6A peaks.

(A) Identification of m6A peaks by MACS and exomePeak algorithms. The numbers of m6A

peaks identified by each or both of these two algorithms from at least two of the four m6A-seq

libraries (i.e., FTO_1/2 and Ctrl_1/2) after normalized by input are shown. (B) A top m6A motif

identified from the 13,278 mRNA fragments with the reliable m6A peaks (8,902/13,278 in the

m6A fragments vs. 754/2,309 in the control fragments, p=1.5e-210) is shown in the left panel,

while the location and frequency of this motif in the positive strands of the mRNA fragments

with the reliable m6A peaks, or in the negative strands of these fragments, or in the positive or

7  

negative strands of control mRNA fragments (i.e., those with no m6A peaks) are shown in the

right panel. (C) Distribution of m6A peaks in exonic or intronic regions of protein-coding genes

or non-coding genes, or in other regions. (D) Distribution of m6A peaks across the length of

mRNA. Each region of 5’untranslated region (5’ UTR), coding region (CDS), and 3’untranslated

region (3’ UTR) were binned into 50 segments, and the percentage of m6A peaks that fall within

each bin was determined. (E) The proportion (upper panel) and enrichment (lower panel) of m6A

peak distribution in the 5’UTR, CDS, stop codon or 3’ UTR region across the entire set of

mRNA transcripts. The enrichment was determined by the proportion of m6A peaks normalized

by the length of the region. (F) Comparison of the abundance of m6A peaks and the expression

level of their associated entire RNA transcripts between human MONOMAC-6 cells with forced

expression of FTO (i.e., FTO_1 and _2) and the control cells (i.e., Ctrl_1 and _2). Identification

of 2,785 hypo- and 3,180 hyper-methylated m6A peaks that showed a significant decrease and

increase (p<0.005; fold change 1.2), respectively, in abundance in FTO_1/2 cells relative to

Ctrl_1/2 cells. The remaining m6A peaks in the list of the 13,278 m6A peaks were termed as

unchanged m6A peaks. IP/INPUT, immunoprecipitation/input, which refers to m6A peak

abundance detected in m6A-seq assay (IP) normalized by that detected in RNA-seq assay

(INPUT). (G) Gene set enrichment analysis (GSEA) of genes with a significant decrease in m6A

levels as well as a significant increase (Hypo-up) or decrease (Hypo-down) in overall transcript

levels in FTO-overexpressing AML cells. (H) GSEA analysis of genes with a significant

increase in m6A levels as well as a significant increase (Hyper-up) or decrease (Hyper-down) in

overall transcript levels in FTO-overexpressing MonoMac-6 AML cells.

8  

Table S1 (Related to Figure 5). Summary of the m6A-seq and RNA-seq of four cell lines

Sample Seq Type Total Reads Aligned Reads

Ctrl_1 RNA-seq 45,887,508 41,424,514 (90.3%)

m6A-seq 30,829,176 27,747,658 (90%)

Ctrl_2 RNA-seq 29,936,502 27,386,859 (91.5%)

m6A-seq 34,024,847 30,946,880 (91.0%)

FTO_1 RNA-seq 83,147,818 77,079,571 (92.7%)

m6A-seq 20,466,641 19,036,992 (93%)

FTO_2 RNA-seq 17,714,000 16,361,563 (92.4%)

m6A-seq 24,903,917 23,322,084 (93.6%)

9  

Table S2 (Related to Figure 5). List of 25 genes that exhibit a significant change between FTO-overexpressing and control MONOMAC-6 AML cells in m6A peak levels, and abundance of the corresponding mRNA transcript, and are also significantly positively or negatively correlated with FTO in expression in four datasets of large cohorts of AMLa

Gene Patternb

m6A level changec mRNA transcript abundance

change

Correlation between the given gene and FTO in expression ind

Peak 1 Peak 2 In-house set

(n=109) TCGA set

(n=177) GSE37642 set

(n=562) GSE14468 set

(n=518)

Fold p Loc Fold p Loc Fold p r p r p r p r p ASB2 Hypo-down 0.58 6.9E-08 3'UTR 0.46 1.8E-09 -0.28 0.0030 -0.22 0 -0.21 3.4E-07 -0.18 3.1E-05KCNG1 Hypo-down 0.21 2.3E-07 CDS 0.40 4.5E-07 -0.31 0.0012 -0.15 0.05 -0.34 5.5E-17 -0.1 0.02 PPARD Hypo-down 0.56 0.0015 3'UTR 0.74 9E-15 StopC 0.62 1.4E-13 -0.36 0.0001 -0.28 0 -0.34 2.1E-16 -0.19 1.0E-05 RAB17 Hypo-down 0.47 9.8E-15 3'UTR 0.30 0.0002 -0.47 2.3E-07 -0.17 0.02 -0.25 3.6E-09 -0.09 0.03 RARA Hypo-down 0.68 4.3E-15 3'UTR 0.80 3E-05 5'UTR 0.66 2.9E-07 -0.37 5.9E-05 -0.26 0.0006 -0.35 5.7E-18 -0.21 2.3E-06 SLC11A1 Hypo-down 0.63 1.0E-08 StopC 0.79 0.001 3'UTR 0.20 0 -0.32 0.0007 -0.34 4.3E-06 -0.37 1.9E-19 -0.28 5.7E-11 SLCO4A1 Hypo-down 0.29 1.2E-14 CDS 0.69 0.004 StopC 0.50 0.0002 -0.47 3.2E-07 -0.28 0 -0.16 0.0001 -0.1 0.03 TBC1D9 Hypo-down 0.34 4.2E-06 CDS 0.02 0 NA NA -0.2 0.01 -0.24 1.3E-08 -0.24 3.2E-08 C21orf59 Hypo-up 0.70 0.0019 CDS 1.27 0.0039 0.64 6.0E-14 0.33 8.2E-06 0.5 1.4E-36 0.25 4.0E-09 MZF1 Hypo-up 0.50 1.9E-52 CDS 1.31 0.0027 0.19 0.0446 0.24 0 0.44 3.3E-28 0.32 9.8E-14 TXLNA Hypo-up 0.74 5.5E-06 3'UTR 1.21 5.6E-16 0.40 1.9E-05 0.21 0.01 0.25 1.9E-09 0.21 1.9E-06 AIM1 Hyper-down 1.49 5.0E-05 CDS 0.43 0 NA NA -0.2 0.01 -0.18 1.6E-05 -0.3 1.6E-12 HLA-F Hyper-down 2.16 0.0007 Intron 0.27 2.8E-11 -0.21 0.0256 -0.32 1.3E-05 -0.37 2.6E-19 -0.33 1.3E-14 RHBDF2 Hyper-down 1.26 0.0046 StopC 1.19 0.028 3'UTR 0.62 0 -0.24 0.0106 -0.35 2.3E-06 -0.23 2.0E-08 -0.35 9.5E-17 SLFN5 Hyper-down 1.23 0.0017 CDS 0.30 0 -0.32 0.0008 -0.17 0.02 -0.39 2.7E-22 -0.12 0.01 CEP78 Hyper-up 1.86 5.9E-07 3'UTR 1.16 0.012 StopC 1.41 0.0015 NA NA 0.36 6.3E-07 0.21 3.0E-07 0.14 0 HIBADH Hyper-up 1.77 0.0012 3'UTR 1.39 9.3E-05 0.54 1.4E-09 0.15 0.04 0.42 9.6E-26 0.13 0 MLF1IP Hyper-up 1.58 0.0028 3'UTR 1.27 0.0022 0.30 0.0018 0.45 4.4E-10 0.3 4.1E-13 0.29 1.7E-11 PPAPDC1B Hyper-up 1.23 1.9E-04 3'UTR 1.69 0 -0.03 0.7421 0.19 0.0136 0.39 1.2E-21 0.28 1.8E-10 RANBP6 Hyper-up 1.77 8.9E-97 CDS 1.43 0 NA NA 0.36 7.7E-07 0.3 2.6E-13 0.32 6.3E-14 S100PBP Hyper-up 1.26 1.9E-07 CDS 1.30 0 0.37 9.3E-05 0.21 0 0.41 1.2E-23 0.24 5.0E-08 ZFP62 Hyper-up 1.37 0.0002 3'UTR 1.14 8E-06 CDS 1.42 0.0003 NA NA 0.37 5.3E-07 0.51 1.2E-38 0.37 7.9E-18 ZNF320 Hyper-up 2.28 0.0014 StopCd 2.33 1E-10 CDS 1.43 6.0E-11 NA NA 0.15 0.05 0.32 3.3E-15 0.27 8.1E-10 ZNF445 Hyper-up 1.21 0.0005 3'UTR 1.32 7E-09 CDS 1.23 0 0.20 0.0373 0.22 0 0.5 3.3E-36 0.26 2.1E-09 ZNF605 Hyper-up 2.33 0.0227 CDS 1.55 1.1E-09 0.31 0.0009 0.16 0.04 0.46 3.8E-31 0.29 3.3E-11 aFold, fold change; Loc, location; 3'UTR, 3' untranslated region; 5'UTR, 5' untranslated region; CDS, coding region; StopC, stop codon; p, p value; r, correlation coefficient. bHypo-down and Hypo-up refer to the gene has a significant decrease in at least one m6A peak and a significant decrease and increase, respectively, in the overall abundance of the mRNA transcript in FTO-overexpressing MonoMac-6 AML cells compared to the control AML cells; Hyper-down and Hyper-up refer to the gene has a significant increase in at least one m6A peak and a significant decrease and increase, respectively, in the overall abundance of the mRNA transcript in FTO-overexpressing MonoMac-6 AML cells compared to the control AML cells. cSix genes have two m6A peaks exhibiting a significant change between the FTO-overexpressing and control MonoMac-6 AML cells. dThe in-house set (n=109) (Huang et al., 2013; Li et al., 2013a; Li et al., 2012b), TCGA set (n=177) (Ley et al., 2013), and GSE14468 set (n=518) (Li et al., 2013a; Taskesen et al., 2011; Wouters et al., 2009) were generated Affymetrix human exon arrays, RNA-seq, Affymetrix U133Plus2.0 arrays, respectively. GSE37642 set (n=562) (Herold et al., 2014; Janke et al., 2014; Li et al., 2013a) were generated by Affymetrix U133Plus2.0 (n=140) and U133A and B (U133A+B) (n=422) arrays. Pearson correlation was used to analyze the correlation between the given gene and FTO in expression in each dataset. NA, expression data is not available in that dataset (the gene was not included in the microarrays).

10  

Table S3 (Related to Figure 5). The patterns of the m6A level and RNA transcript abundance changes of the 11 m6A-Hypo genes (shown in Table S2) in human AML cells with FTO overexpression or knockdown

Gene In FTO-overexpressing MONOMAC-6 cells relative to the control MONOMAC-6 cells

In FTO-knockdown MA9/FLT3-ITD cells relative to the control MA9/FLT3-ITD cells (with expected patterns?)

ASB2 Hypo-down Hyper-Up (Yes) KCNG1 Hypo-down Hyper-Up (Yes) PPARD Hypo-down Hyper-Up (Yes) RAB17 Hypo-down Hyper-Up (Yes) RARA Hypo-down Hyper-Up (Yes) SLC11A1 Hypo-down Hyper-down (No) SLCO4A1 Hypo-down Hyper-up (Yes) TBC1D9 Hypo-down NDa (-) C21orf59 Hypo-up Hyper-down (Yes) MZF1 Hypo-up Hyper-down (Yes) TXLNA Hypo-up Hyper-up (No) aND, with no detectable m6A peaks.

11  

Figure S6 (related to Figure 6). FTO functioned as an oncogene via targeting ASB2 and RARA in AMLs.

(A,B) Quantitative changes of ASB2, RARA and MLL at the protein and mRNA levels in MONOMAC-6 or

NB4 AML cells with forced expression of FTO (or FTO mutant) (A) or FTO knockdown (B), compared to the

corresponding control cells. The relative quantitative information about the protein level changes was derived

from Western blot assays of ASB2, RARA and MLL in MONOMAC-6 or NB4 AML cells with lentivirally

12  

transduced FTO, FTO mutant (H231A and D233A) or control (Ctrl) construct, or with FTO shRNA (shFTO) or

non-specific scrambled shRNA (shNS); the original images of the Western blot were shown in Figure 6A and B.

The mRNA level changes were detected by qPCR. ACTIN was used as the endogenous control for both

Western blot and qPCR assays. (C) Relative expression of RARG and two RARA isoforms in NB4 cells with

forced expression of FTO or FTO-Mut or without (Ctrl), or with knockdown of FTO (shFTO-1) or without

(shNS-1). (D) Relative level of ASB2 and RARA in nucleus and cytoplasm of MONOMAC-6 cell with forced

expression of FTO, FTO-Mut or knockdown of FTO. (E) Western blot assays of ASB2 in MONOMAC-6 AML

cells lentivirally transduced with ASB2 (pmiRNA1+ pLJM1-ASB2), FTO+ASB2 (pmiRNA1-FTO+ pLJM1-

ASB2), FTO (pmiRNA1-FTO+ pLJM1), or control vector (Ctrl; pmiRNA1+ pLJM1). (F, G) Effects of forced

expression of FTO and/or ASB2 on cell growth/proliferation (F) and viability (G) in MONOMAC-6 AML cells.

(H) Western blot assays of RARA in MONOMAC-6 AML cells lentivirally transduced with RARA (pmiRNA1

+ pCDH-RARA), FTO+RARA (pmiRNA1-FTO + pCDH-RARA), FTO (pmiRNAs-FTO + pCDH) or control

vector (Ctrl; pmiRNA1 + pCDH). (I, J) Function of forced expression of FTO and/or RARA on cell

growth/proliferation (I) and viability (J) in MONOMAC-6 cells. *, p<0.05; **, p<0.01; ***, p< 0.001; t-test.

Error bar, mean±SD.

13  

Figure S7 (related to Figure 7). The effects of forced or depleted expression of FTO on the

half-life (t1/2) of ASB2 and RARA mRNA transcripts and the effects of knockdown of m6A

writers on the expression levels of ASB2 and RARA in AML cells. (A,B) Forced and depleted

expression of FTO decreased and increased, respectively, the half-life of ASB2 (A) and RARA

(B) mRNA transcripts in MONOMAC-6 cells. (C) Knockdown of m6A writers, METTL3 or

METTL14, caused down-regulated expression of FTO targets, ASB2 and RARA in AML cells. *,

p<0.05; **, p<0.01; ***, p< 0.001; t-test. Error bar, mean±SD.

14  

SUPPLEMENTAL EXPERIMENTAL PROCEDURES

Leukemic patient samples, cell lines and normal hematopoietic cell samples

The leukemic samples were stored in liquid nitrogen until used. Blasts and mononuclear cells

were purified by use of NycoPrep 1.077A (Axis-Shield, Oslo, Norway). All cell lines were

maintained in the laboratory. Normal CD34+ hematopoietic stem/progenitor cell (HSPC),

CD33+ myeloid cell, and mononuclear cell (MNC) samples were purchased from AllCells, LLC

(Emeryville, CA).

Microarray assays of the 109 human samples.

As described previously (Huang et al., 2013; Li et al., 2013a; Li et al., 2012a), genome-wide

gene expression profiles of a total of 100 human AML with t(11q23)/MLL-rearranged, t(8;21),

inv(16) or t(15;17) and 9 normal control samples (with 3 each of CD34+, CD33+ and MNC cell

samples) were analyzed by use of Affymetrix GeneChip Human Exon 1.0 ST arrays (Affymetirx,

Santa Clara, CA). 1 g total RNA was used for each sample. After hybridization and background

correction according to the standard protocols, the quantified signals were then log2-transformed

and normalized using Robust Multi-array Average (RMA) (Irizarry et al., 2003). The data have

been deposited in the Gene expression Ominibus (GEO) repository with the accession numbers

GSE34184 and GSE30285.

Cell Culture and transfection

K 562 and NB4 cells were grown in RPMI medium 1640 (Invitrogen, Carlsbad, CA) containing

10% FBS (Invitrogen), 1% HEPES and 1% penicillin-streptomycin. MONOMAC-6 cells were

maintained in RPMI 1640 supplemented with 10% FBS, 1% HEPES, 2 mM L-Glutamine,

100×Non-Essential Amino Acid (Invitrogen), 1 mM sodium pyruvate, 9 μg/ml insulin

(Invitrogen) and 1% penicillin-streptomycin. siRNAs (Thermo Scientific, BufferRockford, IL)

and/or plasmids were transfected into MONOMAC-6 cells with Cell Line Nucleofector Kit V

following program T-027 using the Amaxa® Nucleofector® Technology (Amaxa Biosystems,

Berlin, Germany). Experiments were performed 48 hours after transfection.

Apoptosis, cell viability and cell proliferation/growth assays

15  

For apoptosis and cell viability assays, 48 hours after transfection, cells were collected and

seeded with requested concentration, and then apoptosis and viability were assessed using

ApoLive-Glo Multiplex Assay Kit (Promega, Madison, WI) and PE Annexin V apoptosis

Detection Kit 1 (BD Biosciences, San Diego, CA) following the corresponding manufacturer’s

manuals. For cell growth/proliferation assays, 24 hours after transfection, cells were seeded into

96-well plates at the concentration of 10000 cells/ well in triplicates. The cell proliferation was

assessed by MTT (G4000, Promega, Madison, WI) following the manufacturer's instructions.

Briefly, the cells were seeded on 96-well plate at the density of 5000-10000 cells/100 μL, and

then dye solution was added at indicated time points. After incubation at 37°C for 2-4 hours, the

reaction was stopped by adding solubilization/stop. The absorbance at 570nm was read on the

next day.

Chromatin immunoprecipitation (ChIP)-qPCR assays

The procedure for ChIP assays was based on the description in the ChampionChIP One-Day Kit

(Qiagen, Valencia, CA). Briefly, MONOMAC-6 or K-562 cells were crosslinked with

formaldehyde 1% at 37°C for 10 min. The reaction was stopped by adding stop buffer at room

temperature for 5 min. After cell lysis, cross-linked chromatin was sheared using a sonicator.

The 5 μg of anti-FTO (Santa cruze biotechnology, Santa Cruz, CA), anti-MLL N-terminal, anti-

MLL C-terminal, and H3K79me2/3, or control IgG (Abcam, Cambridge, MA) antibodies were

used for immunoprecipitation. After eluting DNA from the precipitated immune complex, DNA

gel electrophoresis or quantitative real-time PCR (qPCR) analysis was performed. For the input

control, 1% of the sonicated DNA was directly purified before immunoprecipitation and

subjected to PCR with the same primers.

Plasmid construction

The wild type FTO-CDS and mutant FTO-CDS (coding region sequence) were amplified from

pcDNA3.1_FTO and pcDNA3.1_mutFTO (the two plasmids were kindly provided by Dr. Chuan

He) by PCR using the following primers: forward 5'-

AGAGCTCTAGAACCACCATGGATTACAAAGATGAC-3' and reverse 5'-

CTAAGATTGCGGCCGCCTAGGGTTTTGCTTCCAGAAGC-3', and then subsequently

cloned into lentivector-based pMIRNA1 (SBI, Mountain View, CA). The RARA-CDS was

16  

transferred from pcDNA Flag-RARalpha, purchased from Addgene (Cambridge, MA), to

lentivector-based pCDH (SBI) with the following primers: forward 5'-TTTTCTAGACCACC

ATGGCCAGCAACAGCAGCT-3' and reverse 5'-TTATGCGGCCGCTCAGGGGA

GTGGGTGG-3'. The ASB2-CDS was amplified from human genomic cDNA with the primers,

forward 5'-TTTTGTCGACCCACCATGGCCACGCAGATCAG-3' and reverse 5'-

CCCTCTAGATTACTGGGTGTTCTCGTATTTCAGGT-3', and directly inserted into pLJM1-

EGFP vector (#19319, Addgene). The pGFP-C-shFTO and pGFP-C-shNS were purchased from

Origene technologies (Rockville, MD). The pTRIPZ shRNA against ASB2 was purchased from

Dharmacon (clone ID: V2THS_97800) and RARA shRNA plasmid was purchased from Santa

Cruz Biotechnology (sc-29465-SH), which have been shown by others to be able to specifically

and effectively targeting ASB2 and RARA, respectively (Marchwicka et al., 2016; Wang et al.,

2012) .

Retrovirus preparation and in vitro colony-forming and replating assay

These assays were conducted as described previously (Huang et al., 2013; Jiang et al., 2012; Li

et al., 2012a; Li et al., 2012c; Li et al., 2013b) with some modifications. Briefly, retroviruses

were produced in 293T cells by co-transfection of individual retroviral construct and the pCL-

Eco packaging vector (IMGENEX, San Diego, CA) with Effectene Transfection Reagent

(Qiagen, Valencia, CA). Bone marrow (BM) cells were collected from 4- to 6-week-old B6.SJL

(CD45.1) mice five days after 5-fluorouracil (5-FU) treatment (150 mg/kg), and BM progenitor

(i.e., lineage negative, Lin-) cells were enriched with the Mouse Lineage Cell Depletion Kit

(Miltenyi Biotec Inc., Auburn, CA). BM progenitor cells were then co-transduced with different

combinations of MSCVneo-based and/or MSCV-PIG-based retroviruses, through two rounds of

“spinoculation”. Thereafter, the retrovirally transduced cells were plated into methylcellulose

medium dishes (with 15,000 cells per dish), supplied with murine recombinant IL-3, IL-6, GM-

CSF and SCF (10 ng/ml for the first three and 30 ng/ml for SCF; R&D Systems, Minneapolis,

MN), along with 1.0 mg/ml of G418 (Gibco BRL, Gaithersburg, MD) and 2.5 g/ml of

puromycin (Sigma, St. Louis, MO). Cultures were incubated at 37C in a humidified atmosphere

of 5% CO2 in air for 7 days. Then, colony cells were collected and replated in methylcellulose

dishes every 7 days with 15,000 cells/dish for 3 passages. Colony numbers were counted and

compared for each passage. In some experiments BM from wildtype or Fto heterozygous

17  

knockout (Fto+/-) (Gao et al., 2010) C57BL/6 (CD45.2) mice was used. Cells were transduced

with MSCVneo-MLL-AF9 through spinoculation and seeded in Methocult M3230

methylcellulose medium as described above with selection of 1.0 mg/mL of G418.

Mouse bone marrow transplantation (BMT; in vivo reconstitution) assays

These assays were conducted as described previously (Huang et al., 2013; Jiang et al., 2012; Li

et al., 2012a; Li et al., 2012c; Li et al., 2013b) with some modifications. Briefly, colony cells

were collected from the colony-forming assays, and after PBS washing twice, the cells were

transplanted via tail vein injection into lethally irradiated (960 cGy, 96 cGy/min, γ-rays) 8- to 10-

week-old C57BL/6 (CD45.2) or B6.SJL (CD45.1) recipient mice. For each recipient mouse, 0.2-

0.3× 106 donor cells from CFA assays and a radioprotective dose of whole bone marrow cells (1

× 106) freshly harvested from a C57BL/6 or B6.SJL (CD45.1) mouse were transplanted.

Leukemic mice were euthanized by CO2 inhalation if they showed signs of systemic illness.

The care and maintenance of animals

C57BL/6 (CD45.2) and B6.SJL (CD45.1) were purchased from the Jackson Lab (Bar Harbor,

ME, USA) or Harlan Laboratories, Inc (Indianapolis, IN, USA). Both male and female mice

were used for the experiments. All laboratory mice were maintained in the animal facility at the

University of Chicago or University of Cincinnati. All experiments on mice in our research

protocol were approved by Institutional Animal Care and Use Committee (IACUC) of the

University of Chicago or University of Cincinnati.

Histolopathology

Leukemic mice were euthanized by CO2 inhalation if they showed signs of systemic illness, and

some negative control recipient mice were also sacrified at same time points (though they did not

develop leukemia) to collect specimens as controls for further analyses. Portions of the spleen

and liver were collected and fixed in formalin, embedded in paraffin, sectioned and stained with

haematoxylin and eosin (H&E) or anti-5hmC antibody (Active Motif, Carlsbad, CA). The

staining was performed by the Tissue Resource Center (TRC) Core Facility at The University of

Chicago. BM Cells were isolated from the tibia and femur. 50,000 cells were washed twice and

were diluted in 200 l of cold MACS Buffer. Each sample was loaded into the appropriate wells

18  

of the cytospin, and then spun at 2000 rpm for 2 min. Blood smear and BM cytospin slides were

stained with Wright-Giemsa.

Lentivirus production, precipitation and infection

Lentivirus for pmiRNA1-FTO, pmiRNA1-FTO-Mut, shFTO-1, shFTO-2, shRARA, pCDH-RARA,

pLJM1-ASB2 as well as their controls were packaged with pMD2.G, pMDLg/pRRE and pRSV-

Rev (purchased from Addgene). Briefly, 0.5 μg pMD2.G, 0.3μg pMDLg/pRRE, 0.7 μgpRSV-

Rev and 1.5 μg construct for overexpression or knockdown of specific genes were co-transfected

into HEK-293T cells in 60mm cell culture dish with Effectene Transfection Reagent (301427,

QIAGEN, Valencia, CA). The pTRIPZ-shASB2 lentivirus was packaged with psPAX2 and

pMD2.G vectors. Briefly, 2.24μg psPAX2, 0.76μg pMD2.G and 1.5μg pTRIPZ-shASB2 were

co-transfected into HEK-293T cells in 60mm dish with Effectene Transfection Reagent. The

lentivirus particles were harvested at 48 and 72 hours after transfection and concentrated with

PEG-itTM Virus Precipitation Solution (# LV810A-1, SBI).Finally, the concentrated lentivirus

particles were directly added into leukemic cells and incubated at 37°C for 24-48 hours before

they were washed out with PBS. For pTRIPZ-shASB2, 1ug/mL Doxycycline (D9891, Sigma, St.

Louis, MO) was added into medium to induce its expression after puromycin selection.

RNA extraction and quantitative RT-PCR analysis

Total RNAs were isolated using the miRNeasy kit (Qiagen, Valencia, CA). For mRNA

expression, 200 ng RNA was reverse-transcribed into cDNA in a total reaction volume of 10 μl

with Qiagen’s RT kit according to the manufacturer’s instructions. Quantitative real-time PCR

analysis was performed with 0.5 μl cDNA using SYBR green PCR master mix (Qiagen,

Valencia, CA) in an AB 7900HT real-time PCR instrument (Applied Biosystems, Foster City,

CA). Gapdh or Actin was used as endogenous control. Each sample was run in triplicate. Nuclear

and cytoplasmic RNA were purified with SurePrep™ Nuclear or Cytoplasmic RNA Purification

Kit (Fisher BioReagents, Fair Lawn, NJ). For quantitative RT-PCR, GAPDH were used as the

endogenous control for cytoplasmic RNA, while 18S RNA was selected as endogenous control

for Nuclear RNA.

19  

Analysis of m6A/A ratio using HPLC-MS/MS

As described previously (Fu et al., 2013), total RNA was isolated from cells using TRIzol®

reagent according to the manufacture procedure. mRNA was then isolated with biotin-oligo(dT)

and streptavidin beads from Promega, following a modified procedure, which eliminated all

heating steps. 8 ug of mRNA was digested with RNase T1 in 20 uL for 15 min at 37 °C,

followed by nuclease P1 (1 U) for 30 min at 37 °C. The solution was diluted 5 times, and 20 μl

of the solution was analyzed by HPLC-QQQ-MS/MS. The nucleosides were quantified using the

nucleoside to base ion mass transitions of 282.1 to 150.1 in MRM positive ESI mode.

m6A dot blot assay

Total RNA was isolated from different cells with miRNeasy Mini Kit (QIAGEN, 217004)

according to the manufacturer's instructions and quantified by UV spectrophotometry. The m6A

dot blot assay was performed following a published protocol (Rapley and Manning, 1998) with

some modifications (Jia et al., 2011). Briefly, the RNA samples were loaded to the Amersham

Hybond-N+ membrane (RPN119B, GE Healthcare, ) with a Bio-Dot Apparatus (#170-6545,

Bio-Rad) and UV crosslinked to the membrane. Then the membrane was blocked with 5%

nonfat dry milk (in 1X PBST) for 1-2 hours and incubated with a specific anti-m6A antibody

(1:2000 dilution, Synaptic Systems, 202003) overnight at 4°C. Then the HRP-conjugated goat

anti-rabbit IgG (sc-2030, Santa Cruz Biotechnology) was added to the blots for 1 hour at room

temperature and the membrane was developed with Amersham ECL Prime Western Blotting

Detection Reagent (RPN2232, GE Healthcare). The relative signal density of each dot was

quantified by Gel-Pro analyzer software (Media Cybernetics) in all experiments.

RNA-seq and m6A-seq assays and data analysis

The m6A-seq procedure was performed as described previously (Dominissini et al., 2013).

Polyadenylated RNA was extracted using FastTrack MAG Maxi mRNA isolation kit (Life

technology). RNA fragmentation Reagents (Ambion) was used to randomly fragment RNA. The

specific anti-m6A antibody (Synaptic Systems) was applied for m6A pull down (i.e., m6A IP).

Both input and m6A IP samples were prepared for next-generation sequencing (NGS). The NGS

library preparation was constructed by TruSeq Stranded mRNA Sample Prep Kit (Illumina) and

was quantified by BioAnalyzer High Sensitivity DNA chip, and then was deeply sequenced on

20  

the Illumina HiSeq 2500. The data have been deposited in the GEO repository with the accession

numbers GSE76414

(http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=avcxukwiztyntud&acc=GSE76414), GSE84944

(http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=mforsiggdtetdah&acc=GSE84944 ), and

GSE85008 ( http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=kbiduyasxlstrib&acc=GSE85008 ). 

For the data analysis, the following pipeline was used to identify m6A peaks. The reads from

mRNA input (RNA-seq) and m6A IP (m6A-seq) sequencing libraries were aligned to hg19/hg38

reference genome using Tophat (Kim et al., 2013). Both MACS/MACS2 (Zhang et al., 2008)

and exomePeak (Meng et al., 2013) were used to call m6A peaks based on the RNA-seq and

m6A-seq bam files. To achieve high specificity, only the m6A peaks called by both

MACS/MACS2 and exomePeak were retained for the further analysis. The m6A peaks were

annotated using an ad hoc perl script. The differentially methylated m6A peaks were identified

according to the procedure described by Schwartz et al (Schwartz et al., 2013). Sequence motifs

enriched in m6A peak regions compared to control regions were identified using DREME

(Bailey, 2011). The RNA-seq reads were normalized using RSEM method (Li and Dewey, 2011)

and Cufflinks (Trapnell et al., 2010). EBSeq (Leng et al., 2013) and Cuffdiff (Trapnell et al.,

2013) were employed to find differentially expressed genes between FTO-overexpressing and

control AML cell lines. DAVID tool (Rapley and Manning, 1998) was used to perform pathway

enrichment analysis.

Gene-specific m6A qPCR

Real-time quantitative PCR (qPCR) was performed to assess the relative abundance of the

selected mRNA in m6A antibody IP samples and input samples between the FTO-overexpressing,

shFTO cell lines and their control cell lines. Here the house-keeping gene HPRT1 was chose as

internal control according to the reason that HPRT1 mRNA did not have m6A peaks from the

m6A profiling data (Wang et al., 2014). Total RNA was isolated by RNeasy kit (QIAGEN) with

additional DNase I on-column digestion then followed polyadenylated RNA extraction using

Dynabeads® mRNA Purification Kit (Life technology). While 50 ng mRNA was saved as input

sample, the rest mRNA was used for m6A-immunoprecipitation as described above in m6A-seq

procedures to obtain m6A pull down portion (m6A IP portion). Briefly, nearly 2 ug mRNA was

21  

incubated with m6A antibody (Synaptic Systems) and diluted into 500 uL IP buffer (150 mM

NaCl, 0.1% NP-40, 10 mM Tris, pH 7.4, 100 U RNase inhibitor). The mixture was rotated at 4

C for 2 hours, following Dynabeads® Protein A (ThermoFisher Scientific) was added into the

solution and rotated for another 2 hours at 4 C. After four times washing by IP buffer, the m6A

IP portion was eluted twice by 50 uL m6A-elute buffer (IP buffer, 6.7mM m6A, 30 U RNase

inhibitor) with incubating and shaking at 4 C for 1 hour. Finally, m6A IP mRNA was recovered

by ethanol precipitation and the RNA concentration was measured with Qubit® RNA HS Assay

Kit (ThermoFisher Scientific). RT-qPCR was performed in Roche LightCycle® 96 system

(Roche) by using the Verso SYBR Green 1-Step qRT-PCR Low ROX Kit (ThermoFisher

Scientific) from 1 ng m6A IP mRNA or input mRNA as the template. The mRNA expression

was determined by the number of amplification cycle (Cq). The relative mRNA expression was

calculated by the value of Cq in m6A IP portion divide by the value of Cq in input portion

(CqIP/Cqinput). And the gene-specific m6A level was determined by the target mRNA relative

expression level between the FTO-overexpressing cell lines or shFTO cell lines with their

control cell lines.

RT-PCR primer used:

ASB2: 5'-CGTGGTGCAGTTCTGTGAGT-3'; 5'-GTGAGCCAGAGGTCTTGGAG-3'

RARA: 5'-CCAGCTCCAACAGAAGCAG-3'; 5'-AGGCCTCTGTCCAAGGAGTC-3'

PPARD: 5'-ACGACATCGAGACATTGTGG-3'; 5'-GGTAACCTGGTCGTTGAGGA-3'

SLC11A1: 5'-GCGAGGTCTGCCATCTCTAC-3'; 5'-GTGTCCACGATGGTGATGAG-3'

TXLNA: 5'-TCCAAAAGCAGCGAGGTATT-3'; 5'-TCCAGCCGTTGGATTTTTAC-3'

RNA stability assays

The mRNA stability in vivo is normally reported as the time for degrading half of the existing

mRNA molecules. FTO-overexpressing cell lines, shFTO cell lines and their control cell lines

were cultured in 6-well plates. Then actinomycin D was added to 5 g/mL at 6 hours, 3 hours

and 0 hour before cell scraping collection. Total RNA was isolated by RNeasy kit (QIAGEN)

with additional DNase I on-column digestion. RT-PCR was conducted to quantify the relative

levels of target mRNA in Roche LightCycle® 96 system (Roche) by using the Verso SYBR

Green 1-Step qRT-PCR Low ROX Kit (ThermoFisher Scientific) from 50 ng total RNA. The

22  

degradation rate of target mRNA was estimated according with the previous published paper

(Liu et al., 2014). With the transcriptional inhibitor actinomycin D, the mRNA transcription was

turned off and the degradation rate of RNA (Kdecay) was estimated by following equation:

In(C/C0)=-Kdecayt

Where C0 is the concentration of mRNA at time 0 hour, which means the mRNA concentration

before decay starts. And t is the transcription inhibition time, meanwhile C is the mRNA

concentration at the time t. Thus the Kdecay can be derived by the exponential decay fitting of

C/C0 versus time t. The half-time (t1/2), which means C/C0=50%/100%=1/2, can be calculated by

the following equation:

In(1/2)= -Kdecayt1/2

Rearrangement of the above equation leads to the mRNA half-life time value, t1/2= In2/Kdecay.

Immunoblotting (Western blot)

Cells were washed twice with ice-cold phosphate-buffered saline (PBS) and ruptured with RIPA

buffer (Pierce, Rockford, IL) containing 5 mM EDTA, PMSF, cocktail inhibitor, and

phosphatase inhibitor cocktail. Cell extracts were microcentrifuged for 20 min at 10000 x g and

supernatants were collected. Cell lysates (20 l) were resolved by SDS-PAGE and transferred

onto PVDF membranes. Membranes were blocked for 1 hour with 5% skim milk in Tris-

buffered saline containing 0.1% Tween 20 and incubated overnight at 4 °C with anti-Fto

antibody (Millipore, Billerica, MA), anti-Actb antibody (Santa Cruz Biotechnology Inc., Santa

Cruz, CA), anti-FTO (#14386, Cell Signal Technology), anti-β-Actin (3700S, Cell Signal

Technology), anti-GAPDH (Santa Cruz), anti-ASB2 (ab13710, Abcam), anti-RARA (616801,

BioLegend) or anti-MLL1 (#14689, Cell Signaling Technology). Membranes were washed 30

min with Tris-buffered saline containing 0.1% Tween 20, incubated for 1 h with appropriate

secondary antibodies conjugated to horseradish peroxidase, and developed using

chemiluminescent substrates.

Flow cytomety analysis

Flow cytometry analysis of mouse BM cells were conducted as described previously (Huang et

al., 2013; Jiang et al., 2012; Li et al., 2012a; Li et al., 2012c; Li et al., 2008) with some

modifications. For in vitro ATRA induction experiments, MONOMAC-6 or NB4 cells were

23  

induced into myeloid differentiation with 100nM or 500nM ATRA (223018, Sigma-Aldrich),

harvested at indicated time points, washed with chilled PBS and stained with PE-labeled anti-

CD11b (101208, BioLegend) and APC-labeled anti-CD14 (17-0149-41, eBioscience) for flow

cytometry analysis. Cells from BM of transplanted mice were harvested for analysis of

immunophenotypes. After washing with PBS, blocking nonspecific binding with affinity-

purified anti–mouse CD16/32 (eBioscience), cells were stained at 4°C with various antibodies

diluted in Flow Cytometry Staining Buffer (eBioscience) for 30 minutes. Subsequently, cells

were washed with PBS and resuspended in IC Fixation Buffer (eBioscience) for flow cytometry

analysis. Antibodies (eBioscience) used for our flow cytometric analysis -include anti–mouse

PE-CD11b (Mac-1; 12-0112), anti–mouse APC-CD117 (17-1172; c-kit), PE anti–mouse CD45.1

(12-0453), and anti–mouse APC-CD45.2 (17-0454).

Dual-Luciferase reporter and mutagenesis assays

The DNA fragments of ASB2-3'UTR, RARA-3'UTR and RARA-5'UTR containing the wild type

m6A motifs as well as mutant motifs (m6A was replaced by T) were directly synthesized from

Integrated DNA Technologies (Coralville, Iowa). Wild type and mutant ASB2-3'UTR and

RARA-3'UTR were inserted into downstream of firefly luciferase of pMIR-REPORT vector

(Luciferase miRNA Expression Reporter Vector, Ambion); while the wild type and mutant

RARA-5'UTR were subcloned into upstream of firefly luciferase of pGL3-basic vector (Promega,

Madison, WI). For dual-luciferase reporter assay, 100ng wild-type or mutant ASB2-3'UTR (or

RARA-3'UTR, or RARA-5'UTR), 200ng pMIRNA1-FTO (or pMIRNA1-FTO-Mut or

pMIRNA1), and 20ng pRL-TK (renilla luciferase control reporter vector) were co-transfected

into HEK-293T cells in 24-well plate. The relative luciferase activities were accessed 48 hours

post transfection by Dual-Luciferase Reporter Assay System (Promega). Each group was

repeated in triplicate.

(1) ASB2-3'UTR with wild-type m6A sites: ATGACTAGACTAGTGGCCACGGGGAGAGAGGAGTAGCCCCTCAGACTCTTCTTACTAAGTCTCAGGACGTCGGTGTTCCCAACTCCAAGGGGACCTGGTGACAGACGAGGCTGCAGGCTGCCTCCCTCTCAGCCTGGACAGCTACCAGGATCTCACTGGGTCTCAGGGCCCAGAGCTTTGGCCAGAGCAGAGAACAGAATGTGTCAAGGAGAAGAATCATTTGTTTACAAACTGATGAGCAGATCCCAGACCTTCTCTACCTTCAGGAATGGCAGAAACCTCTATTCCTGGGGCCAGGGCAGAGCTTGAGGTGTTCTGGGGAAGGTGGTGCTCAGAGCCTTCCCTGTGCCCCTCCACTTGTTCTGGAAAACTCACCACTTGACTTCAGAGCTTTCTCTCCAAAGACTAAGATGAAGACGTGGCCCAAGGTAGGGGGTAGGGGGAGCCTGGAAGCTTGGGATCTA

24  

ASB2-3'UTR with mutant m6A sites: ATGACTAGACTAGTGGCCACGGGGAGAGAGGAGTAGCCCCTCAGTCTCTTCTTACTAAGTCTCAGGACGTCGGTGTTCCCAACTCCAAGGGGACCTGGTGACAGACGAGGCTGCAGGCTGCCTCCCTCTCAGCCTGGTCAGCTACCAGGATCTCACTGGGTCTCAGGGCCCAGAGCTTTGGCCAGAGCAGAGAACAGAATGTGTCAAGGAGAAGAATCATTTGTTTACAATCTGATGAGCAGATCCCAGACCTTCTCTACCTTCAGGAATGGCAGAAACCTCTATTCCTGGGGCCAGGGCAGAGCTTGAGGTGTTCTGGGGAAGGTGGTGCTCAGAGCCTTCCCTGTGCCCCTCCACTTGTTCTGGAAAACTCACCACTTGACTTCAGAGCTTTCTCTCCAAAGTCTAAGATGAAGACGTGGCCCAAGGTAGGGGGTAGGGGGAGCCTGGAAGCTTGGGATCTA

(2) RARA-3'UTR with wild-type m6A sites: ATGACTAGACTAGTCGACCATGTGACCCCGCACCAGCCCTGCCCCCACCTGCCCTCCCGGGCAGTACTGGGGACCTTCCCTGGGGGACGGGGAGGGAGGAGGCAGCGACTCCTTGGACAGAGGCCTGGGCCCTCAGTGGACTGCCTGCTCCCACAGCCTGGGCTGACGTCAGAGGCCGAGGCCAGGAACTGAGTGAGGCCCCTGGTCCTGGGTCTCAGGATGGGTCCTGGGGGCCTCGTGTTCATCAAGACACCCCTCTGCCCAGCTCACCACATCTTCATCACCAGCAAACGCCAGGACTTGGCTCCCCCATCCTCAGAACTCACAAGCCATTGCTCCCCAGCTGGGGAACCTCAACCTCCCCCCTGCCTCGGTTGGTGACAGAGGGGGTGGGACAGGGGCGGGGGGTTCCCCCTGTACATACCCTGCCATACCAACCCCAGGTATTAATTCTCGCTGGAAGCTTGGGATCTA

RARA-3'UTR with mutant m6A sites: ATGACTAGACTAGTCGACCATGTGACCCCGCACCAGCCCTGCCCCCACCTGCCCTCCCGGGCAGTACTGGGGACCTTCCCTGGGGGACGGGGAGGGAGGAGGCAGCGACTCCTTGGTCAGAGGCCTGGGCCCTCAGTGGTCTGCCTGCTCCCACAGCCTGGGCTGACGTCAGAGGCCGAGGCCAGGATCTGAGTGAGGCCCCTGGTCCTGGGTCTCAGGATGGGTCCTGGGGGCCTCGTGTTCATCAAGACACCCCTCTGCCCAGCTCACCACATCTTCATCACCAGCAAACGCCAGGTCTTGGCTCCCCCATCCTCAGATCTCACAAGCCATTGCTCCCCAGCTGGGGAACCTCAACCTCCCCCCTGCCTCGGTTGGTGACAGAGGGGGTGGGACAGGGGCGGGGGGTTCCCCCTGTACATACCCTGCCATACCAACCCCAGGTATTAATTCTCGCTGGAAGCTTGGGATCTA

(3) RARA-5'UTR with wild-type m6A sites: ATGACCGGGGTACCGCCAGCACACACCTGAGCAGCATCACAGGACATGGCCCCCTCAGCCACCTAGCTGGGGCCCATCTAGGAGTGGCATCTTTTTTGGTGCCCTGAAGGCCAGCTCTGGACCTTCCCAGGAAAAGTGCCAGCTCACAGAACTGCTTGACCAAAGGACCGGCTCTTGAGACATCCCCCAACCCACCTGGCCCCCAGCTAGGGTGGGGGCTCCAGGAGACTGAGATTAGCCTGCCCTCTTTGGACAGCAGCTCCAGGACAGGGCGGGTGGGCTGACCACCCAAACCCCATCTGGGCCCAGGCCCCATGCCCCGAGGAGGGGTGGTCTGAAGCCCACCAGAGCCCCCTGCCAGACTGTCTGCCTCCCTTCTGACTCGAGCGGATCTA

RARA-5'UTR with mutant m6A sites: ATGACCGGGGTACCGCCAGCACACACCTGAGCAGCATCACAGGACATGGCCCCCTCAGCCACCTAGCTGGGGCCCATCTAGGAGTGGCATCTTTTTTGGTGCCCTGAAGGCCAGCTCTGGTCCTTCCCAGGAAAAGTGCCAGCTCACAGATCTGCTTGACCAAAGGTCCGGCTCTTGAGACATCCCCCAACCCACCTGGCCCCCAGCTAGGGTGGGGGCTCCAGGAGACTGAGATTAGCCTGCCCTCTTTGGTCAGCAGCTCCAGGTCAGGGCGGGTGGGCTGACCACCCAAACCCCATCTGGGCCCAGGCCCCATGCCCCGAGGAGGGGTGGTCTGAAGCCCACCAGAGCCCCCTGCCAGTCTGTCTGCCTCCCTTCTGACTCGAGCGGATCTA

Luciferase reporter-related gene specific m6A qPCR

To access whether m6A presents on the luciferase reporter constructs and further confirm

whether FTO participates in regulating m6A modification of the constructs bearing ASB2 and

RARA 3'UTR, we conducted luciferase reporter-related gene specific m6A qPCR. Firstly, 1.5ug

luciferase reporter constructs, including ASB2-3'UTR-WT, ASB2-3'UTR-Mut, RARA-3'UTR-WT

25  

and RARA-3'UTR-Mut, were transfected into HEK-293T cells with pmiRNA1-FTO, pmiRNA1-

FTO-Mut or pmiRNA1 in 100mm cell culture dishes. And 48 hours after transfection, samples

were collected, RNAs were extracted, m6A RNA immunoprecipitation was performed with

Magna MeRIP m6A kit (17-10499, Millipore Sigma, Temecula, CA) and qPCR was conducted to

assess m6A modification levels with primers covering the joints between inserted ASB2 and

RARA 3'UTR and pMIR-REPORT vector.

qPCR primers used:

ASB2_WT and Mut_Primer 1_Forward: 5'-GCGGAAAGTCCAAATTGCTC-3'

ASB2_WT_Primer_1_Reverse: 5'-GTAAGAAGAGTCTGAGGGGCTA-3'

ASB2_Mut_Primer_1_Reverse: 5'- GTAAGAAGAGACTGAGGGGCTA-3'

ASB2_WT_Primer 2_Forward: 5'- TCCAAAGACTAAGATGAAGACGTGG-3'

ASB2_Mut_Primer 2_Forward: 5'- TCCAAAGTCTAAGATGAAGACGTGG-3'

ASB2_WT and Mut_Primer_2_Reverse: 5'-CGCATACAAAAAACCAACACACA-3'

RARA_WT and Mut_Primer 1_Forward: 5'- AAAGTCCAAATTGCTCGAGTGAT-3'

RARA_WT_Primer_1_Reverse: 5'-TGGGAGCAGGCAGTCCACTG-3'

RARA_Mut_Primer_1_Reverse: 5'- TGGGAGCAGGCAGACCACTG-3'

RARA_WT_Primer 2_Forward: 5'- CCCCCATCCTCAGAACTCACAA-3'

RARA_Mut_Primer 2_Forward: 5'- CCCCCATCCTCAGATCTCACAA-3'

RARA_WT and Mut_Primer_2_Reverse: 5'- TAAGCTTCCAGCGAGAATTAATACC-3'

26  

SUPPLEMENTAL REFERENCES:

Bailey, T. L. (2011). DREME: motif discovery in transcription factor ChIP-seq data. Bioinformatics 27, 1653-1659.

Fu, Y., Jia, G., Pang, X., Wang, R. N., Wang, X., Li, C. J., Smemo, S., Dai, Q., Bailey, K. A., Nobrega, M. A., et al. (2013). FTO-mediated formation of N6-hydroxymethyladenosine and N6-formyladenosine in mammalian RNA. Nat Commun 4, 1798.

Herold, T., Metzeler, K. H., Vosberg, S., Hartmann, L., Rollig, C., Stolzel, F., Schneider, S., Hubmann, M., Zellmeier, E., Ksienzyk, B., et al. (2014). Isolated trisomy 13 defines a homogeneous AML subgroup with high frequency of mutations in spliceosome genes and poor prognosis. Blood 124, 1304-1311.

Irizarry, R. A., Bolstad, B. M., Collin, F., Cope, L. M., Hobbs, B., and Speed, T. P. (2003). Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res 31, e15.

Janke, H., Pastore, F., Schumacher, D., Herold, T., Hopfner, K. P., Schneider, S., Berdel, W. E., Buchner, T., Woermann, B. J., Subklewe, M., et al. (2014). Activating FLT3 mutants show distinct gain-of-function phenotypes in vitro and a characteristic signaling pathway profile associated with prognosis in acute myeloid leukemia. PLoS ONE 9, e89560.

Kim, D., Pertea, G., Trapnell, C., Pimentel, H., Kelley, R., and Salzberg, S. L. (2013). TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol 14, R36.

Leng, N., Dawson, J. A., Thomson, J. A., Ruotti, V., Rissman, A. I., Smits, B. M., Haag, J. D., Gould, M. N., Stewart, R. M., and Kendziorski, C. (2013). EBSeq: an empirical Bayes hierarchical model for inference in RNA-seq experiments. Bioinformatics 29, 1035-1043.

Ley, T. J., Miller, C., Ding, L., Raphael, B. J., Mungall, A. J., Robertson, A., Hoadley, K., Triche, T. J., Laird, P. W., Baty, J. D., et al. (2013). Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med 368, 2059-2074.

Li, B., and Dewey, C. N. (2011). RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12, 323.

Li, Z., Herold, T., He, C., Valk, P. J., Chen, P., Jurinovic, V., Mansmann, U., Radmacher, M. D., Maharry, K. S., Sun, M., et al. (2013a). Identification of a 24-gene prognostic signature that improves the European LeukemiaNet risk classification of acute myeloid leukemia: an international collaborative study. J Clin Oncol 31, 1172-1181.

Li, Z., Huang, H., Chen, P., He, M., Li, Y., Arnovitz, S., Jiang, X., He, C., Hyjek, E., Zhang, J., et al. (2012a). miR-196b directly targets both HOXA9/MEIS1 oncogenes and FAS tumour suppressor in MLL-rearranged leukaemia. Nat Commun 2, 688.

Li, Z., Huang, H., Li, Y., Jiang, X., Chen, P., Arnovitz, S., Radmacher, M. D., Maharry, K., Elkahloun, A., Yang, X., et al. (2012b). Up-regulation of a HOXA-PBX3 homeobox-gene

27  

signature following down-regulation of miR-181 is associated with adverse prognosis in patients with cytogenetically-abnormal AML. Blood 119, 2314-2324.

Li, Z., Huang, H., Li, Y., Jiang, X., Chen, P., Arnovitz, S., Radmacher, M. D., Maharry, K., Elkahloun, A., Yang, X., et al. (2012c). Up-regulation of a HOXA-PBX3 homeobox-gene signature following down-regulation of miR-181 is associated with adverse prognosis in patients with cytogenetically abnormal AML. Blood 119, 2314-2324.

Li, Z., Lu, J., Sun, M., Mi, S., Zhang, H., Luo, R. T., Chen, P., Wang, Y., Yan, M., Qian, Z., et al. (2008). Distinct microRNA expression profiles in acute myeloid leukemia with common translocations. Proc Natl Acad Sci USA 105, 15535-15540.

Li, Z., Zhang, Z., Li, Y., Arnovitz, S., Chen, P., Huang, H., Jiang, X., Hong, G. M., Kunjamma, R. B., Ren, H., et al. (2013b). PBX3 is an important cofactor of HOXA9 in leukemogenesis. Blood 121, 1422-1431.

Marchwicka, A., Cebrat, M., Laszkiewicz, A., Sniezewski, L., Brown, G., and Marcinkowska, E. (2016). Regulation of vitamin D receptor expression by retinoic acid receptor alpha in acute myeloid leukemia cells. J Steroid Biochem Mol Biol 159, 121-130.

Rapley, R., and Manning, D. L. (1998). RNA Isolation and Characterization Protocols. In, (Totowa, NJ: Humana Press Inc.), pp. 1-261.

Schwartz, S., Agarwala, S. D., Mumbach, M. R., Jovanovic, M., Mertins, P., Shishkin, A., Tabach, Y., Mikkelsen, T. S., Satija, R., Ruvkun, G., et al. (2013). High-resolution mapping reveals a conserved, widespread, dynamic mRNA methylation program in yeast meiosis. Cell 155, 1409-1421.

Taskesen, E., Bullinger, L., Corbacioglu, A., Sanders, M. A., Erpelinck, C. A., Wouters, B. J., van der Poel-van de Luytgaarde, S. C., Damm, F., Krauter, J., Ganser, A., et al. (2011). Prognostic impact, concurrent genetic mutations, and gene expression features of AML with CEBPA mutations in a cohort of 1182 cytogenetically normal AML patients: further evidence for CEBPA double mutant AML as a distinctive disease entity. Blood 117, 2469-2475.

Trapnell, C., Hendrickson, D. G., Sauvageau, M., Goff, L., Rinn, J. L., and Pachter, L. (2013). Differential analysis of gene regulation at transcript resolution with RNA-seq. Nature biotechnology 31, 46-53.

Trapnell, C., Williams, B. A., Pertea, G., Mortazavi, A., Kwan, G., van Baren, M. J., Salzberg, S. L., Wold, B. J., and Pachter, L. (2010). Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nature biotechnology 28, 511-515.

Wouters, B. J., Lowenberg, B., Erpelinck-Verschueren, C. A., van Putten, W. L., Valk, P. J., and Delwel, R. (2009). Double CEBPA mutations, but not single CEBPA mutations, define a subgroup of acute myeloid leukemia with a distinctive gene expression profile that is uniquely associated with a favorable outcome. Blood 113, 3088-3091.