Identification of substituted 3-hydroxy-2-mercaptocyclohex-2-enones as potent inhibitors of human lactate dehydrogenase

Article abstract of DOI:10.1016/j.bmcl.2014.06.076

A novel class of 3-hydroxy-2-mercaptocyclohex-2-enone-containing inhibitors of human lactate dehydrogenase (LDH) was identified through a high-throughput screening approach. Biochemical and surface plasmon resonance experiments performed with a screening hit (LDHA IC₅₀ = 1.7 μM) indicated that the compound specifically associated with human LDHA in a manner that required simultaneous binding of the NADH co-factor. Structural variation of this screening hit resulted in significant improvements in LDHA biochemical inhibition activity (best IC₅₀ = 0.18 μM). Two crystal structures of optimized compounds bound to human LDHA were obtained and explained many of the observed structure-activity relationships. In addition, an optimized inhibitor exhibited good pharmacokinetic properties after oral administration to rats (F = 45%).

Full text of DOI:10.1016/j.bmcl.2014.06.076

Bioorganic & Medicinal Chemistry Letters 24 (2014) 3764–3771
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of substituted 3-hydroxy-2-mercaptocyclohex-2-
enones as potent inhibitors of human lactate dehydrogenase
a
a
b
a
Peter S. Dragovich a, , Benjamin P. Fauber , Jason Boggs , Jinhua Chen , Laura B. Corson ,
Charles Z. Ding b, Charles Eigenbrot a, HongXiu Ge b, Anthony M. Giannetti a, Thomas Hunsaker a,
Sharada Labadie a, Chiho Li b, Yichin Liu a, Yingchun Liu b, Shuguang Ma a, Shiva Malek a, David Peterson a,
Keith E. Pitts a, Hans E. Purkey a, Kirk Robarge a, Laurent Salphati a, Steve Sideris a, Mark Ultsch a,
Erica VanderPorten a, Jing Wang b, BinQing Wei a, Qing Xu b, Ivana Yen a, Qin Yue a, Huihui Zhang b,
Xuying Zhang b, Aihe Zhou a
⇑
a Genentech, Inc., 1 DNA Way, South San Francisco, CA 94080, USA
b WuXi AppTec Co., Ltd, 288 Fute Zhong Road, Waigaoqiao Free Trade Zone, Shanghai 200131, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel class of 3-hydroxy-2-mercaptocyclohex-2-enone-containing inhibitors of human lactate dehy-
drogenase (LDH) was identified through a high-throughput screening approach. Biochemical and surface
plasmon resonance experiments performed with a screening hit (LDHA IC50 = 1.7 lM) indicated that the
compound specifically associated with human LDHA in a manner that required simultaneous binding of
the NADH co-factor. Structural variation of this screening hit resulted in significant improvements in
Received 7 June 2014
Revised 23 June 2014
Accepted 25 June 2014
Available online 1 July 2014
LDHA biochemical inhibition activity (best IC50 = 0.18 lM). Two crystal structures of optimized
Keywords:
compounds bound to human LDHA were obtained and explained many of the observed structure–activity
relationships. In addition, an optimized inhibitor exhibited good pharmacokinetic properties after oral
administration to rats (F = 45%).
Lactate dehydrogenase
X-ray crystal structure
Glycolysis
Ó 2014 Elsevier Ltd. All rights reserved.
Tumor metabolism
The altered metabolism exhibited by many tumors is currently
viewed as an attractive differentiator from normal tissues which
can be exploited for the development of novel anti-cancer thera-
peutics.1 One such example is the utilization of glucose in which
cancer cells metabolize this nutrient primarily via glycolysis as
opposed to the more energy-efficient but oxygen-dependent mito-
chondrial oxidative phosphorylation process.2 Somewhat surpris-
ingly, glycolytic glucose metabolism is known to occur in many
tumors even in the presence of normal oxygen levels.2 In contrast,
normal tissues typically employ glycolysis for energy production
only when oxygen supplies limit oxidative phosphorylation (e.g.,
in strenuously working muscle). This difference suggests that inhi-
bition of ‘aerobic glycolysis’ in cancer cells may preferentially
impede tumor growth and proliferation while having minimal
impacts on normal host tissues.3,4
(Fig. 1).5–7 The enzyme utilizes the reduced form of nicotinamide
adenine dinucleotide (NADH) as a co-factor to stereospecifically
transfer a hydride to the pyruvate ketone moiety. This transforma-
tion can also be accomplished by lactate dehydrogenase B (LDHB;
also known as LDH-H and LDH-1), an alternate LDH isoform,
although this enzyme preferentially catalyzes the reverse reaction
in which lactate is converted to pyruvate.5,7 As a HIF1
a and Myc
target gene, LDHA is induced by hypoxia or mutations in VHL,
FH, SDH, or the RAS/PI3K/AKT signaling pathways.8,9 In addition,
elevated LDHA levels are associated with many cancers and are
often predictive of poor survival in many indications.8 Encourag-
ingly, shRNA-mediated LDHA knock-down in glycolytic cancer cell
lines results in significant inhibition of tumor growth, and more
pronounced effects are observed under hypoxic conditions where
cells rely primarily on glycolytic energy production.9a,c Impor-
tantly, humans lacking LDHA through hereditary deficiency display
mild phenotypes suggesting that inhibition of the enzyme will not
lead to significant intolerable side-effects.10 Collectively, these data
implicate LDHA as an attractive target for the development of new
anti-cancer agents for use against hypoxic tumors and/or those
which display strong glycolytic phenotypes.
The final step in glycolysis is catalyzed by lactate dehydroge-
nase A (LDHA; also known as LDH-M and LDH-5), a homotetramer-
ic enzyme which converts pyruvate to lactate in the cell cytosol
⇑
Corresponding author. Tel.: +1 650 467 6854.
E-mail address: dragovich.peter@gene.com (P.S. Dragovich).
http://dx.doi.org/10.1016/j.bmcl.2014.06.076
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

P. S. Dragovich et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3764–3771
3765
O
O
-
-
NH2
O
O
O
H
H
O
H
HO
O
N
N
Pyruvate
Lactate
N
NH2
N
N
LDHA
LDHB
O
-
OH
OH
HO
HO
O
P
O
O
O
O
O
P
O
O
-
O
H
H
O
H
NADH
NH2
NH2
N
N
RO
RO
O
O
Oxamate
(inhibitor)
H2N
-
O
NADH
NAD
O
HO OH
HO
OH
Figure 1. LDH biochemistry.
O
Cl
CF3
CN
HN
O
H
N
S
N
HN
N
OH
N
OH
OH
H2N
O
S
Cl
O
H2N
N
O
O
111
212
313
HO
O
O
F
F
NH
N
O
S
O
N
H
N
H
N
H
O
OH
OH
S
N
S
O
O
N
H3CO
N
OCH3
O
414
515
F
Cl
OH OH
O
N
O
O
N
S
OH OH
OH
N
H
HO
OCH3
O
616
O
Figure 2. Structures of representative human LDHA inhibitors.
We11,12 and others13–17 have recently reported multiple
examples of human LDHA inhibitors in the scientific and patent lit-
erature (compounds 1–6, Fig. 2). All of these molecules contain
either a carboxylic acid moiety or some other acidic functional
group which can ionize at physiological pH (e.g., the pKa of com-
pound 1 was determined to be 3.9411). In the case of inhibitors 4
and 6, crystallographic studies determined that a carboxylic acid
moiety present in each compound interacted with the basic side-
chain of LDHA residue Arg-168 in a manner related to that
series demonstrated the importance of the ionizable group for
maintaining good LDHA inhibition activity.11,12 Thus, an acidic
moiety that can ionize at physiological pH is an important pharma-
cophore component for the majority of known LDHA inhibitors.
As part of our efforts to identify new LDHA inhibitors, we con-
ducted high-throughput screening of the Roche and Genentech
chemical archives using a fluorescence-based assay which moni-
tored the disappearance of the NADH co-factor during enzymatic
conversion of pyruvate to lactate.20 This screening effort identified
the 3-hydroxy-2-mercaptocyclohex-2-enone 7 as a moderately
potent LDHA inhibitor, and this activity was subsequently con-
firmed in lower-throughput biochemical assessments employing
both spectrophotometric and mass spectrometry detection meth-
observed
for
a
substrate-derived
inhibitor
(oxamate;
Fig. 1).14,16,18,19 Computational modeling suggested that similar
interactions occurred between compound 3 and the LDHA active
site.13 Although contacts between LDHA Arg-168 and the acidic
moieties present in inhibitors 1 and 2 were not observed crystallo-
graphically, structure–activity relationships developed within each
ods (IC50 = 1.7
the importance of acidic moieties in the pharmacophores of known
lM and 0.88 l
M, respectively; Table 1).20–22 Given

3766
P. S. Dragovich et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3764–3771
Table 1
hypothesis and indicated that such ionization under physiological
conditions was indeed possible.
Structure and biological properties of compound 7
OH
S
NO2
Accordingly, additional biochemical and biophysical experi-
ments were performed with 7 to both increase confidence that
the molecule was a bona fide LDHA inhibitor and to further charac-
terize its LDHA inhibition mechanism. Surface plasmon resonance
(SPR) assessments performed in the presence and absence of NADH
determined that the compound specifically associated with LDHA
and that measured KD values were (1) close to the biochemical
IC50 potency and (2) much stronger in the presence of the co-factor
(compare Fig. 3A and B, Table 1).22 Similar SPR results were
observed for several other hits that emerged from our LDHA
screening efforts.11,12 As with these previously characterized inhib-
itors, the SPR findings suggested that binding of 7 to LDHA required
prior association of the NADH co-factor.24 Compound 7 also inhib-
ited the closely related LDHB isoform with an IC50 value similar to
that observed for LDHA (Table 1), and the above characterization
data collectively provided assurance that 7 was a specific and gen-
uine anti-LDHA agent.
To facilitate medicinal chemistry efforts aimed at improving the
inhibition activity of 7, we obtained a crystal structure of the mole-
cule in complex with the LDHA enzyme.25 As shown in Figure 4a, the
compound bound in the LDHA active site in close proximity to the
NADH co-factor that was also present in the structure. This orienta-
tion was consistent with the SPR data described above which indi-
cated that the presence of NADH greatly improved the affinity of 7
for the LDHA protein. The enol moiety of 7 was expected to be depro-
tonated at the neutral pH employed to obtain the structure and this
moiety interacted closely with the (presumably) protonated LDHA
Arg-168 side chain (Figs. 4a and 4b).23,26 This residue is believed to
interact in a related manner with the pyruvate carboxylate moiety
during substrate catalysis by the enzyme.5 One of the nitro-group
oxygen atoms was also observed nearby (ꢀ3.0 Å) the Arg-168 side
chain. However, the orientation of the nitro functionality was not
consistent withthe formation of a robust hydrogen bond. The ketone
carbonyl moiety of 7 was noted crystallographically to form hydro-
gen bonds with the side chains of LDHA residues Asn-137 and
His-192. The protonated form of His-192 is believed to transfer a
proton to the pyruvate carbonyl moiety during substrate turnover.5
Interestingly, the sulfur atom present in 7 was noted in close spatial
proximity (ꢀ3.3 Å) to both the Arg-168 and the His-192 LDHA side
chains. However, the importance of this proximity and whether
the sulfur forms beneficial hydrogen bonds and/or polar interactions
with these residues is not known with certainty. Collectively, these
observations indicated that the ionized form of 7 can function as a
carboxylic acid mimic and thereby effectively interact with several
key catalytic residues in the LDHA active site.
O
7
Assay descriptiona
Result (lM)
LDHA IC50 (UV endpoint)
LDHA IC50 (MS endpoint)
LDHA KD (SPR, +NADH)
LDHA KD (SPR, ꢁNADH)
LDHB IC50 (UV endpoint)
1.7
0.88
3.5
771
12.3
a
See Supplementary data for experimental details associated with each
assessment.
OH
NO2
O
NO2
H
S
S
O
O
7
7a
-H
+H
-H
+H
-
O
NO2
O
NO2
S
S
-
O
O
7b
7c
Scheme 1. Structure of compound 7 along with possible tautomeric and ionized
forms.
LDHA inhibitors, we were intrigued by the possibility that com-
pound 7 might ionize at physiological pH and thereby function as
a carboxylic acid mimic. Specifically, we envisioned deprotonation
of the hydroxy-cyclohexenone present in 7, or the cyclic 1,3-dicar-
bonyl moiety contained in the 7a tautomer, thereby affording the
corresponding ionized forms (structures 7b and 7c; Scheme 1).23
The measured pKa of compound 7 (2.63) was consistent with this
Figure 3. SPR data depicting the binding of compound 7 to LDHA in the presence (A) and absence (B) of NADH. The KD determined from fitting the kinetics is reported in panel
A. In panel B, the KD was estimated relative to a control using methods described in Supplementary data. In both panels, the top concentration is 25 lM with injections related
by a twofold dilution series. See Supplementary data for additional experimental details.

P. S. Dragovich et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3764–3771
3767
Table 2
Structure–activity relationships of 3-hydroxy-2-mercaptocyclohex-2-enone-contain-
ing compounds
R2
R1
OH
S
R3
R5
R4
O
R1
R2
R3
R4
R5
LDHA IC50 (lM)
a
Compd
7
8
9
NO2
H
F
Cl
Br
CH3
CF3
CH2CH3
CH2OH
OH
OCH3
CN
CO2H
CO2CH3
CO2NH2
SO2CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F
F
Cl
Cl
Cl
Cl
Cl
CH3
CH3
CH3
OCH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
CH3
H
H
H
Cl
H
H
H
H
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
CH3
H
H
H
CH3
H
1.7
>100
53
4.6
53
>100
>66
>100
36
>100
>100
1.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4a. Co-crystal structure of inhibitor 7 (cyan) in complex with LDHA (grey).
NADH is also present and is depicted in dark blue. Hydrogen bonds are depicted as
dashed yellow lines. The resolution of the structure is 2.15 Å.
ARG168
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
63
HN
ALA237
-
O
O
H2N
NH2
VAL234
ASP165
GLY193
N
F
Cl
-
O
S
Br
CH3
CF3
OCH3
OCF3
NH2
CO2NH2
SO2NH2
H
H
H
H
H
H
H
H
H
H
H
H
Cl
OCH3
CH3
Cl
CH3
H
Cl
Cl
H
H
H
H
H
H
H
H
ALA237
TYR238
O
LEU164
HN
-
NH
O
NH2
O
O
ILE241
ASP165
HIS192
Cl
Br
CH3
ASN137
Figure 4b. Depiction of key protein–ligand interactions observed in 7-LDHA co-
crystal structure. Hydrogen bonds are indicated as dashed lines (red = questionable/
unknown significance). The depicted protonation states of 7 and various LDHA
residues are inferred. Alternate tautomers and/or resonance forms of 7 and the
LDHA residues are not considered.
CH2CH3
CF3
OH
OCH3
CO2H
SCH3
NH2
NHAc
Cl
OCH3
CH3
H
H
F
H
H
H
H
H
F
CH3
H
H
Further analysis of the 7-LDHA co-crystal structure revealed
that the 2-nitro-phenyl moiety of 7 occupied a narrow cleft located
between the side chains of the Arg-168 and His-192 residues
(Figs. 4a and 4b). Binding of the 2-nitro-phenyl group in this cleft
was facilitated by subtle movements in the side chains and/or
backbone atoms of several of the surrounding LDHA residues rela-
tive to their positions noted in other published crystal structures
(e.g., Asp-165, Arg-168, and Ala-237).11,12 The 2-nitro-phenyl moi-
ety made favorable van der Waals contacts with many of these sur-
rounding residues including Leu-164, Asp-165, Arg-168, Gly-193,
Val-234, and Ala-237. Somewhat surprisingly, the other phenyl
substituent present in 7 was observed crystallographically in an
axial location relative to the 3-hydroxy-cyclohexenone ring. This
orientation resulted in a close edge-to-face interaction between
the axial phenyl group and the 2-nitro-phenyl moiety of the ligand.
The axial location also allowed the phenyl group to contact the
hydrophobic side chains of LDHA residues Ala-237, Tyr-238, and
Ile-241. These favorable protein–ligand and intra-ligand interac-
tions likely compensate for the high-energy geometry adopted by
7 when bound to LDHA relative to the corresponding unbound
conformation.27
12
>100
>100
24
>100
>64
>100
>100
H
OCH3
See Supplementary data for experimental details associated with each assessment.
Biochemical results are reported as the arithmetic mean of at least 2 separate runs.
a
LDHA biochemical inhibition.
replacement of the 2-nitro group with a halogen was only tolerated
when Cl was employed as the 2-substituent (compare 7 with
9–11). Alkyl moieties were not suitable 2-nitro replacements
(12–14), and these included the methyl group which had the
potential to function as a Cl isostere. Analogs of 7 bearing mono-
substituted thio-phenyl rings which incorporated a variety of more
polar 2-substituents were also explored (15–22), but only the
With the above structural information available, we initiated
medicinal chemistry activities aimed at improving the LDHA inhi-
bition activity of compound 7. As shown in Table 2, removal of the
2-nitro group from the thio-phenyl ring present in 7 resulted in
significant loss of anti-LDHA potency (compare 8 with 7). Similarly,

3768
P. S. Dragovich et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3764–3771
Table 3
nitrile 18 retained meaningful LDHA inhibition activity. A large
number of molecules containing other mono-substituted thio-
phenyl rings which lacked a 2-substituent were also examined
(23–49). Unfortunately, all of these compounds exhibited signifi-
cant worsening of anti-LDHA activity relative to the original lead
7. These structure-activity relationships illustrated the importance
of a 2-substituent on the thio-phenyl ring for obtaining good LDHA
inhibition activity.
Structure–activity relationships of 3-hydroxy-2-mercaptocyclohex-2-enone-contain-
ing compounds
OH
X
S
R1
R2
R3
O
R4
Accordingly, we also explored molecules which preserved a
2-substituent on the thio-phenyl ring but also included one
additional moiety (50–60; Table 2). Only two of these compounds
displayed encouraging levels of LDHA inhibition potency (53 and
56), and both incorporated the 2-Cl-substituent identified in our
initial SAR efforts (cf., compound 10). A large number of molecules
were also assessed which replaced the 2-nitro-phenyl moiety pres-
ent in lead 7 with other aliphatic (61–67), benzylic (68–75), aro-
matic (76–77), or heteroaromatic groups (78–99; Supplementary
data, Table S1). However, all of these compounds displayed poor
LDHA inhibition properties. Collectively, the structure activity rela-
tionships depicted in Tables 2 and S1 indicate that the region of
LDHA occupied by the 2-nitro-phenyl moiety of 7 is relatively
intolerant to both subtle changes in the phenyl substitution pat-
tern as well as larger modifications which seek to alter the orienta-
tion of the contacting protein residues. Importantly, this initial SAR
exploration also identified the Cl and CN groups as suitable
replacements for the 2-nitro substituent present in lead 7 that
could subsequently be incorporated into new inhibitor designs.
Having thoroughly explored modification of the 2-nitro-phenyl
moiety contained in 7, we turned our attention to variation of the
other (axially situated) phenyl group present in the inhibitor struc-
ture. As shown in Table 3, this exploration was conducted using Cl
and CN groups in place of the 2-nitro-substituent since use of the
two former moieties facilitated compound synthesis and was per-
ceived to impart more desirable bio-pharmaceutical properties to
the resulting compounds. Systematic introduction of ortho, meta,
and para-Cl groups on the phenyl moiety relative to the
3-hydroxy-cyclohexenone ring afforded minimal changes in LDHA
inhibition activity with the para modification being the most det-
rimental (compounds 100–102). Incorporating a larger para substi-
tuent (103) further worsened anti-LDHA properties, consistent
with the close contact of the axial phenyl group with Tyr-238.
Interestingly, incorporation of a 2,6-di-Cl phenyl moiety into the
inhibitor design resulted in a significant improvement in LDHA
inhibitor potency (104). Several other compounds which contained
various 2,6-disubstituted phenyl groups also exhibited improved
anti-LDHA properties (106, 109, 111, 117 and 120; Table 3)
although not every 2,6-combination afforded such enhancements
(108, 110, 112 and 119, Table 3). A structural explanation for the
potency improvement imparted by the various 2,6-disubstituted
phenyl moieties is provided below. As we previously noted for
other classes of LDHA inhibitors,11,12 the SPR KD’s measured for
selected 3-hydroxy-cyclohexenones described in this work closely
paralleled the corresponding biochemical IC50 values (Table 3).
Similarly, LDHB inhibition activities were determined for the more
active compounds and were ꢀ5–10 fold less potent than the corre-
sponding LDHA potencies (Table 3).
Compd
X
R1
R2 R3
R4
LDHA
IC50
LDHA
LDHB
M) IC50
a
b
c
KD
(l
(
lM)
(lM)
10
100
101
102
103
104
105
106
107
108
109
110
111
112
18
113
114
115
116
117
118
119
120
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
CN
CN Cl
CN
CN
CN Cl
CN Cl
CN Cl
CN CH3
CN Br
H
Cl
H
H
H
Cl
Cl
Cl
Cl
F
Br
CH3
Cl
CF3
H
H
H
Cl
H
H
H
H
H
H
H
H
H
Br
H
H
H
Cl
H
H
H
H
H
H
H
H
H
Cl
OCH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
F
Br
CF3
F
4.6
5.6
5.6
18
4.3
8.2
7.1
27
13
1.8
ND
0.49
0.50
ND
ND
ND
0.37
ND
1.5
ND
ND
19
11
13
27
37
6.9
24
5.3
3.4
50
1.3
15
1.9
>10
7.4
6.6
>10
>10
4.9
2.0
2.8
7.8
0.86
19
0.87
3.7
0.58
1.0
7.4
0.27
Br
CH3 2.3
Cl
CF3
H
H
H
0.50
15
1.6
3.1
6.8
17
0.83
0.45
0.64
H
H
H
Cl
H
H
H
H
ND
Cl
Br
CF3
CH3 2.1
Br 0.18
0.43
0.24
ND
ND
ND
H
H
See Supplementary data for experimental details associated with each assessment.
LDHA biochemical results are reported as the arithmetic mean of at least 2 separate
runs. LDHB and SPR results are n = 1. ND = not determined.
a
LDHA biochemical inhibition.
LDHA dissociation constant as determined by surface plasmon resonance.
LDHB biochemical inhibition.
b
c
Table 4
In vitro properties for representative 3-hydroxy-2-mercaptocyclohex-2-enone-con-
taining compounds
Compd cLogP RLM
HLM
Clb
Rat
Human
PPBd
(fu)
MDCK
(AB)e
x10-6
cm/s
Solubilityf
M)
Cla
PPBc
(fu)
(l
(mL/
min/
kg)
(mL/
min/
kg)
10
18
4.4
3.5
5.7
5.9
6.2
4.8
4.8
14
5.2
27
37
38
24
31
3.9
1.6
14
17
18
12
15
0.002 0.001
0.01 0.001
0.002 0.005
18
4.1
17
ND
ND
ND
14
147
118
128
117
143
117
75
104
107
112
116
117
ND
ND
ND
ND
ND
ND
0.005 0.002
See Supplementary data for experimental details associated with each assessment.
All assay results are reported as the arithmetic mean of at least 2 separate runs
(exception: solubility determinations are n = 1). ND = not determined.
Rat hepatic clearance predicted from rat liver microsomes (stable, moderate,
labile = <17, 17–39, >39 mL/min/kg, respectively).
a
b
Human hepatic clearance predicted from human liver microsomes (stable,
moderate, labile = <6, 6–15, >15 mL/min/kg, respectively).
As described above, 2,6-disubstitution of the 5-phenyl moiety
present in the 3-hydroxy-cyclohexenone-containing inhibitors
afforded significant improvements in LDHA inhibition properties.
We therefore obtained a crystal structure of one such molecule
(104) in complex with LDHA to better understand the nature of
these improvements.25 As shown in Figure 5, the 3-hydroxy-
cyclohexenone ring of 104 bound to and interacted with LDHA in
a manner that was nearly identical to that observed for compound
7 (c.f., Figs. 4a and 4b). However, the 2,6-dichlorophenyl moiety
of 104 was situated in an equatorial location relative to the
c
Rat plasma protein binding; fraction unbound (fu).
Human plasma protein binding; fraction unbound (fu).
Apparent permeability coefficients (Papp,A–B) across MDCK cell monolayers.
Aqueous solubility.
d
e
f
3-hydroxy-cyclohexenone ring in contrast to the axial orientation
previously observed for the unsubstituted phenyl group present
in compound 7. In addition, the plane of the 2,6-disubstituted phe-
nyl ring was rotated 90° relative to the cyclohexenone ring. This

P. S. Dragovich et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3764–3771
3769
tolerant of most compound alterations. The LDHA-104 crystal struc-
ture illustrates that chemically similar inhibitors may adopt signif-
icantly different binding modes when complexed with the protein
and therefore suggested that it would be prudent to continue to
use crystallography to guide future compound optimization.
A subset of the newly-identified compounds was profiled in var-
ious in vitro and in vivo assessments to better characterize the
DMPK properties of this novel inhibitor series. As shown in Table 4,
the tested compounds exhibited stabilities toward rat and human
liver microsomes which generally correlated with their lipophilici-
ties (lower lipophilicity = more stable). All tested molecules also
displayed moderate or high permeability properties in MDCK
assessments with the most polar of the compounds (18) providing
the lowest value. The aqueous solubilities measured for the com-
pounds in Table 4 were all favorable, and these characteristics were
likely imparted by the acidic moieties contained within the mole-
cules. Similarly, all tested compounds exhibited very high protein
binding in plasma from humans and rats. In addition, all of the com-
pounds in Table 4 did not display meaningful CYP P450 inhibition
Figure 5. Co-crystal structure of inhibitor 104 (cyan) in complex with LDHA (grey).
NADH is also present and is depicted in dark blue. Hydrogen bonds are depicted as
dashed yellow lines. The resolution of the structure is 2.00 Å.
activity (IC50 >5 lM 3A4, 2D6, 2C9, 1A2; data not shown). As
depicted in Figure 6 and Table 5, compound 104 exhibited good
exposures and bioavailability in rats following oral administration
with plasma levels persisting for extended time periods. The latter
result was likely due to the compound’s high plasma protein bind-
ing properties which also afforded much lower free concentrations
of the molecule. The PK data provide an initial assessment of how
the in vitro DMPK results depicted in Table 4 may translate to
in vivo outcomes and also highlight areas for possible subsequent
compound improvement (e.g., increase free exposures).
The described inhibitors were synthesized as depicted in
Scheme 2.28 Condensation of acetone with benzaldehyde or various
substituted analogs afforded olefins of general structure 121.29 The
two-step conversion of these intermediates to the corresponding 3-
hydroxy-cyclohexenones 122 was subsequently accomplished
using literature methods.30 Reaction of these cyclohexenones with
N-bromosuccinimide afforded the bromides 123, and these entities
were exposed to various thiols under slightly basic conditions to
provide the desired products (general structure 124).
Figure 6. Pharmacokinetic profile of 104 in rats following oral and iv administra-
tion. Error bars depict standard deviations (n = 3).
orientation (1) avoided steric clashes between the 2,6-dichloro-
substituents and the cyclohexenone ring, (2) afforded good van
der Waals contacts between the phenyl ring and the side chain
of LDHA residue Ile-241, and (3) positioned one of the Cl atoms
in proximity to the NADH co-factor which was also observed in
the co-crystal structure. The other Cl substituent contacted the side
chain of LDHA residue Tyr-238, which moved closer to inhibitor
104 relative to its location in the LDHA-7 crystal structure.
Interestingly, although inhibitor 104 contained a 2-Cl-thiophenyl
moiety in place of the 2-nitro-thiophenyl group present in 7, the
LDHA residues surrounding this portion of both inhibitors were
crystallographically observed in virtually identical locations. These
results were consistent with the structure–activity relationships
described above which suggested that the cleft occupied by the
2-nitro/Cl-phenyl moiety was not conformationally flexible or
In summary, a novel class of 3-hydroxy-2-mercaptocyclohex-2-
enone-containing LDHA inhibitors was identified using a high
throughput screening approach. The crystal structure of a repre-
sentative compound in complex with the protein indicated that
these inhibitors bound in the LDHA active site in a manner that
mimicked the pyruvate substrate. Structural modification of the
initial high throughput screening hit afforded >10-fold improve-
ment in LDHA biochemical potency. An optimized molecule (104)
was not active against the structurally-related31 malate dehydro-
genase 1 and malate dehydrogenase 2 enzymes (IC50 >100 lM)
suggesting that the 3-hydroxy-2-mercaptocyclohex-2-enone
inhibitor class would not indiscriminately interact with other
dehydrogenase enzymes. The ability of some of the more potent
compounds (104, 107, and 117) to inhibit the production of lactate
in the HCC1954 cell line was also assessed.22 Unfortunately, none
Table 5
Pharmacokinetic parameters for compound 104 following iv and po administration to rats (n = 3)
c
d
e
Route
Dosea (mg/kg)
AUCb
(
lM h)
Clplasma (mL/min/kg)
Vss (L)
T1/2 (h)
Ff (%) (
lm)
iv
po
0.5
2.0
60.3
142
0.35
NA
0.39
NA
13.2
10.7
NA
45
See Supplementary data for experimental descriptions.
a
Formulations: IV = 10:50:40 EtOH/PEG400/50 mM citrate pH = 3 (solution); PO = MCT (suspension).
b
c
d
e
f
Area under pharmacokinetic curve (0–24 h).
Clearance.
Volume of distribution (steady-state).
Half-life.
Bioavailability. NA = not applicable.

3770
P. S. Dragovich et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3764–3771
O
OH
O
CH3
H
i
ii, iii
H
O
R
R
R
121
122
iv
OH
OH
S
Br
R'
v
O
O
R
R
124
123
Scheme 2. Synthesis of 3-hydroxy-2-mercaptocyclohex-2-enone-containing compounds. Reagents and conditions: (i) acetone, 1% aq NaOH, 65 °C, 2 h; (ii) CH3ONa, diethyl
malonate, CH3OH, reflux, 4 h; (iii) 6 N aq NaOH, THF, 80 °C, 1 h, then 12 M HCl, reflux, 1 h; (iv) NBS, CH3CN, 25 °C, 2 h; (v) R0SH, piperidine, CH2Cl2, 0–25 °C, 12 h.
7. LDHA and LDHB are each homotetramers comprised of M and H subunits,
of the molecules displayed activity in this assay when tested up to
the 50 lM concentration level. The reasons for this lack of cell-
respectively. LDH heterotetramers containing both M and H subunits are also
known. For more information, see Ref. 5.
based activity are currently not known with certainty but may
include: (1) insufficient biochemical potency, and/or (2) high
protein binding. Additional efforts to improve the biochemical
potency of the described inhibitors and to optimize their ability
to inhibit LDHA in cell-culture experiments will be reported in
due course.
8. (a) Kolev, Y.; Uetake, H.; Takagi, Y.; Sugihara, K. Ann. Surg. Oncol. 2008, 15,
2336; (b) Koukourakis, M. I.; Giatromanolaki, A.; Sivridis, E.; Gatter, K. C.;
Harris, A. L. J. Clin. Oncol. 2006, 24, 4301; (c) Koukourakis, M. I.; Giatromanolaki,
A.; Sivridis, E.; Bougioukas, G.; Didilis, V.; Gatter, K. C.; Harris, A. L. Br. J. Cancer
2003, 89, 877.
9. (a) Seth, P.; Grant, A.; Tang, J.; Vinogradov, E.; Wang, X.; Lenkinski, R.;
Sukhatme, V. P. Neoplasia 2011, 13, 60; (b) Qing, G.; Skuli, N.; Mayes, P. A.;
Pawel, B.; Martinez, D.; Maris, J. M.; Simon, M. C. Cancer Res. 2010, 70, 10351;
(c) Fantin, V. R.; St.-Pierre, J.; Leder, P. Cancer Cell 2006, 9, 425.
Acknowledgments
10. (a) Kanno, T.; Sudo, K.; Maekawa, M.; Nishimura, Y.; Ukita, M.; Fukutake, K.
Clin. Chim. Acta 1988, 173, 89; (b) Kanno, T.; Sudo, K.; Takeuchi, I.; Kanda, S.;
Honda, N.; Nishimura, Y.; Oyama, K. Clin. Chim. Acta 1980, 108, 267.
11. Dragovich, P. S.; Fauber, B.; Corson, L. B.; Ding, C. Z.; Eigenbrot, C.; Ge, H.;
Giannetti, A. M.; Hunsaker, T.; Labadie, S.; Liu, Y.; Malek, S.; Pan, B.; Peterson,
D.; Pitts, K.; Purkey, H. E.; Sideris, S.; Ultsch, M.; VanderPorten, E.; Wei, B.; Xu,
Q.; Yen, I.; Yue, Q.; Zhang, H.; Zhang, X. Bioorg. Med. Chem. Lett. 2013, 23, 3186.
12. Fauber, B. P.; Dragovich, P. S.; Chen, J.; Corson, L. B.; Ding, C. Z.; Eigenbrot, C.;
Giannetti, A. M.; Hunsaker, T.; Labadie, S.; Liu, Y.; Liu, Y. c.; Malek, S.; Peterson,
D.; Pitts, K.; Sideris, S.; Ultsch, M.; Vander Porten, E.; Wang, J.; Wei, B. Q.; Yen,
I.; Yue, Q. Bioorg. Med. Chem. Lett. 2013, 23, 5533.
13. Granchi, C.; Roy, S.; Giacomelli, C.; Macchia, M.; Tuccinardi, T.; Martinelli, A.;
Lanza, M.; Betti, L.; Giannaccini, G.; Lucacchini, A.; Funel, N.; León, L. G.;
Giovannetti, E.; Peters, G. J.; Palchaudhuri, R.; Calvaresi, E. C.; Hergenrother, P.
J.; Minutolo, F. J. Med. Chem. 2011, 54, 1599.
14. Ward, R. A.; Brassington, C.; Breeze, A. L.; Caputo, A.; Critchlow, S.; Davies, G.;
Goodwin, L.; Hassall, G.; Greenwood, R.; Holdgate, G. A.; Mrosek, M.; Norman,
R. A.; Pearson, S.; Tart, J.; Tucker, J. A.; Vogtherr, M.; Whittaker, D.; Wingfield, J.;
Winter, J.; Hudson, K. J. Med. Chem. 2012, 55, 3285.
We thank Drs. Krista Bowman and Jiansheng Wu and their
respective Genentech research groups for protein expression and
purification activities. We also thank Dr. Peter Jackson for many
helpful discussion regarding LDHA. In addition, we thank Crystallo-
graphic Consulting, LLC for diffraction data collection. We
acknowledge use of synchrotron X-ray sources at the Advanced
Light Source and the Stanford Synchrotron Radiation Lightsource
supported by the Department of Energy’s Office of Science under
contracts
DE-AC02-05CH11231
and
DE-AC02-76SF00515,
respectively.
Supplementary data
15. (a) Billiard, J.; Dennison, J. B.; Briand, J.; Annan, R. S.; Chai, D.; Colón, M.;
Dodson, C. S.; Gilbert, S. A.; Greshock, J.; Jing, J.; Lu, H.; McSurdy-Freed, J. E.;
Orband-Miller, L. A.; Mills, G. B.; Quinn, C. J.; Schneck, J. L.; Scott, G. F.; Shaw, A.
N.; Waitt, G. M.; Wooster, R. F.; Duffy, K. J. Cancer Metab. 2013, 1, 19. http://
dx.doi.org/10.1186/2049-3002-1-19; (b) Xie, H.; Hanai, J.-i.; Ren, J.-G.; Kats, L.;
Burgess, K.; Bhargava, P.; Signoretti, S.; Billard, J.; Duffy, K. J.; Grant, A.; Wang,
X.; Lorkiewicz, P. K.; Schatzman, S.; Bousamra, M.; Lane, A. N.; Higashi, R. M.;
Fan, T. W. M.; Pandolfi, P. P.; Sukhatme, V. P.; Seth, P. Cell Metab. 2014. http://
dx.doi.org/10.1016/j.cmet.2014.03.003; (c) Brown, K. K.; Chai, D.; Dodson, C. S.;
Duffy, K. J.; Shaw, A. N. WO 2013/096151.; (d) Brown, K. K.; Chai, D.; Dodson, C.
S.; Duffy, K. J.; Shaw, A. N. WO 2013/096153.; (e) Chai, D.; Colon, M.; Dodson,
C.; Duffy, K. J.; Shaw, A. N. WO 2012/061557.
16. Kohlmann, A.; Zech, S. G.; Li, F.; Zhou, T.; Squillace, R. M.; Commodore, L.;
Greenfield, M. T.; Lu, X.; Miller, D. P.; Huang, W.-S.; Qi, J.; Thomas, R. M.; Wang,
Y.; Zhang, S.; Dodd, R.; Liu, S.; Xu, R.; Xu, Y.; Miret, J. J.; Rivera, V.; Clackson, T.;
Shakespeare, W. C.; Zhu, X.; Dalgarno, D. C. J. Med. Chem. 2013, 56, 1023.
17. Inhibitors of Plasmodium falciparum LDH with potential for use as anti-
malarials have also been described: (a) Choi, S.-r.; Pradham, A.; Hammond, N.
L.; Chittiboyina, A. G.; Tekwani, B. L.; Avery, M. A. J. Med. Chem. 2007, 50, 3841;
(b) Choi, S.-r.; Beeler, A. B.; Pradham, A.; Watkins, E. B.; Rimoldi, J. M.; Tekwani,
B.; Avery, M. A. J. Comb. Chem. 2007, 9, 292; (c) Cameron, A.; Read, J.; Tranter,
R.; Winter, V. J.; Sessions, R. B.; Brady, R. L.; Vivas, L.; Easton, A.; Kendrick, H.;
Croft, S. L.; Barros, D.; Lavandera, J. L.; Martin, J. J.; Risco, F.; García-Ochoa, S.;
Gamo, F. J.; Sanz, L.; Leon, L.; Ruiz, J. R.; Gabarró, R.; Mallo, A.; Gómez de la
Heras, F. J. Biol. Chem. 2004, 279, 31429.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2014.06.
076.
References and notes
1. (a) Galluzzi, L.; Kepp, O.; Vander Heiden, M. G.; Kroemer, G. Nat. Rev. Drug Disc.
2013, 12, 829; (b) Ward, P. S.; Thompson, C. B. Cancer Cell 2012, 21, 297; (c)
Vander Heiden, M. G. Nat. Rev. Drug Disc. 2011, 10, 671; (d) Zhao, Y.; Liu, H.;
Riker, A. I.; Fodstad, O.; Ledoux, S. P.; Wilson, G. L.; Tan, M. Front. Biosci. 2011,
16, 1844; (e) Kaelin, W. G., Jr.; Thompson, C. B. Nature 2010, 465, 562; (f)
Tennant, D. A.; Durán, R. V.; Gottlieb, E. Nat. Rev. Cancer 2010, 10, 267.
2. (a) Vander Heiden, M. G.; Cantley, L. C.; Thompson, C. B. Science 2009, 324,
1029; (b) Hsu, P. P.; Sabatini, D. M. Cell 2008, 134, 703.
3. (a) Warburg, O. Science 1956, 123, 309; (b) Bensinger, S. J.; Christofk, H. R.
Semin. Cell Dev. Biol. 2012, 23, 352; (c) Koppenol, W. H.; Bounds, P. L.; Dang, C.
V. Nat. Rev. Cancer 2011, 11, 325.
4. (a) Hamanaka, R. B.; Chandel, N. S. J. Exp. Med. 2012, 209, 211; (b) Jones, N. P.;
Schulze, A. Drug Discovery Today 2011, 17, 232; (c) Pelicano, H.; Martin, D. S.;
Xu, R.-H.; Huang, P. Oncogene 2006, 25, 4633.
5. Granchi, C.; Bertini, S.; Macchia, M.; Minutolo, F. Curr. Med. Chem. 2010, 17, 672.
6. Salaway, J. G. Metabolism at a Glance, 3rd ed.; Blackwell: Malden, 2004. pp 10–
25.

P. S. Dragovich et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3764–3771
3771
18. The LDH residue numbering scheme used herein follows the convention of PDB
entries 1I10 and 4AJP which does not account for the initiating methionine in
the gene sequence reflected in UniProt entry P00338. This numbering scheme
differs by one residue from that described elsewhere: Chung, F. Z.; Tsujibo, H.;
Bhattacharyya, U.; Sharief, F. S.; Li, S. S. Biochem. J. 1985, 231, 537.
19. For the crystal structure of oxamate in complex with LDHA, see PDB entry 1I10
Read, J. A.; Winter, V. J.; Eszes, C. M.; Sessions, R. B.; Brady, R. L. Proteins 2001,
43, 175.
25. See Supplementary data for experimental details associated with the described
co-crystal structure. All crystallographic descriptions are based on analysis of
the ligand chain ‘W’/protein chain ‘D’ LDHA protomer. Two other LDHA
protomers containing bound ligand were also observed in the tetramer
asymmetric unit. The protein–ligand interactions noted in these other
protomers were similar to those described for the ligand chain ‘W’/protein
chain ‘D’ complex, although some subtle differences were apparent. See
deposited pdb files for additional structural details: compound 7 = 4QO7;
compound 104 = 4QO8.
20. VanderPorten, E.; Frick, L.; Turincio, R.; Thana, P.; LaMarr, W.; Liu, Y. Anal.
Biochem. 2013, 441, 115.
26. The described protein–ligand interactions can also be depicted using alternate
tautomers and/or resonance forms of 7 and the corresponding LDHA side
chains.
21. For the use of related dicarbonyl-containing compounds as inhibitors of HIV
protease and HCV NS5B polymerase, see: (a) Li, H.; Tatlock, J.; Linton, A.;
Gonzalez, J.; Borchardt, A.; Dragovich, P.; Jewell, T.; Zhou, R.; Blazel, J.; Parge,
H.; Love, R.; Hickey, M.; Doan, C.; Shi, S.; Duggal, R.; Lewis, C.; Fuhrman, S.
Bioorg. Med. Chem. Lett. 2006, 16, 4834; (b) Vara Prasad, J. V. N.; Boyer, F. E.;
Domagala, J. M.; Ellsworth, E. L.; Gajda, C.; Hamilton, H. W.; Hagen, S. E.;
Markoski, L. J.; Steinbaugh, B. A.; Tait, B. D.; Humblet, C.; Lunney, E. A.;
Pavlovsky, A.; Rubin, J. R.; Ferguson, D.; Graham, N.; Holler, T.; Hupe, D.;
Nouhan, C.; Tummino, P. J.; Urumov, A.; Zeikus, E.; Zeikus, G.; Gracheck, S. J.;
Saunders, J. M.; VanderRoest, S.; Brodfuehrer, J.; Iyer, K.; Sinz, M.; Gulnik, S. V.;
Erickson, J. W. Bioorg. Med. Chem. 1999, 7, 2775.
27. Quantum mechanical calculations were used to predict the preferred
conformation of 5-phenyl of compound 7 in an unbound state [DFT-B3LYP
with 6-31G⁄⁄ basis set either in gas phase or with a Poisson–Boltzmann
continuum solvation model (PBF) as implemented in Jaguar (Schrödinger)]. In
vacuum, the phenyl group of 7 was found to prefer an equatorial location over
the axial alternative by 1.3 kcal/mol in conformational energy. This equatorial
preference remained the same when solvation energy was taken into account,
although the calculated energy difference was reduced to 0.6 kcal/mol.
28. Compound 7 can be purchased from eMolecules.
22. See Supplementary data for details regarding enzyme, SPR, and cell culture
assay methods, associated errors, and controls.
29. The condensation with benzaldehyde itself was carried out at 0 °C instead of at
elevated temperature.
23. No attempts were made to identify which, if any, compound tautomers
predominated in organic and/or aqueous solutions. All new molecules described
in this work are arbitrarily drawn and described as 3-hydroxy-cyclohexenones.
24. The LDHA inhibition mechanism of 7 may mimic substrate catalysis in which
binding of NADH to the protein precedes binding of pyruvate. See Ref. 5 for
additional details.
30. Tamura, Y.; Yoshimoto, Y.; Kunimoto, K.; Tada, S.; Tomita, T.; Wada, T.; Seto, E.;
Murayama, M.; Shibata, Y.; Nomura, A.; Ohata, K. J. Med. Chem. 1977, 20, 709.
31. Human LDHA and human MDH-2 protein sequences share 20% similarity. X-
ray structures of their protomers can be superimposed using 173 pairs of C
atoms with an rms deviation of 1.7 Å. C atoms within 5 Å of bound NADH can
be superimposed with an rms deviation of 0.7 Å.
a
a