Synthesis and pharmacological evaluation of carboxycoumarins as a new antitumor treatment targeting lactate transport in cancer cells

Article abstract of DOI:10.1016/j.bmc.2013.09.010

Under hypoxia, cancer cells consume glucose and release lactate at a high rate. Lactate was recently documented to be recaptured by oxygenated cancer cells to fuel the TCA cycle and thereby to support tumor growth. Monocarboxylate transporters (MCT) are the main lactate carriers and therefore represent potential therapeutic targets to limit cancer progression. In this study, we have developed and implemented a stepwise in vitro screening procedure on human cancer cells to identify new potent MCT inhibitors. Various 7-substituted carboxycoumarins and quinolinone derivatives were synthesized and pharmacologically evaluated. Most active compounds were obtained using a palladium-catalyzed Buchwald-Hartwig type coupling reaction, which proved to be a quick and efficient method to obtain aminocarboxycoumarin derivatives. Inhibition of lactate flux revealed that the most active compound 19 (IC ₅₀ 11 nM) was three log orders more active than the CHC reference compound. Comparison with warfarin, a conventional anticoagulant coumarin, further showed that compound 19 did not influence the prothrombin time which, together with a good in vitro ADME profile, supports the potential of this new family of compounds to act as anticancer drugs through inhibition of lactate flux.

Full text of DOI:10.1016/j.bmc.2013.09.010

Bioorganic & Medicinal Chemistry 21 (2013) 7107–7117
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and pharmacological evaluation of carboxycoumarins
as a new antitumor treatment targeting lactate transport in
cancer cells
Nihed Draoui a,b, Olivier Schicke b, Antony Fernandes b, Xavier Drozak b, Fady Nahra b, Amélie Dumont c,
Jonathan Douxfils d, Emmanuel Hermans c, Jean-Michel Dogné d, Romu Corbau e, Arnaud Marchand e,
b,
Patrick Chaltin e,f, Pierre Sonveaux a, Olivier Feron a, , Olivier Riant
⇑
⇑
a UCL/SSS/IREC/FATH, Pole of Pharmacology and Therapeutics (FATH), Institut de Recherche Expérimentale et Clinique (IREC), UCLouvain, Avenue E. Mounier 53, B1.53.09, B-1200
Brussels, Belgium
b Molecules, Solids and Reactivity (MOST), Institute of Condensed Matter and Nanosciences (IMCN), UCLouvain, Place Louis Pasteur 1, 1348 Louvain-la-Neuve, Belgium
c Cellular and Molecular Division (CEMO), Institute of NeuroScience (IoNS), UCLouvain, Avenue Hippocrate 54, 1200 Brussels, Belgium
d Namur Thrombosis and Hemostasis Center (NTHC), University of Namur, Rue de Bruxelles 61, 5000 Namur, Belgium
e CISTIM Leuven vzw, Bio-incubator 2, Gaston Geenslaan 2, 3001 Leuven, Belgium
f Centre for Drug Design and Discovery (CD3), KULeuven, Bio-incubator 2, Gaston Geenslaan 2, 3001 Leuven, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 12 June 2013
Revised 30 August 2013
Accepted 4 September 2013
Available online 13 September 2013
Under hypoxia, cancer cells consume glucose and release lactate at a high rate. Lactate was recently
documented to be recaptured by oxygenated cancer cells to fuel the TCA cycle and thereby to support
tumor growth. Monocarboxylate transporters (MCT) are the main lactate carriers and therefore represent
potential therapeutic targets to limit cancer progression. In this study, we have developed and
implemented a stepwise in vitro screening procedure on human cancer cells to identify new potent
MCT inhibitors. Various 7-substituted carboxycoumarins and quinolinone derivatives were synthesized
and pharmacologically evaluated. Most active compounds were obtained using a palladium-catalyzed
Buchwald–Hartwig type coupling reaction, which proved to be a quick and efficient method to obtain
aminocarboxycoumarin derivatives. Inhibition of lactate flux revealed that the most active compound
19 (IC50 11 nM) was three log orders more active than the CHC reference compound. Comparison with
warfarin, a conventional anticoagulant coumarin, further showed that compound 19 did not influence
the prothrombin time which, together with a good in vitro ADME profile, supports the potential of this
new family of compounds to act as anticancer drugs through inhibition of lactate flux.
Keywords:
Cancer
Monocarboxylate transporter
Lactate
Carboxycoumarins
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
endothelial cells10,11 as well as between tumor-associated
fibroblasts and oxidative tumor cells.12,13 These observations have
Cancer cells consume large amounts of glucose to survive and
proliferate.1 The high turnover of the glycolytic pathway uncou-
pled from mitochondrial respiration accounts for the high concen-
tration of lactate (up to 10 mM) in tumors,2,3 that is proposed to be
associated with tumor invasiveness and poor patient outcomes.4–7
Lactate however does not merely represent a waste metabolite or a
biomarker of tumor aggressiveness. It can indeed be captured by
oxidative tumor cells and reconverted into pyruvate to be used
in the TCA cycle.2 A synergy actually takes place between glycolytic
tumor cells exporting lactate, and oxygenated cells importing it to
feed their metabolism.8,9 In addition, lactate shuttle has been re-
ported to occur in tumors between glycolytic tumor cells and
led investigators to focus on the expression and the regulation of
the monocarboxylate transporters (MCT) in tumors since these
proteins represent the major path for inward and outward lactate
fluxes.14
The family of MCT (also named SLC16 solute carrier) is com-
posed of 14 members.15,16 Among them, only four isoforms
(MCT1-4) have been documented to act as proton-linked trans-
porters that can carry short chain monocarboxylates such as lac-
tate and pyruvate across cell membranes.17,18 In cancer cells,
MCT1 and MCT4 are the most widely expressed.16,19 MCT1 shows
a better affinity for L
-lactate than MCT4,16 but MCT4 has a higher
turnover rate than MCT1.20 These differences are consistent with
their respective roles in tumors.18,21 With a high affinity for lactate,
MCT1 enables lactate entry into oxidative tumor cells whereas low
affinity MCT4 is mainly expressed in glycolytic tumor cells
and tumor-associated fibroblasts that export lactate. The
⇑
Corresponding authors. Tel.: +32 2 7645264; fax: +32 2 7645269 (O. Feron).
E-mail addresses: olivier.feron@uclouvain.be (O. Feron), olivier.riant@uclouvain.
be (O. Riant).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.09.010

7108
N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
complementarity between MCT1 and MCT4 to drive lactate shut-
tle(s) in tumors, represents an attractive target for new anticancer
drugs. MCT1 blockade can indeed prevent oxygenated tumor cells
to use lactate and therefore force them to consume glucose more
avidly.8 Consequently, hypoxic tumor cells that are essentially
dependent on glucose and have limited or no access to replace-
ment fuels die from glucose deprivation.8,9 As an alternative ther-
apeutic strategy, interference with tumor pH regulation through
the inhibition of carbonic anhydrases, anion exchangers but also
MCT4 can lead to a cytosolic acidification of glycolytic tumor cells
up to a level incompatible with cell survival.22
alkyl or benzylic halide in DMF, or the Mitsunobu etherification
with the corresponding alcohol as the alkylating agent. The methyl
ester group was then hydrolyzed with lithium hydroxide in THF/
water, and the desired carboxylic acids were isolated in good over-
all yields and suitable purity after acidification and crystallization
in ethanol.
We then investigated the replacement of the oxygen atom of
the lactone ring by a substituted or un-substituted nitrogen atom
(Scheme 2). This aimed to examine the impact of the hydrogen
bond donor –NH– (instead of the H-bond acceptor –O–) but also
of diverse nitrogen substitutions on the ability of the compounds
to more efficiently inhibit lactate transporters. Protected 3-hydro-
xy aniline bearing an NH-Boc ortho-directing group31,32 was easily
prepared in two-step starting from commercially available 3-ami-
no-phenol. ortho-Lithiation was then carried out with tert-butyl
lithium (2.4 equiv) in diethyl ether at À40 °C. The lithio intermedi-
ate was trapped with DMF, and the desired protected amino alde-
a
-Cyano-4-hydroxycinnamate (CHC) was historically reported
as the first MCT inhibitor.23 CHC, however, is usually described to
be active in the upper M range and lacks specificity since in some
l
experimental setups, it can also inhibit the mitochondrial pyruvate
carrier.24 More recently, AR-C155858, a highly potent MCT1/MCT2
inhibitor was disclosed by Astra-Zeneca.25 This compound was
demonstrated to strongly and selectively block MCT1 and MCT2
activity in activated T-lymphocytes, obstructing lactate efflux and
thereby acting as a powerful immunosuppressive drug.26 The ther-
apeutic effects of this compound in tumors is however limited by
the compensatory effects of MCT4 that can take the lead to facili-
tate lactate efflux when the high affinity MCT1 transporter is
blocked.20
hyde was isolated in
a 44% yield after chromatographic
purification. Deprotection of the TBS and benzylation of the phenol
with benzyl bromide was carried out in a one-pot procedure by
using potassium fluoride in DMF. Cyclisation to the corresponding
unsubstituted lactam ring with diethyl malonate and followed by
alkylation of the nitrogen atom and final saponification of the ester
group gave the desired acids 10–13 in good overall yields.
Here, we designed a stepwise screening procedure to identify
MCT inhibitors able to prevent lactate influx into tumor cells
expressing both MCT1 and MCT4. Rational design led us to identify
monocarboxylate-containing coumarins as a potential scaffold en-
dowed with inhibitory effects on lactate transport. The Knoevena-
gel reaction27 and Palladium-coupling methods28–30 were the main
routes of synthesis. SAR studies confirmed the ability of synthe-
sized compounds to efficiently block lactate transport in the nano-
molar range, leading to anti-proliferative effects.
The impact of substitution on the position 4 of the coumarin
ring was also investigated (Scheme 3) and we designed the prepa-
ration derivatives 14–16 starting from 2,4-dihydroxy acetophe-
none. Regioselective alkylation occurred on position 4 to yield
the mono-alkylated acetophenones. The methyl ketone being
much less electrophilic compared to the corresponding aldehyde,
no reaction occurred when the ketones were directly reacted with
either methyl malonate or Meldrum acid using various conditions.
It was found that the desired coumarins could however be ob-
tained by prior activation of the methyl ketone by ammonia which
gave the corresponding imine derivatives which were not isolated
and directly reacted with Meldrum acid.
2. Results
We then focused on the direct introduction of various amino
substituents of the position 7 starting from the corresponding
7-triflyl or 7-iodoalkylcarboxycoumarin derivatives. In our hands,
all the attempts aiming at the substitution of either the triflate or
iodo35,36 derivatives by an amino group via palladium-catalyzed
cross-coupling33 as well as copper (I)-catalyzed Buchwald amina-
tion failed to produce the desired pure aminocarboxycoumarins
(Scheme 4). We then adopted and optimized a second route that
started by a palladium-catalyzed Buchwald–Hartwig type coupling
reaction, which proved to be a quick and efficient method to obtain
aminocarboxycoumarin derivatives.28–30 Indeed, after careful
selection of the palladium precatalyst [Pd(AcO)2] associated with
2.1. Chemistry
We used the Knoevenagel condensation to obtain substituted
carboxycoumarins for the first in vitro structure–activity relation-
ship studies. We then carried out further modulation on the
coumarin scaffold using simple nucleophilic substitutions or
Mitsunobu reactions to optimize the activity of hits by varying
the substituent on the position 7 (Scheme 1). The various coumarin
derivatives 1–9 were easily prepared starting from 2,4-dihydroxy
benzaldehyde. Knoevenagel condensation with dimethyl malonate
gave the common precursor in excellent yield. Alkylation of the
phenol group was then carried out using either the corresponding
RO
O
O
O
HO
OH
O
HO
O
O
O
i
ii
OH
O
1-4, 7 and 8
iii
R = 1 -CH3
2 -CH2-C6H5
3 -CH2-CH2-C6H5
4 -geranyl
RO
O
O
5 -2-indanyl
OH
6 -N-benzotriazolyl
7 -m-fluorobenzyl
O
5, 6 and 9
8 -3,5-bis-trifluoromethylbenzyl
9 -p-fluorobenzyl
Scheme 1. General synthesis of compounds 1–9. Reagents and conditions: (i) Dimethyl malonate, (piper/AcOH) cat in EtOH, 3 h reflux. (ii) R-X, Cs2CO3 or K2CO3 in DMF, O/N
reflux followed by LiOH in THF/H2O, 1 h reflux. (iii) R-OH, DTBAD/PPH3 in THF, O/N at RT followed by LiOH in THF/H2O, 1 h reflux.

N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
7109
HO
NH2
HO
NHBoc
ii TBSO
NHBoc
HO
NHBoc
O
i
iii
iv
R
N
R = 10 -H
11 -CH3
12 -CH2-CH3
13 -CH2-C6H5
O
O
COOH
O
NHBoc
v,vi
O
Scheme 2. General synthesis of compounds 10–13. Reagents and conditions: (i) di-tert-butyl dicarbonate in THF, O/N reflux. (ii) imidazole, TBSCl in DMF, O/N at RT.
(iii) t-BuLi, DMF in Et2O, À40 °C 2 h and 0 °C 1 h. (iv) Benzylbromide, KF in DMF, O/N at RT (v) Diethyl malonate, (piper/AcOH)cat in EtOH, O/N reflux (vi) R-I or R-Br, K2CO3
in DMSO, O/N refux followed by LiOH in THF/H2O, 1 h reflux.
RO
OH
O
HO
OH
O
RO
O
O
O
i
ii, iii
OH
14 R= -H
15 R= -CH2-C6H5
16 R= -CH3
Scheme 3. General synthesis of compounds 14–16. Reagents and conditions: (i) R-X, Cs2CO3 in DMF, O/N reflux. (ii) NH3 7N in MetOH, O/N at RT. (iii) Meldrum acid in EtOH,
5 h reflux.
R1
2.2. Pharmacological evaluation
R1
NH
R2
2.2.1. Carboxycoumarin derivatives inhibit the proliferation of
X
O
O
N
O
O
R2
Pd or Cu
lactate-consuming human tumor cells with consistent SAR
The activity of the different compounds was first evaluated by
comparing their cytotoxic effects on human cancer cells in either
glucose- or lactate-containing medium after 72 h of treatment.
This assay was based on preliminary experiments documenting
that the human cervix carcinoma cell line SiHa can switch from a
glycolytic to an oxidative metabolism when lactate is available as
an energy substrate.8 This primary assay led to the identification
of compounds which selectively inhibited tumor cell proliferation
in the presence of lactate but failed to exert toxic effects in the
presence of glucose (Table 1). Concentration-dependent inhibition
of cell proliferation was compared with CHC as a reference com-
pound (see examples in Fig. 1a). Of note, in SiHa cells, lactate up-
take primarily depends on the high affinity MCT1 transporter.8
catalysed
coupling
COOR
COOR
no reaction, degradation or mixture
X=OTf, I
Scheme 4. Palladium- and copper (I)-catalyzed amination reaction.
the strong Verkade base as ligand and LiHMDS as a base in toluene,
we successfully substituted 3-bromophenol with various second-
ary amines to introduce the desired amino group which consti-
tuted the main building block. Formylation using classical
Vilsmeier–Haack reaction followed by cyclisation with Meldrum
acid allowed us to obtain the pure acids 17–23 without further
purification steps (Scheme 5).
R1
R1
N
OH
Br
OH
N
OH
O
R2
R2
i
ii
iii
R = 17 -N,N-Diethyl
18 -Morpholino
19 -N-Methylbenzyl
20 -Piperidino
R1
N
O
O
R2
21 -N,N-Dimethyl
22 -Pyrrolidino
OH
23 -Juloidino
17-23
O
Scheme 5. General synthesis of compounds 17–23. Reagents and conditions: (i) secondary amine, Pd(OAc)2, verkade base, LiHMDS in toluene, 24 h reflux. (ii) POCl3 in DMF,
1 h 90 °C. (iii) Meldrum acid, (piper/AcOH)cat in EtOH, 3 h reflux.

7110
N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
Indeed, although SiHa cells express both MCT1 and MCT4,34 the
higher affinity for lactate of MCT1 (Km: 3–6 mM) versus MCT4
(Km: 25–30 mM)16 indicates that in the presence of 10 mM lactate
in the culture medium (to mimic lactate concentration reported in
tumors),2 MCT1 functions at maximal rate contrary to MCT4. Thus,
the contribution of MCT4 to lactate entry in our experimental mod-
el should in theory be minimal and if any, compounds unable to
block this compensatory mechanism should not be selected in this
primary assay. Furthermore, we found that under hypoxia, the
expression of MCT4 was significantly increased in SiHa cells
(Supplementary Fig. 1a) and that, as long as glucose was present,
our compounds failed to show any cytotoxicty in these conditions
(Supplementary Fig. 1b).
a
125
100
75
CHC
Cpd 17
Cpd 19
50
25
0
-8
-7
-6
-5
-4
-25
Log[Drug] M
Several structure–activity relationships were identified (Table 1).
First, O-benzyl and secondary amino group on the position 7 of the
carboxycoumarin scaffold drastically improved compound toxicity
in lactate-containing medium without affecting the survival of
SiHa cells in the presence of glucose with or without pyruvate
(Fig. 1b). The carboxylic acid function was necessary to keep a
potent effect since the ester analogues failed to demonstrate any
toxicity (data not shown). However, methyl substitution on the
position 4 of the carboxycoumarin led to a complete loss of activity
(1, 2 vs 14, 15, 16), as well as the replacement of the lactone by a
lactam moiety (substituted or not) (2 vs 10, 11, 12, 13). These data
showed that the oxygen of the carboxycoumarin structure and the
freedom of the position 4 played key roles to impair lactate entry
or further intracellular metabolism. The primary SAR of these
new compounds is summarized in Scheme 6. Compounds 17 and
19 were also used to document the occurrence of apoptosis after
48 h exposure, as indicated by caspase-3 cleavage in tumor cells
incubated in lactate-containing medium (Fig. 1c).
b
CHC
Cpd 17
Cpd 19
100
*
50
0
**
**
r
Py
r
r
e
at
ct
Py
Py
Gluc
Gluc
Gluc
Gluc
Gluc
Gluc
ctate
La
ctate
La
+
+
+
La
2.2.2. Carboxycoumarin derivatives block lactate uptake in
human cancer cells
c
We then developed a secondary assay to test whether our se-
lected compounds functionally blocked lactate entry into cancer
cells. Using an enzymatic assay to measure the remaining lactate
concentration in culture medium after 24 h of SiHa cell treatment
with the compounds, we validated previous SAR conclusions (see
Table 1). As shown in Figure 2a, we observed a dose-dependent
inhibition of lactate consumption and a shift to the left of the sig-
moid dose-response curves of compounds 17 and 19. To further
confirm these observations, we measured [14C]-lactate uptake by
SiHa cells on a shorter time frame (12 min). In particular, we vali-
dated compounds 17 and 19 as potent inhibitors of lactate uptake
with IC50 values equal to ꢀ250 and ꢀ10 nM, respectively, in com-
Figure 1. Cytotoxic effects of compounds 17 and 19. (a) Human SiHa cancer cell
proliferation was measured after 72 h of incubation in lactate-containing medium
in the presence of increasing concentrations of the indicated compounds (n = 18).
parison to CHC which exhibited an IC50 ꢀ10
lM in this assay
(b) Bar graph represents the effect of 10 lM of the indicated compounds on SiHa
cell proliferation in glucose-containing medium with or without pyruvate (vs
lactate-containing medium); ⁄P < 0.05, ⁄⁄P < 0.01, n = 18. (c) Caspase-3 activation in
SiHa cells treated during 48 h in lactate-containing medium with compounds 17 or
(Fig. 2b).
2.2.3. In vitro ADME properties and in vivo PK profile of
compounds 17 and 19
19 (10 lM) and camptothecin (10 lM) used as positive control (n = 3).
We then evaluated the general in vitro ADME profile of com-
pounds 17 and 19. Both compounds showed a moderate-to-good
aqueous solubility, an excellent chemical stability in simulated
gastric (SGF) and intestinal (SIF) fluids, a good apparent permeabil-
ity coefficient (Papp) through Caco-2 monolayer and a high meta-
bolic stability on mouse (MLM) and human liver microsomes
(HLM) as well as on human hepatocytes (Table 2). They also
showed an excellent stability after 1 h incubation in mouse plasma,
and a moderate stability (17 > 19) in human plasma. Also, the
extent of human plasma protein binding (HPPB) of 17 and 19
supports a good potential for drug distribution. Finally, the intra-
peritoneal administration of 19 (3 mg/kg) to mice led to a Cmax of
2.2.4. Carboxycoumarins 17 and 19 do not show anticoagulant
properties
Since hit compounds had a coumarinic scaffold, we finally
aimed to exclude any major anticoagulant side effect. We first
tested the effects of compounds 17 and 19 on mouse survival in
comparison with WarfarinÒ, a well-known anticoagulant coumari-
nic drug. Compounds 17 or 19 did not induce any mortality at the
daily dose of 3 mg/kg while WarfarinÒ induced hemorrhages and
mouse death at doses starting from 1.5 mg/kg (Fig. 3a). We next
performed a prothrombin time test (PT time) on the plasma of
treated mice. We found that while 17 and 19 did not influence
the extrinsic pathway of coagulation, WarfarinÒ dramatically pro-
longed the clotting time (Fig. 3b).
1246 ng/ml (4
lM) in a very short time (Tmax = 10 min) associated
with a plasma half-life of 4.5 h (Table 2).

N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
7111
Table 1
Pharmacological evaluation of synthesized compounds
Compd
Substitutent (X)
Cytotoxicity
Lactate uptake inhibition
EC50
10.7
9.8
(
lM)
IC50
43.5
2.3
(lM)
CHC
Reference compound
1
2
3
9.1
5.6
1.9
2.4
9
3.9
1.25
10
>30
>30
11
12
>30
>30
>30
>30
13
14
>30
>30
>30
>30
15
>30
>30
>30
16
17
>30
1.8
0.86
18
19
17.5
7.9
0.22
0.059
1.4
21
22
2.9
14.3
>30
23
1.2
0.73
EC50: compound concentration that reduces cell proliferation by 50%.
IC50: compound concentration that reduces lactate uptake by 50%.
Compounds 4–8 did not exert any significant cytotoxicity.
thus in the security of use regarding healthy tissues. Our SAR study
documented that 7-alkylamino substituents on the 3-carboxy-
coumarin scaffold significantly improved the anti-proliferative ef-
fects of the compounds, with submicromolar EC50 values. We
previously reported that lactate can support cancer cell prolifera-
tion: upon uptake and intracellular oxidation into pyruvate, it sup-
ports TCA-cycle-dependent ATP synthesis and cataplerosis.8,9 To
3. Discussion
In this study, various 7-substituted carboxycoumarins and quin-
olinone derivatives were synthesized and pharmacologically evalu-
ated for their capacity (i) to kill lactate-fueled tumor cells and (ii) to
spare the same tumor cells when fueled by glucose. This double fil-
ter was designed to gain in tumor selectivity of our compounds and

7112
N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
Activity when X = -O-
No activity when X = -NH
or substituted N
Good activity with
R1
R2
Y
X
O
R
O
( )
R3
n
R = COOH
No activity when R = COOR
R1
N
R2
Loss of activity
when subtitued
Scheme 6. SAR study of the synthesized carboxycoumarins as inhibitors of lactate flux.
measuring inhibition of [14C]-lactate influx: the IC50 value for com-
pound 19 amounted to 11 nM in comparison to 11 M for CHC.
Compounds 17 and 23 showed an intermediary profile with IC50
ꢀ50–60 fold more favorable than CHC.
a
l
125
100
75
CHC
Cpd 17
Cpd 19
Interestingly, Supuran and colleagues have reported that cou-
marins could act as inhibitors of carbonic anhydrases, including
CAIX and CAXII.22,35,36 In our experimental model, we could docu-
ment that CAIX was not expressed in normoxic SiHa cells but could
be induced in response to hypoxia while CAXII was expressed inde-
pendently of the pO2 levels (Supplementary Fig. 1c). We further
found that in the presence of glucose, our compounds failed to ex-
ert any cytotoxic effects in normoxia (Fig. 1b) but also under hy-
poxia (Supplmentary Fig. 1b). These data support that in our
experimental model, compounds 17 and 19 exert their antiprolif-
erative activity through an inhibition of lactate uptake and not of
CAIX/CAXII. Still, the above observations suggest that some cou-
marinic compound could have the capacity to inhibit two of the
most critical pH regulators in tumors,22 namely carbonic anhydras-
es and MCT. Such possibility is supported by the recent work of
Bonneau et al who reported that the CA hydrolysis product of 3-cy-
ano-7-hydroxy-coumarin could lead to a CHC derivative.35
Finally, compounds 17 and 19 exhibited good in vitro ADME
profiles and 19 led to significant systemic exposure following in-
tra-peritoneal mouse administration demonstrating its potential
for further in vivo studies. Contrary to WarfarinÒ, the compounds
did not increase the prothrombin time (despite their coumarinic
scaffold) and did not lead to fatal hemorrhagic accidents.
50
25
0
-9
-8
-7
-6
-5
-4
-25
Log[Drug] M
b
125
CHC
Cpd 17
Cpd 19
100
75
50
25
0
-9
-8
-7
-6
-5
-4
Log[Drug] M
4. Conclusions
Compound
CHC
17
[IC50]
10.7 µM
0.25 µM
11.0 nM
We have identified a new family of inhibitors of lactate flux
through monocarboxylate transporters in human cancer cells.
The most active compounds are 7-alkylamino 3-carboxycouma-
rins, among which compound 19 has an IC50 of ꢀ10 nM. The lack
of toxicity in cells using glucose (instead of lactate) as a preferen-
tial energy fuel (which is the most common situation for healthy
tissues) together with the lack of anticoagulant activity allows to
anticipate a safe profile and a selective antitumor action of com-
pound 19.
19
Figure 2. Inhibition of lactate uptake by compounds 17 and 19. (a) Lactate uptake
by human SiHa cancer cells was determined by measuring the remaining lactate
concentration in the extracellular medium after a 24 h incubation in the presence of
increasing concentrations of compounds 17 and 19 (n = 3–6). (b) Radiolabeled
lactate influx was determined by measuring the amount of 14C-lactate captured by
SiHa cells after a 12 min incubation with compounds 17 and 19 (n = 3–5); in these
experiments, CHC was used as reference compound.
5. Experimental section
5.1. Chemistry
discriminate between compounds interfering with lactate trans-
port or metabolism, we used additional assays evaluating extracel-
lular lactate consumption and intracellular lactate uptake. Thus, the
secondary assay measuring lactate clearance from cell supernatant
was used to evaluate the most active compounds (i.e., cytotoxic hits
in the presence of lactate) using CHC as a reference compound. This
assay identified compound 19 as a potent inhibitor of lactate uptake
with an IC50 = 59 nM, that is, 3log orders more active than CHC
5.1.1. General procedure for the synthesis of 1–4, 7 and 8
(Scheme 1)
The Knoevenagel condensation allowed us to obtain important
coumarin derivatives for our stepwise synthesis. 2,4-dihydroxy-
benzaldehyde (1.0 equiv) was dissolved in ethanol before the di-
methyl malonate (1.2 equiv) was slowly added to the solution.
(IC50 = 43.5 lM). This was confirmed in the tertiary assay

N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
7113
Table 2
a
In vitro ADME and PK profiles of compounds 17 and 19
100
Compound
Compound
17
19
75
50
25
0
In vitro
M)a
199
113
109
50
>60
>120
108
58
105
91
>60
>60
>120
103
Vehicle
Aqueous solubility (
l
% Remaining after 1 h in SGFb
% Remaining after 4 h in SIFc
MLMd (T1/2 in min)
Cpd 17; 3 mg/kg
Cpd 19; 3 mg/kg
Warfarin; 1.5 mg/kg
Warfarin; 3 mg/kg
HLMe (T1/2 in min)
Human hepatocytes (T1/2 in min)
% Remaining after 1H incubation in mouse
plasma
% Remaining after 1H incubation in Human
plasma
69
44
0
1
2
3
4
5
6
In vivo mouse PK (3 mg/kg, intra-peritoneal)f
tmax (min)
Days of treatment
n.d.
n.d.
n.d.
n.d.
10
1246
270
Cmax (ng/ml)
t1/2 (min)
P < 0.01
AUCinf (min ng/ml)g
56109
50
b
N.d. = nondetermined.
40
30
20
10
0
a
Solubility measured in PBS (pH 7.4) with 2% DMSO.
Simulated gastric fluid.
Simulated intestinal fluid.
Mouse liver microsome.
Human liver microsome.
b
c
d
e
f
Injection of 50 ll in pure DMSO. Results are given as average of three mice
(serial sampling).
g
Area under the concentration–time curve from the time of dosing, extrapolated
to infinity.
Piperidine and acetic acid were added dropwise to catalyze the
reaction (one drop for 3 mmol).The reaction mixture was stirred
and heated under reflux 3 h. After cooling, the resulting precipitate
was filtered, dried and used for the nucleophilic substitution. The
coumarin product (1.0 equiv) was dissolved in DMF (3.5 ml per
mmol). The halogenoalkane (bromide or chloride derivative) was
then added to the solution (1.1 equiv) before addition of a weak
base (1.0 equiv of Cs2CO3 or K2CO3) to the mixture. The reaction
mixture was stirred and heated under reflux overnight. After cool-
ing, an equivalent volume of distilled water was added to the DMF
solution before liquid extraction with ethyl acetate. The organic
layers were merged and washed with a saturated solution of Li2SO4
to ensure residual DMF traces elimination. The final organic layer
was dried over Na2SO4, filtered and concentrated with the rotary
evaporator. The resulting solid was recrystallized from EtOH, fol-
lowed by hydrolysis with LiOH (10 equiv) in a THF/water solution
(1:1) heated 1 h under reflux. After cooling, the mixture was con-
centrated by rotary evaporation. The resulting precipitate was sol-
ubilized in a minimum volume of NH4OHcc/water (1:4). HCl 36%
was carefully added dropwise to the stirred solution until precipi-
tation at low acidic pH (control with pH paper). The resulting pre-
cipitate was filtered and recrystallized from EtOH to give the final
product.
Cpd 17
Cpd 19
Vehicle
Warfarin
No vehicle
Figure 3. Evaluation of the anti-coagulant effects of compounds 17 and 19 in mice.
(a) Survival curves of C57Bl6/J mice daily treated with vehicle, compounds 17 and
19 (3 mg/kg), or Warfarin (1.5 and 3 mg/kg) (10 mice per group). (b) Prothrombin
time using mouse plasma collected after 5 days of daily intraperitoneal injection
with vehicle (DMSO), 17 (3 mg/kg), 19 (3 mg/kg) or Warfarin (1.5 mg/kg).
2H) 13C NMR (DMSO-d6) d ppm 164.62 (C), 164.06 (C), 157.64
(C), 157.23 (C), 149.40 (CH), 136.45 (C), 132.05 (CH), 129.03 (CH),
128.69 (CH), 128.50 (CH), 114.56 (C), 114.28 (CH), 112.26 (C),
101.61 (CH), 70.64 (CH2) MS (APCI): m/z = 296.93 (MH+) HRMS:
m/z calcd for [C17H13O5] 297.07629, measured 297.07594.
Global yield: 31.8%.
5.1.1.3. 2-Oxo-7-phenethoxy-2H-chromene-3-carboxylic acid
(3).
1H NMR (DMSO-d6) d ppm 8.70 (s, 1H), 7.79 (d, J = 8.7,
1H), 7.33–7.22 (m, 5H), 7.02 (s, 1H), 6.98 (d, J = 8.7, 1H), 4.34 (t,
J = 6.9, 2H), 3.07 (t, J = 6.6, 2H) 13C NMR (DMSO-d6) d ppm 164.62
(C), 164.21 (C), 157.70 (C), 157.29 (C), 149.48 (CH), 138.36 (C),
132.03 (CH), 129.44 (CH), 128.82 (CH), 126.86 (CH), 114.29 (C),
114.01 (CH), 112.09 (C), 101.22 (CH), 69.55 (CH2), 35.02 (CH2)
MS (APCI): m/z = 310.96 (MH+) HRMS: m/z calcd for [C18H15O5]
311.0919, measured 311.0925.
5.1.1.1. 7-Methoxy-2-oxo-2H-chromene-3-carboxylic acid
(1). 1H NMR (300 MHz, DMSO): d ppm 8.69 (s, 1H), 7.80 (d, J =
8.4, 1H), 7.00 (s, 1H), 6.98 (d, J = 9.3, 1H), 3.88 (s, 3H).
13C NMR (75 MHz, DMSO): d ppm 165.04 (C), 164.69 (C), 157.69
(C), 157.30 (C), 149.29 (CH), 131.96 (CH), 114.60 (C), 113.72 (CH),
112.09 (C), 100.71 (CH), 56.68 (CH3).
Global yield: 9.6%.
5.1.1.4. (E)-7-(3,7-Dimethylocta-2,6-dienyloxy)-2-oxo-2H-chro-
mene-3-carboxylic acid (4).
1H NMR (DMSO-d6) d ppm
12.98 (s, 1H), 8.72 (s, 1H), 7.81 (d, J = 8.7, 1H), 7.01 (s, 1H), 6.98
(d, J = 9, 1H), 5.44 (t, J = 6, 1H), 5.04 (t, J = 5.7, 1H), 4.68 (d, J = 6.6,
2H), 2.09 (s, 4H), 1.72 (s, 3H), 1.59 (s, 3H), 1.55 (s, 3H) 13C NMR
(DMSO-d6) d ppm 164.61 (C), 164.33 (C), 157.69 (C), 157.29 (C),
149.58 (CH), 142.03 (C), 131.95 (CH), 131.53 (C), 124.14 (CH),
119.10 (CH), 114.26 (CH), 114.07 (C), 111.96 (C), 101.32 (CH),
65.93 (CH2), 39.68 (CH2), 26.19 (CH2), 25.87 (CH3), 17.99 (CH3),
16.85 (CH3) MS (APCI): m/z = 343.16 (MH+) HRMS: m/z calcd for
[C20H23O5] 343.15454, measured 343.15420.
MS (APCI): m/z = 221.09 (MH+).
HRMS: m/z calcd for [C11H9O5] 221.04500, measured
221.04433.
Global yield: 56.6%.
5.1.1.2. 7-(Benzyloxy)-2-oxo-2H-chromene-3-carboxylic acid
(2).
1H), 7.50–7.36 (m, 5H), 7.14 (s, 1H), 7.09 (d, J = 9, 1H), 5.26 (s,
1H NMR (DMSO-d6) d ppm 8.72 (s, 1H), 7.84 (d, J = 8.7,
Global yield: 13.9%.

7114
N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
5.1.1.5. 7-(3-Fluorobenzyloxy)-2-oxo-2H-chromene-3-carbox-
ylic acid (7).
1H NMR (DMSO-d6) d ppm 13.02 (s, 1H), 8.73
1H), 7.38 (s, 1H), 7.18 (d, J = 8.7, 1H), 6.98 (s, 2H) 13C NMR
(DMSO-d6) d ppm 164.49 (C), 161.16 (C), 157.27 (C), 156.75 (C),
149.12 (CH), 145.76 (C), 133.19 (C), 132.22 (CH), 129.04 (CH),
125.32 (CH), 119.98 (CH), 115.62 (C), 114.45 (CH), 113.93 (C),
111.15 (CH), 102.92 (CH), 74.04 (CH2) MS (APCI): m/z = 337.97
(MH+) HRMS: m/z calcd for [C17H12N3O5] 338.0777, measured
338.0777.
(s, 1H), 7.85 (d, J = 8.7, 1H), 7.47 (dd, J = 7.8, 1H), 7.34 (s, 1H),
7.33 (d, J = 7.5, 1H), 7.23–7.09 (m, 3H), 5.29 (s, 2H) 13C NMR
(DMSO-d6) d ppm 164.59 (C), 164.27 (C), 163.82 (C), 157.58 (C),
157.21 (C), 149.47 (CH), 139.40 (C), 132.11 (CH), 131.10 (CH),
124.33 (CH), 115.32 (CH), 114.89 (CH), 114.55 (C), 114.24 (CH),
112.37 (C), 101.66 (CH), 69.72 (CH2) MS (APCI): m/z = 314.84
(MH+) HRMS: m/z calcd for [C17H12O5F] 315.0669, measured
315.0663.
Global yield: 21.3%.
5.1.2.3. 7-(4-Fluorobenzyloxy)-2-oxo-2H-chromene-3-carbox-
Global yield: 45%.
ylic acid (9).
1H NMR (DMSO-d6) d ppm 13.02 (s, 1H), 8.72
(s, 1H), 7.84 (d, J = 8.7, 1H), 7.56 (d, J = 5.7, 1H), 7.53 (d, J = 5.7,
1H), 7.27–7.06 (m, 4H), 5.23 (s, 2H) 13C NMR (DMSO-d6) d ppm
164.59 (C), 163.97 (C), 160.82 (C), 157.62 (C), 157.24 (C), 149.50
(CH), 132.71 (C), 132.07 (CH), 130.92 (CH), 130.81 (CH), 116.01
(CH), 115.73 (CH), 114.42 (C), 114.26 (CH), 112.27 (C), 101.59
(CH), 69.91 (CH2) MS (APCI): m/z = 315.28 (MH+) HRMS: m/z calcd
for [C17H12O5F] 315.06687, measured 315.06703.
5.1.1.4. 7-(3,5-Bis(trifluoromethyl)benzyloxy)-2-oxo-2H-chro-
mene-3-carboxylic acid (8).
1H NMR (DMSO-d6) d ppm
13.04 (s, 1H), 8.73 (s, 1H), 8.22 (s, 2H), 8.13 (s, 1H), 7.87 (d,
J = 8.7, 1H), 7.18 (s, 1H), 7.15 (d, J = 8.7, 1H), 5.45 (s, 2H) 13C NMR
(DMSO-d6) d ppm 164.58 (C), 163.48 (C), 157.16 (C), 157.52 (C),
149.34 (CH), 140.03 (C), 132.16 (CH), 131.13 (C), 130.69 (C),
129.12 (CH), 125.54 (C),122.52 (CH), 122.42 (CH), 121.92 (C),
114.86 (C), 114.13 (CH), 112.58 (C), 101.66 (CH), 68.91 (CH2) MS
(APCI): m/z = 433.01 (MH+) HRMS: m/z calcd for [C19H11O5F6]
433.0511, measured 433.0509.
Global yield: 19.9%.
5.1.3. General procedure for the synthesis of 10–13 (Scheme 2)
To a solution of 3-aminophenol (1 g, 9.16 mmol, 1.0 equiv) in
THF (35 mL) was added a solution of di-tert-butyl dicarbonate
(2.41 g, 11.1 mmol, 1.2 equiv) in THF (10 mL). The reaction mixture
was refluxed overnight. Then the solvent was evaporated under
vacuum to afford a brown residue. The residue was dissolved in
EtOAc and washed with H2O, saturated NaHCO3 and brine, dried
over MgSO4 and concentrated in vacuo. This procedure provided
the carbamate derivative as a grey solid with in a quantitative
yield. Imidazole (1.37 g, 20.1 mmol, 2.0 equiv) was slowly added
at 0 °C to a solution of the solid previously synthesized (2.11 g,
10.0 mmol, 1.0 equiv) in DMF (15 ml) followed by a solution of
TBSCl (1.75 g, 11.6 mmol, 1.15 equiv) in DMF (5 ml). The reaction
mixture was stirred at room temperature overnight, partitioned
between half saturated NaHCO3 solution and AcOEt. The organic
layer was washed successively with half-saturated NH4Cl solution,
saturated NaHCO3 solution, and brine, then dried over MgSO4 and
concentrated in vacuo. This procedure provided the diprotected
compound as a white solid (2.89 g, 89%) with enough purity for
the next step. This compound (500 mg, 1.54 mmol, 1.0 equiv) in
solution in freshly distilled diethyl ether (15 ml) was added at
Global yield: 43.2%.
5.1.2. General procedure for the synthesis of 5, 6 and 9
(Scheme 1)
The Knoevenagel condensation allowed us to obtain important
coumarin derivatives for our stepwise synthesis. 2,4-dihydroxy-
benzaldehyde (1.0 equiv) was dissolved in ethanol before the di-
methyl malonate (1.2 equiv) was slowly added to the solution.
Piperidine and acetic acid were added dropwise to catalyze the
reaction (one drop for 3 mmol).The reaction mixture was stirred
and heated under reflux 3 h. After cooling, the resulting precipitate
was filtered, dried and used for the Mitsunobu reaction. The cou-
marin derivative (1.0 equiv) was dissolved in THF (8 ml per mmol).
The alcohol derivative (1.0 equiv) was added to the mixture. After
DTBAD (1.2 equiv) and triphenylphosphine (1.2 equiv) addition,
the reaction mixture was stirred overnight at room temperature.
We performed a liquid extraction with DCM/HCl 1 M before the or-
ganic layer was dried over Na2SO4 and concentrated with the ro-
tary evaporator. The resulting powder was recrystallized from
EtOH followed by hydrolysis with LiOH (10 equiv) in a THF/water
solution (1:1) heated 1 h under reflux. After cooling, the mixture
was concentrated by rotary evaporator. The resulting precipitate
was solubilized in a minimum volume of NH4OHcc/water (1:4).
HCl 36% was carefully added dropwise to the stirred solution until
precipitation at low acidic pH (control with pH paper). The result-
ing precipitate was filtered and recrystallized from EtOH to give
the final product.
À40 °C
a solution of tert-butyllithium in pentane [1.6 M]
(2.31 ml, 3.71 mmol, 2.4 equiv). The reaction mixture was stirred
at À40 °C during 2 h. DMF (0.95 ml, 12.36 mmol, 8.0 equiv) was in-
jected to the reaction mixture which was warmed to 0 °C and stir-
red during 1 h. The reaction mixture was partitioned between
water and diethyl ether. The aqueous layer was extracted three
times with diethyl ether. The combined organic layers were
washed two times with brine, dried over MgSO4 and concentrated
in vacuo.
5.1.2.1. 7-(2,3-Dihydro-1H-inden-2-yloxy)-2-oxo-2H-chromene-
3-carboxylic acid (5).
The residue was purified by column chromatography (petro-
leum ether/diethyl ether, 99:1) to give the desired formylated
product as a white powder (239 mg, 44%). Potassium fluoride
(140.4 mg, 2.42 mmol, 2.2 equiv) was added to a solution of C
(386.2 mg, 1.099 mmol, 1.0 equiv) in DMF (30.9 ml) followed by
benzyl bromide (0.17 ml, 1.43 mmol, 1.3 equiv). The reaction mix-
ture was stirred at room temperature overnight then diluted in
water and AcOEt. The aqueous layer was extracted two times with
AcOEt. The combined organic layer were washed two times with a
solution of HCl 0.5 M, one time with water, two times with a solu-
tion of LiCl (20%), and brine. The mixture was then dried over
MgSO4 and concentrated in vacuo to afford quantitatively a white
powder (351.3 mg). To a solution of the previously obtained
O-benzyl derivative (351.3 mg, 1.073 mmol, 1.0 equiv) in ethanol
(4.3 ml) was added diethyl malonate (0.4 ml, 2.68 mmol,
2.5 equiv), piperidine (0.26 ml, 2.7 mmol, 2.5 equiv) and acetic acid
1H NMR (DMSO-d6) d ppm 13.01 (s,
1H), 8.71 (s, 1H), 7.81 (d, J = 8.7, 1H), 7.27–7.15 (m, 4H), 7.06 (s,
1H), 6.96 (d, J = 8.7, 1H), 5.39 (s, 1H), 3.43 (d, J = 12, 2H), 3.05 (d,
J = 15, 2H) 13C NMR (DMSO-d6) d ppm 164.61 (C), 163.14 (C),
157.70 (C), 157.30 (C), 149.56 (CH), 140.86 (C), 132.17 (CH),
127.09 (CH), 125.09 (CH), 114.58 (CH), 114.18 (C), 112.05 (C),
101.91 (CH), 79.17 (CH), 39.52 (CH2) MS (APCI): m/z = 323.20
(MH+) HRMS: m/z calcd for [C19H15O6] 323.0919, measured
323.0911.
Global yield: 17.7%.
5.1.2.2. 7-((1H-Benzo[d][1,2,3]triazol-1-yl)methoxy)-2-oxo-2H-
chromene-3-carboxylic acid (6).
1H NMR (DMSO-d6) d ppm
13.11 (s, 1H), 8.72 (s, 1H), 8.13 (d, J = 8.4, 1H), 8.03 (d, J = 8.4,
1H), 7.87 (d, J = 8.7, 1H), 7.67 (dd, J = 8.1, 1H), 7.49 (dd, J = 7.8,

N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
7115
(0.006 ml, 0.107 mmol, 0.1 equiv). The mixture was stirred under
reflux overnight. The mixture was then cooled to À5 °C, the result-
ing precipitate was filtered and washed with cold ethanol (À20 °C)
to afford the desired quinolinone product (198 mg, 57%). K2CO3
(2 equiv) and LiI (20–25 mol %) were added to a solution of the
quinolinone derivative (1 equiv) in DMSO (1.5 ml/mmol) followed
by the alkyl halide (1.2 equiv). The reaction mixture was heated
under reflux overnight, then partitioned between water and
DCM. The aqueous layer was extracted three times with DCM.
The combined organic layers were washed with water, dried over
MgSO4 and concentrated in vacuo. To a solution of the ester prod-
uct (1 equiv) in THF (6 ml/mmol) was added a solution of LiOH
(10 equiv) in water (6 ml/mmol). The mixture was then vigorously
stirred and heated under reflux for 1h40. THF was removed by dis-
tillation under vacuum. The pH was then lowered to two by adding
the proper amount of HCl (1 M). The white precipitate was diluted
in AcOEt. The aqueous layer was extracted three times with AcOEt.
The combined organic layers were washed with water, dried over
MgSO4 and concentrated in vacuo. The crude product was then
recrystallized from ethanol to obtain a sample with enough purity
for biological test.
5.1.4. General procedure for the synthesis of 14–16 (Scheme 3)
We used the nucleophilic substitution as first step of synthesis.
2,4-dihydroxyacetophenone (1.0 equiv) was dissolved in DMF
(1.5 ml per mmol). The halogenoalkane (bromide or iodide deriva-
tive) was then added to the solution (1.1 equiv) before addition of
the weak base Cs2CO3 (1 equiv) to the mixture. The reaction mix-
ture was stirred and heated under reflux overnight. After cooling,
an equivalent volume of distilled water was added to the DMF
solution before liquid extraction with ethyl acetate. The organic
layers were merged and washed with a saturated solution of LiSO4
to ensure residual DMF traces elimination. The final organic layer
was dried over Na2SO4, filtered and concentrated with the rotary
evaporator. The resulting powder was recrystallized from EtOH.
The ketone function is then overnight activated into ketimine in
a solution of NH3 7 N in methanol at room temperature before syn-
thesis of the coumarin scaffold by the knoevenagel condensation.
The ketimine derivative (1 equiv) was dissolved in ethanol (4 ml
per mmol) before addition of Meldrum acid (1.2 equiv). The reac-
tion mixture was stirred 5 h under reflux to obtain the desired final
pure coumarinic product.
5.1.4.1. 7-Hydroxy-4-methyl-2-oxo-2H-chromene-3-carboxylic
acid (14).
5.1.3.1. 7-(Benzyloxy)-1,2-dihydro-2-oxoquinoline-3-carboxylic
acid (10).
1H NMR (DMSO-d6) d ppm 8.16 (s, 1H), 7.50 (d,
1H NMR (DMSO-d6) d ppm 13.02 (s, 1H), 8.88 (s,
J = 8.7, 1H), 6.79 (d, J = 2.1, 1H), 6.76 (s, 1H), 2.27 (s, 3H) 13C NMR
(DMSO-d6) d ppm 168.29 (C), 161.02 (C), 159.44 (C), 153.87 (C),
142.52 (C), 126.63 (CH), 126.307 (C), 113.37 (CH), 112.55 (C),
102.45 (CH), 16.09 (CH3) MS (APCI): m/z = 220.91 (MH+) HRMS:
m/z calcd for [C11H9O5] 221.04500, measured 221.04522.
Global yield: 3.7%.
1H), 7.97 (d, J = 9, 1H), 7.49–7.35 (m, 5H), 7.11 (dd, J = 8.7, 1H),
7.02 (d, J = 2.4, 1H), 5.23 (s, 2H) 13C NMR (DMSO-d6) d ppm
165.49 (C), 165.05 (C), 163.62 (C), 146.74 (CH), 142.34 (C),
136.53 (C), 132.84 (CH), 129.17 (CH), 128.84 (CH), 128.63 (CH),
114.91 (CH), 114.50 (C), 99.88 (C), 99.84 (CH), 70.53 (CH2)HRMS:
m/z (MH+) calcd for [C17H14NO4] 296.09173, measured 296.09176.
Global yield: 29%.
5.1.4.2. 7-(Benzyloxy)-4-methyl-2-oxo-2H-chromene-3-carbox-
ylic acid (15).
1H NMR (DMSO-d6) d ppm 7.62 (d, J = 9.6, 1H),
5.1.4.2. 7-(Benzyloxy)-1-methyl-2-oxo-1,2-dihydroquinoline-3-
carboxylic acid (11).
1H NMR (CDCl3) d ppm 14.47 (s, 1H),
7.49–7.34 (m, 5H), 7.01 (s, 1H), 6.98 (d, J = 7.5, 1H), 5.20 (s, 2H),
2.29 (s, 3H) 13C NMR (DMSO-d6) d ppm 167.88 (C), 160.61 (C),
159.27 (C), 153.65 (C), 141.88 (C), 136.92 (C), 128.97 (CH),
128.51 (CH), 128.36 (CH), 126.68 (CH), 114.36 (C), 112.84 (CH),
101.78 (CH), 70.21 (CH2), 16.11 (CH3) MS (APCI): m/z = 311.01
(MH+) HRMS: m/z calcd for [C18H15O5] 311.0919, measured
311.0916.
8.81 (s, 1H), 7.72 (d, J = 8.8 Hz, 1H), 7.50–7.34 (m, 5H), 7.08 (dd,
J = 8.8, 2.2 Hz, 1H), 6.95 (d, J = 2.0 Hz, 1H), 5.25 (s, 2H), 3.77 (s,
3H) 13C NMR (CDCl3) d ppm 165.58 (C), 164.32 (C), 163.84 (C),
145.58 (CH), 142.80 (C), 135.33 (C), 133.21 (CH), 128.91 (CH),
128.68 (CH), 127.58 (CH), 114.40 (CH), 113.21 (CH), 99.95 (CH),
70.83 (CH2), 30.22 (CH3) FTMS (pos, ESI): m/z = 310.1 (MH+) HRMS:
calculated for [C18H16O4N1], 310.1074, measured 310.1077.
Global yield: 28%.
Global yield: 10.8%.
5.1.4.3. 7-Methoxy-4-methyl-2-oxo-2H-chromene-3-carboxylic
acid (16).
1H NMR (DMSO-d6) d ppm 7.62 (d, J = 9.3, 1H),
5.1.3.3.
carboxylic acid (12).
7-(Benzyloxy)-1-ethyl-2-oxo-1,2-dihydroquinoline-3-
1H NMR (CDCl3) d ppm 14.53 (s, 1H),
6.92 (d, J = 6, 1H), 6.91 (s, 1H), 3.84 (s, 3H), 2.29 (s, 3H). 13C NMR
(DMSO-d6) d ppm 167.95 (C), 161.58 (C), 159.3 (C), 153.73 (C),
141.73 (C), 127.75 (C), 126.61 (CH), 114.19 (C), 112.23 (CH),
100.82 (CH), 56.20 (CH3), 16.12 (CH3) MS (APCI): m/z = 234.88
(MH+) HRMS: m/z calcd for [C12H11O5] 235.0606, measured
235.0602.
8.81 (s, 1H), 7.73 (d, J = 8.7 Hz, 1H), 7.56–7.32 (m, 5H), 7.08 (dd,
J = 8.7, 2.0 Hz, 1H), 6.95 (s, 1H), 5.26 (s, 2H), 4.37 (q, J = 7.1 Hz,
2H), 1.34 (t, J = 7.1 Hz, 3H) 13C NMR (75 MHz, CDCl3) d 165.64
(C), 164.73 (C), 163.91 (C), 145.53 (CH), 141.76 (C), 135.43 (C),
133.37 (CH), 128.96 (CH), 128.67 (CH), 127.47 (CH), 114.46 (C),
113.03 (CH), 99.95 (CH), 70.85 (CH2), 38.57 (CH2), 12.34 (CH3)
FTMS (pos, ESI): m/z = 324.1 (MH+) HRMS: calcd for [C19H18O4N1]
324.1230, measured 324.1233.
Global yield: 6.4%.
5.1.5. General procedure for the synthesis of 17–23 (Scheme 5)
The first step is a palladium-catalyzed coupling reaction:
3-bromophenol (1.0 equiv) and substituted secondary amine
(1.2 equiv) were stirred together under strictly dry conditions
and under inert atmosphere (argon). After 10 min, Pd(OAc)2
(2 mol %) was added to the reaction mixture. The Verkade base
was then added dropwise to the reaction (4 mol %). LiHMDS
(2.3 equiv, 1 M in THF) was carefully added, followed by freshly
distilled toluene (3.5 ml for 1 mmol of A). The reaction mixture
was stirred and heated 24 h under reflux at 85 °C. After cooling,
the mixture was extracted with a toluene/water solution, the or-
ganic phase was finally dried on Na2SO4, filtrated and concen-
trated for column chromatography purification (AcOEt/EP
20:80). This step was used for the compounds 19, 20 and 22.
Global yield: 15%.
5.1.3.4. 1-Benzyl-7-(benzyloxy)-2-oxo-1,2-dihydroquinoline-3-
carboxylic acid (13).
1H NMR (CDCl3) d 14.37 (s, 1H), 8.89
(s, 1H), 7.72 (d, J = 8.8 Hz, 1H), 7.46–7.28 (m, 10H), 7.19 (d,
J = 6.3 Hz, 2H), 7.03 (dd, J = 8.8, 2.2 Hz, 1H), 6.87 (d, J = 2.0 Hz,
1H), 5.55 (s, 2H), 5.06 (s, 2H) 13C NMR (75 MHz, CDCl3) d 165.54
(C), 164.47 (C), 163.66 (C), 146.11 (CH), 142.35 (C), 135.26 (C),
133.02 (CH), 129.15 (CH), 128.85 (CH), 127.93 (CH), 127.42 (CH),
126.52 (CH), 114.60 (C), 114.32 (C), 113.73 (CH), 100.46 (CH),
70.62 (CH2), 46.92 (CH2) FTMS (pos, ESI): m/z = 386.1 (MH+) HRMS:
calcd for [C24H20O4N1] 386.13868, measured 386.1387.
Global yield: 26%.

7116
N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
We performed the Vilsmeier–Haack reaction to add the alde-
Global yield: 22.7%.
hyde function to the 2-aminophenol derivative. The phosphoryloxy
chloride POCl3 (1.2 equiv in 3.0 equiv of anhydrous DMF) was care-
fully added dropwise to the previous synthesis product (also in
5.1.5.6.
acid (22).
2-Oxo-7-(pyrrolidin-1-yl)-2H-chromene-3-carboxylic
1H NMR (DMSO-d6) d ppm 8.58 (s, 1H), 7.64 (d,
anhydrous DMF: 200
l
l for 1 mmol) under argon at 0 °C. The reac-
J = 9, 1H), 6.66 (d, J = 9, 1H), 6.43 (s, 1H), 3.39 (s, 4H), 1.99 (s, 4H)
13C NMR (DMSO-d6) d ppm 165.04 (C), 159.91 (C), 157.96 (C),
152.78 (C), 149.99 (CH), 132.10 (CH), 111.30 (CH), 108.04 (C),
107.81 (C), 96.95 (CH), 48.30 (CH2), 25.33 (CH2) MS (APCI): m/
z = 259.97 (MH+) HRMS: m/z calcd for [C14H14NO4] 260.0923, mea-
sured 260.0928.
tion mixture was stirred 15 min at 0 °C, then 15 min at room tem-
perature, 15 min at 37 °C and finally 30 min at 80–90 °C. After
cooling, addition of ice and Na2CO3 in the reaction mixture and
stirring, we obtain a precipitate (not pure but the cyclisation se-
lects the desired coumarinic end product). This reaction step was
used for the compounds 18–22. The previously synthesized salicyl-
aldehyde (1.0 equiv) was dissolved in EtOH (10 ml for 1 mmol).
Meldrum acid (1.2 equiv) was added to the stirred solution. Piper-
idine and acetic acid were added dropwise to catalyse the reaction
(six drops for 1 mmol). The solution was stirred and heated 3 h un-
der reflux. After cooling, the yellow to orange precipitate was fil-
tered and obtained pure. This final reaction step was used for all
the compounds (the only synthesis step of 17 and 23).
Global yield: 22.1%.
5.1.5.7. 7-(Juloidino)-2-oxo-2H-chromene-3-carboxylic acid
(23).
1H NMR (DMSO-d6) d ppm 8.44 (s, 1H), 7.22 (s, 1H),
3.35 (s, 4H), 2.71 (d, J = 5.7, 4H), 1.88 (d, J = 5.4, 4H). 13C NMR
(DMSO-d6) d ppm 165.13 (C), 160.97 (C), 156.81 (C), 149.65 (CH),
149.15 (C), 127.95 (CH), 119.97 (C), 107.70 (C), 105.61 (C),
105.18 (C), 50.11 (CH2), 49.57(CH2), 27.22 (CH2), 26.97 (CH2),
20.93(CH2), 19.96 (CH2) MS (APCI): m/z = 285.95 (MH+) HRMS:
m/z calcd for [C16H16NO4] 286.1079, measured 286.1089.
Global yield: 24%.
5.1.5.1.
acid (17).
7-(Diethylamino)-2-oxo-2H-chromene-3-carboxylic
1H NMR (DMSO-d6) d ppm 12.52 (s, 1H), 8.58 (s,
1H), 7.63 (d, J = 9, 1H), 6.78 (d, J = 9.1 1H), 6.56 (s, 1H), 3.48 (d,
J = 6.9, 4H), 1.14 (m, 6H) 13C NMR (DMSO-d6) d ppm 164.96 (C),
160.00 (C), 158.35 (C), 153.35 (C), 149.9 (CH), 132.31 (CH),
110.50 (CH), 107.84 (C), 107.58 (C), 96.37 (CH), 44.86 (CH2),
12.78 (CH3) MS (APCI): m/z = 261.91 (MH+) HRMS: m/z calcd for
[C14H16NO4] 262.1079, measured 262.1079.
5.2. Pharmacology
5.2.1. Cytotoxicity assay
SiHa cells (human cervix squamous carcinoma) were routinely
cultured in DMEM containing serum and antibiotics as previously
described.8 Two distinct media were used for the cytotoxicity as-
Global yield: 59%.
say, either DMEM containing 25 mM
or DMEM containing 10 mM -lactate (without glucose). SiHa cells
were seeded in flat-bottom 96-well plates in normal DMEM. After
6 h incubation, the culture medium was replaced by 100 l of glu-
cose- (with or without 1 mM pyruvate) or lactate-containing med-
ium and incubated overnight for metabolic adaptation. SiHa cells
were treated with the synthesized compounds at a concentration
D-glucose (without pyruvate)
5.1.5.2. 7-Morpholino-2-oxo-2H-chromene-3-carboxylic acid
(18).
L
1H NMR (DMSO-d6) d ppm 8.59 (s, 1H), 7.68 (d, J = 9,
1H), 7.01 (d, J = 9, 1H), 6.83 (s, 1H), 3.72 (t, J = 4.5, 4H), 3.41 (t,
J = 4.8, 4H) 13C NMR (DMSO-d6) d ppm 164.88 (C), 159.03 (C),
157.77 (C), 155.72 (C), 149.57 (CH), 131.82 (CH), 111.78 (CH),
110.73 (C), 109.49 (C), 98.96 (CH), 66.17 (CH2), 47.01 (CH2) MS
(APCI): m/z = 275.93 (MH+) HRMS: m/z calcd for [C14H14NO5]
276.0872, measured 276.0863.
l
range of 0.01–100
ium was removed and replaced by freshly prepared MTT in PBS
(1 mg/ml, 100 l/well). After a 3 h incubation at 37 °C, the plates
were centrifuged (1000g, 10 min at 4 °C) and the supernatant
was removed before addition of DMSO (100 l/well). Plates were
lM. After a 72 h incubation at 37 °C, cell med-
Global yield: 18.9%.
l
5.1.5.3.
carboxylic acid (19).
7-(N-Benzyl-N-methylamino)-2-oxo-2H-chromene-3-
1H NMR (DMSO-d6) d ppm 8.59 (s,
l
kept for 10 min in the dark before reading at the spectrophotome-
1H), 7.64 (d, J = 9, 1H), 7.35–7.20 (m, 5H), 6.86 (d, J = 9, 1H), 6.62
(s, 1H), 4.78 (s, 2H), 3.20 (s, 3H) 13C NMR (DMSO-d6) d ppm
164.92 (C), 159.47 (C), 157.89 (C), 154.74 (C), 149.88 (CH),
137.93 (C), 131.99 (CH), 129.41 (CH), 127.56 (CH), 126.99 (CH),
110.85 (CH), 108.91 (C), 108.33 (C), 97.25 (CH), 56.49 (CH2),
40.24 (CH3) MS (APCI): m/z = 309.98 (MH+) HRMS: m/z calcd for
[C18H16NO4] 310.1079, measured 310.1085.
ter (Victor X4).
5.2.2. Lactate consumption assay
SiHa cells were seeded on flat-bottom 24-well plates
(500,000 cells/well) in normal DMEM. After 6 h of incubation, the
culture medium was replaced by 1 ml of DMEM containing
10 mM lactate and incubated overnight for metabolic adaptation.
Cells were then treated for 24 h with increasing concentrations
Global yield: 12.2%.
of compounds (0.1, 1, 10 and 100 lM) in lactate-containing
5.1.5.4.
acid (20).
2-Oxo-7-(piperidin-1-yl)-2H-chromene-3-carboxylic
1H NMR (DMSO-d6) d ppm 8.04 (s, 1H), 7.48 (d,
medium at 37 °C. The supernatants were then centrifuged using
deproteinizing columns (15 min, 10000g at 4 °C) and lactate
concentration was determined using the enzymatic assay commer-
cialized by CMA Microdialysis AB on a CMA600 analyzer (Aurora
Borealis).
J = 9, 1H), 6.92 (d, J = 9, 1H), 6.72 (s, 1H), 3.34 (s, 6H), 1.59 (s, 6H)
13C NMR (DMSO-d6): DMSO-d6 saturation at higher cc, no 13C
NMR possible MS (APCI): m/z = 273.97 (MH+) HRMS: m/z calcd
for [C15H16NO4] 274.1079, measured 274.1087.
Global yield: 11.6%.
5.2.3. [14C]-lactate uptake assay
SiHa cells were seeded on flat-bottom 24-well plates previously
5.1.5.5. 7-(Dimethylamino)-2-oxo-2H-chromene-3-carboxylic
acid (21).
coated with poly-L-lysine (500,000 cells/well). After 6 h of incuba-
1H NMR (DMSO-d6) d ppm 8.60 (s, 1H), 7.66 (d,
tion, the culture medium was replaced by 1 ml of DMEM contain-
ing 10 mM lactate and incubated overnight for metabolic
adaptation. Cells were first rinsed with a modified Krebs solution
J = 9, 1H), 6.81 (d, J = 9, 1H), 6.58 (s, 1H), 3.09 (s, 6H). 13C NMR
(DMSO-d6) d ppm 164.97 (C), 157.94 (C), 155.41 (C), 149.99 (CH),
149.96 (C), 131.98 (CH), 131.92 (C), 110.71 (C), 108.40 (CH),
108.06 (CH), 96.91 (C), 40.53 (CH3) MS (APCI): m/z = 233.93
(MH+) HRMS: m/z calcd for [C12H12NO4] 234.0766, measured
234.0777.
(containing 10
lM L-lactate, without glucose) and pre-exposed to
vehicle or increasing concentrations of compounds (0.1, 1, 10 and
100
of 2
l
l
M) at 37 °C in the modified Krebs solution before addition
M [14C]-lactate for 12 min. Cells were then rinsed with a

N. Draoui et al. / Bioorg. Med. Chem. 21 (2013) 7107–7117
7117
7. Yang, Y.; Li, C.; Nie, X.; Feng, X.; Chen, W.; Yue, Y.; Tang, H.; Deng, F. J. Proteome
Res. 2007, 6, 2605.
ice-cold
D-lactate-containing Krebs solution (10 lM D-lactate,
without glucose) and lysed with 0.1 M NaOH. Sample aliquots
were then incubated with liquid scintillation solution (Microscint
40) into a 96-well plate (Optiplate). After a 1 h agitation, radioac-
tivity was measured (PerkinElmer Topcount); cpm values were
normalized per protein amounts.
8. Sonveaux, P.; Vegran, F.; Schroeder, T.; Wergin, M. C.; Verrax, J.; Rabbani, Z. N.;
De Saedeleer, C. J.; Kennedy, K. M.; Diepart, C.; Jordan, B. F.; Kelley, M. J.; Gallez,
B.; Wahl, M. L.; Feron, O.; Dewhirst, M. W. J. Clin. Invest. 2008, 118, 3930.
9. Boidot, R.; Vegran, F.; Meulle, A.; Le, B. A.; Dessy, C.; Sonveaux, P.; Lizard-Nacol,
S.; Feron, O. Cancer Res 2012, 72, 939.
10. Vegran, F.; Boidot, R.; Michiels, C.; Sonveaux, P.; Feron, O. Cancer Res 2011, 71,
2550.
11. Sonveaux, P.; Copetti, T.; De Saedeleer, C. J.; Vegran, F.; Verrax, J.; Kennedy, K.
M.; Moon, E. J.; Dhup, S.; Danhier, P.; Frerart, F.; Gallez, B.; Ribeiro, A.; Michiels,
C.; Dewhirst, M. W.; Feron, O. PLoS One 2012, 7, e33418.
12. Whitaker-Menezes, D.; Martinez-Outschoorn, U. E.; Lin, Z.; Ertel, A.;
Flomenberg, N.; Witkiewicz, A. K.; Birbe, R. C.; Howell, A.; Pavlides, S.;
Gandara, R.; Pestell, R. G.; Sotgia, F.; Philp, N. J.; Lisanti, M. P. Cell Cycle 2011, 10,
1772.
13. Fiaschi, T.; Marini, A.; Giannoni, E.; Taddei, M. L.; Gandellini, P.; De, D. A.;
Lanciotti, M.; Serni, S.; Cirri, P.; Chiarugi, P. Cancer Res 2012, 72, 5130.
14. Harguindey, S.; Arranz, J. L.; Wahl, M. L.; Orive, G.; Reshkin, S. Anticancer Res.
2009, 29, 2127.
15. Hirschhaeuser, F.; Sattler, U. G.; Mueller-Klieser, W. Cancer Res. 2011, 71, 6921.
16. Draoui, N.; Feron, O. Dis. Model. Mech. 2011, 4, 727.
17. Jackson, V. N.; Halestrap, A. P. J. Biol. Chem. 1996, 271, 861.
18. Halestrap, A. P. Mol. Aspects Med. 2013, 34, 337.
19. Pinheiro, C.; Longatto-Filho, A.; Azevedo-Silva, J.; Casal, M.; Schmitt, F. C.;
Baltazar, F. J. Bioenerg. Biomembr. 2012, 44, 127.
5.2.4. Western blotting
Caspase-3 antibodies were from Cell Signaling. CAIX and CAXII
antibodies were from R&D and Abcam, respectively. MCT antibod-
ies were produced and validated in our lab.9,10 Immunoblotting
was performed as previously described,9,10 b-actin was used as a
loading control.
5.2.5. ADME, pharmacokinetics and prothrombin time assay
In vitro ADME and in vivo pharmacokinetics profiles were per-
formed by Cerep (France). PT time assay was performed using
InnovinÒ reagents according to the manufacturer’s instructions;
plasma was collected at day 5 on citrated tubes (3.2%) from mice
(four mice/group) daily ip injected with vehicle or compounds.
20. Le Floch, R.; Chiche, J.; Marchiq, I.; Naiken, T.; Ilc, K.; Murray, C. M.; Critchlow,
S. E.; Roux, D.; Simon, M. P.; Pouyssegur, J. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,
16663.
Acknowledgments
21. Dimmer, K. S.; Friedrich, B.; Lang, F.; Deitmer, J. W.; Broer, S. Biochem. J. 2000,
350(Pt 1), 219.
This work was supported by Grants from the Fonds de la
Recherche Scientifique FRS-FNRS, the Télévie, the Belgian Founda-
tion against cancer, the J. Maisin Foundation, IUAP Research
Program #UP7-03 from the Belgian Science Policy Office (Belspo),
an Action de Recherche Concertée (ARC 09/14-020), and a starting
Grant from the European Research Council (ERC No. 243188
TUMETABO to P.S.). O.F. is an honorary Research Director and P.S.
a Research Associate of the F.R.S.-FNRS. O.R., O.F. and P.S. are
co-senior authors.
22. Neri, D.; Supuran, C. T. Nat. Rev. Drug Disc. 2011, 10, 767.
23. Wang, X.; Levi, A. J.; Halestrap, A. P. Am. J. Physiol 1996, 270, H476.
24. Hildyard, J. C.; Halestrap, A. P. Biochem. J. 2003, 374, 607.
25. Ovens, M. J.; Davies, A. J.; Wilson, M. C.; Murray, C. M.; Halestrap, A. P. Biochem.
J. 2010, 425, 523.
26. Murray, C. M.; Hutchinson, R.; Bantick, J. R.; Belfield, G. P.; Benjamin, A. D.;
Brazma, D.; Bundick, R. V.; Cook, I. D.; Craggs, R. I.; Edwards, S.; Evans, L. R.;
Harrison, R.; Holness, E.; Jackson, A. P.; Jackson, C. G.; Kingston, L. P.; Perry, M.
W.; Ross, A. R.; Rugman, P. A.; Sidhu, S. S.; Sullivan, M.; Taylor-Fishwick, D. A.;
Walker, P. C.; Whitehead, Y. M.; Wilkinson, D. J.; Wright, A.; Donald, D. K. Nat.
Chem. Biol. 2005, 1, 371.
27. Song, B.; Ang, X.; Am, K. S. Tetrahedron Lett. 2003, 44, 1755.
28. Shen, Q.; Ogata, T.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 6586.
29. Lundgren, R. J.; Sappong-Kumankumah, A.; Stradiotto, M. Chemistry 2010, 16,
1983.
Supplementary data
30. Surry, D. S.; Buchwald, S. L. Chem. Sci. 2011, 2, 27.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2013.09.010.
31. Gould, S. J.; Eisenberg, R. J. J. Org. Chem. 1991, 56, 6666.
32. Fraley, M. E.; Hoffman, W. F.; Rubino, R. S.; Hungate, R. W.; Tebben, A. J.;
Rutledge, R. Z.; McFall, R. C.; Huckle, W. R.; Kendall, R. L.; Coll, K. E.; Thomas, K.
A. Bioorg. Med. Chem. Lett. 2002, 12, 2767.
33. Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J.; Buchwald, S. L. J. Org. Chem. 2000,
65, 1158.
34. De Saedeleer, C. J.; Copetti, T.; Porporato, P. E.; Verrax, J.; Feron, O.; Sonveaux, P.
PLoS One 2012, 7, e46571.
35. Bonneau, A.; Maresca, A.; Winum, J. Y.; Supuran, C. T. J. Enzyme Inhib. Med.
Chem. 2013, 28, 397.
References and notes
1. Gillies, R. J.; Robey, I.; Gatenby, R. A. J. Nucl. Med. 2008, 49(Suppl. 2), 24S.
2. Feron, O. Radiother. Oncol. 2009, 92, 329.
3. Dhup, S.; Dadhich, R. K.; Porporato, P. E.; Sonveaux, P. Curr. Pharm. Des. 2012,
18, 1319.
4. Schwickert, G.; Walenta, S.; Sundfor, K.; Rofstad, E. K.; Mueller-Klieser, W.
Cancer Res. 1995, 55, 4757.
5. Walenta, S.; Salameh, A.; Lyng, H.; Evensen, J. F.; Mitze, M.; Rofstad, E. K.;
Mueller-Klieser, W. Am. J. Pathol. 1997, 150, 409.
36. Maresca, A.; Temperini, C.; Vu, H.; Pham, N. B.; Poulsen, S. A.; Scozzafava, A.;
Quinn, R. J.; Supuran, C. T. J. Am. Chem. Soc.1 2009, 131, 3057.
6. Yokota, H.; Guo, J.; Matoba, M.; Higashi, K.; Tonami, H.; Nagao, Y. J. Magn. Reson.
Imaging 2007, 25, 992.