Discovery of selective ATP-competitive eIF4A3 inhibitors

Article abstract of DOI:10.1016/j.bmc.2017.02.035

Eukaryotic initiation factor 4A3 (eIF4A3), an ATP-dependent RNA helicase, is a core component of exon junction complex (EJC). EJC has a variety of roles in RNA metabolism such as translation, surveillance, and localization of spliced RNA. It is worthwhile to identify selective eIF4A3 inhibitors with a view to investigating the functions of eIF4A3 and EJC further to clarify the roles of the ATPase and helicase activities in cells. Our chemical optimization of hit compound 2 culminated in the discovery of ATP-competitive eIF4A3 inhibitor 18 with submicromolar ATPase inhibitory activity and excellent selectivity over other helicases. Hence, compound 18 could be a valuable chemical probe to elucidate the detailed functions of eIF4A3 and EJC.

Full text of DOI:10.1016/j.bmc.2017.02.035

Accepted Manuscript
Discovery of Selective ATP-competitive eIF4A3 Inhibitors
Masahiro Ito, Misa Iwatani, Yusuke Kamada, Satoshi Sogabe, Shoichi Nakao,
Toshio Tanaka, Tomohiro Kawamoto, Samuel Aparicio, Atsushi Nakanishi,
Yasuhiro Imaeda
PII:
S0968-0896(16)31483-3
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.02.035
BMC 13565
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
22 December 2016
15 February 2017
16 February 2017
Please cite this article as: Ito, M., Iwatani, M., Kamada, Y., Sogabe, S., Nakao, S., Tanaka, T., Kawamoto, T.,
Aparicio, S., Nakanishi, A., Imaeda, Y., Discovery of Selective ATP-competitive eIF4A3 Inhibitors, Bioorganic &
Medicinal Chemistry (2017), doi: http://dx.doi.org/10.1016/j.bmc.2017.02.035
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Discovery of Selective ATP-competitive eIF4A3
Inhibitors
,†
†
†
†
†
Masahiro Ito,* Misa Iwatani, Yusuke Kamada, Satoshi Sogabe, Shoichi Nakao, Toshio
†
†
‡
†
,†
Tanaka, Tomohiro Kawamoto, Samuel Aparicio, Atsushi Nakanishi, Yasuhiro Imaeda*
†
Pharmaceutical Research Division, Takeda Pharmaceutical Company Limited, 26-1,
Muraoka-Higashi 2-chome, Fujisawa, Kanagawa 251-8555, Japan
‡
Department of Molecular Oncology, BC Cancer Agency, 675 W10th Avenue, Vancouver, BC,
V5Z 1L3, Canada

Corresponding Author
*
(M.I.) phone: +81-466-32-1196. E-mail: masahiro.ito@takeda.com; (Y.I.) phone:
81-466-32-4112. E-Mail: yasuhiro.imaeda@takeda.com
+
Abstract: Eukaryotic initiation factor 4A3 (eIF4A3), an ATP-dependent RNA helicase, is a
core component of exon junction complex (EJC). EJC has a variety of roles in RNA metabolism
such as translation, surveillance, and localization of spliced RNA. It is worthwhile to identify
selective eIF4A3 inhibitors with a view to investigating the functions of eIF4A3 and EJC further
to clarify the roles of the ATPase and helicase activities in cells. Our chemical optimization of
hit compound 2 culminated in the discovery of ATP-competitive eIF4A3 inhibitor 18 with
submicromolar ATPase inhibitory activity and excellent selectivity over other helicases. Hence,
1

compound 18 could be a valuable chemical probe to elucidate the detailed function of eIF4A3
and EJC.
Graphical abstract
Keywords: RNA helicase, eukaryotic initiation factor 4A3 (eIF4A3) inhibitors, exon junction
complex (EJC), ATP-competitive type inhibitors, isothermal titration calorimetry (ITC), surface
plasmon resonance (SPR).
Introduction
Eukaryotic initiation factor 4A3 (eIF4A3) is not only a member of the ATP-dependent RNA
1
–3
helicase family but also a core component of the splicing-dependent multiprotein exon junction
4
complex (EJC). It is known that EJC plays a variety of roles in translation, surveillance and
4
–11
localization of spliced RNA. Among the functions of EJC, nonsense-mediated RNA decay
5
–7
(
NMD), an important quality control system of mRNA, has been studied intensively. In
1
0,11
addition, other roles including regulation of RNA splicing were also reported recently. Since
5
–7,10,11
the aforementioned research was conducted by using RNA interference (RNAi),
it is
worthwhile to identify selective small molecule eIF4A3 inhibitors with a view to investigating
the functions of eIF4A3 and EJC further and clarify the roles of the ATPase/helicase activities in
2

cells. Although a marine natural product hippuristanol 1 was reported as a pan-eIF4A inhibitor
12–14
(
Figure 1),
no selective eIF4A3 inhibitors have been reported to date.
In order to explore selective eIF4A3 inhibitors, we executed a high through-put screening
HTS) campaign and chemical optimization of some hit compounds, which led to the discovery
(
of several types of selective eIF4A3 inhibitors. This report focuses on identification of
ATP-competitive type of eIF4A3 inhibitors as typified by compound 18.
Figure 1. Structure of known pan-eIF4A inhibitor hippuristanol (1).
Chemistry
Compounds 2–7 and 11 are commercially available. Synthetic routes of the other compounds
were described as follows. Methyl ester 8 was synthesized by esterification of 2. Fluorination of
1
5,16
8
and subsequent saponification provided compound 12. Amidation of 2 afforded
carboxamide 9, then dehydration with POCl followed by treatment with NaN gave tetrazole 10.
3
3
3

a
Scheme 1. Synthesis of Compounds 8–10 and 12
a
Reagents and conditions: (a) conc. H SO , MeOH, reflux, 54%; (b) Selectfluor, MgO, MeCN, rt; (c) NaOH, THF,
2
4
rt, 6% for 2 steps; (d) EDC·HCl, HOBt·NH , DIPEA, 2 M NH in iPrOH, DMF, rt, 65%; (e) POCl , 90 °C; (f) NaN ,
3
3
3
3
NH Cl, DMF, rt, 18% for 2 steps.
4
17
Compounds 13–17 were synthesized as depicted in Scheme 2. Suzuki–Miyaura coupling of
bromides 20a–d with borate 21, subsequent condensation with ethyl azidoacetate using sodium
1
8
methoxide, indole ring formation in a thermal condition afforded indoles 22a–f. In the case of
starting materials 20c and 20d, regioisomeric fluoro derivatives 22c and 22e and chloro
derivatives 22d and 22f were obtained, respectively. Following reduction of the isopropenyl
moieties, hydrolysis of the ester groups in compounds 22a–c, 22e, and 22f led to compounds
1
3–17.
4

a
Scheme 2. Synthesis of Compounds 13–17
a
Reagents and conditions: (a) 21, Na CO , Pd(PPh ) Cl , dioxane–H O, 60 °C; (b) ethyl azidoacetate, NaOMe,
2
3
3
2
2
2
MeOH, 0 °C to rt; (c) toluene, microwave, 150 °C, 4–31% for 3 steps; (d) Pd/C, H , MeOH, rt; (e) NaOH, THF, rt,
2
2
4–59% for 2 steps.
Compound 18 was synthesized as shown in Scheme 3. Aldehyde 23 was condensed with
methyl azidoacetate in a basic condition, then heating of the azidoester intermediate provided
1
9
indole 24. Suzuki coupling with borate 21 and reduction of the resulting isopropenyl moiety
followed by hydrolysis of the ester group afforded compound 18.
a
Scheme 3. Synthesis of Compound 18
a
Reagents and conditions: (a) Methyl azidoacetate, NaOMe, MeOH, −10 °C to rt; (b) xylene, 160 °C, 15% for 2
steps; (c) 21, Na CO , Pd(PPh ) Cl , dioxane–H O, 100 °C, 57%; (d) Pd/C, ZnBr , H , MeOH, rt; (e) LiOH·H O,
2
3
3
2
2
2
2
2
2
THF–H O, rt, 42% for 2 steps.
2
Results & Discussion
5

Through HTS of an in-house compound library, indole-2-carboxylic acid derivative 2 was
identified as a hit compound. Our design strategy based on compound 2 is illustrated in Figure 2.
First, we investigated the necessity of the 6-isopropyl group (A), the 2-carboxyl group (B), and
the NH moiety in the indole ring (C) of 2. Second, introduction of a small functional group on
the indole ring was executed (D). Finally, combinations of the promising functional groups were
examined to maximize potency.
Figure 2. Design strategy for compound 2.
Biological evaluation was conducted using the ADP-Glo assay system (Promega) in which the
ATPase activity of eIF4A3 is detected, and the results are expressed as IC values. As an initial
5
0
SAR exploration, the effect of the 6-isopropyl group of compound 2 was investigated (Table 1).
-Methyl (3), 6-ethyl (4), and 6-chloro (5) derivatives showed diminished inhibitory activity,
6
which suggested that sufficient hydrophobicity and bulkiness would be necessary at this position,
and the electron withdrawing character of the chloro group would not have any positive effect
for inhibitory activity. Additionally, 5-isopropyl (6) and 7-isopropyl (7) derivatives did not show
any inhibitory activity up to 100 M in contrast to compound 2, which emphasized the
importance of the alkyl group at the 6-position.
Table 1. SAR of Compounds 2–7
6

eIF4A3 a
compound
IC (M)
5
0
2
3
4
5
6
7
27 (23–32)
>100
>100
>100
>100
>100
a
n = 2, 95% confidence intervals shown in parentheses.
Next, modification of the 2-carboxyl moiety was conducted (Table 2). Decreased activity of
ester 8 implied a salt bridge or hydrogen bonding interaction between the carboxyl group of 2
and eIF4A3 protein. Carboxamide 9 did not show any inhibitory activity up to 100 M, whereas
tetrazole 10 exhibited the same level of inhibitory activity as carboxylic acid 2. Results indicate
the importance of an acidic proton for the potency increase, which is most likely based on an ion
bonding to the protein.
Table 2. SAR of Compounds 8–10
7

eIF4A3 a
compound
R
IC (M)
5
0
8
9
COOMe
CONH2
>100
>100
1
0
5-tetrazolyl
30 (18–47)
a
n = 2, 95% confidence intervals shown in parentheses.
For the sake of further increasing the activity of the compounds, we explored additional
substituents to compound 2 and their proper positions. To that end, small functional groups such
as a methyl and fluoro group were introduced into the remaining positions on the indole ring
(
Table 3). No inhibitory activity of 1-methylindole derivative (11) up to 100 M indicated the
importance of the NH hydrogen in the indole ring for the inhibitory activity. Fluoro derivatives
12, 13, 15, and 16) showed more potent or similar inhibitory activity as compared to compound
. Considering the increased potency in the cases of 4-fluoro (13) and 7-fluoro (16) derivatives, a
(
2
more hydrophobic chloro group was substituted at the indole 4- and 7-postions to maximize the
potential hydrophobic interaction with eIF4A3 protein. Intriguingly, the 4-chloro derivative (14)
exhibited slightly improved potency, while introduction of a 7-chloro group (17) was not
tolerated. The binding pocket of eIF4A3 around the indole 7-position may be narrower compared
with that around the indole 4-postion. One more possibility is that the 7-fluoro substituent may
restrict the rotation of the 6-isopropyl group to fit into the binding pocket of eIF4A3. As a final
step, the SAR information was incorporated into a design of the tetra-substituted indole
8

derivative, which led to the identification of the most potent inhibitor 18 of the indole
-carboxylic acid series.
2
Table 3. SAR of Compound 11–18
eIF4A3 a
compound
R
1-Me
3-F
IC (M)
5
0
1
1
1
1
1
1
1
1
1
2
3
4
5
6
7
8
>100
20 (14–31)
5.0 (4.2–5.9)
1.9 (1.4–2.5)
30 (25–37)
6.9 (5.5–8.7)
>100
4-F
4-Cl
5-F
7-F
7-Cl
4-Cl, 7-F
0.97 (0.76–1.2)
a
n = 2, 95% confidence intervals shown in parentheses.
9

Selectivity over other helicases is shown in Table 4. Interestingly, compounds 14, 16, and 18
showed high eIF4A3 selectivity against closely related protein eIF4A1. The most potent
compound 18 was subjected to further selectivity profiling. As a result, 18 exhibited excellent
selectivity over another eIF4A family protein eIF4A2, and other helicases such as DHX29 and
BRR2. On the other hand, pan-eIF4A inhibitor 1 was selective for eIF4A1 and eIF4A2 rather
1
3
than eIF4A3 as previously reported by the Pelletier group.
Table 4. Selectivity of Compounds 14, 16, 18, and 1
eIF4A3
eIF4A1 a
eIF4A2 b
DHX29 b
BRR2
compound
b
IC (M)
IC (M)
IC (M)
IC (M)
IC (M)
5
0
50
50
50
50
c
1
1
1
4
6
8
1.9 (1.4–2.5)
6.9 (5.5–8.7)
0.97 (0.76–1.2)
40 (30–53)
>100
>100
>100
NT
NT
NT
NT
NT
NT
>100
>100
>30
>100
>30
1
3.1 (1.4–7.0) 18 (12–29)
a
b
c
n = 2. n = 4. 95% confidence intervals shown in parentheses. NT = not tested.
In order to investigate the inhibitory mechanism of compound 18, the effect of the ATP
concentration against the ATPase inhibitory activity was examined (Figure 3). Compound 18
showed an unambiguous pattern of competitive inhibition with respect to the ATP substrate, as
demonstrated by an ascending, linear dependence of the IC value on the ATP substrate
50
concentration. The fitted data by competitive inhibition model of Cheng–Prusoff equation
yielded a Ki value of 0.43 (0.38–0.49) µM, consistent with the estimated IC values at K . On
the other hand, the compound 18 did not show a theoretical competitive inhibition against RNA
Supplementary figure 3). Taken together, these results indicated that 18 would be an
ATP-competitive inhibitor.
m
5
0
(
1
0

Figure 3. IC values of 18 were plotted as a function of [ATP]/K (n = 4).
m
50
For the further clarification of the binding mode of compound 18 with eIF4A3, biophysical
evaluation using isothermal titration calorimetry (ITC) and surface plasmon resonance (SPR)
biosensing assay was performed. The ITC experiment of compound 18 revealed a dissociation
constant (K ) of 0.062 μM (Figure 4). The binding stoichiometry was close to one, reflecting the
D
binding of one inhibitor molecule to one monomer protein. The experiment also provided the
thermodynamic signature, characterized by a favorable change in enthalpy (ΔH = −26.0 kcal/mol)
and an unfavorable entropic contribution (−TΔS = 16.1 kcal/mol). This enthalpy-driven binding
implies the formation of strong polar interactions with the binding pocket, which is consistent
with the SAR such as requirement of a carboxyl group and NH of the indole ring.
1
1

Figure 4. Representative ITC experiment of compound 18 titrated into eIF4A3. Left: titration
isotherms (upper panel) and integrated reaction heats (lower panel). The solid line represents the
best fitting curve. Right: summary table of the thermodynamic parameters including equilibrium
binding constants.
Binding affinity and kinetics of compounds (2, 18, and 1) were analyzed using an SPR
biosensing assay (Table 5). The K value of 18 was close to the data measured in the ITC
D
analysis. Binding affinities of these compounds appeared to be correlated with the ATPase
inhibitory activity, whereas k and k values showed different trends among compounds.
on
off
Compounds 2 and 18 showed faster k rather than 1, and the k value of 18 would be slowed
on
off
through chemical optimization of 2. Comparing compound 1 with ATP-competitive inhibitor 2
1
2

bearing similar enzymatic activity and closer K value, compound 1 exhibited slower k and k
D
on
off
and longert than compound 2. Since 1 was reported to bind an allosteric site of eIF4A1 and
1
/2
mutations of specific amino acids in the corresponding region of eIF4A3 enhanced sensitivity
13
toward 1, 1 could interact with a similar allosteric site in eIF4A3 to that in eIF4A1. In that case,
the difference of the binding sites (ATP site vs allosteric site) would be considered to affect those
binding kinetics.
Table 5. Binding Kinetics of Compounds 2, 18, and 1 Measured in an SPR Biosensing Assay.
-
1 -1
eIF4A3
eIF4A3
k (M s )/
a
on
compound
-1 a
t (s)
a
1/2
IC (M)
SPR K (M)
koff (s )
5
0
D
5
-1
-2
-2
2
27
5.4
0.20
50
1.8 × 10 /9.9 × 10
0.7
8.9
5
1
8
0.97
40
4.0 × 10 /7.8 × 10
3
1
1.1 × 10 /5.7 × 10
12.2
a
Sensorgrams of each compound were attached in Supplementary figure 1.
As eIF4A3 is an ATP-dependent RNA helicase, ATPase inhibition of eIF4A3 by the
2
0
compounds should induce inhibition of the helicase activity. Inhibition of eIF4A3 helicase
2
1
activity by compound 18 was evaluated in a gel electrophoresis assay (Figures 5A, 5B).
Compound 18 showed significant inhibition of eIF4A3 helicase activity at 300 M, whereas the
negative control 9 did not show inhibitory activity at the same concentration. The large
discrepancy between ATPase and helicase inhibitory activities of 18 would be partially explained
by the different concentrations of ATP in each assay. A physiologically relevant high ATP
concentration (1 mM) was employed for the helicase assay, whereas the same concentration as
K
m
value of eIF4A3 (35 M) was used for the ATPase assay. As compound 18 is an
1
3

ATP-competitive inhibitor, further enhancement of ATPase inhibitory activity might be required
for this chemical series to show helicase inhibitory activity at lower compound concentration.
A
B
Figure 5. Inhibition of eIF4A3 helicase activity by compound 18. (A) Helicase activity of
eIF4A3 was monitored using fluorescence-labeled RNA in gel electrophoresis experiments.
Double-stranded RNA (dsRNA), eIF4A3, and ATP were incubated with compounds (9 or 18) at
1
4

0
–300 M, and the fluorescence of remaining dsRNA and unwound single-stranded RNA
(
ssRNA) were detected. The dsRNA and ssRNA were used as controls, respectively (lane 1 and
lane 2). ATP(−) represents a control reaction without ATP and compound (lane 3), and Heat
Denatured represents denatured dsRNA by heating at 95°C (lane 11). The data shown is
representative of three independent experiments. (B) Mean band intensity of dsRNA. The y-axis
represents mean band intensity of remaining dsRNA relative to ATP(−). Error bars represent the
SD of triplicated experiments. ***; P ≤0.0005 by William's test.
Conclusion
In summary, we executed chemical optimization of hit compound 2 and discovered the first
ATP-competitive eIF4A3 inhibitor 18 with submicromolar ATPase inhibitory activity and
excellent selectivity over other helicases. The binding mode of 18 with eIF4A3 was
characterized using ITC and SPR analyses. The results support the validity of the compound as
an eIF4A3 inhibitor, and thus 18 could be a valuable tool to elucidate the detailed function of
eIF4A3 and EJC.
1
5

Experimental section
1
Proton nuclear magnetic resonance ( H NMR) spectra were recorded on Bruker AVANCE-300
(
300 MHz) and Bruker AVANCE-400 (400 MHz) instruments in CDCl or DMSO-d solution.
3
6
Chemical shifts are given in parts per million (ppm) with tetramethylsilane as an internal
standard. Abbreviations are used as follows: s = singlet, d = doublet, t = triplet, spt = septet, m =
multiplet, dd = doublet of doublets, brs = broad singlet. Coupling constants (J values) are given
in hertz (Hz). Elemental analyses were carried out by Elemental Microanalysis Team in the
Medicinal Chemistry Research Laboratories, Sumika Chemical Analysis Service, or Toray
Research Center and were within 0.4% of the theoretical values. Low-resolution mass spectra
(
MS) were acquired using an Agilent LC/MS system (Agilent1200SL/Agilent6130MS,
Agilent1200SL/Agilent1956MS, or Agilent1200SL/Agilent6110MS) or Shimadzu UFLC/MS
Prominence UFLC high pressure gradient system/LCMS-2020) operating in an electron spray
(
ionization mode (ESI+). The column used was an L-column 2 ODS (3.0 × 50 mm I.D., 3 μm,
CERI) with a temperature of 40 °C and a flow rate of 1.2 or 1.5 mL/min. Condition 1: Mobile
phases A and B under an acidic condition were 0.05% TFA in water and 0.05% TFA in MeCN,
respectively. The ratio of mobile phase B was increased linearly from 5 to 90% over 0.9 min,
9
0% over the next 1.1 min. Condition 2: Mobile phases A and B under a neutral condition were a
mixture of 5 mmol/L AcONH and MeCN (9/1, v/v) and a mixture of 5 mmol/L AcONH and
4
4
MeCN (1/9, v/v), respectively. The ratio of mobile phase B was increased linearly from 5 to 90%
over 0.9 min, 90% over the next 1.1 min. High-resolution mass spectrometric analysis (HRMS)
was carried on a Bruker Autoflex Speed (MALDI-TOF Mass). The purities of all the compounds
tested in biological systems were assessed as being > 95% using elemental analysis or analytical
HPLC. Purity data were collected by HPLC with NQAD (Nano Quality Analyte Detector) or
Corona CAD (Charged Aerosol Detector). The column was an L-column 2 ODS (30 × 2.1 mm
1
6

I.D., CERI) or a Capcell Pak C18AQ (50 mm × 3.0 mm I.D., Shiseido) with a temperature of 50
C and a flow rate of 0.5 mL/min. Mobile phases A and B under a neutral condition were a
°
mixture of 50 mmol/L AcONH , water and MeCN (1/8/1, v/v/v) and a mixture of 50 mmol/L
4
AcONH and MeCN (1/9, v/v), respectively. The ratio of mobile phase B was increased linearly
4
from 5 to 95% over 3 min, 95% over the next 1 min. Reagents and solvents were obtained from
commercial sources and used without further purification. Reaction progress was determined by
thin layer chromatography (TLC) analysis on Merck Kieselgel 60 F254 plates or Fuji Silysia NH
plates. Chromatographic purification was carried out on silica gel columns ((Merck Kieselgel 60,
7
0‒230 mesh or 230‒400 mesh, Merck), (Chromatorex NH-DM 1020, 100‒200 mesh, Fuji Silysia
Chemical), (Inject column and Universal column, YAMAZEN,
http://yamazenusa.com/products/columns/)), or (Purif-Pack Si or NH, Shoko Scientific,
http://shoko-sc.co.jp/english2/)). Preparative TLC was carried out on Merck Kieselgel 60 PLC
plates. Preparative HPLC was acquired as follows: [Method A] Gilson Preparative HPLC
System (GX-281) with UV detector (214 and 254 nm). The column used was a SunFirePrep C18
OBD (250 × 19 mm I.D., 10 μm, Waters) and a flow rate of 25 mL/min. Mobile phases A and B
under an acidic condition were 0.05% TFA in water and MeCN, respectively. The ratio of
mobile phase B was increased linearly over 7.8 min. [Method B] Gilson Preparative HPLC
System (GX-281) with UV detector (214 and 254 nm). The column used was an XBridgePrep
C18 OBD (100 × 30 mm I.D., 5 μm, Waters) and a flow rate of 20 mL/min. Mobile phases A and
B under basic conditions were 10 mM aqueous NH HCO and MeCN, respectively. The ratio of
4
3
mobile phase B was increased linearly over 9 min. [Method C] MassLynx UV Prep System with
UV detector (220 and 254 nm). The column used was a L-Column 2 ODS (150 × 20 mm I.D., 5
μm, CERI) and a flow rate of 20 mL/min. Mobile phases A and B under an acidic condition were
0
.1% TFA in water and 0.1% TFA in MeCN, respectively. The ratio of mobile phase B was
1
7

increased linearly over 7 min. [Method D] Gilson Preparative HPLC System (GX-D) with UV
detector (220 and 254 nm). The column used was a Symmetrix ODS-R (150 × 30 mm I.D., 5 μm,
Boston Analytics) and a flow rate of 25 mL/min. Mobile phases A and B under an acidic
condition were 0.225% formic acid in water and MeCN, respectively. The ratio of mobile phase
B was increased linearly over between 10 min. Yields were not optimized.
22
Compound 1 was purchased from Centaurus biopharm (fee-for-service). Compounds 2−7
and 11 are commercially available. 2 was purchased from Enamine, 3 from ChemDiv, 4 from
UOS, 5 from ASINEX, 6 from ACB blocks, 7 from ArkPharm, 11 from EVO blocks.
Methyl 6-isopropyl-1H-indole-2-carboxylate (8). To a mixture of 2 (35 mg, 0.17 mmol) in
MeOH (4 mL) was added concentrated H SO (0.018 mL, 0.34 mmol) at room temperature. The
2
4
mixture was refluxed overnight. The reaction mixture was concentrated, neutralized with
saturated aqueous NaHCO solution at room temperature, and extracted with EtOAc. The organic
3
layer was separated, washed with water and brine, dried over MgSO , and concentrated. The
4
residue was purified by NH-silica gel column chromatography (hexane/EtOAc = 95/5 to 20/80)
1
to give 8 (20 mg, 54 %) as a white solid. H NMR (300 MHz, CDCl ) δ1.31 (d, J = 6.9 Hz, 6H),
3
3
1
.02 (spt, J = 6.9 Hz, 1H), 3.93 (s, 3H), 7.06 (dd, J = 8.4, 1.2 Hz, 1H), 7.18 (dd, J = 2.1, 1.0 Hz,
1
H), 7.23-7.25 (m, 1H), 7.60 (d, J = 8.4 Hz, 1H), 8.83 (brs, 1H). ( H NMR spectrum was
+
matched with authentic sample purchased from ACB-blocks.) MS m/z 218 (M+H) . Anal. Calcd
for C H NO : C, 71.87; H, 6.96; N, 6.45. Found: C, 71.71; H, 6.92; N, 6.38.
13
15
2
3
-Fluoro-6-isopropyl-1H-indole-2-carboxylic acid (12). To a mixture of 8 (400 mg, 1.84
1
6
mmol) and MgO (148 mg, 3.68 mmol) in MeCN (10 mL) was slowly added Selectfluor (783
mg, 2.21 mmol) at room temperature, then the resulting mixture was stirred at room temperature
overnight. The reaction mixture was concentrated, and the residue was purified by silica gel
1
8

preparative TLC (petroleum ether/EtOAc = 5/1) to afford 19 (81 mg). To a solution of 19 (81
mg) in THF (8 mL) was added NaOH (41 mg, 1.02 mmol), then the reaction mixture was stirred
at room temperature overnight. The mixture was concentrated, and to the residue was added
water. The mixture was adjusted to pH 3–4 with 1 M HCl and extracted with EtOAc. The
organic layer was separated, washed with brine, dried over Na SO , and concentrated. The
2
4
residue was purified by preparative HPLC (method A) to give 12 (22 mg, 6% for 2 steps) as a
1
white solid. H NMR (400 MHz, DMSO-d ) δ1.23 (t, J = 6.8 Hz, 6H), 2.89–3.10 (m, 1H), 7.04
6
(
dd, J = 8.4 Hz, 1.2 Hz, 1H), 7.19 (s, 1H), 7.53 (d, J = 8.4 Hz, 1H), 11.37 (s, 1H), 13.12 (brs,
+
1
H). MS m/z 222 (M+H) . Anal. Calcd for C H FNO ·0.4H O: C, 63.09; H,5.65; N, 6.13.
12
12
2
2
Found: C, 63.15; H, 5.40; N, 6.10.
6
-Isopropyl-1H-indole-2-carboxamide (9). A mixture of 2 (100 mg, 0.492 mmol),
HOBt·NH (374 mg, 2.46 mmol), EDC·HCl (472 mg, 2.46 mmol), and DIPEA (0.430 mL, 2.46
3
mmol) in DMF (5 mL) was stirred at room temperature overnight. To the mixture was added 2 M
NH in iPrOH (0.74 mL, 1.48 mmol) at room temperature, and the resulting mixture was stirred
3
at room temperature overnight. The reaction mixture was poured into water at room temperature
and extracted with EtOAc. The organic layer was separated, washed with water and brine, dried
over MgSO , and concentrated. The residue was purified by NH-silica gel column
4
chromatography (hexane/EtOAc = 95/5 to 0/100) and crystallized from hexane–EtOAc to give 9
1
(
(
65 mg, 65%) as a white solid. H NMR (300 MHz, CDCl ) δ1.31 (d, J = 6.9 Hz, 6H), 2.93–3.11
3
m, 1H), 5.86 (brs, 2H), 6.82–6.92 (m, 1H), 7.07 (dd, J = 8.3, 1.4 Hz, 1H), 7.27 (s, 1H), 7.58 (d,
+
J = 8.3 Hz, 1H), 9.22 (brs, 1H). MS m/z 203 (M+H) . Anal. Calcd for C H N O·0.2H O: C,
1
2
14
2
2
7
0.01; H,7.05; N, 13.61. Found: C, 70.10; H, 6.93; N, 13.49.
6
-Isopropyl-2-(1H-tetrazol-5-yl)-1H-indole (10). A solution of 9 (600 mg, 2.98 mmol) in
1
9

POCl (3 mL) was heated at 90 °C for 3 h. The reaction mixture was concentrated, diluted with
3
ice water, and extracted with EtOAc. The separated organic layer was washed with brine, dried
over Na SO ,and concentrated. The residue was purified by silica gel column chromatography
2
4
(
petroleum ether/EtOAc = 1/1) to give 6-isopropyl-1H-indole-2-carbonitrile (400 mg) as a crude
product. A mixture of the crude (100 mg), NaN (113 mg, 1.74 mmol), and NH Cl (1.74 mmol)
3
4
in DMF (10 mL) was stirred at room temperature for 6 h. The reaction mixture was poured into
water and extracted with EtOAc. The separated organic layer was washed with brine, dried over
Na SO , and concentrated. The residue was purified by preparative HPLC twice (method B and
2
4
1
C) to give 10 (30 mg, 18% for 2 steps) as a yellow oil. H NMR (300 MHz, DMSO-d ) δ1.26 (d,
6
J = 6.9 Hz, 6H), 2.89–3.09 (m, 1H), 7.01 (d, J = 8.3 Hz, 1H), 7.10 (d, J = 1.3 Hz, 1H), 7.31 (s,
+
+
1
H), 7.58 (d, J = 8.3 Hz, 1H), 11.98 (brs, 1H). MS m/z 228 (M+H) . HRMS m/z (M+H) . Calcd
+
for C H N : 228.1244. Found: 228.1234. Purity 97.4% (HPLC).
12
14
5
Methyl 4-fluoro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate (22a). To a mixture of
-bromo-2-fluorobenzaldehyde 20a (2.00 g, 9.85 mmol),
,4,5,5-tetramethyl-2-(prop-1-en-2-yl)-1,3,2-dioxaborolane 21 (2.50 g, 14.8 mmol), and Na CO
4
4
2
3
(
3.10 g, 29.6 mmol) in dioxane (10 mL) and water (5 mL) was added Pd(PPh ) Cl (344 mg,
3
2
2
0
.490 mmol). The resulting mixture was degassed and then heated at 60 °C for 3 h under N2
atmosphere. The reaction mixture was diluted with water and extracted with EtOAc. The organic
layer was separated, washed with brine, dried over Na SO , and concentrated. The residue was
2
4
purified by silica gel column chromatography (petroleum ether/EtOAc = 50/1) to give
-fluoro-4-(prop-1-en-2-yl)benzaldehyde (1.30 g) as a crude product. The crude (1.30 g) and
2
NaOMe (1.30 g, 23.8 mmol) were dissolved in MeOH (30 mL), and the solution was added
dropwise to ethyl 2-azidoacetate (3.10 g, 23.8 mmol) at 0 °C. The resulting mixture was stirred
at room temperature overnight. The reaction mixture was concentrated, diluted with water, and
2
0

extracted with EtOAc. The organic layer was separated, washed with brine, dried over Na SO ,
2
4
and concentrated. The residue was purified by silica gel column chromatography (petroleum
ether/EtOAc = 30/1) to give methyl 2-azido-3-(2-fluoro-4-(prop-1-en-2-yl)phenyl)acrylate (1.20
g) as a crude product. Then, the crude (400 mg) in toluene (3 mL) was stirred under microwave
irradiation at 150 °C for 50 min. The reaction mixture was filtrated, and the filtrate was
concentrated. The residue was purified by silica gel preparative TLC (petroleum ether/EtOAc =
1
3
2
7
/1) to afford 22a (130 mg, 17% for 3 steps) as a white solid. H NMR (300 MHz, DMSO-d ) δ
6
.15 (s, 3H), 3.89 (s, 3H), 5.16 (s, 1H), 5.51 (s, 1H), 7.11 (d, J = 1.2 Hz, 1H), 7.13–7.19 (m, 1H),
+
.33 (s, 1H), 12.29 (brs, 1H). MS m/z 234 (M+H) .
4
-Fluoro-6-isopropyl-1H-indole-2-carboxylic acid (13). A mixture of 22a (130 mg, 0.557
mmol) and 10% Pd/C (13 mg) in MeOH (5 mL) was hydrogenated (0.20 MPa) at room
temperature overnight. The mixture was filtered, and the filtrate was concentrated to afford
methyl 4-fluoro-6-isopropyl-1H-indole-2-carboxylate (120 mg) as a crude product. To the crude
(
120 mg) in THF (10 mL) was added NaOH (57.6 mg, 1.44 mmol), and the reaction mixture was
stirred at room temperature overnight. The mixture was concentrated, and the residue was poured
into water. The mixture was then adjusted to pH 3–4 with 1 M HCl and extracted with EtOAc.
The organic layer was washed with brine, dried over Na SO , and concentrated. The residue was
2
4
purified by preparative HPLC (method A) to give 13 (30 mg, 25% for 2 steps) as a white solid.
1
H NMR (400 MHz, CD OD) δ1.30 (d, J = 6.8 Hz, 6H), 2.97–2.98 (m, 1H), 6.67 (d, J = 11.6 Hz,
3
+
1
H), 7.08 (d, J = 7.2 Hz, 2H). 2H were not observed. MS m/z 222 (M+H) . Anal. Calcd for
C H FNO : C, 65.15; H,5.47; N, 6.33. Found: C, 65.04; H, 5.52; N, 6.28.
1
2
12
2
Methyl 4-chloro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate (22b). Compound 22b was
prepared from compound 20b in a similar manner to that described for the synthesis of
2
1

1
compound 22a and obtained as a white solid (31% for 3 steps). H NMR (300 MHz, CDCl ) δ
3
2
7
.19 (s, 3H), 3.96 (s, 3H), 5.15 (s, 1H), 5.43 (s, 1H), 7.28 (s, 1H), 7.35 (brs, 1H), 7.36 (brs, 1H),
+
.28 (s, 1H). MS m/z 250 (M+H) .
4
-Chloro-6-isopropyl-1H-indole-2-carboxylic acid (14). Compound 14 was prepared from
compound 22b in a similar manner to that described for the synthesis of compound 13 and
1
obtained as a white solid (41% for 2 steps). H NMR (400 MHz, DMSO-d ) δ1.24 (d, J = 6.8 Hz,
6
6
1
5
H), 2.95–3.04 (m, 1H), 6.99 (d, J = 1.2 Hz, 1H), 7.09 (s, 1H), 7.22 (s, 1H), 12.01 (brs, 1H),
+
3.13 (brs, 1H). MS m/z 238 (M+H) . Anal. Calcd for C H ClNO ·0.5H O: C, 58.43; H,5.31; N,
12
12
2
2
.68. Found: C, 58.35; H, 5.12; N, 5.61.
Methyl
5-fluoro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate
(22c)
and
methyl
7
-fluoro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate (22e). Compounds 22c and 22e were
prepared from compound 20c in a similar manner to that described for the synthesis of
compound 22a. Final indole ring formation reaction gave regioisomers 22c (4% for 3 steps) as a
1
white solid and 22e (8% in 3 step) as a white solid. 22c: H NMR (300 MHz, CDCl ) δ2.19 (s,
3
3
1
2
2
H), 3.95 (s, 3H), 5.26 (s, 2H), 7.14 (brs, 1H), 7.29 (d, J = 4.1 Hz, 1H), 7.32 (s, 1H), 8.82 (brs,
+
1
H). MS m/z 234 (M+H) . 22e: H NMR (300 MHz, CDCl ) δ2.22 (s, 3H), 3.96 (s, 3H), 5.29 (s,
3
H), 7.08 (t, J = 7.5 Hz, 1H), 7.17–7.23 (m, 1H), 7.39 (d, J = 8.3 Hz, 1H), 8.99 (brs, 1H). MS m/z
+
34 (M+H) . Structures of 22c and 22e were confirmed by X-ray crystal structure analysis (see
Supporting Information).
5
-Fluoro-6-isopropyl-1H-indole-2-carboxylic acid (15). Compound 15 was prepared from
compound 22c in a similar manner to that described for the synthesis of compound 13 and
1
obtained as a white solid (59% for 2 steps). H NMR (400 MHz, DMSO-d ) δ1.25 (d, J = 6.8 Hz,
6
6
H), 3.17–3.24 (m, 1H), 7.01 (d, J = 1.6 Hz, 1H), 7.29 (d, J = 6.8 Hz, 1H), 7.35 (d, J = 11.6 Hz,
2
2

+
1
H), 11.73 (brs, 1H), 12.97 (brs, 1H). MS m/z 222 (M+H) . Anal. Calcd for C H FNO ·0.2H O:
1
2
12
2
2
C, 64.11; H,5.56; N, 6.23. Found: C, 64.08; H, 5.42; N, 6.19.
7
-Fluoro-6-isopropyl-1H-indole-2-carboxylic acid (16). Compound 16 was prepared from
compound 22e in a similar manner to that described for the synthesis of compound 13 and
1
obtained as a white solid (24% for 2 steps). H NMR (400 MHz, DMSO-d ) δ1.26 (d, J = 6.8 Hz,
6
6
H), 3.29–3.36 (m, 1H), 6.99–7.03 (m, 1H), 7.11 (t, J = 2.6 Hz, 1H), 7.40 (d, J = 8.0 Hz, 1H),
+
1
2.08 (s, 1H), 13.01 (s, 1H). MS m/z 222 (M+H) . Anal. Calcd for C H FNO ·0.3H O: C, 63.55;
1
2
12
2
2
H,5.61; N, 6.18. Found: C, 63.52; H, 5.42; N, 6.16.
Methyl
5-chloro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate
(22d)
and
methyl
7
-chloro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate (22f). Compounds 22d and 22f were
prepared from compound 20d in a similar manner to that described for the synthesis of
compound 22a. Final indole ring formation reaction gave regioisomers 22d (14% for 3 steps) as
1
an off-white solid and 22f (16% for 3 steps) as a white solid. 22d: H NMR (300 MHz,
DMSO-d ) δ2.09 (s, 3H), 3.88 (s, 3H), 4.96 (s, 1H), 5.26 (s, 1H), 7.12 (s, 1H), 7.29 (s, 1H), 7.75
6
+
1
(
s, 1H), 12.09 (brs, 1H). MS m/z 250 (M+H) . 22f: H NMR (300 MHz, DMSO-d6) δ 2.11 (s,
3
8
H), 3.88 (s, 3H), 4.99 (s, 1H), 5.33 (s, 1H), 7.02 (d, J = 8.3 Hz, 1H), 7.25 (s, 1H), 7.61 (d, J =
+
.2 Hz, 1H), 11.99 (s, 1H). MS m/z 250 (M+H) .
7
-Chloro-6-isopropyl-1H-indole-2-carboxylic acid (17). Compound 17 was prepared from
compound 22f in a similar manner to that described for the synthesis of compound 13 and
1
obtained as a white solid (47% for 2 steps). H NMR (400 MHz, DMSO-d ) δ1.24 (d, J = 6.8 Hz,
6
6
H), 3.41–3.51 (m, 1H), 6.97 (s, 1H), 7.09 (d, J = 8.4 Hz, 1H), 7.55 (d, J = 8.0 Hz, 1H), 11.10
2
3

+
(
brs, 1H). 1H was not observed. MS m/z 238 (M+H) . Anal. Calcd for C H ClNO ·0.3H O: C,
1
2
12
2
2
5
9.29; H, 5.22; N, 5.76. Found: C, 59.10; H, 5.02; N, 5.58.
4
-Chloro-7-fluoro-6-isopropyl-1H-indole-2-carboxylic acid (18). To a solution of sodium
460 mg, 20.0 mmol) in MeOH (20 mL) was added solution of
-bromo-2-chloro-5-fluorobenzaldehyde 23 (1.19 g, 5.01 mmol) and methyl 2-azidoacetate (2.30
(
a
4
g, 20.0 mmol) in MeOH (10 mL) dropwise at −10 °C under N atmosphere. The resulting
2
mixture was warmed to room temperature and stirred for 12 h at room temperature. The reaction
mixture was poured into ice water and extracted with EtOAc. The organic layer was separated,
washed with brine, dried over Na SO , and concentrated. The residue was purified by silica gel
2
4
column
chromatography
(petroleum
ether)
to
give
methyl
2
-azido-3-(4-bromo-2-chloro-5-fluorophenyl)acrylate (400 mg) as a crude product. Then, the
crude (340 mg) was dissolved in xylene (10 mL), and the solution was stirred at 160 °C for 3 h.
The reaction mixture was concentrated in vacuo and the residue was solidified from petroleum
1
ether to give 24 (200 mg, 15% for 2 steps) as a white solid. H NMR (400 MHz, CDCl ) δ4.00 (s,
3
3
H), 7.28–7.32 (m, 2H), 9.24 (brs, 1H).
A mixture of 24 (200 mg, 0.652 mmol), 21 (110 mg, 0.654 mmol), Na CO (138 mg, 1.30
2
3
mmol), and Pd(dppf)Cl (48 mg, 0.066 mmol) in dioxane (50 mL) and water (5 mL) was
2
degassed and then heated to 100 °C for 12 h under N atmosphere. The reaction mixture was
2
poured into water and extracted with EtOAc. The organic layer was separated, washed with
brine, dried over Na SO , and concentrated. The residue was purified by silica gel column
2
4
chromatography
(petroleum
ether/EtOAc
=
10/1)
to
afford
methyl
4
-chloro-7-fluoro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate (100 mg, 57%) as a light yellow
2
4

1
solid. H NMR (400 MHz, CDCl ) δ2.21 (s, 3H), 3.99 (s, 3H), 5.31–5.36 (m, 2H), 7.11 (d, J =
3
6
.0 Hz, 1H), 7.28–7.30 (m, 1H), 9.19 (brs, 1H).
To a mixture of methyl 4-chloro-7-fluoro-6-(prop-1-en-2-yl)-1H-indole-2-carboxylate (100
mg, 0.373 mmol) and 10% Pd/C (30 mg) in MeOH (20 mL) was added ZnBr (112 mg, 0.497
2
mmol) at room temperature. The resulting mixture was hydrogenated (0.28 MPa) for 12 h. The
reaction mixture was filtered, and the filtrate was concentrated to give methyl
4
-chloro-7-fluoro-6-isopropyl-1H-indole-2-carboxylate as a crude product. The crude and
LiOH·H O (31.3 mg, 0.746 mmol) were dissolved in THF (20 mL) and H O (5 mL), and the
2
2
solution was stirred at room temperature for 12 h. Then the pH of the mixture was adjusted to <
6
with 4 M HCl and the mixture was concentrated. The residue was purified by preparative
1
HPLC (method D) to give 18 (40 mg, 42% for 2 steps) as a white solid. H NMR (400 MHz,
CDCl ) δ1.33 (d, J = 6.8 Hz, 6H), 3.35–3.45 (m, 1H), 7.06 (d, J = 5.2 Hz, 1H), 7.42–7.43 (m,
3
1
3
1
H), 9.11(brs, 1H). 1H was not observed. C NMR (151 MHz, CDCl ) δppm 23.0 (2C), 27.0 (d,
3
J = 2.2 Hz, 1C), 109.5, 119.5 (d, J = 3.3 Hz, 1C), 122.5 (d, J = 3.3 Hz, 1C), 126.8, 127.0 (d, J =
1
6.6 Hz, 1C), 127.2 (d, J = 5.0 Hz, 1C), 131.7 (d, J = 12.2 Hz, 1C), 145.4 (d, J = 243 Hz, 1C),
+
1
64.1. MS m/z 256 (M+H) . Anal. Calcd for C H ClFNO ·0.3H O: C,55.21; H,4.48; N, 5.36.
12
11
2
2
Found: C, 55.29; H, 4.36; N, 5.40.
Preparation of enzymes. The human recombinant proteins, full-length eIF4A3,
MLN51(residues 137–283), full-length eIF4A1, full-length eIF4B, and eIF4G (residues
7
12–1451) were expressed in Escherichia coli BL21(DE3) as fusion proteins with 6 × histidine
His)-small ubiquitin-like modifier (SUMO) or His tags followed by a tobacco etch virus (TEV)
protease cleavage site at the N-terminus and purification by Ni-NTA superflow affinity column
QIAGEN) and Superdex 200 gel-filtration column (GE Healthcare). The His-SUMO or His tags
(
(
2
5

were cleaved with SUMO protease or TEV protease. Human recombinant full-length DHX29
and full-length BRR2 were expressed in Sf-9 insect cells as fusion proteins with FLAG-His or
His tags at the N terminus, respectively, using the BaculoDirect C-term Baculovirus expression
system (Thermo Fisher Scientific). FLAG-His-DHX29 and His-BRR2 were purified by Ni-NTA
superflow affinity column and Superdex 200 gel-filtration column. Protein concentrations were
determined using a BCA Protein Assay Kit (Thermo Fisher Scientific) with bovine serum
albumin as a standard. Recombinant human His-eIF4A2 protein was purchased from
GeneCopoeia.
RNA-dependent ATPase assay. The RNA-dependent ATPase assay was performed using the
ADP-Glo assay system (Promega). Single-stranded RNA poly(U) was purchased from MP
Biomedicals. The assay buffer comprised 20 mM Tris-HCl (pH7.5), 2.5 mM MgCl , 100 mM
2
KCl, 1 mM dithiothreitol (DTT), and 0.01% (v/v) Tween20. To enhance ATPase activity for
eIF4A, the equivalent molar concentration of MLN51 for 150 nM eIF4A3 or eIF4B and eIF4G
for 100 nM eIF4A1 or eIF4A2 were added. Regarding the ATPase assays for DHX29 or BRR2,
the optimal concentrations were 6.4 nM and 6.25 nM, respectively. Concentrations of ATP or
RNA were set at the Km value of each substrate for each enzyme as follows: 35 μM ATP and 1.5
μg/mL poly(U) for eIF4A1 and eIF4A3; 20 μM ATP and 3.0 μg/mL poly(U) for eIF4A2; 30 μM
ATP and 1.8 μg/mL poly(U) for DHX29; and 20 μM ATP and 2.5 μg/mL poly(U) for BRR2.
After the addition of the substrates and test compounds, the ATPase reactions were started by the
addition of the enzymes. They were incubated at room temperature for 30 min for eIF4A3,
DHX29, and BRR2 or 40 min for eIF4A1 and eIF4A2. The enzymatic reactions were terminated
by ADP-Glo reagent, and then ADP-Glo detection reagent was added to detect ADP.
Luminescent signals were measured using an EnVision 2102 Multilabel Plate Reader
(
PerkinElmer). We defined the luminescent signals of the reaction without enzyme as 100%
2
6

inhibitory activity and those of the complete reaction mixture as 0% inhibitory activity. Curve
fittings and calculations of IC values were performed using the program XLfit version 5 (ID
50
Business Solutions) with the maximum and minimum of the curve constrained to 100 and 0,
respectively.
Substrate competitive assay. ATP or RNA competition assay for eIF4A3 were performed
using the ADP-Glo Max assay system (Promega). The enzymatic reactions were performed
under identical conditions to those used for RNA-dependent ATPase assay, except for substrate
concentrations. ATP or RNA concentrations were set at the Km values— 35 μM or 1.5 μg/mL,
respectively—and 3.5–175 μM ATP or 60 μg/mL poly (U) were used to examine ATP or RNA
competitive inhibition. The enzymatic reactions were performed at room temperature for 105
min, except for 45 min at lower concentration than 44 μM ATP and 1.5 μg/mL poly(U), and for
3
0 minat 35 μM ATP and 60 μg/mLpoly(U). Curve fittings and calculations of IC values were
5
0
performed using the program GraphPad Prism version 5.03 (GraphPad Software, Inc.). Ki was
calculated using following formulas: IC =Ki (1+ [ATP]/K ), where [ATP] represents ATP
is the Michaelis constant, and Ki demonstrates the inhibitor constant.
m
5
0
concentration, K
m
Isothermal titration calorimetry (ITC). All measurements were performed using a Microcal
iTC200 calorimeter (Malvern) at 25 °C. The syringe was loaded with the compound; the sample
cell was filled with the protein. The protein solution was exchanged into the stock buffer (50 mM
Tris pH 7.6, 0.15 M NaCl) and concentrated to 10–20 μM using ultrafiltration (Merck Millipore).
The final concentration of compounds was 100–200 μM. Binding isotherms were analyzed by
nonlinear regression using the software Origin 7.0 (OriginLab). Prior to curve fitting, the heat of
dilution was corrected by subtracting the average heat of injection after saturation. A one-site
binding model was employed to yield the stoichiometry n, the dissociation constant K and the
D
2
7

enthalpy change ΔH. Standard free energy of binding and entropic contribution were obtained,
respectively, as ΔG = −RT ln (1/K ) and −TΔS = ΔG − ΔH.
D
Surface plasmon resonance (SPR) biosensing assay. SPR biosensing experiments were
performed on BiacoreS200 and a BiacoreT100 instrument equipped with CM7 sensor chips (GE
healthcare). Immobilization of eIF4A3 onto CM7 sensor surface was performed following the
standard amine coupling procedure according to the manufacturer's instructions. HBS-P+ (10
mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.05% Surfactant P20) supplemented with 1 mM DTT
was used as the running buffer for immobilization. Typical immobilization levels ranged from
5
000 to 10000 RUs. All binding experiments were performed in 20 mM Tris-HCl, pH 8.0, 150
mM NaCl, 0.2 mM TCEP, 0.02% Surfactant P20 (v/v), and 1–2% DMSO (v/v) at 25 °C.
Differently concentrated compound solutions (7–10 points) were injected for 60–120 sec at a
flow rate of 50 μL/min and the dissociation was thereafter followed for up to 120 sec. Data
processing and analyzing were performed using BiacoreS200 and BiacoreT100 evaluation
software (GE healthcare). Sensorgrams were double referenced prior to global fitting the
concentration series to 1:1 binding model. Dissociative half-lives t were calculated form the
1
/2
dissociation rate constants k (t = ln2/k ).
off
1/2
off
Helicase assay. Helicase assays were performed in assay buffer containing 40 mM Tris-HCl
(
pH 7.5), 2 mM MgCl , 50 mM NaCl, 3 mM DTT and 0.01% Tween-20. dsRNAs were
2
generated with fluorescence labeled 12-mer (5'-(FITC)-GCUUUACGGUGC-3') and 24-mer
5'-GGGGAGAAAAACAAAACAAAACUAGCACCGUAAAGCACGC-3') oligo RNAs
purchased from GeneDesign. The RNAs in assay buffer (30 nM) were annealed by heating at 90
C for 5 min and slowly cooling to 4 °C at 0.1 °C/sec. Unwinding reaction was performed with
0 nM dsRNAs, 1.5 µM unlabeled 12-mer capture oligo RNA (5'-GCUUUACGGUGC-3')
(
°
3
2
8

(
Integrated DNA Technologies), 2 µM eIF4A3, 2 µM MLN51, and 2 mM ATP at 37 °C for 30 min.
Negative control compound 9 or eIF4A3 inhibitor 18 were also incubated at indicated
concentrations. Reactions were terminated with 40 mM EDTA, 3.2% (w/v) SDS, and 40% (v/v)
glycerol. The reaction solutions were electrophoresed with 15% polyacrylamide gel (ATTO) at
1
0.5 mV for 45 min under native running buffer containing 25 mM Tris and 192 mM glycine.
The fluorescence-labeled RNAs were visualized by ImageQuant LAS 4000 (GE healthcare) and
its fluorescence intensity was analyzed by Multi Gauge v3.1 (FUJI FILM). Remaining dsRNA
was calculated with following formula;
Remaining dsRNA = I (sample) × 100 / I (without ATP)
Where I (sample) and I (without ATP) are defined as fluorescence intensity of each sample and sample
without ATP, respectively. Statistical analysis was performed by Williams's multiple
comparison.
Acknowledgement
We thank Yasumi Kumagai and Motoo Iida for their assistance with NMR spectroscopic
experiments and analysis, Mitsuyoshi Nishitani for acquiring and determining X-ray crystal
structures, Yumi Imai and Haruyuki Nishida for supervising the structure determination. We
thank Jyun-ichi Sakamoto for support of SPR analysis, Takashi Ito for preparation of proteins,
and Shinji Moriyama, Shin-ichi Matsumoto, Shigeru Kondo, and Akihiko Naotsuka for
conducting high-throughput screening. We would like to express our appreciation to Douglas R.
Cary, Nao Morishita, Satoshi Endo, Daisuke Nakata, Shinsuke Araki, Ryo Mizojiri, and
Hironobu Maezaki. We thank Ryosuke Tokunoh, Tomohiko Suzaki, Yuuko Hitomi, Atsuo Baba,
2
9