N-Hydroxyindole-based inhibitors of lactate dehydrogenase against cancer cell proliferation

Article abstract of DOI:10.1016/j.ejmech.2011.08.046

Current cancer research is being increasingly focused on the study of distinctive characters of tumour metabolism, resulting in a switch from oxidative phosphorylation to glycolysis (Warburg effect). Isoform 5 of human lactate dehydrogenase (hLDH5), which

Full text of DOI:10.1016/j.ejmech.2011.08.046

European Journal of Medicinal Chemistry 46 (2011) 5398e5407
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
N-Hydroxyindole-based inhibitors of lactate dehydrogenase against cancer cell
proliferation
Carlotta Granchia, Sarabindu Roya, Alessio De Simonea, Irene Salvettia, Tiziano Tuccinardia,
Adriano Martinellia, Marco Macchiaa, Mario Lanzab, Laura Bettib, Gino Giannaccinib, Antonio Lucacchinib,
,
Elisa Giovannettic, Rocco Sciarrilloc, Godefridus J. Petersc, Filippo Minutoloa
*
a Dipartimento di Scienze Farmaceutiche, Università di Pisa, Via Bonanno 6, 56126 Pisa, Italy
b Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Università di Pisa, Via Bonanno 6, 56126 Pisa, Italy
c Department of Medical Oncology, VU University Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Current cancer research is being increasingly focused on the study of distinctive characters of tumour
metabolism, resulting in a switch from oxidative phosphorylation to glycolysis (Warburg effect). Isoform
5 of human lactate dehydrogenase (hLDH5), which catalyzes the final step in the glycolytic cascade
(pyruvate to lactate), constitutes a relatively new and untapped anti-cancer target. In this study, careful
design and synthesis of a selected series of aryl-substituted N-hydroxyindole-2-carboxylates (NHIs) has
led to several hLDH5-inhibitors, showing “first-in-class” potency and isoform selectivity. Enzyme kinetics
studies indicated that these inhibitors exhibit a competitive mode of inhibition. Some representative
examples were tested against two human pancreatic carcinoma cell lines, and displayed a good anti-
proliferative activity, which was even more evident under hypoxic conditions.
Received 22 July 2011
Received in revised form
29 August 2011
Accepted 31 August 2011
Available online 7 September 2011
Keywords:
Lactate dehydrogenase
Glycolysis
Warburg effect
Cancer
Ó 2011 Elsevier Masson SAS. All rights reserved.
Hypoxia
Enzyme inhibitors
1. Introduction
those constituting invasive tumour tissues. Furthermore, cancer
cells are often exposed to hypoxic conditions, since oxygenation of
Neoplastic proliferation in hypoxia is characterized by an altered
metabolism, which involves several crucial hallmarks of cancer,
including a peculiar sugar metabolism [1]. Cell growth requires
glucose as the most important source of energy, to generate
adenosine-50-triphosphate (ATP) molecules. Consequently, glucose
consumption is generally increased in rapidly dividing cells, such as
tumours by vascularisation is usually less efficient. A metabolic
switch from normal oxidative phosphorylation (OXPHOS) to
anaerobic glycolysis, also known as the Warburg Effect [2], is very
often found in invasive tumours. This adaptation allows malignant
cells to produce ATP from glucose even in the absence of oxygen, by
promoting the glycolytic process leading to the production of lactic
acid. Therefore, the “glycolytic phenotype” confers cancer cells with
an adaptation to otherwise hostile hypoxic environments, although
the efficiency of this metabolic pathway is much lower when
compared to that of OXPHOS [3]. For that reason, cancer cells
exhibit an exaggerated avidity of glucose, as shown by the
increased uptake of labelled glucose analogue 18F-fluorodeox-
yglucose (FDG), which is used to visualize in vivo invasive tumours
[4]. Hence, this unique metabolic feature of invasive and hypoxic
tumours, that does not occur in healthy tissues, is currently being
considered as an ideal target for cancer-selective therapy [5].
There are several key-steps of the glycolytic process which are
promoted by enzymes or transporters that have already been
considered as suitable targets for an “anti-glycolytic“ cancer therapy
(Fig. 1).
Abbreviations: LDH, lactate dehydrogenase; NHI, N-hydroxyindole-2-
carboxylate; ATP, adenosine-50-triphosphate; OXPHOS, oxidative phosphorylation;
FDG, 18F-flurordeoxyglucose; GLUT-1, glucose membrane transporter; HK, hexoki-
nase; GPI, glucose-6-phosphate isomerase; PFK, phospho-fructokinase; ALDO,
aldolase; TPI, triosephosphate isomerase; GAPDH, glyceraldehyde-3-phosphate
dehydrogenase; PGK, phosphoglycerate kinase; PGM, phosphoglycerate mutase;
ENO, enolase; PK, pyruvate kinase; PDK, dehydrogenase kinase; DCA, dichlor-
oacetate; PDH, pyruvate dehydrogenase; IDH, isocitrate dehydrogenase; NADH/
NADþ, nicotinamide adenine dinucleotide cofactor; TBAB, tetrabutylammonium
bromide; DMF, N,N-dimethylformamide; MD, molecular dynamics; rmsd, root-
mean-square deviation; DMSO, dimethylsulfoxide; HPLC, high performance liquid
chromatography; TLC, thin-layer chromatography; NMR, nuclear magnetic reso-
nance; EI, electron impact; MS, mass spectroscopy; tR, retention time; RT, room
temperature; CG, conjugated gradient.
* Corresponding author. Tel.: þ39 050 2219557; fax: þ39 050 2219605.
E-mail address: filippo.minutolo@farm.unipi.it (F. Minutolo).
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.08.046

C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
5399
induced gene expression of LDH-A subunit in cancer cell cultures
media [14]. A validation of hLDH5 as an anti-cancer target derived
from studies conducted on LDH-A-deprived cancer cell lines, which
demonstrated that gene repression of the production of the LDH-A
subunit in cancer cells by shRNA seriously compromised the
tumour growth and invasiveness of these LDH-A-deficient cells
under hypoxia [15]. In addition, studies on naturally occurring
knock-out human “models”, constituted by individuals affected by
hereditary LDH-A deficiency, show that the lack of this enzyme
does not cause any symptoms under ordinary circumstances,
whereas muscle damage and myoglobinuria are observed only after
intense anaerobic exercise [16]. These evidences would further
confirm the high possibility for hLDH5 to be a safe target for anti-
tumour agents.
Fig. 1. Targets of glucose metabolism considered for anti-cancer treatment.
Some enzymes/transporters promoting the degradation of
glucose to pyruvate have been targeted by perspective anti-cancer
agents: glucose membrane transporter (GLUT-1), hexokinase (HK),
glucose-6-phosphate isomerase (GPI), phospho-fructokinase (PFK),
aldolase (ALDO), triosephosphate isomerase (TPI), glyceraldehyde-
3-phosphate dehydrogenase (GAPDH), phosphoglycerate kinase
(PGK), phosphoglycerate mutase (PGM), enolase (ENO), pyruvate
kinase (PK) [5d]. However, inhibitors of these factors are also ex-
pected to interfere with the metabolism of normal cells, because
they are placed before the “bifurcation” point of pyruvate, whose
destiny depends on the cellular context (tumour vs. healthy cells),
as well as on the amount of oxygen available. An interesting
exception is represented by selective inhibitors of isoform M2 of PK,
which is only found in foetal and cancer cells [6]. Nevertheless, the
real role of PK-M2 in either tumour suppression or carcinogenesis is
currently being debated, because paradoxical effects were reported
as associated to PK-M2 inhibition/activation, so it is still not clear if
and when it promotes or suppresses tumours [7]. Enzymes oper-
ating inside mitochondria were also considered as suitable targets
for cancer chemotherapy. One of them is pyruvate dehydrogenase
kinase (PDK), whose inhibition by dichloroacetate (DCA) restores
the activity of pyruvate dehydrogenase (PDH) and re-activates
OXPHOS in cancer cells, thus leading them to apoptosis [8].
Another attractive target in this context is isocitrate dehydrogenase
(IDH), whose mutated form present in several tumours is currently
being considered for therapeutic opportunities [9]. Within this
picture, we focused our attention on lactate dehydrogenase (LDH).
In fact, the metabolic position of this enzyme, right after the
“bifurcation” of the pyruvate destinies, would be ideal for a selec-
tive intervention against cancer, since its inhibition would only
involve cells where OXPHOS is mostly compromised, such as
tumour cells. This means that a block of the glucose metabolism at
this stage should not be expected to present relevant toxic effects to
most normal cells, which exploit the normal mitochondrial
degradation of pyruvate.
Not until very recent times do we find examples of efficient
LDH-inhibitors, because only very few clinical applications were so
far envisioned [17]. For example, reference LDH-inhibitor oxamic
acid proved to reduce cellular growth in human hepatocellular
carcinoma cell lines [18]. Furthermore, one of the very few efficient
hLDH5-inhibitors reported in the literature, such as FX-11 (1, Fig. 2),
was initially designed as an antimalarial agent [19]. FX-11 is a 2,3-
dihydroxynaphtalen-1-carboxylate derivative, displaying a Ki value
of 8 mM in the inhibition of hLDH5 and a >10-fold selectivity over
the hLDH1 isoform, which proved to inhibit tumour progression by
inducing an oxidative stress to the cancer cells [20]. Furthermore,
a “bifunctional” butyric acid derivative (2, Fig. 2) was reported to
inhibit hLDH5 with an IC50 value of 14.8 mM [21], although no data
relative to cancer cell proliferation assays were reported.
Recently, we developed a new class of N-hydroxyindole-2-
carboxylates (NHIs) [22], and those containing a phenyl ring at
either 5 or 6 position (3, Fig. 2) displayed efficient and isoform-
selective inhibition of hLDH5, with Ki values reaching the low
micromolar range [22a]. These compounds also proved to sensibly
decrease cellular lactate production and caused a marked reduction
of cancer cell proliferation, especially under hypoxic conditions
[22a].
In the light of these results, we decided to further investigate
and expand this series of 5/6-phenyl-substituted NHIs by intro-
ducing chemical variability at the aryl substituents of the NHI
central core. Details of the parallel synthetic routes that were
developed, along with analysis of the activity of the analogues
generated on isolated hLDH5 and hLDH1 isoforms, and against
cancer cells in vitro are reported.
LDH is an oxidoreductase which catalyzes the interconversion of
pyruvate and lactate by using a nicotinamide adenine dinucleotide
cofactor (NADH/NADþ). Human LDH is a tetrameric enzyme which
may be composed by three different monomeric subunits: LDH-A,
LDH-B and LDH-C. While the C subunit takes only part of a homo-
tetrameric enzyme, hLDHX (LDH-C4), which plays a role in male
fertility [10], the A and B subunits are found to be combined in five
tetrameric isoforms (hLDH1-5) constituted of homotetramers, such
as hLDH1 (LDH-B4) and hLDH5 (LDH-A4), and heterotetramers
made of the other three possible combinations of A and B subunits
(hLDH2-4) [11]. Localization of hLDH1 is mostly limited to the heart,
where it generally catalyzes the oxidation of lactate to pyruvate,
whereas isoform hLDH5 preferentially catalyzes the reduction of
pyruvate to lactate and is the main form present in skeletal muscles
and liver [12]. Overexpression of isoform 5 was detected in highly
invasive and hypoxic carcinomas [13]. and oxygen deprivation
Fig. 2. Structures of hLDH5-inhibitors: FX-11 (1), a “bifunctional” butyric acid deriv-
ative (2), and 5/6-phenyl-substituted NHIs (3).

5400
C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
2. Chemistry
The introduction of the R2-substituent was obtained in two
different ways: in some cases 4-bromo-1-methyl-2-nitrobenzene (5)
was reacted with boronic acids R2B(OH)2, whereas in other cases we
found it convenient to react 4-methyl-3-nitrophenylboronic acid (6)
with the aryl bromides R2-Br, as reported in details in the Experi-
mental Section. In particular, compounds 8fen were obtained by
reacting the appropriate R2B(OH)2 with bromo-aryl 5. Most of these
reactions were carried out under ligand-free conditions (method B),
in the presence of either Na2CO3 upon microwave irradiation [24], to
get compounds 8jel,n, or K3PO4 under conventional heating [25], to
produce intermediates 8h,i. Carboxy-substituted intermediate 8m
could only be obtained in the presence of a phosphine ligand, by
generating in situ the Pd(PPh3)4 catalyst (method A). A particular care
was necessary during the preparation of 4-chlorophenyl- and 2,4-
dichlorophenyl-substituted intermediates 8f and 8g, respectively.
In fact, when we employed microwave conditions to the preparation
of these compounds, we obtained complex reaction mixtures,
probably deriving from over-reactions of the chloro-aryl portions
under those harsh conditions. Therefore, we had to switch back to
classical thermal Pd(PPh3)4ecatalyzed Suzuki conditions, so that, in
spite of prolonged reaction times (24 h) required to get satisfactory
conversions, 8f and 8g could be easily obtained and purified. Nitro-
toluene derivatives 8o,q–s were, instead, obtained starting from
boronic acid 6, which was reacted with the required aryl bromide R2-
Br. In this case, all the cross-coupling reactions could be efficiently
carried out under the same ligand-free microwave-assisted condi-
tions [24], to produce the required intermediates. Finally, the
contemporary insertion of two phenyl ring in the structure of 1,2-
dichloro-4-methyl-5-nitrobenzene (7) [26] was obtained under
phosphine-free thermal cross coupling conditions in the presence of
K3PO4 [25].
A series of 5- and 6-arylsubstituted NHIs (3aet) were synthe-
sized, by utilizing a general pathway which is displayed in Scheme 1.
In each case, different conditions were chosen on the basis of
preliminary optimizations of the reaction outcome, which proved to
be highly dependent on the precursor structure. This synthesis
started with the initial formation of aryl-substituted nitrotoluene
intermediates 8aet, which were obtained by cross-coupling reac-
tions performed under various conditions (methods A and B, Scheme
1). In details, 4-chloro-2-methyl-1-nitrobenzene (4) efficiently
underwent Pd-catalyzed cross-coupling reactions with the appro-
priate arylboronic acid R1B(OH)2 under either classical Suzuki
conditions [23], or phosphine-free conditions comprising the use of
tetrabutylammonium bromide (TBAB) as the phase-transfer agent,
in the presence of sodium carbonate upon microwave irradiation
[24]. It should be noted that 4, in spite of being an aryl chloride
which is notoriously less reactive under the conditions reported in
methods A and B, it possesses a nitro-group in para-position to the
reactive chloro atom which activates this compound toward the
cross-coupling reaction. This way, we were able to efficiently insert
the R1-substituent and produce intermediates 8aee.
The subsequent sequence of reactions involved
a base-
promoted condensation of 8a–t with dimethyloxalate, followed
by a reductive cyclization, initiated by the partial reduction of the
nitro- to hydroxylamine-group, to generate the N-hydroxyindole
core (9aet) [27]. While the first part of this sequence worked nicely
under the same conditions for all the substrates, consisting in
a deprotonation with sodium hydride in anhydrous DMF and
condensation with (COOMe)2, which led to the formation of a a-
ketoester intermediate [27], the subsequent reductive-cyclization
step required a careful optimization in each case and proved to
be highly dependent on the substrate. In fact, the main problem we
encountered in this step was the possible over-reduction of the
nitro group, which led to the formation of variable amounts of NeH
indole side products, which were often difficult to separate from
their desired N-OH analogues. One of the most commonly used
methods reported in the literature for this purpose (method C),
involving stannous chloride as the reducing agent [27,28], proved
to be efficacious only in a few examples of ours (9d,p), whereas in
most cases we had to turn to the addition of additives, such as
thiophenol and triethylamine in acetonitrile (method D) [29], which
allowed us to efficiently obtain compounds 9aec,fel,n,o,qes. This
procedure gave a very quick (5 min) full conversion of starting
material, but it required an additional treatment with a catalytic
amount of para-toluenesulfonic acid (p-TsOH) and several cycles of
dissolution in toluene and evaporation under vacuum of the
toluene/water azeotrope, in order to obtain an exhaustive
consumption of hydrated precursors of N-hydroxyindole deriva-
tives 9. In some restricted cases, these conditions did not guarantee
the selective partial reduction of the nitro-group. Fortunately, the
replacement of the tin-based reducing agent with either sodium
hypophosphite in the presence of palladium on charcoal (method E)
[30], or lead with tetraethylammonium formate in refluxing
methanol (method F) [31], efficiently produced compounds 9e,m
and 9t, respectively.
Scheme 1. Synthesis of NHIs 3aet. Reagents and conditions: a) method A: R1B(OH)2,
Pd(OAc)2, PPh3, aq 2M Na2CO3, toluene, EtOH, 100 ꢀC, 24 h (8a,e); method B: R1B(OH)2,
Pd(OAc)2, TBAB, Na2CO3, H2O,
PPh3, aq 2M Na2CO3, toluene, EtOH, 100 ꢀC, 24 h (8f,g,m) or
R2B(OH)2, Pd(OAc)2, TBAB, Na2CO3, H2O, W, 20 min (8jel,n), or K3PO4, H2O 125 ꢀC,
24 h (8h,i); c) R2Br, Pd(OAc)2, TBAB, Na2CO3, H2O,
W, 5 min (8o,qes); d) PhB(OH)2,
Pd(OAc)2, TBAB, K3PO4, H2O, 125 ꢀC (sealed vial), 48
(8t); e) (COOMe)2, NaH,
m
W, 5 min (8bed); b) method A: R2B(OH)2, Pd(OAc)2,
mW, 5 min (8p); method B:
m
m
h
DMF, ꢁ15 ꢀC to RT, 2e48 h; f) method C: SnCl2$2H2O, anhydrous DME, 4 Å MS, 0 ꢀC to
RT, 6e16 h (9d,p); method D: 1) SnCl2$2H2O, PhSH, Et3N, CH3CN, RT, 5 min, 2) pTsOH
(cat.), toluene, evaporation under vacuum (9aec,fel,n,o,qes); method E:
H2PO2Na$3H2O, PdeC, H2O/THF (1:1), RT; 24 h (9e,m); method F: Pb, TEAF, MeOH,
55 ꢀC, 12 h (9t); g) aq 2 N LiOH, THF/MeOH (1:1), RT, 1e16 h.

C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
5401
Hydrolysis of methyl-ester precursors 9aet, with lithium
hydroxide in a 1:1 mixture of tetrahydrofuran and methanol,
afforded final products 3aet.
although this compound competes more efficiently than 3a with
M). When the aryl ring in position 5 contains
a 4-fluoro- (3b), 4-trifluoromethyl- (3c), or a 4-methoxy-group
(3d), the inhibitory activity is lost.
pyruvate (Ki ¼ 27
m
In contrast, a generally better inhibition of hLDH5 is obtained
with 6-monoaryl-substituted NHIs, especially under pyruvate-
saturating conditions (competition vs. NADH). Among these deriv-
atives, para-Cl-phenyl-substituted NHI 3f shows a potent NADH-
3. hLDH5 and hLDH1 inhibition assays
The inhibitory activities of compounds 3aet on the two purified
human enzyme isoforms hLDH5 and hLDH1 were measured by
standard enzyme kinetics experiments. We first verified in
a preliminary screening the percent inhibition of hLDH5 relative to
competitive inhibition with a Ki as low as 5.3
a less efficient Ki of 56 M vs. pyruvate. The insertion of a second
chlorine atom, as in 2,4-dichlorophenyl-substituted derivative 3g,
still preserves a good inhibitory activity vs. NADH (Ki ¼ 9.8 M) and
contemporarily improves the competitive inhibition against the
M). A slightly lower potency is associated to 4-
mM, together with
m
control at a compound concentration of 125
mM in the presence of
m
25 M NADH. Enzyme activity was determined by measuring the
m
absorbance decrease at 340 nm, due to the consumption of NADH.
Only compounds displaying a >50% inhibition at that concentration
were further evaluated for a full enzyme kinetics characterization,
to verify their type of inhibition versus both NADH and pyruvate,
together with their isoform selectivity (hLDH5 vs. hLDH1). Hence,
we initially evaluated the apparent MichaeliseMenten constants
(Km) relative to the cofactor (NADH) and the substrate (pyruvate)
for hLDH5, measured from LineweavereBurk plots. Afterwards, we
substrate (Ki ¼ 28
m
fluoro- and 4-trifluoromethyl-substituted compounds 3h,i,
whereas biphenyl-derivative 3j turns out to be very effective, with Ki
values of 4.5 mM (NADH) and 14 mM (pyruvate), although its isoform
selectivity is spoiled by a certain inhibition of hLDH1, as mentioned
above. Naphthyl-substituted compounds 3k,l have very similar
inhibitory properties, with a-naphthyl-substituted 3k being slightly
more potent. A loss of activity is observed with 3m, bearing a car-
boxyethyl-substituents in the para-position of the aryl ring,
whereas both 4- and 3-trifluoromethoxy-substituted NHIs 3n and
3o, respectively, show a very efficient inhibition, especially vs.
0
measured the apparent Km values in the presence of the selected
inhibitors (concentration range ¼ 25e100
mM). From the values of
0
Km so obtained for each compound showing >50% inhibition at
125
mM, Ki values were determined by double-reciprocal Line-
NADH, with Ki values in the 7.6e8.5 mM range. Substitution with
weavereBurk plots and their values are reported in Table 1.
The first general observation deriving from these studies was
that all the analyzed compounds are unable to inhibit hLDH1, since
% inhibition of this isoform in the presence of 125 mM solutions of
these compounds is always below 3%, with the single exception
alkoxy-groups (3p,q) affords inactive inhibitors, in agreement to the
5-substituted series (see 3d), whereas the introduction of a gem-
difluorobenzodioxolane portion (3r) or of a 3-furyl group (3s)
produces inhibition levels that are more pronounced vs. pyruvate
(Ki ¼ 17e20
m
M) than vs. NADH (Ki ¼ 52e54
mM). Finally, compound
constituted by biphenyl-derivative 3j, displaying a 44% inhibition of
3t, containing phenyl rings in both positions 5 and 6, shows excel-
lent levels of inhibition in both competition assays, with Ki values of
hLDH1 at 125
mM. Isoform hLDH5 is instead sensitive to the
inhibitory action operated by many of these derivatives, as
described below. This preference for isoform-5 had already been
observed with 4- and 5-phenyl-substituted NHIs [22a].
5.4
m
M (vs. NADH) and 7.5
mM (vs. pyruvate).
The competitive behaviour of these inhibitors was demon-
strated by the fact that different concentrations of each active
compound does not change the Vmax of the enzyme kinetics, but it
only changes the K’m value. This is also graphically displayed by the
LineweavereBurk plot of a representative example constituted by
compound 3g (Fig. 3), where the competitive behaviour vs. both the
cofactor (panel a) and the substrate (panel b) is demonstrated.
Among the 5-monoaryl-substituted NHIs, only two compounds
display appreciable levels of hLDH5-inhibition. Compound 3a,
bearing a para-chlorophenyl group, shows an efficient inhibition, at
least in competition with NADH (Ki ¼ 16
mM), comparable to the
activity of its previously reported non-chlorinated counterpart
[22a], whereas its competitive inhibition vs. pyruvate is consider-
ably weaker (Ki ¼ 100
mM). A similar level of inhibition vs. NADH is
associated to 3e, containing a 5-(3-carboxyphenyl) substituent,
4. Molecular modelling
In order to further characterize the interaction of these new
compounds with the hLDH5 active site, one of the most potent
inhibitors of this series (3g) was docked into the minimized average
structure of LDH-A chain, which was recently obtained by us [22]
and then subjected to 10 ns of molecular dynamics (MD) simula-
tions. The analysis of the root-mean-square deviation (rmsd) from
Table 1
hLDH5 inhibition data for selected NHI derivatives (3).
Entry
Compound
Ki [
m
M]a (vs. NADHb
Ki [m
M]a (vs. Pyr)c
1
2
3
4
5
6
7
8
3a
3e
3f
16
19
5.3
9.8
67
42
4.5
10
15
7.6
8.5
52
54
5.4
100
27
56
28
17
25
14
the initial model of the
a carbons of the proteins showed that, after
3g
3h
3i
an initial increase, the rmsd remained approximately constant
around 1.4 Å during the last 7 ns (Supplementary Data, Fig. S1).
Fig. 4 shows the minimized structure of the average of the last 7 ns
of the MD simulation. In agreement with our previously reported
binding hypothesis [22], the carboxylic group of compound 3g
shows polar interactions with R169 and T248, and the N-hydroxy
group displays an H-bond with the backbone nitrogen atom of
T248, as well as with a water molecule that mediates the interac-
tion with H193.
As shown in Table 2, the H-bond analysis of the MD simulation
confirms the stability of these interactions, since during the last
7 ns all the indicated interactions are highly conserved during the
MD simulation (49e99% occupied populations).
3jd
3k
3l
7.1
9
13
40
33
17
20
10
11
12
13
14
3n
3o
3r
3s
3t
7.5
a
Values are reported as the average of three or more measurements; the error in
these values is within ꢂ30% of the average.
b
Saturating concentration (2.0 mM) of sodium pyruvate and competitive
increasing concentrations (12.5 O 150
m
M) of NADH.
M) of NADH and competitive increasing
M O 1.0 mM) of sodium pyruvate.
This compound also showed a 44% inhibition of isoform hLDH1 at 125
c
Saturating concentration (200
concentrations (25
m
The indole central scaffold is placed in a cleft mainly delimited
by N138, A238, V241, T248, I252 whereas the aryl group is inserted
m
d
m
M.

5402
C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
a
control
3g (5 µM)
3g (10 µM)
0.08
0.06
0.04
0.02
3g (20 µM)
-0.05
0.00
0.05
0.10
0.15
0.20
1/NADH (µM)
0.175
0.150
0.125
0.100
0.075
0.050
0.025
control
b
3g (12.5 µM)
3g (25 µM)
3g (50 µM)
Fig. 4. MD simulation results for the complex of LDH-A with 3g: main proteineligand
interactions (A) and surface analysis (B, hydrophobic regions are coloured green). (For
interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.)
0.00
0.01
0.02
1/Pyruvate (
0.03
M)
0.04
0.05
µ
Fig. 3. LineweavereBurk plots showing the mode of inhibition of a representative NHI
(3g). Linear competitive inhibition with respect to: a) NADH; b) pyruvate.
tested with a normal immortalized cell line of human foreskin
fibroblast cells Hs27. During these experiments, cancer cells were
exposed to both normoxic and hypoxic (1% of oxygen) conditions.
Treatment of cells with LDH-inhibitors for 48 h resulted in a dose-
dependent decline in cell viability and the effects of these
in a lipophilic cleft (see Fig. 4B) and shows interactions with V31,
T95, V136 and Y247. The analysis of the volume of the cavities
placed around 3g in the enzyme active site reveals that there is
a large empty space near the p-chloro-atom of the aryl substituent
(region A of Fig. 5). Similarly, there is a certain space in corre-
spondence of the 5 position of the indole ring (region B of Fig. 5)
and this evidence is in accord with the fact that a few 5-aryl-
substituted (3a,e) or 5,6-diphenyl-substituted (3t) NHI-derivatives
give appreciable levels of LDH-inhibition. Nevertheless, these
modelling results support the evidence that larger groups can be
profitably tolerated in position 6, since this position points toward
the enzyme entrance channel, and this is also consistent with the
general higher activity associated to 6-aryl-substituted NHIs.
Overall, these results confirm that the NHI-based inhibitors
herein reported may occupy both the substrate pocket, delimited by
R169 and T248 and, in part, the cofactor pocket of the enzyme active
site, placed around the enzyme entrance channel. This is in fine
agreement with the experimental enzyme inhibition data, which
indicate that they are competitive with both pyruvate and NADH.
Table 2
Hydrogen bonds analysis of the enzyme-3g interactions during the last 7 ns of MD
simulation.
H
N+ H2
N H1
-
O2
O1
Cl
N
HN
R169
N
O3
H
H1
Cl
H
H2
O
O
H1
H2
NH
N
T248
H193
donora
acceptor
LIG@H
WAT@H1/2
T248@H1
R169@H1
R169@H2
T248@H2
acceptor
distance (Å)
% occupied
WAT@O
H193@N
LIG@O1
LIG@O1
LIG@O2
LIG@O3
LIG@O3
WAT@O
T248@OG1
R169@NH1
R169@NH2
T248@N
2.7
2.8
2.7
2.8
2.8
2.9
99
84
79
76
71
49
5. Antiproliferative activities against cancer cells
The growth inhibitory effect of some selected NHI derivatives
was assayed on two human pancreatic carcinoma cell lines, PANC-1
and MIA PaCa-2, whereas possible unspecific cytotoxicity was
a
WAT ¼ water, LIG ¼ ligand 3g.

C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
5403
of our knowledge, some of the compounds herein described
represent the most potent and selective hLDH5-inhibitors ever
reported in the literature. Their activity against invasive pancreatic
cancer cell lines was confirmed under both normoxic and hypoxic
conditions, although it is still too early to make any prediction
about in vivo activities of these compounds. Overall, these findings
contribute to shed light on the main pharmacophore characteristics
needed for an efficient inhibition of this relatively new anti-cancer
target. Moreover, they constitute important starting points for
further molecular optimization in an attempt to develop new anti-
tumour agents targeting the peculiar glucose metabolism of cancer
cells.
7. Experimental section
7.1. Chemistry
Fig. 5. Cavities (light blue) around 3g and LDH active site. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of
this article.)
7.1.1. General
Commercially available chemicals were purchased from Sig-
maeAldrich or Alfa Aesar and used without further purification.
NMR spectra were obtained with a Varian Gemini 200 MHz spec-
Table 3
Antiproliferative activities (IC50) of selected NHI derivatives on two pancreatic
cancer cell lines.
trometer. Chemical shifts (d) are reported in parts per million (ppm)
downfield from tetramethylsilane and referenced from solvent
references (DMSO-d6: dH ¼ 2.50, dC ¼ 39.52 ppm; acetone-d6:
dH ¼ 2.05, dC ¼ 29.84 ppm). Electron impact (EI, 70 eV) mass spectra
were obtained on a Thermo Quest Finningan (TRACE GCQ plus)
mass spectrometer. Purity was routinely measured by HPLC on
Compound
PANC-1 (IC50
normoxiab
,
m
M)a
hypoxiac
MIA PaCa-2 (IC50
normoxiab
, m
M)a
hypoxiac
3f
3g
3i
3o
3r
3t
11.9
12.0
8.8
10.2
11.3
11.3
3.5
2.0
2.5
2.3
1.9
2.3
8.3
6.8
8.9
5.8
6.8
7.8
3.6
1.8
1.6
2.2
2.0
1.9
a Waters SunFire RP 18 (3.0 ꢃ 150 mm, 5
mm) column (Waters,
Milford, MA, www.waters.com) using a Beckmann SystemGold
instrument consisting of chromatography 125 Solvent Module and
a 166 UV Detector. Mobile phases: 10 mM ammonium acetate in
Millipore purified water (A) and HPLC grade acetonitrile (B). A
gradient was formed from 5% to 80% of B in 10 min and held at 80%
for 10 min; flow rate was 0.7 mL/min and injection volume was
a
IC50, Inhibitory concentration causing a 50% reduction in cell growth, in mM;
mean values of at least 2 triplicate cytotoxicity sulphorhodamine-B (SRB) experi-
ments are reported; the error in these values is within ꢂ30% of the average.
b
Normoxic conditions: 20e21% (v/v) O2.
Hypoxic conditions: 1% (v/v) O2.
c
30 mL. HPLC purity of final compounds was determined by moni-
toring at 254 and 300 nm and was found to be ꢄ98%; retention
times (HPLC, tR) are given in minutes (min). Chromatographic
separations were performed on silica gel columns by flash (Kie-
selgel 40, 0.040e0.063 mm; Merck) or gravity column (Kieselgel
60, 0.063e0.200 mm; Merck) chromatography. Reactions were
followed by thin-layer chromatography (TLC) on Merck aluminum
silica gel (60 F254) sheets that were visualized under a UV lamp.
Evaporation was performed in vacuo (rotating evaporator). Sodium
sulphate was always used as the drying agent. Microwave assisted
reaction were run in a CEM or Biotage microwave synthesizer.
Characterization data of non-key compounds are reported in the
Supplementary Data.
compounds on cell growth in both normoxic and hypoxic envi-
ronments are reported in Table 3 as IC50 values.
All the tested LDH-inhibitors are active against these pancreatic
cancer cells. Under normoxic conditions, the IC50 values against
PANC-1 are in the 8.8e12.0
slightly more sensitive to these compounds, showing IC50 values
which are always below 10 M (5.8e8.9 M). Most importantly, the
mM range, whereas MIA PaCa-2 cells are
m
m
anti-proliferative activity against both cancer cell lines turns out to
be enhanced in hypoxia, with IC50s generally dropping at around
2 mM for almost all the tested compounds. This phenomenon had
already been observed with previously reported LDH inhibitors
[22a], and is consistent with the fact that cancer cells rely more on
LDH activity upon oxygen deprivation than under normally
oxygenated conditions. This “addiction” to glycolysis may be
exploited in a strategy recently defined as “synthetic lethal inter-
actions” [32], where the two concurring factors leading to a coop-
erative cytotoxicity are hypoxia and blockade of glycolysis.
7.1.2. General procedure for the formation of nitrotoluene
derivatives 8a-e
Method A A solution of Pd(OAc)2 (25.2 mg, 0.112 mmol) and
triphenylphosphine (147 mg, 0.560 mmol) in absolute ethanol
(4 mL) and anhydrous toluene (4 mL) was stirred at RT under
nitrogen for 10 min. After that period, commercially available 5-
chloro-2-nitrotoluene 4 (646 mg, 3.76 mmol), 4 mL of 2M
aqueous Na2CO3, and the appropriate boronic acid R1B(OH)2
(6.03 mmol) were sequentially added. The resulting mixture was
heated at 100 ꢀC in a sealed vial under nitrogen overnight. After
being cooled to RT, the mixture was diluted with water and
extracted with EtOAc. The combined organic phase were dried and
concentrated. The crude product was purified by flash chroma-
tography over silica gel column using n-Hex/EtOAc or CHCl3/MeOH
mixtures as the eluent. Method B Commercially available 5-chloro-
2-nitrotoluene 4 (429 mg, 2.50 mmol) was placed in a vial together
with the appropriate boronic acid R1B(OH)2 (2.50 mmol), sodium
The selective cytotoxicity against cancer cells is supported by
the fact that all these compounds showed IC50 values greater than
100
tuted by 3i, which displays a residual toxicity on these cells
(IC50 ¼ 68 M).
mM on Hs27 fibroblast cells, with the single exception consti-
m
6. Conclusions
Following our initial discovery of a new set of NHI-based hLDH5-
inhibitors [22], in this study, we have further developed and opti-
mized this class, by appropriately modifying the position and the
type of the aryl-substituents present in the NHI scaffold. To the best

5404
C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
carbonate (794 mg, 7.50 mmol), Pd(OAc)2 (2.2 mg, 0.010 mmol),
tetrabutylammonium bromide (805 mg, 2.50 mmol) and water
(5 mL). The vial was sealed and heated under stirring at 175 ꢀC in
a microwave reactor for 5e10 min. The reaction mixture was then
diluted with water and repeatedly extracted with EtOAc. The
combined organic phase was dried over anhydrous sodium
sulphate and evaporated to afford a crude residue that was purified
by column chromatography over silica gel using n-Hex/EtOAc
mixtures as the eluent.
Evaporation under vacuum of the organic solvent afforded a crude
product which was quickly filtered through a short column of silica
gel using n-hexane/EtOAc mixtures as the eluent. The resulting a-
ketoester intermediate was directly dissolved in anhydrous DME
(2 mL) and the resultant solution was added dropwise to a cooled
(0 ꢀC) solution of SnCl2$2H2O (860 mg, 3.80 mmol) in DME (2 mL)
containing 4Å molecular sieves, previously activated for 18 h at
130 ꢀC in oven and cooled to RT in a desiccator over anhydrous
calcium chloride. The reaction mixture was stirred under nitrogen
at RT for 5 h, then it was monitored by TLC until disappearance of
starting material (6e16 h). In some cases additional portions (1 eq.
each) of SnCl2$2H2O were required to promote substrate conver-
sion. The mixture was then diluted with water and repeatedly
extracted with EtOAc. The combined organic phase was dried over
anhydrous sodium sulphate and evaporated to afford a crude
residue that was purified by column chromatography over silica gel
using n-Hex/EtOAc or CHCl3/MeOH mixtures as the eluent. Method
D. The nitrotoluene precursor 8 (0.30 mmol) was submitted to the
same initial treatment with sodium hydride and dimethyl oxalate
7.1.3. General procedure for the formation of nitrotoluene
derivatives 8fem, p.
Commercially available 4-bromo-2-nitrotoluene (5) and the
appropriate boronic acid of general formula R2B(OH)2 were
submitted to the same procedure described above for “Method A”,
under either conventional heating at 100 ꢀC, or microwave irradi-
ation. In two cases (8h,i, Scheme 1), the use of potassium phosphate
(3.93 mmol, for 2.31 mmol of 5) as the base in water (0.5 mL) and
conventional heating at 125 ꢀC for 48 h in a sealed vial was
preferred.
as described above in “Method C”. Then the resulting a-ketoester
intermediate was transferred to a yellow suspension, previously
generated by adding triethylamine (0.18 mL, 0.13 g, 1.3 mmol)
dropwise to a stirred solution of SnCl2$2H2O (88.3 mg, 0.391 mmol)
and thiophenol (0.12 mL, 0.13 g, 1.2 mmol) in acetonitrile (2.1 mL) at
room temperature for 5 min. Upon consumption of starting mate-
rial (typically within 5 min), the mixture was diluted with water
and repeatedly extracted with EtOAc. The combined organic phase
was dried over anhydrous sodium sulphate, diluted with toluene
and treated with a catalytic amount of p-toluensulfonic acid. Then it
was evaporated to promote the azeotropic dehydration of small
amounts of hydrated precursors of NHIs 9, affording a crude residue
that was purified by column chromatography over silica gel using
n-Hexane/EtOAc or CHCl3/MeOH mixtures 7:3 as the eluent.
Method E. The nitrotoluene precursor 8 (0.30 mmol) was submitted
to the same initial treatment with sodium hydride and dimethyl
7.1.4. General procedure for the formation of nitrotoluene
derivatives 8n,o,qes
Commercially available aryl bromide of general formula R2Br
(1.76 mmol) was placed in a vial together with 4-methyl-3-
nitrobenzeneboronic acid
6 (350 mg, 1.94 mmol), sodium
carbonate (560 mg, 5.28 mmol), Pd(OAc)2 (1.6 mg, 0.0070 mmol),
tetrabutylammonium bromide (567 mg, 1.76 mmol) and water
(5.5 mL). The vial was sealed and heated under stirring at 150 ꢀC in
a microwave reactor for 5e10 min. The reaction mixture was then
diluted with water and repeatedly extracted with EtOAc. The
combined organic phase was dried over anhydrous sodium
sulphate and evaporated to afford a crude residue that was purified
by column chromatography over silica gel using n-Hex/EtOAc
mixtures as the eluent.
oxalate as described above in “Method C”. Then the resulting
a-
7.1.5. Procedure for the formation of nitrotoluene derivative 8t
1,2-Dichloro-4-methyl-5-nitrobenzene (7) [25] (460 mg,
2.23 mmol) was placed in a vial together with phenylboronic acid
(1.09 g, 8.93 mmol), potassium phosphate (2.08 g, 9.81 mmol),
Pd(OAc)2 (13 mg, 0.059 mmol), tetrabutylammonium bromide
(1.12 g, 34.6 mmol) and water (5 mL). The vial was sealed and
heated under stirring at 125 ꢀC for 48 h. The reaction mixture was
cooled to RT and then diluted with water. The water phase was
acidified with 1 N aqueous HCl and repeatedly extracted with
EtOAc. The combined organic phase was dried over anhydrous
sodium sulphate and evaporated to afford a crude residue that was
purified by column chromatography over silica gel using n-hexane
as the eluent, to give pure 8t (587 mg, 2.03 mmol, 91% yield).
ketoester intermediate was dissolved in THF (0.5 mL) and the
resulting solution was added to an aqueous solution (0.5 mL) of
sodium hypophosphite monohydrate (88.4 g, 0.834 mmol). Then
2.7 mg of 10% palladium over charcoal was added and the resulting
suspension was stirred at RT. Then it was monitored by TLC and
additional portions of H2PO2NaH2O (88.4 g, 0.834 mmol) and Pd/C
(2.7 mg) were added, if necessary. After disappearance of the
precursor (TLC), typically after 24 h, the mixture was concentrated,
diluted with water and repeatedly extracted with EtOAc. The
combined organic phase was dried over anhydrous sodium
sulphate and evaporated to afford a crude residue that was purified
by column chromatography over silica gel using n-Hex/EtOAc or
CHCl3/MeOH mixtures as the eluent. Method F. The nitrotoluene
precursor 8 (0.40 mmol) was submitted to the same initial treat-
ment with sodium hydride and dimethyl oxalate as described
7.1.6. General procedure for the formation of N-hydroxyindole
methyl-esters 9
above in “Method C”. Then the
a-ketoester intermediate
Method C A stirred suspension of sodium hydride (60% disper-
sion in mineral oil, 263 mg, 6.58 mmol) in 4 mL of anhydrous DMF
cooled to ꢁ15 ꢀC under nitrogen was treated dropwise with
a solution containing the appropriate nitrotoluene precursor 8
(1.64 mmol) and dimethyl oxalate (968 mg, 8.20 mmol) in 3 mL of
anhydrous DMF. The mixture was left under stirring at the same
temperature for 10 min and then it was slowly warmed to room
temperature. Stirring at RT was continued until the disappearance
of the precursor was verified by TLC. Then, the reaction mixture was
slowly poured in an ice-water mixture; the water phase was acid-
ified with 1 N aqueous HCl and extracted several times with EtOAc.
The combined organic phase was washed with saturated NaHCO3
solution and brine, then dried over anhydrous sodium sulphate.
(0.39 mmol) was dissolved in 4 mL of MeOH, and the resulting
solution was treated with lead powder (400 mg, 1.93 mol) and
0.4 mL of triethylammonium formate. The mixture was stirred at
55 ꢀC for 12 h, cooled to RT, and filtered over a pad of Celite. The
solvent was removed under reduced pressure and the residue
purified by column chromatography over silica gel using n-hexane/
EtOAc 1:1 as the eluent.
7.1.7. General procedure for the formation of N-hydroxyindole-2-
carboxylates 3
Methyl esters 9 (0.16 mmol) were dissolved in a 1:1 mixture of
THF/methanol (3 mL) and treated with 0.5 mL of 2 N aqueous
solution of LiOH. The reaction was kept under stirring in the dark at

C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
5405
RT and it was monitored by TLC. Upon consumption of starting
material, the reaction mixture was concentrated, diluted with
water and washed with Et2O. The aqueous phase was treated with
1 N aqueous HCl and extracted several times with EtOAc. The
combined ethyl acetate organic phase was dried over anhydrous
sodium sulphate and evaporated to afford the desired N-hydrox-
yindol-carboxylic acid products 3.
122.45, 123.94, 125.48 (q, J ¼ 269.7 Hz), 126.54 (q, 2C, J ¼ 3.7 Hz),
127.60, 128.62 (2C), 129.35 (q, J ¼ 34.8 Hz), 137.36, 137.38, 146.10,
161.99; MS (EI, 70 eV) m/z (%): 321 (100) [Mþ], 305 (18) [Mþ eO];
HPLC, tR ¼ 10.4 min.
7.1.7.7. 6-(Biphenyl-4-yl)-1-hydroxy-1H-indole-2-carboxylic acid
(3j). Yield 93%; 1H NMR (200 MHz, acetone-d6)
d (ppm): 7.16 (d,
1H, J ¼ 0.7 Hz), 7.34e7.56 (m, 3H), 7.72e7.90 (m, 9H); 13C NMR
7.1.7.1. 5-(4-Chlorophenyl)-1-hydroxy-1H-indole-2-carboxylic acid
(50 MHz, acetone-d6) d (ppm): 106.10, 108.23, 121.37, 122.12,
(3a). Yield 99%; 1H NMR (200 MHz, DMSO-d6)
d
(ppm): 7.07 (s, 1H),
123.78, 126.29, 127.65 (2C), 128.18, 128.26 (2C), 128.51 (2C),
129.77 (2C), 137.54, 138.85, 140.84, 141.33, 141.44, 162.40; MS (EI,
70 eV) m/z (%): 329 (10) [Mþ], 313 (100) [Mþ eO], 267 (41) [Mþ
eOH eCOOH], 190 (27) [Mþ eOH eCOOH eC6H5]; HPLC,
tR ¼ 10.4 min.
7.44e7.56 (m, 3H), 7.63 (dd, 1H, J ¼ 8.7, 1.6 Hz), 7.71 (AA0/XX0, 2H,
JAx ¼ 8.6 Hz, JAA /xx ¼ 1.5 Hz), 7.93 (d, 1H, J ¼ 0.8 Hz); 13C NMR
0
0
(50 MHz, DMSO-d6)
d (ppm): 105.11, 110.24, 120.07, 121.44, 124.08,
127.43, 128.38 (2C), 128.78 (2C), 131.42, 135.46, 139.70, 161.10; MS
(EI, 70 eV) m/z (%): 289 (40) [37Cl, Mþ], 287 (100) [35Cl, Mþ], 271 (85)
[
37Cl, Mþ eH2O, 35Cl, Mþ eO]; HPLC, tR ¼ 9.9 min.
7.1.7.8. 1-Hydroxy-6-(naphthalen-1-yl)-1H-indole-2-carboxylic
acid (3k). Yield 82%; 1H NMR (200 MHz, DMSO-d6)
d (ppm): 7.11
7.1.7.2. 5-(3-Carboxyphenyl)-1-hydroxy-1H-indole-2-carboxylic acid
(3e). Yield 84%; 1H NMR (200 MHz, acetone-d6)
(ppm): 7.18 (d,
(d, 1H, J ¼ 0.9 Hz), 7.21 (dd, 1H, J ¼ 8.2, 1.6 Hz), 7.45e7.64 (m, 5H),
7.77 (dd, 1H, J ¼ 8.2, 0.6 Hz), 7.86 (dd, 1H, J ¼ 7.9, 1.3 Hz), 7.97 (d,
1H, J ¼ 8.6 Hz), 8.02 (dd, 1H, J ¼ 7.9, 1.6 Hz); 13C NMR (50 MHz,
d
1H, J ¼ 0.9 Hz), 7.58 (td, 1H, J ¼ 7.5, 0.4 Hz), 7.62 (dt, 1H, J ¼ 8.8,
0.7 Hz), 7.71 (dd, 1H, J ¼ 8.6, 1.6 Hz), 7.91 (dd, 1H, J ¼ 2.0, 1.3 Hz),
7.94e8.00 (m, 2H), 8.31 (t, 1H, J ¼ 1.6 Hz); 13C NMR (50 MHz,
DMSO-d6)
d (ppm): 104.65, 110.28, 120.26, 121.97, 122.84, 125.30,
125.50, 125.83, 126.24, 126.99, 127.34, 127.48, 128.28, 130.94,
133.38, 135.99, 136.66, 139.86, 161.06; MS (EI, 70 eV) m/z (%): 303
(53) [Mþ], 287 (87) [Mþ eO], 241 (100) [Mþ eOH eCOOH]; HPLC,
tR ¼ 10.3 min.
acetone-d6)
d (ppm): 106.66, 111.07, 121.52, 123.01, 125.70, 128.60,
128.80, 129.89, 132.10, 132.21, 133.81, 136.50, 139.10, 142.78, 162.00,
167.49; MS (EI, 70 eV) m/z (%): 297 (10) [Mþ], 281 (60) [Mþ eO], 248
(100) [Mþ eO2 eOH]; HPLC, tR ¼ 5.8 min.
7.1.7.9. 1-Hydroxy-6-(naphthalen-2-yl)-1H-indole-2-carboxylic
7.1.7.3. 6-(4-Chlorophenyl)-1-hydroxy-1H-indole-2-carboxylic acid
acid (3l). Yield 99%; 1H NMR (200 MHz, acetone-d6)
d (ppm): 7.17
(3f). Yield 93%; 1H NMR (200 MHz, DMSO-d6)
d
(ppm): 7.04 (d, 1H,
(d, 1H, J ¼ 0.9 Hz), 7.51e7.57 (m, 2H), 7.63 (dd, 1H, J ¼ 8.5, 1.6 Hz),
J ¼ 0.8 Hz), 7.41 (dd, 1H, J ¼ 8.4, 1.4 Hz), 7.52 (AA’XX’, 2H,
7.81 (dd, 1H, J ¼ 8.4, 0.6 Hz), 7.92e7.97 (m, 3H), 7.99e8.06 (m, 2H),
JAX ¼ 8.4 Hz, JAA’/XX’ ¼ 2.0 Hz), 7.65 (s, 1H), 7.68e7.82 (m, 3H); 13C
8.29 (d, 1H, J ¼ 1.5 Hz); 13C NMR (50 MHz, DMSO-d6)
d (ppm):
NMR (50 MHz, DMSO-d6)
d
(ppm): 104.49, 107.16, 119.71, 120.58,
104.52, 107.40, 120.09, 120.40, 120.49, 122.68, 125.23 (2C), 125.95,
126.30, 127.37, 128.14, 128.39, 132.09, 133.35, 136.50, 136.68, 137.86,
161.08; MS (EI, 70 eV) m/z (%): 303 (44) [Mþ], 287 (100) [Mþ eO],
241 (57) [Mþ eOH eCOOH]; HPLC, tR ¼ 10.1 min.
122.77, 127.52, 128.58 (2C), 128.85 (2C), 132.04, 135.53, 136.31,
139.39, 161.08; MS (EI, 70 eV) m/z (%): 289 (15) [37Cl, Mþ], 287 (36)
[
35Cl, Mþ] 273 (22) [37Cl, Mþ eO], 271 (56) [35Cl, Mþ eO], 190 (100)
[Mþ eCl eH2O eCO2]; HPLC, tR ¼ 10.2 min.
7.1.7.10. 1-Hydroxy-6-(4-(trifluoromethoxy)phenyl)-1H-indole-2-
carboxylic acid (3n). Yield 55%; 1H NMR (200 MHz, acetone-d6)
7.1.7.4. 6-(2,4-Dichlorophenyl)-1-hydroxy-1H-indole-2-carboxylic
acid (3g). Yield 99%; 1H NMR (200 MHz, acetone-d6)
d
(ppm): 7.17
d
(ppm): 7.16 (d,1H, J ¼ 0.9 Hz), 7.42e7.50 (m, 3H), 7.75e7.80 (m, 2H),
(d, 1H, J ¼ 0.9 Hz), 7.20 (dd, 1H, J ¼ 8.4, 1.6 Hz), 7.48 (dd, 1H, J ¼ 8.3,
1.7 Hz), 7.53 (dd,1H, J ¼ 8.4, 0.7 Hz), 7.58 (dt,1H, J ¼ 1.5, 0.7 Hz), 7.64
(dd, 1H, J ¼ 1.7, 0.6 Hz), 7.76 (dd, 1H, J ¼ 8.4, 0.7 Hz); 13C NMR
7.88 (AA0XX0, 2H, JAX ¼ 8.9 Hz, JAA0/XX0 ¼ 2.6 Hz); 13C NMR (50 MHz,
acetone-d6)
d
(ppm): 106.02, 108.48,121.23,121.43 (q, J ¼ 256.5 Hz),
122.06,122.23 (2C),123.81,127.34,129.66 (2C),137.36,137.60,141.42,
149.21, 161.98; MS (EI, 70 eV) m/z (%): 337 (100) [Mþ], 321 (74) [Mþ
eO], 293 (7) [Mþ eCO2], 275 (22) [Mþ eCO2 eH2O], 249 (26) [Mþ
eCO2 eH2O eC2H2]; HPLC, tR ¼ 10.3 min.
(50 MHz, acetone-d6)
d (ppm): 106.06, 111.12, 122.05, 122.98,
123.25, 127.45, 128.24, 130.20, 133.72, 133.90, 134.14, 136.10, 136.61,
140.58, 161.98; MS (EI, 70 eV) m/z (%): 325 (5) [37/37Cl2, Mþ], 323
(23) [37/35Cl2, Mþ], 321 (36) [35/35Cl2, Mþ], 309 (6) [37/37Cl2, Mþ
eO], 307 (29) [37/35Cl2, Mþ eO], 305 (44) [35/35Cl2, Mþ eO], 226 (37)
7.1.7.11. 1-Hydroxy-6-(3-(trifluoromethoxy)phenyl)-1H-indole-2-
carboxylic acid (3o). Yield 83%; 1H NMR (200 MHz, acetone-d6)
[
37Cl, Mþ eCl eOH eCOOH]; 224 (100) [35Cl, Mþ eCl eOH eCOOH];
189 (74) [Mþ e2Cl eOH eCOOH]; HPLC, tR ¼ 11.3 min.
d
(ppm): 7.16 (d, 1H, J ¼ 0.7 Hz), 7.31e7.38 (m, 1H), 7.49 (dd, 1H,
J ¼ 8.4, 1.6 Hz), 7.63 (t, 1H, J ¼ 7.9 Hz), 7.68e7.70 (m, 1H), 7.77e7.83
(m, 3H); 13C NMR (50 MHz, acetone-d6)
(ppm): 105.95, 108.72,
7.1.7.5. 6-(4-Fluorophenyl)-1-hydroxy-1H-indole-2-carboxylic acid
d
(3h). Yield 93%; 1H NMR (200 MHz, acetone-d6)
d
(ppm): 7.15 (d,
1H, J ¼ 0.6 Hz), 7.18e7.32 (m, 2H), 7.42 (dd, 1H, J ¼ 8.6, 1.6 Hz),
7.67e7.86 (m, 4H); 13C NMR (50 MHz, acetone-d6)
(ppm): 106.19,
120.34, 120.55, 121.23, 121.62 (q, J ¼ 253.4 Hz), 122.41, 123.94,
126.94, 127.53, 131.48, 137.31, 137.49, 144.81, 150.62, 162.27; MS (EI,
70 eV) m/z (%): 337 (76) [Mþ], 321 (100) [Mþ eO], 293 (8) [Mþ
eCO2], 275 (13) [Mþ eCO2 eH2O], 249 (21) [Mþ eCO2 eH2O
eC2H2]; HPLC, tR ¼ 10.4 min.
d
108.17, 116.32 (d, 2C, J ¼ 21.0 Hz), 121.26, 121.77, 123.68, 127.13,
129.79 (d, 2C, J ¼ 8.2 Hz), 137.50, 138.14, 138.52 (d, J ¼ 3.7 Hz),
161.90, 163.15 (d, J ¼ 244.5 Hz); MS (EI, 70 eV) m/z (%): 271 (100)
[Mþ], 255 (33) [Mþ eO], 208 (55) [Mþ eCO2 eF]; HPLC,
tR ¼ 9.7 min.
7.1.7.12. 6-(Furan-3-yl)-1-hydroxy-1H-indole-2-carboxylic
acid
(3s). Yield 97%; 1H NMR (200 MHz, acetone-d6)
d
(ppm): 7.01 (dd,
1H, J ¼ 1.8, 0.9 Hz), 7.10 (d, 1H, J ¼ 0.9 Hz), 7.42 (dd, 1H, J ¼ 8.4,
7.1.7.6. 1-Hydroxy-6-(4-(trifluoromethyl)phenyl)-1H-indole-2-
carboxylic acid (3i). Yield 89%; 1H NMR (200 MHz, acetone-d6)
1.5 Hz), 7.65e7.72 (m, 3H), 8.13 (dd, 1H, J ¼ 1.5, 0.9 Hz); 13C NMR
(50 MHz, acetone-d6)
d (ppm): 106.24, 106.75, 109.61, 120.33,
d
(ppm): 7.17 (d, 1H, J ¼ 0.7 Hz), 7.52 (dd, 1H, J ¼ 8.4, 1.6 Hz), 7.81
121.52, 123.59, 127.75, 129.64, 130.64, 137.34, 140.02, 144.86, 162.14;
MS (EI, 70 eV) m/z (%): 243 (56) [Mþ], 227 (100) [Mþ eO], 180 (26)
[Mþ eCO2 eH2O]; HPLC, tR ¼ 8.6 min.
(dd, 1H, J ¼ 8.4, 0.7 Hz), 7.82e7.90 (m, 3H), 7.96e8.03 (m, 2H); 13C
NMR (50 MHz, acetone-d6)
d (ppm): 106.00, 108.83, 121.23,

5406
C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
7.1.7.13. 1-Hydroxy-5,6-diphenyl-1H-indole-2-carboxylic
(3t). Yield 80%; 1H NMR (200 MHz, acetone-d6)
(ppm): 7.06e7.28
(m, 11H), 7.54 (s, 1H), 7.70 (s, 1H); 13C NMR (50 MHz, acetone-d6)
(ppm): 106.28, 111.80, 121.88, 124.85, 126.80, 127.18, 127.51, 128.49
(2C), 128.56 (2C), 130.73 (2C), 130.84 (2C), 135.32, 136.47, 139.65,
142.93, 143.02, 162.01; MS (EI, 70 eV) m/z (%): 330 (15) [M þ Hþ],
268 (100) [Mþ eCO2 eH2O], 254 (31) [Mþ eCO2 eNOH]; HPLC,
tR ¼ 10.3 min.
acid
activities were measured using commercially available purified
human lactate dehydrogenase isoform 5 (hLDH5, LDH-A4, LDH-A
Lee Biosolution, Inc.) and isoform 1 (hLDH1, LDH-B4, LDH-B,
ERMÒ - AD453/IFCC from Sigma). The reaction velocity of purified
human hLDH5 and hLDH1, and the values of optical density were
determined at 30-s intervals for 5 min, by measuring the decrease
in absorbance at 340 nm of NADH at 37 ꢀC. The apparent Michae-
liseMenten constants (Km) of hLDH5 and hLDH1 were measured by
non-linear regression analysis with the GraphPad Prism 3.0 from
LineweavereBurk plots. The Km values of hLDH5 for NADH and
pyruvate were determined in saturating conditions as described
below. In 100 mM sodium phosphate buffer (pH 7.4), 0.005 units of
hLDH5 were combined with 2 mM sodium pyruvate (pyruvate-
d
d
7.2. Molecular modelling
Compound 3g was built using Maestro 9.0 [33] and was mini-
mized using the conjugate gradient method until a convergence
value of 0.05 kcal/(molꢅÅ) was reached. The minimization was
carried out in a water environment model (generalized-Born/
surface-area model) using the MMFFs force field and a distance-
saturated) and 12.5
saturated) and 25
m
m
Me150
mM NADH, or 200 mM NADH (NADH-
M to 1 mM sodium pyruvate, and these
conditions were used for all assays. Under these conditions NADH
showed a Km of 20.0 M and a Vmax of 62.5 mol/min/mg, whereas
pyruvate showed a Km of 120 M and a Vmax of 62.5 mol/min/mg.
dependent dielectric constant of 1.0. The human muscle
L-LDH M
m
m
chain was extracted from the minimized average structure of the
LDH obtained by us through molecular dynamic simulations [22].
Automated docking was carried out by means of the GOLD
program, version 4.1.1 [34]. The “allow early termination” option
was deactivated, the remaining GOLD default parameters were
used, and the ligand was submitted to 30 genetic algorithm runs by
applying the ChemScore fitness function. The best docked confor-
mation was taken into account. The so obtained complex was
energy minimized using AMBER 11 [35]. The complex was placed in
a rectangular parallelepiped water box, an explicit solvent model
for water (TIP3P) was used, and the complex was solvated with
a 10 Å water cap. Chlorine ions were added as counterions to
neutralize the system. Two steps of minimization were then carried
out. In the first stage, we kept the complex fixed with a position
restraint of 500 kcal/(molꢅÅ2) and we solely minimized the posi-
tions of the water molecules. In the second stage, we minimized the
entire system through 20 000 steps of steepest descent followed by
conjugate gradient until a convergence of 0.05 kcal/(molꢅÅ) was
m
m
MichaeliseMenten constants for substrates were determined from
initial rate measurements at 37 ꢀC by non-linear regression analysis
with the GraphPad Prism 3.0. Briefly, the synthesized compounds
were dissolved in stock solutions of DMSO (concentrations of
DMSO during the initial rate measurements did not exceed 0.5%).
Using 25 mM NADH and 2 mM sodium pyruvate we initially eval-
uated the percentual inhibition of the compounds. Then, we eval-
0
uated the apparent Km values in the presence of these inhibitors
0
(concentration range ¼ 5e100
m
M). From the values of Km so ob-
tained, Ki values for each single inhibitor were determined using
double-reciprocal plots. The effect of the same compounds on
hLDH1 enzyme activity was measured as described before for
hLDH5, only in pyruvate-saturating conditions. Under these
conditions NADH showed a Km of 27
mg. The inhibition percentage of these compounds at a fixed
concentration of 125 M was determined as described above.
mM and a Vmax of 65 mmol/min/
m
attained. All the
a
carbons of the protein were blocked with
7.3.2. Cell culture and cytotoxicity in normoxic and hypoxic
exposure
a harmonic force constant of 10 kcal/(molꢅÅ2). Ten nanoseconds of
MD simulation were then carried out. The time step of the simu-
lations was 2.0 fs with a cutoff of 10 Å for the non-bonded inter-
action, and SHAKE was employed to keep all bonds involving
hydrogen atoms rigid. Constant-volume periodic boundary MD was
carried out for 400 ps, during which the temperature was raised
from 0 to 300 K. Then 9.6 ns of constant pressure periodic boundary
MD was carried out at 300 K using the Langevin thermostat to
maintain constant the temperature of our system. In the first 1.2 ns,
Human pancreatic tumour cells PANC-1 (CRL-1469) and MIA
PaCa-2 (CRL-1420), as well as skin fibroblasts Hs27 (CRL-1634),
were obtained from the American Type Culture Collection (ATCC,
Manassas, VA, USA). Cells were maintained in RPMI 1640 medium,
supplemented with fetal calf serum (10%), glutamine (2 mM), and
1% penicillin-streptomycin (100 00U/ml), at 37 ꢀC under an atmo-
sphere of 5% CO2. The cells were maintained in 75 cm2 culture flasks
(Greiner-Bio-One, Frickenhausen, Germany) and harvested with
trypsin-EDTA (Invitrogen, Paisley, UK) at their exponential phase of
growth phase. Chemosensitivity tests were performed using the
SRB assay as described in the NCI protocol, with slight modifica-
tions. Selected compounds 3f,g,i,o,r,t were dissolved in sterile
DMSO at 2.5 mM and stored at ꢁ20 ꢀC, in the dark. They were
diluted in sterile culture medium immediately before their use.
Cells were plated at 5$104 cells/well and each compound was tested
all the
a carbons of the protein were blocked with a harmonic force
constant, which decreased during these 1.2 ns from 10 to 1 kcal/
(molꢅÅ2), while in the last 7.8 ns, no constraints were applied.
General Amber force field (GAFF) parameters were assigned to the
ligand, while partial charges were calculated using the AM1-BCC
method as implemented in the Antechamber suite of AMBER 11.
The final structure of the complex was obtained as the average of
the last 7 ns of MD minimized by the CG method until a conver-
gence of 0.05 kcal/(molꢅÅ). The average structure was obtained
using the ptraj program implemented in AMBER 11.
in triplicate at different concentrations (1e100 mM), using the SRB
assay as described previously [22]. Briefly, control cells were
exposed to an equivalent concentration of DMSO (0.25% v/v,
negative control). At the end of drug incubation (48 h), growth
inhibition was expressed as the percentage of control absorbance
(corrected for absorbance before drug addition). The 50% inhibitory
concentration of cell growth (IC50) was calculated by non-linear
least squares curve fitting (GraphPad PRISM, Intuitive Software
for Science, San Diego, CA). These experiments were also performed
in cells under hypoxia conditions, using the incubator IncuSafe,
(Labo Equipment Sanyo, UK). Hypoxic cells were maintained in
a controlled atmosphere chamber with a gas mixture containing 1%
(v/v) O2, 5% (v/v) CO2, and 94% (v/v) N2 at 37 ꢀC.
7.3. Biological methods
7.3.1. Enzyme kinetics experiments
The LDH inhibitory activities of the new compounds were
measured by standard enzyme kinetics experiments, by means of
spectrophotometric measurement (Lambda 25, PerkinElmer) of the
oxidation of NADH at 340 nm (ε ¼ 6.2 mMꢁ1 cmꢁ1), which is
a direct measure of the reduction of pyruvate to lactate. The effect
of the perspective inhibitors on hLDH5 and hLDH1 enzyme

C. Granchi et al. / European Journal of Medicinal Chemistry 46 (2011) 5398e5407
5407
[13] (a) M.I. Koukorakis, A. Giatromanolaki, C. Simopoulos, A. Polychronidis,
E. Sivridis, Clin. Experim. Metast 22 (2005) 25e30;
Acknowledgment
(b) M.I. Koukorakis, M. Pitiakoudis, A. Giatromanolaki, A. Tsarouha,
A. Polychronidis, E. Sivridis, C. Simopoulos, Cancer Sci. 97 (2006) 1056e1060;
(c) M.I. Koukourakis, A. Giatromanolaki, E. Sivridis, K.C. Gatter, T. Trarbach,
G. Folprecht, M.M. Shi, D. Lebwohl, T. Jalava, D. Laurent, G. Meinhardt,
A.L. Harris, Clin. Cancer Res. 17 (2011) 4892e4900.
The research leading to these results has received funding (to
FM and SR) from the European Union’s Seventh Framework Pro-
gramme (FP7-PEOPLEeIIFe2008) under grant agreement n.
235016 (“NOXYCANCERSTARV”). CG thanks the Italian Ministry for
University and Research (MIUR) for a PhD fellowship. Siena Biotech
S. p. A. - Italy, is gratefully acknowledged for a research contribution
(FM). This work was also supported by a grant from the EORTC-
PAMM group to FM and GJP. We thank Dr. Giorgio Placanica and Dr.
Caterina Orlando for technical assistance in the analysis of the
chemical products.
[14] (a) B.S. Sørensen, J. Hao, J. Overgaard, H. Vorum, B. Honoré, J. Alsner,
M.R. Horsman, Radiother. Oncol. 76 (2005) 187e193;
(b) B.S. Sørensen, J. Alsner, J. Overgaard, M.R. Horsman, Radiother. Oncol. 83
(2007) 362e366.
[15] (a) V.R. Fantin, J. St-Pierre, P. Leder, Cancer Cell. 9 (2006) 425e434;
(b) Z. -Y. Wang, T. Y. Loo, J. -G. Shen, N. Wang, D. -M. Wang, D. -P. Yang, S. -L.
Mo, X. -Y. Guan, J. -P. Chen, Breast Cancer Res. Treat. 2011, in press.
[16] T. Kanno, K. Sudo, M. Maekawa, Y. Nishimura, M. Ukita, K. Fukutake, Clin.
Chim. Acta 173 (1988) 89e98.
[17] C. Granchi, S. Bertini, M. Macchia, F. Minutolo, Curr. Med. Chem. 17 (2010)
672e697.
[18] L. Fiume, M. Manerba, M. Vettraino, G. Di Stefano, Pharmacology 86 (2010)
157e162.
[19] L.M. Deck, R.E. Royer, B.B. Chamblee, V.M. Hernandez, R.R. Malone, J.E. Torres,
L.A. Hunsaker, R.C. Piper, M.T. Makler, D.L. Vander Jagt, J. Med. Chem. 41
(1998) 3879e3887.
Appendix. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ejmech.2011.08.046.
[20] A. Le, C.R. Cooper, A.M. Gouw, R. Dinavahi, A. Maitra, L.M. Deck, R.E. Royer,
D.L. Vander Jagt, G.L. Semenza, C.V. Dang, Proc. Natl. Acad. Sci. USA 107 (2010)
2037e2042.
References
[21] A.D. Moorhouse, C. Spiteri, P. Sharma, M. Zloh, J.E. Moses, Chem. Commun. 47
(2011) 230e232.
[1] D. Hanahan, R.A. Weinberg, Cell 144 (2011) 646e674.
[2] O. Warburg, Science 123 (1956) 309e314.
[22] (a) C. Granchi, S. Roy, C. Giacomelli, M. Macchia, T. Tuccinardi, A. Martinelli,
M. Lanza, L. Betti, G. Giannaccini, A. Lucacchini, N. Funel, L.G. León,
E. Giovannetti, G.J. Peters, R. Palchaudhuri, E.C. Calvaresi, P.J. Hergenrother,
F. Minutolo, J. Med. Chem. 54 (2011) 1599e1612;
[3] (a) M.G. Vander Heiden, L.C. Cantley, C.B. Thompson, Science 324 (2009)
1029e1033;
(b) R.A. Gatenby, R.J. Gillies, Nat. Rev. Cancer 4 (2004) 891e899;
(c) R.A. Cairns, I.S. Harris, T.W. Mak, Nat. Rev. Canc. 11 (2011) 85e95.
[4] S. Ben-Haim, P. Ell, J. Nucl. Med. 50 (2009) 88e99.
(b) C. Granchi, S. Roy, C. Del Fiandra, T. Tuccinardi, M. Lanza, L. Betti,
G. Giannaccini, A. Lucacchini, A. Martinelli, M. Macchia, F. Minutolo, Med.
Chem. Commun. 2 (2011) 638e643.
[5] (a) C.V. Dang, M. Hamaker, P. Sun, A. Le, P. Gao, J. Mol. Med. 89 (2011)
205e212;
[23] N. Miyaura, A. Suzuki, Chem. Rev. 95 (1995) 2457e2483.
[24] N.E. Leadbeater, M. Marco, Org. Lett. 4 (2002) 2973e2976.
[25] R.B. Bedford, M.E. Blake, C.P. Butts, D. Holder, Chem. Commun. (2003)
466e467.
(b) G. Kroemer, J. Pouyssegur, Cancer Cell 13 (2008) 472e482;
(c) D.A. Tennant, R.V. Durán, E. Gottlieb, Nat. Rev. Canc. 10 (2010) 267e277;
(d) R. Scatena, P. Bottoni, A. Pontoglio, L. Mastrototaro, B. Giardina, Exp. Opin.
Invest. Drugs 17 (2008) 1533e1534;
[26] I. Shimada, K. Maeno, K. Kazuta, H. Kubota, T. Kimizuka, Y. Kimura,
K. Hatanaka, Y. Naitou, F. Wanibuchi, S. Sakamoto, S. Tsukamoto, Bioorg. Med.
Chem. 16 (2008) 1966e1982.
[27] K.C. Nicolaou, A.A. Estrada, G.C. Freestone, S.H. Lee, X. Alvarez-Mico, Tetra-
hedron 63 (2007) 6088e6114.
[28] W. Dong, L.S. Jimenez, J. Org. Chem. 64 (1999) 2520e2523.
[29] (a) M. Bartra, P. Romea, F. Urpí, J. Vilarrasa, Tetrahedron 46 (1990) 587e594;
(b) D. Sawant, R. Kumar, P.R. Maulik, B. Kundu, Org. Lett. 8 (2006) 1525e1528.
[30] I.D. Entwistle, T. Gilkerson, R.A.W. Johnstone, R.P. Telford, Tetrahedron 34
(1978) 213e215.
[31] A. Wong, J.T. Kuethe, I.W. Davies, J. Org. Chem. 68 (2003) 9865e9866.
[32] D.A. Chan, A.J. Giaccia, Nat. Rev. Drug Discov. 10 (2011) 351e364.
[33] Maestro, Version 9.0. Schrödinger Inc, Portland, OR, 2009.
[34] M.L. Verdonk, J.C. Cole, M.J. Hartshorn, C.W. Murray, R.D. Taylor, Proteins 52
(2003) 609e623.
[35] D.A. Case, T.A. Darden, T.E. Cheatham III, C.L. Simmerling, J. Wang, R.E. Duke,
R. Luo, R.C. Walker, W. Zhang, K.M. Merz, B. Roberts, B. Wang, S. Hayik,
A. Roitberg, G. Seabra, I. Kolossvai, K.F. Wong, F. Paesani, J. Vanicek, J. Liu,
X. Wu, S.R. Brozell, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-
J. Hsieh, G. Cui, D.R. Roe, D.H. Mathews, M.G. Seetin, C. Sagui, V. Babin,
T. Luchko, S. Gusarov, A. Kovalenko, P.A. Kollman, AMBER 11. University of
California, San Francisco, 2010.
(e) S.P. Mathupala, Recent Pat. Anti-Canc. Drug Discov. 6 (2011) 6e14.
[6] (a) H.R. Christofk, M.G. Vander Heiden, M.H. Harris, A. Ramanathan,
R.E. Gerszten, R. Wei, M.D. Fleming, S.L. Schreiber, L.C. Cantley, Nature 452
(2008) 230e233;
(b) M.G. Vander Heiden, H.R. Christofk, E. Schuman, A.O. Subtelny, H. Sharfi,
E.E. Harlow, J. Xian, L.C. Cantley, Biochem. Pharmacol. 79 (2010) 1118e1124.
[7] (a) J.-K. Jiang, M.B. Boxer, M.G. Vander Heiden, M. Shen, A.P. Skoumbourdis,
N. Southall, H. Veith, W. Leister, C.P. Austin, H.W. Park, J. Inglese, L.C. Cantley,
D.S. Auld, C.J. Thomas, Bioorg. Med. Chem. Lett. 20 (2010) 3387e3393;
(b) M.B. Boxer, J.-K. Jiang, M.G. Vander Heiden, M. Shen, A.P. Skoumbourdis,
N. Southall, H. Veith, W. Leister, C.P. Austin, H.W. Park, J. Inglese, L.C. Cantley,
D.S. Auld, C.J. Thomas, J. Med. Chem. 53 (2010) 1048e1055.
[8] S. Bonnet, S.L. Archer, J. Allalunis-Turner, A. Haromy, C. Beaulieu, R. Thompson,
C.T. Lee, G.D. Lopaschuk, L. Puttagunta, S. Bonnet, G. Harry, K. Hashimoto,
C.J. Porter, M.A. Andrade, B. Thebaud, E.D. Michelakis, Cancer Cell 11 (2007)
37e51.
[9] K.E. Yen, M.A. Bittinger, S.M. Su, V.R. Fantin, Oncogene 29 (2010) 6409e6417.
[10] E. Goldberg, Exp. Clin. Immunogenet. 2 (1985) 120e124.
[11] J.A. Read, V.J. Winter, C.M. Eszes, R.B. Sessions, R.L. Brady, Proteins 43 (2001)
175e185.
[12] G.A. Brooks, H. Dubouchaud, M. Brown, J.P. Sicurello, C.E. Butz, Proc. Natl.
Acad. Sci. USA 96 (1999) 1129e1134.