Identification of A Novel Small-Molecule Binding Site of the Fat Mass and Obesity Associated Protein (FTO)

Article abstract of DOI:10.1021/acs.jmedchem.5b00702

N-(5-Chloro-2,4-dihydroxyphenyl)-1-phenylcyclobutanecarboxamide (N-CDPCB, 1a) is found to be an inhibitor of the fat mass and obesity associated protein (FTO). The crystal structure of human FTO with 1a reveals a novel binding site for the FTO inhibitor and defines the molecular basis for recognition by FTO of the inhibitor. The identification of the new binding site offers new opportunities for further development of selective and potent inhibitors of FTO, which is expected to provide information concerning novel therapeutic targets for treatment of obesity or obesity-associated diseases.

Full text of DOI:10.1021/acs.jmedchem.5b00702

Article
pubs.acs.org/jmc
Identification of A Novel Small-Molecule Binding Site of the Fat Mass
and Obesity Associated Protein (FTO)
Wu He,†,▽ Bin Zhou,‡,§,▽ Weijia Liu,‡,§,▽ Meizi Zhang,# Zhenhua Shen,† Zhifu Han,‡,§ Qingwei Jiang,†
Qinghua Yang,†,⊥ Chuanjun Song,† Ruiyong Wang,† Tianhui Niu,‡,§ Shengna Han,∥ Lirong Zhang,∥
Jie Wu,† Feima Guo,# Renbin Zhao,# Wenquan Yu,† Jijie Chai,*,‡,§ and Junbiao Chang*,†,⊥
†College of Chemistry and Molecular Engineering, Zhengzhou University, Zhengzhou 450001, PR China
‡School of Life Sciences, Tsinghua University, Beijing 100084, PR China
§Tsinghua-Peking Center for Life Sciences, Beijing 100084, PR China
∥Basic Medical College, Zhengzhou University, Zhengzhou 450001, PR China
⊥Collaborative Innovation Center of New Drug Research and Safety Evaluation, Henan Province, Zhengzhou 450001, PR China
#Space Biology Research and Technology Center, Engineering Research Center of Space Biology, China Academy of Space
Technology, Beijing 100190, PR China
S
* Supporting Information
ABSTRACT: N-(5-Chloro-2,4-dihydroxyphenyl)-1-phenylcyclobutanecarboxa-
mide (N-CDPCB, 1a) is found to be an inhibitor of the fat mass and obesity
associated protein (FTO). The crystal structure of human FTO with 1a reveals a
novel binding site for the FTO inhibitor and defines the molecular basis for
recognition by FTO of the inhibitor. The identification of the new binding site
offers new opportunities for further development of selective and potent inhibitors
of FTO, which is expected to provide information concerning novel therapeutic
targets for treatment of obesity or obesity-associated diseases.
obesity,11 but whether it is a valid drug target remains
1. INTRODUCTION
unknown. FTO-specific inhibitors will be critical not only to
In mammals, the Fe(II)/α-ketoglutarate (αKG)-dependent
answer this question but also to probe the biology of FTO.12
dioxygenases of the AlkB family function as critical mediators
FTO inhibitors such as αKG analogues and other nonspecific
of nucleic acid methylation and play pivotal roles in regulating
inhibitors have been reported,12,13 and structural studies
development and diseases.1 Originally identified in Escherichia
indicated that most of these inhibitors target the conserved
coli,2,3 AlkB members possess both DNA and RNA
3meU- or αKG-recognizing site of FTO. Identification of
demethylation activities. Subsequent studies have identified
nonconserved small-molecule binding sites of FTO is expected
homologues of AlkB members with similar activities in
to offer new opportunities for the design of inhibitors with
mammals. The fat mass and obesity associated protein
more specificity and potency.
(FTO) was predicted and biochemically confirmed to belong
to the AlkB family of proteins with (αKG)-dependent single-
2. CHEMISTRY
stranded (ss) DNA/RNA demethylases which prefer 3meA/
3meU-containing substrates.4 Our previous structural study
The synthesis of compound 1 (Scheme 1) started from
substituted 2-phenylacetonitrile 2. Reaction of 2 with 1,4-
dibromobutane gave intermediate 3 by introducing a four-
member ring (B ring). Hydrolysis of 3 by potassium hydroxide
formed acid 4, coupling of which with 5-chloro-2,4-dimethox-
yaniline followed by demethylation with boron tribromide
(BBr3) afforded products 1a−t bearing mono- or disubstituents
showed that the L1 loop unique to FTO enables it to
distinguish single stranded DNA and RNA from double-
stranded nucleic acids,5 a mechanism for the selection of
ssDNA/RNA substrates that appears to be conserved in other
AlkB members.6−9 N6-Methyladenosine (m6A) in both DNA
and RNA was later found to be another substrate of FTO but
with 50-fold preference for the enzyme as compared to
3meU.10 It is currently still not entirely clear whether one or
both of these modified bases, m6A and 3meU, are in vivo
substrate(s) of FTO. Numerous studies have linked FTO to
1
on the A rings. Their structures were fully characterized by H
Received: May 6, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
NMR, 13C NMR, and HRMS, and analytical HPLC indicated
that all of the tested compounds possessed a purity of at least
95%.
Scheme 1. Synthesis of Compounds 1
3. RESULTS AND DISCUSSION
3.1. Identification of N-(5-Chloro-2,4-dihydroxyphen-
yl)-1-phenylcyclobutanecarboxamide (N-CDPCB, 1a) as
a Novel Small-Molecule FTO Inhibitor. Initial virtual
screening identified many derivatives of o-catechol (1,2-
benzenediol) that exhibit strong inhibition of demethylation
of 3meU by FTO, but these compounds are strong iron
chelators14 and are probably nonselective. In the crystal
structure of FTO complexed with the mononucleotide 3meT,
two carbonyl oxygen atoms in FTO which form hydrogen
a
(a) Br(CH2)4Br, KOH, DMSO, 0 °C ∼ rt; (b) KOH, glycol, reflux;
(c) 5-chloro-2,4-dimethoxyaniline, EDCl, DMAP, CH2Cl2, rt; (d)
BBr3, CH2Cl2, − 78 °C ∼ rt.
Table 1. Inhibitory Activity of Compounds 1 on FTO Demethylation of the Substrate 3meT
a
Demethylation ratio by FTO (%) = AT peak/(AT peak + A3meT peak) × 100%.
B
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
bonds with 3meT are critical for binding of the nucleotide to
FTO. We reasoned that derivatives of benzene-1,3-diol may
mimic the mononucleotide and thereby inhibit FTO activity.
To identify selective FTO inhibitors, we designed and
synthesized 20 compounds containing this structural motif
and tested their effect on the demethylation activity of FTO
using the 3meT substrate. The FTO activity was measured as
described previously.5 In the absence of the inhibitor (control
group), 71.9% of the substrate was demethylated by FTO after
incubation at 16 °C for 17 h. Addition of compound 1a at 80
μM significantly inhibited the demethylation ratio, with only
14.4% of the substrate demethylated. Most analogues of 1a
bearing substituted A rings also showed inhibitory activity on
FTO. The SAR study indicated that introduction of the
chloro-/bromo-group to the A ring (1b−g) slightly decreased
the inhibitory activity of the compound. Analogues with the
halogen group at the ortho-/meta-position showed higher
activity than those at the para-position (1b−c vs 1d; 1e−f vs
1g). Incorporation of strong electron-withdrawing groups (e.g.,
CF3 and NO2, 1h−i) or electron-donating groups (1j−n) to
the meta- or para-position of the A ring compromised the
inhibitory activity of 1a to different degrees (Table 1). Among
the disubstituted derivatives, 2,6-difluoro (1o) and 2,6-dichloro
analogues (1p) displayed similar inhibitory activity on FTO to
1a. Interestingly, the β-naphthyl substituted derivative (1t) was
more potent than its α-isomer (1s). Overall, ten analogues
(1a−c, 1f, 1j, 1m−p, and 1t) possessing potent inhibitory
activity on FTO (demethylation ratio of 3meT: < 20%) were
identified, with the parent compound 1a being the optimal one.
It was also successfully cocrystallized with FTO (see section
3.4), and we therefore focus on the complex structure in the
following discussions.
that it is a competitive or allosteric inhibitor of FTO. To further
validate the interaction between FTO and compound 1a, we
determined its binding affinity using the isothermal titration
calorimetry (ITC) technique. The ITC results showed that it
interacts with FTO with a dissociation constant of ∼13 μM
(Figure 2), supporting the biochemical assay showing that
compound 1a displays strong inhibitory activity of demethy-
lation of the ssRNA by FTO.
3.2. Inhibitory Activity on FTO Demethylation of the
15-mer ssRNA. To further verify the inhibitory effect of 1a on
the enzymatic activity of FTO, we used the 15-mer ssRNA (5′-
CUUGUCA(m6A)CAGCAGA-3′) as the substrate and de-
tected the demethylation activity of FTO in the presence of 1a
as described previously.10 As shown in Figure 1, this compound
possesses an IC50 of 4.95 μM for the reduction of FTO
demethylation of the ssRNA (Figure 1) under the conditions
tested. The dose-dependent inhibition of FTO by 1a suggests
Figure 2. Measurement of the binding affinity between FTO and
compound 1a by isothermal titration calorimetry (ITC). Data fitting
revealed a binding dissociation constant of ∼13 μM between
compound 1a and FTO.
3.3. Inhibition by Compound 1a of Demethylation of
m6A Modified mRNA in Cells. To investigate if compound
1a inhibits FTO-mediated m6A modification of mRNA in cells,
we transfected preadipocytes (3T3-L1 cells)15 in the presence
or absence of compound 1a (10 μM) with a pcDNA 3.1 vector,
wild type (WT)-FTO, or the FTO mutant R96Q plasmid and
then quantified m6A of mRNA using a method similar to that
described previously.10 Treatment with compound 1a at
different concentrations had no detectable effect on the
morphology of the cells, suggesting that the compound is not
cytotoxic.16 Consistent with previous data derived from HeLa
cells,10 overexpression of WT FTO significantly decreased the
level of m6A of mRNA in preadipocytes (Figure 3). In contrast,
mRNA of the FTO-overexpressing preadipocytes treated with
compound 1a (10 μM) had much higher levels of m6A when
compared to that of the nontreated cells, suggesting that this
compound is specific for FTO in the cells. Cells overexpressing
the FTO mutant R96Q which has a much reduced demethylase
activity5 displayed a higher level of m6A than those expressing
WT FTO. The level of m6A of mRNA in the FTO mutant-
expressing cells was further increased by treatment with
compound 1a but remains less than that in the WT FTO-
Figure 1. Dose-dependent inhibition of FTO demethylase activity by
compound 1a. The 15-mer ssRNA, 5′-CUUGUCA(m6A)CAGCAGA-
3′, was used to determine FTO demethylase activity in the presence of
different concentrations of 1a (0.2 μM, 0.8 μM, 2.0 μM, 4.0 μM, 10.0
μM, 20.0 μM, and 100.0 μM). FTO activity (%, vertical axis) was
plotted against the logarithm of concentrations of 1a (horizontal axis).
The molecular formula of compound 1a is given. Numbers indicate
the positions of the two hydroxyl groups. The inhibitions of FTO
demethylation m6A in ssRNA are assayed in triplicate.
C
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. Compound 1a enhances m6A of mRNA of preadipocytes.
mRNA samples were isolated from preadipocytes overexpressing WT
FTO or the mutant R96Q in the absence and presence of compound
1a (10 μM). The levels of m6A were quantified by a dot blot assay (see
Supporting Information). Vertical axis represents the integrated optical
density (IOD) of m6A relative level in mRNA. Error bar, mean SEM
for n = 4 experiments. **P < 0.01 vs control.
overexpressing cells. Together, these results provide further
evidence for the specific inhibition of FTO by the compound in
cells. Notably, the level of m6A in the empty vector-expressing
cells were also sensitive to compound 1a, suggesting that
preadipocytes have a high level of FTO expression, which
further supports a previous observation.17
Figure 4. (A) Mesh shown around 1a (in stick form) is omit electron
density “Fo − Fc” (pink) in the finally refined structure contoured at
3.0 sigma. (B) Surface and cartoon representations of compound 1a
bound to FTO. The 1a interacting L1 loop, β-sheet, and the bound
αKG (α-ketoglutarate) are labeled.
3.4. Structural Basis for the Recognition of Com-
pound 1a by FTO. To reveal the structural basis for the
recognition of compound 1a by FTO, we soaked the native
FTO crystals with 1a for 24 h prior to data collection. The
structure of the soaked FTO crystals was solved with
isomorphous replacement using the coordinates of FTO
(PDB code: 3LFM) as the initial model (Table S1 in
Supporting Information). In the structure, the FTO-bound
compound 1a (PDB code: 5DAB) is well-defined in the
electron density map and deeply anchored in a narrow FTO
cleft that has a generally basic surface (Figure 4). The
compound-binding pocket is a continuance of αKG-binding,
suggesting that the FTO substrates, ssRNA/DNA, may extend
to the substrate-binding site. Further supporting this possibility,
structural comparison between FTO and ABH2 showed that 1a
completely overlaps with the methylated strand in the dsDNA
bound by ABH218 (Figure 5A). However, more studies are
needed in the future to assess this possibility. The binding
results in a buried surface area of 373 Å2. Surprisingly, the
small-molecule is located far away from the αKG-binding site of
FTO with a minimum distance of 8.5 Å (Figure 4B). The
bound inhibitor (1a) is sandwiched between an antiparallel β-
sheet and the L1 loop of FTO (Figure 4B). Structural
comparison indicates that the loop of FTO that interacts with
compound 1a is not conserved in other mammalian AlkB
members (Figure 5B), implying that it might be an FTO-
specific inhibitor. Compound 1a partially overlaps the recently
reported FTO-specific inhibitor meclofenamic acid (MA),19
though the two molecules have completely different chemical
structures (Figure 5C−E). The FTO L1 loop has been shown
to be important for selection by FTO of ssDNA/ssRNA.5 The
1a-bound and 3meT-bound FTO proteins have nearly identical
structures (Figure 5D), indicating that binding of 1a does not
induce conformational changes in FTO, the 1a-binding site
being preformed. This structural observation supports the idea
that the compound 1a is a competitive inhibitor of FTO.
Specific recognition of compound 1a by FTO is through a
combination of hydrogen bonds and hydrophobic and van der
Waals contacts. One half of compound 1a containing the
cyclobutane ring and the phenyl ring establishes extensive
hydrophobic contacts with the nonconserved antiparallel β-
sheet of FTO (Figure 5F). The side chains of Leu90 and
Thr92, and the main chain atoms of Leu91 are located less than
3.7 Å from the four-carbon ring of 1a, whereas Ile85 and
Leu109 pack tightly against the phenyl ring. More specific and
robust interactions between 1a and FTO appear to stem from
hydrophobic and polar contacts of the phenolic ring at the
other side of 1a with the nonconserved long loop of FTO
(Figure 5F). The phenolic ring stacks tightly against the side
chain of Pro93. In contrast to the phenyl hydroxyl group at C2
that is solvent exposed, the phenyl hydroxyl group at C4 forms
a strong bidentate hydrogen bond with the amide nitrogen of
Lys216 (3.0 Å) and the carbonyl oxygen of Met226 (2.8 Å).
The chloride of compound 1a further strengthens binding of
the compound to FTO by making van der Waals contacts with
Met226, Ala227, and Ser229. The 1a-FTO interaction is further
strengthened by a water-mediated hydrogen bond formed
between Ser229 and the carbonyl oxygen of compound 1a (also
see Figure S3).
4. CONCLUSIONS
We have identified a small-molecule FTO inhibitor (1a) whose
structure is entirely different from those reported previ-
ously.12,13 To analyze the structural basis for their interaction,
we solved the crystal structure of human FTO in complex with
this compound. The structure revealed a novel binding site for
FTO inhibitors and defines the molecular basis for FTO
D
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 5. (A) Structural comparison between compound 1a-bound FTO and dsDNA-bound ABH2. Color codes are indicated. (B) The compound
1a binding loop is unique to FTO. A structural comparison of the 1a-bound FTO with ABH3 (PDB code: 2IUW),20 ABH5 (PDB code: 4NRP),6
and ABH8 (PDB code: 3THP)21 is shown. The position of the 1a binding loop is indicated by a red arrow. For clarity, the C-terminal side of FTO is
not shown. (C) Structural comparison of compound 1a- and meclofenamic acid-bound FTO. 1a and MA (meclofenamic acid; PDB code, 4QKN)19
are shown in stick form. Colors are indicated. (D) Compound 1a binding induces no conformational changes in FTO. The structural alignment of
3meT-bound FTO with 1a-bound FTO is shown. Colors are indicated. (E) The structural alignment of 3meT-bound FTO with MA-bound FTO is
shown. Colors are indicated. (F) Recognition mechanism of compound 1a by FTO. The side chains of FTO are shown in yellow (ball-and-stick).
Hydrogen bonds and van der Waals interactions are indicated by red and gray dashed lines, respectively. Red sphere represents water molecule.
referred to the internal standard of TMS (tetramethylsilane). The peak
patterns are indicated as follows: s, singlet; d, doublet; t, triplet; q,
quadruplet; m, multiplet; dd, doublet of doublets and dt, doublet of
triplets. The coupling constants (J) are reported in Hz. Flash column
chromatography was performed over silica gel 200−300 mesh. High-
resolution mass spectra (HRMS-ESI) were obtained on a Q-TOF
Mass Spectrometer. All of the tested compounds possess a purity of at
least 95%. Analytical HPLC was run on a Thermo Scientific Accela
HPLC instrument, equipped with a Waters XTerra MS C18, 3.5 μM,
2.1 × 100 mm column and UV detection at 234 nm. The eluent
system was A (MeOH) and B (H2O); flow rate = 0.2 mL/min.
Method: 10% A (2 min), 10% A to 90% A (4 min), 90% A (4 min),
90% A to 10% A (0.1 min), 10% A (4.9 min). All of the target
compounds 1a−t (for their characterization data, see the Supporting
Information.) were prepared according to the following representative
procedure for the synthesis of compound 1a.
recognition of the inhibitor. Although the novel small-molecule
binding site is formed by nonconserved amino acids of FTO,
we cannot rule out the possibility that this FTO inhibitor also
interacts with other proteins in cells. However, the identi-
fication of the new binding site offers new opportunities for
further development of selective and potent inhibitors of FTO.
It remains unknown whether FTO is a viable pharmaceutical
target. Nonetheless, selective FTO inhibitors will be a valuable
tool for probing the biology of FTO, which is expected to allow
for the definition of novel therapeutic targets for the treatment
of obesity or obesity-associated diseases.
5. EXPERIMENTAL SECTION
5.1. Synthesis and Characterization of Compounds 1.
1
5.1.1. General Information. H NMR spectra were recorded on a
400 MHz spectrometer, and 13C NMR spectra were recorded on a 100
MHz spectrometer. Chemical shift values are given in ppm and
5.1.2. 1-Phenylcyclobutanecarbonitrile (3a). To a suspension of
potassium hydroxide (6.1 g, 34.76 mmol) in DMSO (15 mL) was
E
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
added dropwise a solution of 2-phenylacetonitrile (2a, 1.0 g, 8.54
mmol) and 1,3-dibromopropane (1.7 g, 8.54 mmol) in DMSO (1.5
mL). After being stirred at room temperature for 5 h, the reaction
mixture was quenched with H2O (30 mL), then extracted with EtOAc
(20 mL × 3). The combined organic layers were dried over anhydrous
Na2SO4, concentrated, and then purified by silica gel column
chromatography (10% EtOAc in petroleum ether as eluent) to give
Pharmacia) and gel-filtration chromatography (Superdex-200, Phar-
macia). Protein purification was performed at 4 °C. For crystallization,
the purified protein was concentrated to approximately 8.0 mg/mL.
5.3. Assessment of Inhibitory Activity on FTO Demethyla-
tion of the Substrate 3-meT. Preliminary screening of the FTO
demethylation inhibiting activity of the small-molecule inhibitor was
performed in 100 μL of reaction mixture containing 3 μL of FTO, 20
μL of 3meT (0.1 mg/mL), 5 μL of MES (1 M), 2 μL of FeSO4·7H2O
(10 mM), 10 μL of αKG (10 mM), 10 μL of L-ascorbic acid (20 mM),
2 μL of the inhibitor (4 mM in DMSO) (2 μL of DMSO for the
control experiment), and H2O (48 μL). The reaction was incubated
for 17 h at 16 °C before being centrifuged. The resulting mixture was
analyzed on a HPLC system equipped with a Star Technologic LIC
Polars C18 column (250 × 4.6 mm, 5 μm) gradient eluted with mobile
phase A (H2O), mobile phase B (CH3OH) with a flow rate of 1 mL/
min at room temperature. The LC system was held at 10% B for 0.01
min followed by nonlinear gradient profiles from 10% B to 80% B for 6
min, 80% B to 80.2% B for 0.2 min, 80.2% B to 100% B for 5.8 min.
The LC system was then held at 100% B for 5 min. The injection
volume of 10 μL was used. The detection wavelength was set at 270
1
product 3a as a colorless oil (805 mg, 60%); H NMR (400 MHz,
CDCl3): δ 7.43−7.25 (m, 5H), 2.85−2.78 (m, 2H), 2.66−2.58 (m,
2H), 2.42 (m, 1H), 2.07 (m, 1H). 13C NMR (100 MHz, CDCl3): δ
139.8, 129.0, 127.9, 125.6, 124.5, 40.2, 34.7, 17.2.
5.1.3. 1-Phenylcyclobutanecarboxylic Acid (4a). A reaction
mixture of 1-phenylcyclobutanecarbonitrile (3a, 805 mg, 5.12 mmol)
and potassium hydroxide (2.87 g, 51.2 mmol) in ethylene glycol (8
mL) was refluxed for 2 h. After cooling to room temperature, the
mixture was partitioned between EtOAc (15 mL) and H2O (10 mL).
The separated aqueous layer was extracted with EtOAc (15 mL × 2).
The organic layers were combined, dried over anhydrous Na2SO4,
concentrated, and then purified through silica gel column chromatog-
raphy (50% EtOAc in petroleum ether as eluent) to give product 4a as
1
nm. Demethylation ratio by FTO (%) = AT peak/(AT peak + A3meT peak
× 100%.
)
a white solid (766 mg, 85%); mp. 104−105 °C. H NMR (400 MHz,
CDCl3): δ 7.36−7.23 (m, 5H), 2.89−2.83 (m, 2H), 2.58−2.51 (m,
2H), 2.08 (m, 1H), 1.88 (m, 1H). 13C NMR (100 MHz, CDCl3): δ
182.7, 143.1, 128.3, 126.8, 126.4, 52.2, 32.3, 16.6.
5.4. Assessment of Inhibitory Activity on FTO Demethyla-
tion of the 15-mer ssRNA. In vitro enzymatic activity assays were
performed as a modified version of the one described previously.11
The 15-mer ssRNA (5′-CUUGUCA(m6A)CAGCAGA-3′) was used
as a substrate to evaluate the demethylase activity of FTO. All of the
reactions were carried out in 50 μL of buffer containing 50 mM Tris-
HCl at pH 7.5, 2.0 μM ssRNA, 2.0 nM FTO, 1.0 mM αKG, 280 μM
(NH4)2Fe(SO4)2, and 2 mM L-ascorbic acid. Compound 1a at varying
concentrations (0.2 μM, 0.8 μM, 2.0 μM, 4.0 μM, 10.0 μM, 20.0 μM,
and 100.0 μM) was added to each reaction mixture and incubated at
room temperature for 0.5 h. The reactions were quenched by heating
for 5 min at 95 °C. The ssRNA was digested by nuclease P1 (1 Unit)
and NH4OAc (100 μM, 5 μL) at 42 °C for 4 h, followed by the
addition of NH4HCO3 (1.0 M, 5 μL), and alkaline phosphatase (0.5
Unit). After further incubation at 37 °C for 3 h, 6-Cl-G (10 μg/mL, 5
μL) was added as an internal reference, then the solution was diluted
to 100 μL, and 10 μL of the solution was injected into LC-MS/MS.
The nucleosides were separated by reverse phase high-performance
liquid chromatography on a C18 column, with online mass
spectrometry detection using Thermo TSQ Quantum Ultra LC/MS
in a positive electrospray ionization mode. The nucleosides were
quantified using the nucleoside to base ion mass transitions of 282 to
150 (m6A) and 268 to 136 (A). Quantification was performed by
comparison with the standard curve obtained from pure nucleosides
running on the same batch of samples.
5.5. Isothermal Titration Calorimetry (ITC). The ITC experi-
ment was carried out at 25 °C with a Microcal ITC200 isothermal
titration calorimeter (GE Healthcare). Purified FTO protein stock
solution in a buffer containing 25 mM HEPES (pH 7.5), 100 mM
NaCl, and 10 mM β-mercaptoethanol was used for the ITC
experiment. The protein solution was prepared from the stock
solution to contain 10% DMSO and 140 μM FTO. The titrant
contained the same buffer plus 1.5 mM compound 1a. To correct the
thermal effect from mixing and dilution, a control experiment was
performed by injecting the titrant into the buffer without protein. The
0.2 mL sample cell was filled with the protein solution and stirred
constantly at 750 rpm. The titrant was titrated into the sample cell
with one 0.4 μL injection followed by 2.0 μL injections at 150 s
intervals. After the background dilution heats were subtracted from the
experimental data, the net titration data were analyzed with the
Microcal ORIGIN V7.0 software (Microcal Software, Northampton,
MA).
5.1.4. N-(5-Chloro-2,4-dimethoxyphenyl)-1-phenylcyclobutane-
carboxamide (5a). A mixture of 1-phenylcyclobutanecarboxylic acid
(4a, 0.77 g, 4.35 mmol), 5-chloro-2,4-dimethoxyaniline (0.82 g, 4.35
mmol), EDCI (1.67 g, 8.70 mmol), and DMAP (0.27 g, 2.18 mmol) in
anhydrous dichloromethane (20 mL) was stirred at room temperature
for 6 h and then quenched with H2O (10 mL). The organic layer was
separated, and the aqueous layer was extracted with dichloromethane
(15 mL × 2). The combined organic layers were dried over anhydrous
Na2SO4, concentrated, and then purified through silica gel column
chromatography (25% EtOAc in petroleum ether as eluent) to give
1
product 5 as a white solid; mp 115−116 °C. H NMR (400 MHz,
CDCl3): δ 8.40 (s, 1H), 7.41−7.24 (m, 5H), 6.36 (s, 1H), 3.79 (s,
3H), 3.63 (s, 3H), 2.94−2.87 (m, 2H), 2.57−2.50 (m, 2H), 2.20 (m,
1H), 1.91 (m, 1H). 13C NMR (100 MHz, CDCl3): δ173.7, 151.0,
147.6, 144.1, 128.8, 127.0, 126.3, 121.8, 120.9, 113.8, 96.7, 56.6, 56.1,
53.8, 32.1, 16.6.
5.1.5. N-(5-Chloro-2,4-dihydroxyphenyl)-1-phenylcyclobutane-
carboxamide (1a). To a solution of N-(5-chloro-2,4-dimethoxyphen-
yl)-1-phenylcyclobutanecarboxyamide (5a, 1.21 g, 3.48 mmol) in
anhydrous dichloromethane (15 mL) at −78 °C was added dropwise a
solution of BBr3 (1 M) in dichloromethane (10 mL). After addition,
the resulting mixture was stirred at −78 °C for 0.5 h and then allowed
to warm to room temperature over 1 h. The reaction mixture was
quenched with water (20 mL). The organic layer was separated, and
the aqueous layer was extracted with ethyl acetate (15 mL × 3). All of
the organic extracts were combined, dried over Na2SO4, concentrated,
and then purified through silica gel column chromatography (50%
ethyl acetate in petroleum ether as eluent) to give product 1a as a
lavender solid (0.99 g, 72% from acid 4a); mp 174−175 °C; IR (KBr)
3523, 3354, 3158, 1644, 1613, 1536, 1517, 1503, 1437, 1366, 1244,
1
1197, 1151, 1124. H NMR (400 MHz, CD3OD): δ 7.71 (s, 1H),
7.43−7.36 (m, 4H), 7.27 (m, 1H), 6.42 (s, 1H), 2.91−2.84 (m, 2H),
2.59−2.52 (m, 2H), 2.08 (m, 1H), 1.91 (m, 1H). 13C NMR (100
MHz, CD3OD): δ176.7, 151.3, 148.9, 145.1, 129.9, 128.0, 127.2,
123.3, 120.4, 111.3, 104.6, 55.1, 33.3, 17.2; HRMS (m/z) [M + Na]+
calcd for C17H1635ClNO3Na 340.0716, found 340.0712. HPLC purity:
97.8% (tR = 9.16 min).
5.2. Protein Expression and Purification. The FTO protein
(residues 31−505) was expressed as previously described.5 Briefly,
cells (BL21 (DE3)) expressing FTO were induced with 1 mM
isopropyl β-D-1-thiogalactopyranoside (IPTG) for 12 h at room
temperature. The cells were then harvested, pelleted, and resuspended
in buffer A (50.0 mM Tris, pH 8.0, and 100.0 mM NaCl). The cells
were lysed by sonication, and the lysate was then centrifuged at 14,000
r.p.m for 1 h. The soluble proteins were first purified using Ni2+-resin
(Novagen) and then further by ion exchange (Source-15Q,
5.6. Inhibition by Compound 1a of the Demethylation of
m6A Modified mRNA in Cells. Preadipocytes (3T3-L1 cells) stably
transfected with the pcDNA 3.1 vector, R96Q-FTO or WT-FTO
plasmid, were cultured in high glucose DMEM supplemented with
10% fetal bovine serum (Hyclone) and 1% penicillin−streptomycin
(Gibco) in a 5% CO2 humidified atmosphere. For induced
F
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
differentiation, post-confluent preadipocytes were incubated with a
cocktail of insulin (1 μg/mL), dexamethasone (1 μM), and 3-isobutyl-
1-methylxanthine (0.5 μM) in DMEM supplemented with 10% fetal
bovine serum for 48 h, followed by a culture of DMEM supplemented
with 10% fetal bovine serum, insulin (1 μg/mL), and 10 μM
compound 1a for another 48 h. Total cellular RNA was isolated from
3T3-L1 cells (2 × 107) using the TRIZOL reagent (Roche, USA).
Cellular mRNA were purified with Oligotex Direct mRNA Kits
(Qiagen, Germany), followed by rRNA removal using RiboMinus
Transcriptome Isolation Kit (invitrogen, USA) in accordance with the
manufacturer’s instructions. The total RNA and mRNA concentrations
were quantified with a NanoDrop 2000c UV−vis Spectrophotometer
(Thermo, USA). The purified mRNA was detected by agarose gel
electrophoresis.
5.7. Crystallization, Data Collection, Structure Determina-
tion, and Refinement. Crystallization experiments were performed
with hanging-drop vapor-diffusion methods by mixing equal volumes
of protein and reservoir solution at 18 °C. Crystals of the FTO protein
were generated in buffer and 10% (w/v) polyethylene glycol (PEG)
3350, 0.1 M trisodium citrate dehydrate at pH 5.6, and 3.0 mM N-
oxalylglycine (NOG). Crystals of FTO were then transferred to and
soaked in the crystallization buffer plus 1.0 mM compound 1a for 24 h.
For data collection, the crystals were equilibrated in a cryoprotectant
buffer containing reservoir buffer plus 20% (v/v) glycerol. Diffraction
data were collected on the BL17U1 beamline of the Shanghai
Synchrotron Research Facility (SSRF). The data sets were processed
using HKL2000.22 The structure of FTO complexed with compound
1a was determined using isomorphous replacement with the
coordinates of free FTO (PDB code: 3LFM) as an initial model.
After refinement of the initial model, 1a was built into the electron
density in COOT.23 The structure of the 1a-FTO complex was finally
refined to a resolution of 2.10 Å with Rfactor = 19.8% and Rfree = 23.6%.
5.8. Statistical Analysis. All values are presented as the mean
SEM. Statistical significance was analyzed using independent sample t
tests and ANOVA for the comparison of multiple means. P < 0.05 was
considered to be statistically significant.
REFERENCES
■
(1) Shen, L.; Song, C. X.; He, C.; Zhang, Y. Mechanism and function
of oxidative reversal of DNA and RNA methylation. Annu. Rev.
Biochem. 2014, 83, 585−614.
(2) Falnes, P. O.; Johansen, R. F.; Seeberg, E. AlkB-mediated
oxidative demethylation reverses DNA damage in Escherichia coli.
Nature 2002, 419, 178−182.
(3) Trewick, S. C.; Henshaw, T. F.; Hausinger, R. P.; Lindahl, T.;
Sedgwick, B. Oxidative demethylation by Escherichia coli AlkB directly
reverts DNA base damage. Nature 2002, 419, 174−178.
(4) Gerken, T.; Girard, C. A.; Tung, Y. C.; Webby, C. J.; Saudek, V.;
Hewitson, K. S.; Yeo, G. S.; McDonough, M. A.; Cunliffe, S.; McNeill,
L. A.; Galvanovskis, J.; Rorsman, P.; Robins, P.; Prieur, X.; Coll, A. P.;
Ma, M.; Jovanovic, Z.; Farooqi, I. S.; Sedgwick, B.; Barroso, I.; Lindahl,
T.; Ponting, C. P.; Ashcroft, F. M.; O’Rahilly, S.; Schofield, C. J. The
obesity-associated FTO gene encodes a 2-oxoglutarate-dependent
nucleic acid demethylase. Science 2007, 318, 1469−1472.
(5) Han, Z.; Niu, T.; Chang, J.; Lei, X.; Zhao, M.; Wang, Q.; Cheng,
W.; Wang, J.; Feng, Y.; Chai, J. Crystal structure of the FTO protein
reveals basis for its substrate specificity. Nature 2010, 464, 1205−1209.
(6) Feng, C.; Liu, Y.; Wang, G.; Deng, Z.; Zhang, Q.; Wu, W.; Tong,
Y.; Cheng, C.; Chen, Z. Crystal structures of the human RNA
demethylase Alkbh5 reveal basis for substrate recognition. J. Biol.
Chem. 2014, 289, 11571−11583.
(7) Xu, C.; Liu, K.; Tempel, W.; Demetriades, M.; Aik, W.; Schofield,
C. J.; Min, J. Structures of human ALKBH5 demethylase reveal a
unique binding mode for specific single-stranded N6-methyladenosine
RNA demethylation. J. Biol. Chem. 2014, 289, 17299−17311.
(8) Aik, W.; Scotti, J. S.; Choi, H.; Gong, L.; Demetriades, M.;
Schofield, C. J.; McDonough, M. A. Structure of human RNA N6-
methyladenine demethylase ALKBH5 provides insights into its
mechanisms of nucleic acid recognition and demethylation. Nucleic
Acids Res. 2014, 42, 4741−4754.
(9) Chen, W.; Zhang, L.; Zheng, G.; Fu, Y.; Ji, Q.; Liu, F.; Chen, H.;
He, C. Crystal structure of the RNA demethylase ALKBH5 from
zebrafish. FEBS Lett. 2014, 588, 892−898.
(10) Jia, G.; Fu, Y.; Zhao, X.; Dai, Q.; Zheng, G.; Yang, Y.; Yi, C.;
Lindahl, T.; Pan, T.; Yang, Y. G.; He, C. N6-methyladenosine in
nuclear RNA is a major substrate of the obesity-associated FTO. Nat.
Chem. Biol. 2011, 7, 885−887.
(11) Loos, R. J.; Yeo, G. S. The bigger picture of FTO: the first
GWAS-identified obesity gene. Nat. Rev. Endocrinol. 2014, 10, 51−61.
(12) Aik, W.; Demetriades, M.; Hamdan, M. K.; Bagg, E. A.; Yeoh, K.
K.; Lejeune, C.; Zhang, Z.; McDonough, M. A.; Schofield, C. J.
Structural basis for inhibition of the fat mass and obesity associated
protein (FTO). J. Med. Chem. 2013, 56, 3680−3688.
(13) Chen, B.; Ye, F.; Yu, L.; Jia, G.; Huang, X.; Zhang, X.; Peng, S.;
Chen, K.; Wang, M.; Gong, S.; Zhang, R.; Yin, J.; Li, H.; Yang, Y.; Liu,
H.; Zhang, J.; Zhang, H.; Zhang, A.; Jiang, H.; Luo, C.; Yang, C. G.
Development of cell-active N6-methyladenosine RNA demethylase
FTO inhibitor. J. Am. Chem. Soc. 2012, 134, 17963−17971.
(14) Sedo, J.; Saiz-Poseu, J.; Busque, F.; Ruiz-Molina, D. Catechol-
based biomimetic functional materials. Adv. Mater. 2013, 25, 653−701.
(15) For a recent study about the demethylase activity of FTO in
preadipocytes, see: Zhang, M.; Zhang, Y.; Ma, J.; Guo, F.; Cao, Q.;
Zhang, Y.; Zhou, B.; Chai, J.; Zhao, W.; Zhao, R. The demethylase
activity of FTO (fat mass and obesity associated protein) is required
for preadipocyte differentiation. PLoS One 2015, 10, e0133788.
(16) A preliminary cytotoxicity study indicated that after the
treatment with compound 1a at different concentrations (1 μM, 10
μM, and 20 μM) for 8 days, the cells were still healthy (see Supporting
Information).
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jmed-
chem.5b00702.
Characterization data, 1H, and 13C NMR spectra of
compounds 1a−t; preliminary cytotoxicity study of 1a,
raw dot blot assay and images, data collection statistics of
compound 1a-FTO, and schematic diagram showing
putative interactions between FTO and rhein (PDF)
Compound ID and SMILES data (CSV)
AUTHOR INFORMATION
■
Corresponding Authors
*(J. Chang) E-mail: changjunbiao@zzu.edu.cn.
*(J. Chai) E-mail: chaijj@tsinghua.edu.cn.
Author Contributions
▽W.H., B.Z., and W.L. contributed equally to this work.
Notes
The authors declare no competing financial interest.
(17) Friebe, D.; Loffler, D.; Schonberg, M.; Bernhard, F.; Buttner, P.;
̈
̈
̈
Landgraf, K.; Kiess, W.; Korner, A. Impact of metabolic regulators on
̈
ACKNOWLEDGMENTS
■
the expression of the obesity associated genes FTO and NAMPT in
human preadipocytes and adipocytes. PLoS One 2011, 6, e19526.
(18) Yang, C. G.; Yi, C.; Duguid, E. M.; Sullivan, C. T.; Jian, X.; Rice,
P. A.; He, C. Crystal structures of DNA/RNA repair enzymes AlkB
and ABH2 bound to dsDNA. Nature 2008, 452, 961−965.
We thank F. Yu and J. He at Shanghai Synchrotron Radiation
Facility (SSRF). The research was funded by the National
Natural Science Foundation of China (No. 81330075 to J.
Chang).
G
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(19) Huang, Y.; Yan, J.; Li, Q.; Li, J.; Gong, S.; Zhou, H.; Gan, J.;
Jiang, H.; Jia, G. F.; Luo, C.; Yang, C. G. Meclofenamic acid selectively
inhibits FTO demethylation of m6A over ALKBH5. Nucleic Acids Res.
2015, 43, 373−384.
(20) Sundheim, O.; Vagbø, C. B.; Bjøras, M.; Sousa, M. M.; Talstad,
̊
̊
V.; Aas, P. A.; Drabløs, F.; Krokan, H. E.; Tainer, J. A.; Slupphaug, G.
Human ABH3 structure and key residues for oxidative demethylation
to reverse DNA/RNA damage. EMBO J. 2006, 25, 3389−3397.
(21) Pastore, C.; Topalidou, I.; Forouhar, F.; Yan, A. C.; Levy, M.;
Hunt, J. F. Crystal structure and RNA binding properties of the RNA
recognition motif (RRM) and AlkB domains in human AlkB homolog
8 (ABH8), an enzyme catalyzing tRNA hypermodification. J. Biol.
Chem. 2012, 287, 2130−2143.
(22) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data
collected in oscillation mode. Methods Enzymol. 1997, 276, 307−326.
(23) Emsley, P.; Cowtan, K. Coot: Model-building tools for
molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,
60, 2126−2132.
H
DOI: 10.1021/acs.jmedchem.5b00702
J. Med. Chem. XXXX, XXX, XXX−XXX