Fragment growing and linking lead to novel nanomolar lactate dehydrogenase inhibitors

Article

pubs.acs.org/jmc

Fragment Growing and Linking Lead to Novel Nanomolar Lactate

Dehydrogenase Inhibitors

Anna Kohlmann,* Stephan G. Zech, Feng Li, Tianjun Zhou, Rachel M. Squillace, Lois Commodore,

Matthew T. Greenﬁeld, Xiaohui Lu,^†David P. Miller, Wei-Sheng Huang, Jiwei Qi, R. Mathew Thomas,

Yihan Wang, Sen Zhang, Rory Dodd, Shuangying Liu, Rongsong Xu, Yongjin Xu, Juan J. Miret,^‡

Victor Rivera, Tim Clackson, William C. Shakespeare, Xiaotian Zhu, and David C. Dalgarno

ARIAD Pharmaceuticals, Inc., 26 Landsdowne Street, Cambridge, Massachusetts 02139, United States

S

* Supporting Information

ABSTRACT: Lactate dehydrogenase A (LDH-A) catalyzes the inter-

conversion of lactate and pyruvate in the glycolysis pathway. Cancer

cells rely heavily on glycolysis instead of oxidative phosphorylation

to generate ATP, a phenomenon known as the Warburg eﬀect.

The inhibition of LDH-A by small molecules is therefore of interest

for potential cancer treatments. We describe the identiﬁcation and

optimization of LDH-A inhibitors by fragment-based drug discovery.

We applied ligand based NMR screening to identify low aﬃnity frag-

ments binding to LDH-A. The dissociation constants (K_d) and enzyme

inhibition (IC₅₀) of fragment hits were measured by surface plasmon

resonance (SPR) and enzyme assays, respectively. The binding modes

of selected fragments were investigated by X-ray crystallography.

Fragment growing and linking, followed by chemical optimization,

resulted in nanomolar LDH-A inhibitors that demonstrated stoichiometric binding to LDH-A. Selected molecules inhibited lactate

production in cells, suggesting target-speciﬁc inhibition in cancer cell lines.

Several lines of evidence suggest that LDH-A plays a critical

role in cancer cell proliferation. Up-regulation of LDH-A mRNA

and protein expression have been reported across numerous

cancer types.¹⁰⁻¹⁵Regulation of the LDH-A gene and other

genes involved in promoting glycolysis is under control of

potentially oncogenic transcription factors such as c-Myc and

HIF-1α.¹⁶⁻¹⁹Aberrant activation of these oncogenes results in

elevated uptake of glucose and increased production of lactate.

Reduction of LDH-A by expression of targeted siRNA in cancer

lines produces inhibition of cell growth under hypoxic conditions

and dramatically reduces tumor formation in mouse models.^14,15

Knockdown of LDH-A also increases oxygen consumption, stim-

ulates mitochondrial respiration, and induces oxidative stress

while adding back LDH-A protein reverses these phenotypes.^20,21

Finally, a potential safety concern of inhibiting LDH-A for cancer

treatment is addressed by the observation that individuals who

naturally lack LDH-A do not show any symptoms under normal

conditions except some degree of muscle rigidity and

myoglobinuria after intense exercise.²²Therefore, LDH-A is a

potentially important and relatively safe metabolic target for

inhibition of the highly activated glycolysis pathway in cancer cells.

LDH-A is a highly conserved protein that contains approx-

imately 330 amino acids in a bilobal structure.⁹Although the

INTRODUCTION

■

Otto Warburg’s historic discovery showing that cancer cells pref-

erentially derive ATP through glycolysis rather than oxidative

phosphorylation provided the ﬁrst tangible connection between

energy metabolism and cancer.¹Despite the low eﬃciency of

generating ATP from glucose by utilizing glycolysis rather than

oxidative phosphorylation, the increased rate of glycolysis in

cancer cells is believed to ensure rapid generation of energy and

biomass needed for cellular proliferation.²⁻⁴Dependence on

glycolysis may also provide an advantage for cancer cells to

survive and grow under hypoxic conditions that characterize the

environment of solid tumor. Altered cancer cell energy metab-

olism has been categorized as an emerging hallmark of cancer,^5,6

and thus, inhibition of metabolic processes associated with cancer

cell growth constitutes a promising approach in the search for new

cancer therapies.

One molecular target in the glycolytic pathway is the enzyme

lactate dehydrogenase (LDH) which catalyzes the interconver-

sion of lactate and pyruvate as the last step of glycolysis. In

mammalian cells, there are two LDH isoforms, LDH-A and

LDH-B, with the A form having higher intrinsic activity to

catalyze pyruvate transformation to lactate.^7,8LDH exists in

tetramer form in human cells, comprising combinations of the A

and B isoforms. Speciﬁcally, LDH-5, composed of four LDH-A

monomers, is mostly expressed in liver and muscle and favors the

conversion of pyruvate to lactate.⁹

Received: October 12, 2012

A

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

LDH enzyme exists as a tetramer, each LDH-A monomer inde-

pendently possesses full catalytic function. The large domain

adopts a Rossmann fold²³and forms the NADH cofactor binding

site, while the small domain, with a mixed α/β structure, constitutes

the substrate binding domain. Mechanistic studies have revealed

an ordered sequence of cofactor and substrate binding events

in LDH-A.^24,25The cofactor NADH binds ﬁrst and prepares

the binding site for the substrate binding, which is followed by

the closure of the catalytic loop (residues 98−107), bringing the

catalytic Arg105 into contact with substrate, followed by hydride

transfer. The crystal structure of human LDH-A containing both

NADH and the LDH inhibitor oxamate,⁹a pyruvate mimic,

shows that NADH binds to LDH-A in an extended conforma-

tion with the nicotinamide and adenosine portions of NADH

occupying the two ends of the cofactor active site separated by a

distance of approximately 20 Å.⁹The substrate analogue oxamate

binds parallel to the nicotinamide while forming a bifurcated

hydrogen bond to the guanidine group of Arg168 of LDH-A, and

hydrogen bonds to catalytic residues His192 and Arg105. Both

the NADH and pyruvate substrate binding sites present potential

opportunities for developing small molecule LDH-A inhibitors.

However, the polar nature of the active site, small size of the

substrate site, and the large distance between the adenine and

nicotinamide sites pose a challenge to the discovery of potent

small molecule inhibitors.

Although LDH-A is well characterized, only a few LDH-A

inhibitors are reported in the literature. Most published LDH-A

inhibitors have relatively poor inhibitory activities with low μM

IC₅₀in enzymatic assays. Among them, oxamate, a pyruvate

substrate analogue, has been shown to have K_i≈ 26 μM against

LDH-A.⁹Several chemical series were designed and synthesized

based on oxamate.^26,27These compounds have high molecular

weight and diverse functional groups; however, they exhibit

relatively poor inhibitory activities. In addition to the oxamate

analogues, a natural product named gossypol and the gossypol

analogue FX11 were reported as LDH-A inhibitors with K_ivalues

in the low micromolar range.²⁰Recently a new chemical series of

N-hydroxyindole-based LDH-A inhibitors was published with

reported enzymatic inhibitory activity in the low micromolar

range.²⁸Finally, a recent publication by Ward et al. described

fragment-based discovery of LDH-A inhibitors.²⁹This study was

published as our research program was concluded. The authors

report a similar fragment-linking workﬂow described here; inter-

estingly both groups achieved nanomolar potency with chemi-

cally diﬀerent linked structures.

Figure 1. Fragment based drug discovery workﬂow employs multiple

conﬁrmatory assays. The asterisk (∗) indicates that enzyme assays are

run in the presence of detergent to rule out aggregation.

in two subpockets of the LDH-A catalytic active site. Then a

linking strategy was deployed to improve the inhibitory activities

and led to the identiﬁcation of a novel chemical series that

inhibits LDH-A with IC₅₀in the nanomolar range in the enzyme

assay. Some of the compounds demonstrated inhibition of lactate

production in cellular assays, suggesting that they inhibit the

LDH-A target in cancer cell lines.

RESULTS AND DISCUSSION

■

Structural Analysis. To investigate the binding site, we

solved the X-ray crystal structure of rabbit LDH-A in complex

with NADH and oxamate and the structure of rabbit LDH-A

in complex with AMP. The rabbit LDH-A has 93% sequence

similarity to human LDH-A³⁴and was the optimal system for

obtaining well-diﬀracting cocrystals with AMP. The LDH-A/

AMP system was required for soaking the fragments, as discussed

below. The structures exhibit a similar fold to the human LDH-

A/NADH/oxamate ternary structure that was previously reported

(PDB code 1i10⁹). However, in our structures, the active site loop

(residues 98−110) remained disordered. To map binding site

properties in LDH-A, we applied SiteMap^35,36and WaterMap.³⁷

SiteMap highlighted the largely hydrophilic nature of the binding

cavity, with hydrophobic patches at the nicotinamide and the

adenosine sites that are far removed from each other. Analysis of

the hydration sites generated by WaterMap conﬁrmed this ﬁnding,

with high-energy water located in the adenosine binding pocket,

the nicotinamide site, and the substrate site. The computational

investigation of LDH-A indicated the presence of a large, hydro-

philic site of challenging druggability. The buried energetic hot

spots are far apart, indicating that a linking strategy connecting two

sites might be necessary to achieve optimal potency against LDH-A.

Fragment Screening. To identify novel fragments binding

to LDH-A, we used NMR spectroscopy as the primary screening

method to detect and validate protein−ligand interaction. A

library of 735 diverse Ro3 compliant small molecules (May-

bridge) was investigated using STD-NMR experiments. As

shown previously,³²this magnetization transfer method is well

suited to identify weakly binding fragments with aﬃnities ranging

from tens of micromolar to millimolar K_d.

In search of novel LDH-A inhibitors, we took a fragment-based

approach to identify low molecular weight chemical compounds

that bind to LDH-A. As illustrated in Figure 1, we also performed

a virtual screen of commercially available compounds from the

ZINC³⁰database on LDH-A, in parallel to our fragment based

eﬀorts. Although several virtual screening hits showed activity in

the enzyme assay (300−800 μM), the activity was not conﬁrmed

by our secondary screens. The hits displayed aggregator prop-

erties and were not followed up for optimization.

In the search of small, speciﬁc fragments, a chemically diverse,

rule-of-three (Ro3)³¹compliant fragment library was screened

in the LDH-A binding assay monitored by saturation transfer

diﬀerence NMR (STD-NMR)³²and followed by surface plasmon

resonance (SPR)³³K_dmeasurement experiments. Crystal struc-

tures of the identiﬁed fragment hits bound in LDH-A were

determined, and the structural information was used to guide

fragment evolution and optimization. These fragment-based

eﬀorts led to the identiﬁcation of novel LDH-A inhibitors bound

The NMR screen returned a total of 38 hits (5.2% hit rate), 19

of which were carboxylic acids. The high hit rate for carboxylic

B

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

acids (∼50% of hits compared to 16% of total compounds in

the library) implies a preference for small, negatively charged

molecules in the binding site, potentially mimicking the pyruvate

binding pattern. Similar ﬁndings have been previously

reported.^38,39NMR competition binding experiments with the

natural cofactor NADH were performed in order to (i) further

validate the hits, (ii) conﬁrm the site of binding, and (iii)

distinguish between speciﬁc and unspeciﬁc binding of fragments.

As an example, Figure 2 shows the reference NMR spectrum of

fragment 1 in the presence of LDH-A (panel A) along with the

for early hits using enzyme assays, and necessitated implementa-

tion of orthogonal biophysical techniques, such as SPR.

Investigation of the initial fragment hits by SPR revealed only

weak interaction with LDH-A, with K_dvalues ranging between 1

and 5 mM for most compounds. To obtain guidance on chemical

optimization of the fragments, crystals of LDH-A with AMP,

back-soaked to remove AMP, were utilized for soaking the

fragment hits and subsequent crystallography. Clear electron

density was visible for one of the soaked compounds (fragment 1,

Figure 3). An unambiguous binding mode was observed. The

Figure 3. Binding of fragment 1 in the active site of LDH-A.

fragment occupies the oxamate and part of the nicotinamide

binding sites, with the carboxylic acid making a salt bridge to the

catalytic residue Arg168. An additional hydrogen bond is formed

between one of the carboxylic oxygen atoms and the hydroxyl

group of the catalytic Thr247. The pyridine ring is in contact with

the hydrophobic side chains of Ile251 and Val30 and has addi-

tional favorable interactions with Asn137. The phenyl ring makes

productive interactions with Val30, Val135, Thr94, and Ser136.

Superposition of the X-ray crystal structure with that of the LDH-

A ternary complex (1i10) shows that the fragment extends in the

direction of the pyrophosphate group of NADH, with the phenyl

ring occupying the position of the sugar moiety. Although

fragment 1 was inactive in the enzyme assay (IC₅₀> 2 mM), a K_d

of 2.3 mM was detected in the SPR assay.

Fragment 1 was selected as the starting point for exploration

because of the ease of synthetic modiﬁcation and the availability

of a crystal structure to guide the optimization. After analyzing

available vectors for fragment growing, we modiﬁed the fragment

at the para position of the phenyl ring (C4) to access the open

cofactor-binding pocket. Meta-substitutions on the phenyl ring

(positions C3 and C5) were also explored to engage protein

interactions. Speciﬁcally, a small hydrophobic pocket was

observed in the crystal structure, formed between two strands

at the ends of the β sheets by residues Ser136, Val135, and Thr94.

Introduction of ﬂuorine at position C5 to ﬁll that pocket

improved activity. At position C3, hydrophilic chains were added

to facilitate hydrogen bonding interactions with the main chain of

the protein. Although C4-substitutions yielded no improvement

in IC₅₀, combining modiﬁcations at positions C5 and C3 yielded

compounds with measurable IC₅₀in the low millimolar range

(for example, see compounds 2 and 3 in Table 1). Finally,

we attempted fragment growing into the nicotinamide site by

Figure 2. (A) Structure of compound 1 and reference NMR spectrum.

(B) STD-NMR spectrum of 1 in the presence of 20 μM rabbit LDH-A.

(C) STD-NMR spectrum of 1 in the presence of 20 μM rabbit LDH-A

and 250 μM NADH. All spectra were obtained at 2 mM 1. Addition of

NADH results in a reduction of STD signal intensity by about 50%,

indicating competition between NADH and compound 1.

STD-NMR spectrum in panel B. Upon addition of NADH

(panel C), the intensity of STD signals of compound 1 is reduced

by a factor of 2, demonstrating NADH competitiveness. The

remaining STD-NMR signal of 1 in the presence of NADH can

be explained by (i) incomplete saturation of the NADH binding

site due to its limited potency (K_M≈ 10 μM) or (ii) alternative

binding sites for compound 1 including unspeciﬁc binding at the

protein surface. However, a reduction of STD signal of about

50% or more in the presence of NADH was suﬃcient to con-

ﬁrm compounds binding to the substrate/cofactor sites for the

majority of early fragment hits. The competition of most frag-

ments with NADH is consistent with the observation of a very

low hit rate in an NMR screen run in the presence of NADH.

Apparently, most fragments in our library were too large to ﬁt

into the small substrate binding site in the presence of NADH or

were not potent enough to displace NADH.

Unfortunately, even though binding to the substrate/cofactor

site was demonstrated by NMR competition experiments, most

of the initial fragments did not show signiﬁcant inhibition of the

enzyme reaction (IC₅₀> 1 mM), making it diﬃcult to follow SAR

C

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Table 1. Structures and Activity of Selected Fragments

a

LE computed as pK_d/(no. of heavy atoms).

Figure 4. SPR sensograms for compound 4 (A) and compound 6 (B) binding to the surface with immobilized LDH-A (1), CA-II (2), or an empty

surface (3). The sensograms were obtained at 20 °C using a 2-fold dilution series with an upper concentration of 1 mM. Compounds in the millimolar

aﬃnity range were found to have signiﬁcant oﬀ-target aﬃnity as demonstrated by the binding to CA-II. Elaborated hits, such as compound 14, not only

showed better aﬃnity to LDHA but also improved selectivity.

introducing C3 and C4 substitutions onto the pyridine ring;

however, these derivatives showed only minor aﬃnity improve-

ments (for example, see compound 4, with K_d≈ 1.8 mM) and

still displayed signiﬁcant oﬀ-target eﬀects to the SPR reference

protein (Figure 4A).

Adenosine Site. In the course of screening elaborated anal-

ogues of compound 1, compound 5 was identiﬁed (Figure 5A).

With an IC₅₀of 770 μM in enzyme assay and a K_dof 137 μM

measured by SPR, it was the strongest binder derived from the

initial fragments thus far. However, when compound 5 was

soaked in the LDH-A crystal, it bound in the adenosine site

rather than in the oxamate site analogous to its parent compound

1. In the observed binding mode, the phenyl ring occupies a

lipophilic pocket normally ﬁlled by the adenosine moiety of

NADH (Figure 5B). The phenyl is sandwiched between the side

chains of Ile115 and Ile119 on the bottom and Val52 on the top.

The 1-methoxy group is bound through van der Waals contacts

to Ile115 and Phe118. The 4-Cl substituent is in contact with

the backbone atoms of Asp51 and residues Gly26 and Val25.

The ﬂexible linker lies on the surface of the protein and makes

favorable interactions with Ile115 and the side chains of Ile119

and Phe118. Finally, the pyridine ring interacts with Asn114 and

Ile115. The carboxylic acid forms a salt bridge with the side chain

of Arg111. An internal hydrogen bond between the amide

nitrogen of the linker and the pyridine N stabilizes the bound

conformation. Replacement or deletion of the phenyl ring sub-

stituents or linker modiﬁcations resulted mostly in reduced potency.

However, we were successful in introducing substitutions at position

D

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 6. Cosoaked structures of compounds 2 and 6 in the active site of

LDH-A. The side chain of Arg98 was removed for an unobstructed view

of the binding site.

Figure 5. (A) Compound 5. (B) The crystal structure reveals compound

5 bound in the adenosine binding site of LDH-A. Arg168 is shown for

reference to indicate the oxamate binding site.

compounds 9 and 8 with the protein were similar and consistent

with those described above for the cosoaked parent fragments.

The eight-atom linker spanned the solvent-exposed region con-

necting the two binding sites, with the hydroxyl groups making

direct and water-mediated hydrogen bonds to the backbone

atoms of the protein. Speciﬁcally, three crystallographic water

molecules were detected in the linker region. These waters

form a network of hydrogen bonds, connecting 1-hydroxyl and

3-hydroxyl to the backbone nitrogen atoms of Val30, Ala29, and

Gly31. One additional water molecule is wedged between the

phenyl ring of the inhibitor and the backbone carbonyl atom of

Thr247. The carbonyl atom of the S-containing linker in the

adenosine site forms a hydrogen bond with the side chain of

Tyr82, mediated by the ﬁfth crystallographic water. Interestingly,

the side chain of Arg111 in all these structures was disordered.

Clearly, the exact positioning of the four hydroxyl groups was key

to activity. The removal of these groups leads to a marked loss

of potency, as did the modiﬁcation of the stereochemistry (see

compound 11). We also explored an amide-containing linker

(compound 10) in an attempt to engage water-mediated hydro-

gen bonds without introducing hydroxyl groups; however, this

resulted in loss of activity.

Linking of fragments in adenosine and pyruvate binding

pockets yielded compounds with greatly improved potency and

selectivity. Figure 8A shows the dose response curves for enzyme

inhibition of compound 8 in the presence and absence of deter-

gent. The results are reproducible within the margin of experi-

mental error (2-fold). The SPR sensogram of compound 7 is

shown in Figure 8B. The SPR response can be well ﬁtted with a

simple 1:1 binding model. Both kinetic and equilibrium K_d

analysis are in agreement with an aﬃnity of 0.9 μM in the enzyme

assay for this compound. No aﬃnity to the SPR reference surface

containing immobilized carbonic anhydrase II (CA-II) was

detected. Finally, the potency gain in linking compounds was

achieved without a loss in LE compared to the parent fragments

(Tables 1 and 2).

3 of the phenyl ring, with the goal of extending the molecule into the

binding site of LDH-A. To mimic the interactions that the sugar

moiety of NADH makes with the protein, we added an oxypropane-

1,2-diol group and thereby gained some potency (compound 6;

Figure 4B; IC₅₀= 342 μM, K_d= 360 μM). An additional hydrogen

bond was formed between the 3-hydroxy group of the molecule and

the side chain of Asp51. This hydrogen bond is also observed in the

LDH-A/NADH/oxamate structure, where the adenosine sugar

interacts with the side chain of Asp51 via one of its hydroxyl groups.

Further substitutions at the 1 and 4 positions of the phenyl ring,

as well as in the linker, and the replacement of carboxylic acid failed

to increase potency. However, adenosine site binders showed

improved potency with K_dvalues in the 200−400 μM range and

improved selectivity for LDH-A, compared to oxamate site binders.

To quantify the potency improvement, we computed the ligand

eﬃciency (LE) of the elaborated fragments. We saw either no

improvement or loss of LE in the process of fragment growing

(Table 1), with both initial fragment hits and elaborated fragments

exhibiting fairly low LE. The diﬃculty in improving LE is a

consequence of the low druggability of the LDH-A binding site. We

therefore concluded that for signiﬁcant potency gain we would need

to adopt a fragment-linking approach: bridging the gap between

potent oxamate and adenosine fragments with an appropriately

spaced chemical linker.

Linked Compounds. Cosoaking the two molecules in the

LDH-A crystal, we determined the structures of compounds 2

and 6 in the LDH-A binding site (Figure 6). The two molecules

assume binding modes identical to the modes in the individual

crystal structures. The distance between the terminal carbons of

these compounds is 3.3 Å. Flexible linkers of seven to nine atoms

were introduced to bridge the space between the two molecules.

Initially, a simple ether linkage was used (compound 7),

with resulting IC₅₀of 59 μM for the linked compound. To

explore functional groups and linker lengths, compounds with

functional groups such as ester, amide, NH, and OH in the linker

were synthesized and tested (Table 2). An eight-atom linker

presented the optimal geometry for connecting the two binding

sites. The most active molecule synthesized was compound 9,

with an eight-atom linker containing four hydroxyl groups.

The X-ray crystal structure was solved by soaking compound 9

with LDH-A; however, a structure of an analogue compound 8

had stronger electron density (Figure 7). The interactions of

Finally, anticipating potential permeability issues in a cellular

assay, we synthesized analogues without one or both terminal

carboxylic acids (Table 3, compounds 12, 13, 14). However, the

removal of the carboxylic acid leads to at least a 20-fold potency

loss (see compound 12 relative to reference compound 9). The

SAR highlights the importance of both salt bridges observed in

the crystal structure.

Chemistry. Compound 1. Compound 1 was purchased from

Maybridge as part of the fragment screening collection.

E

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Table 2. SAR for Selected Linked Compounds

a

LE computed as pK_d/(no. of heavy atoms).

Figure 7. X-ray crystal structure of linked compound 8 in LDH-A with hydrogen bonding network shown.

Compound 2. Compound 2 was synthesized by a Suzuki

coupling of 6-bromonicotinic acid and 3-(carboxymethoxy)-

5-ﬂuorophenylboronic acid without protection as shown in

Scheme 1. HPLC puriﬁcation aﬀorded the desired product in

14% yield.

Compound 3. Synthesis of compound 3 proceeded via Suzuki

coupling of methyl 6-chloronicotinate and 3-amino-5-ﬂuoro-

phenylboronic acid to give intermediate compound 15. Hydro-

lysis with LiOH in MeOH/H₂O furnished the target compound

3 as depicted in Scheme 2.

F

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 8. Biochemical and biophysical evaluation of linked compounds. (A) Compound 8 displayed enzyme inhibition virtually independent of

detergent. (B) SPR sensogram (black) of compound 7 obtained with “fast-step” concentration ramping (dotted green line). The data were ﬁtted

(orange line) to a 1:1 binding model and yielded a K_dof 0.9 μM, in agreement with an IC₅₀of 60 μM.

Compound 6. Compound 6 was synthesized as outlined

in Scheme 4. 1-Chloro-2-ﬂuoro-4-methoxy-5-nitrobenzene (19)

was synthesized from commercially available 4-chloro-3-ﬂuoro-

phenol utilizing a known procedure.⁴⁰Displacement of ﬂuoro-

containing 19 by 1,2-isopropylidenglycerin and subsequent

reduction of the corresponding nitro derivative 20 by iron in

acetic acid furnished aniline 21. Acylation of 21 with chloroacetic

chloride at 0 °C followed by S_n2 displacement with 6-

mercaptonicotinic acid at 100 °C in a single pot led to

compound 22. Finally, protecting group removal with HCl

yielded the desired product 6.

Table 3. Modiﬁcation of Lead Compound 9 To Improve

Cellular Permeability

compd

R₁

R₂

IC₅₀, μM

K_d, μM

9

COOH

H

COOH

H

0.120

2.4

0.019

0.222

2.95

na

12

13

14

Compound 25. Compound 25 was synthesized by Suzuki

coupling as shown in Scheme 5.

COOH

CO₂Me

13

CO₂Me

>10000

Compound 7. Compound 7 was synthesized as shown in

Scheme 6. 3,3′-Dihydroxylipropyl ether (23) was prepared from

commercially available 3,3′-oxidipropionitrile utilizing a known

procedure.⁴¹S_NAr reaction of ﬂuoro-containing 19 by 23 in THF

in the presence of NaH furnished compound 24 in 64% yield.

The Mitsunobu coupling of alcohol 24 and phenol 25 using

DIAD and triphenylphosphine aﬀorded ether 26. Subsequent

nitro group reduction by hydrogenation in the presence of

Pd−C yielded aniline 27. Similar to the synthesis of compound 6,

treatment of 27 with chloroacetic chloride at 0 °C followed

by displacement with 6-mercaptonicotinic acid methyl ester

Compound 4. The synthesis of target compound 4 was

initiated via esteriﬁcation of 4,6-dichloronicotinic acid to aﬀord

ester 16. Selective Suzuki coupling at the 4-position of com-

pound 16 gave compound 17. A subsequent Negishi reaction

introduced a 6-methyl group to furnish compound 18, which was

then hydrolyzed with LiOH in EOH/H₂O to yield 4 as shown in

Scheme 3.

Compound 5. Compound 5 was purchased from a

commercial vendor.

a

Scheme 1. Synthesis of Compound 2

a

Reagents and conditions: (a) Pd(PPh₃)_4,K₂CO_3,dioxane, microwave, 120 °C/20 min.

G

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 2. Synthesis of Compound 3

a

Reagents and conditions: (a) Pd(PPh₃)₄, K₂CO₃, dioxane, 80 °C; (b) LiOH, MeOH/H₂O.

a

Scheme 3. Synthesis of Compound 4

a

Reagents and conditions: (a) SOCl₂and then NaOEt; (b) Pd(PPh₃)₄, K₂CO₃, DME; (c) Pd(PPh₃)₄, THF; (d) LiOH, EtOH/H₂O.

a

Scheme 4. Synthesis of Compound 6

a

Reagents and conditions: (a) K₂CO₃, DMF, 60 °C; (b) Fe, AcOH, reﬂux; (c) Et₃N, THF, 0 °C; (d) Et₃N, THF, 100 °C; (e) HCl, MeOH/H₂O, 50 °C.

Scheme 5. Synthesis of Compound 25

at 100 °C led directly to diester 28. Subsequent hydrolysis with

K₂CO₃in MeOH/H₂O at 80 °C furnished compound 7.

Compound 8. 2,3:4,5-Bis-O-(1-methylethylidene)-^D-manni-

tol (29) was prepared from 1,2:5,6-bis-O-(1-methylethylidene)-

D-chiromannitol by NaIO₄diol cleavage followed NaBH₄

reduction of the corresponding dialdehyde. With alcohol 29

in hand, compound 8 was synthesized in a similar manner to

compound 7, with the exception of an additional actonidie-

deprotection after nitro reduction of compound 31 (Scheme 7).

Compound 9. PDC oxidation of intermediate alcohol 30

gave the corresponding aldehyde which was then converted to

semiacetal 34 during workup and column puriﬁcation in the

presence of methanol. Reductive amination of 34 with aniline 15

in the presence of molecular sieves worked well to aﬀord com-

pound 35 in 70% yield. Finally, compound 9 was synthesized by

following the same procedure as described in the synthesis of

compound 8 (Scheme 8).

was treated with KOH in DMSO/H₂O at 100 °C to give phenol

37 in good yield. Nitro group reduction of 37 resulted in aniline

38, which was used in the next step without further puriﬁcation.

In parallel, compound 39 was made by heating a mixture of ethyl

6-chloro-3-pyridinecarboxylate and mercaptoacetic acid neat at

100 °C. Subsequent DCC coupling of aniline 38 and acid 39 gave

compound 40 in 46% yield.

Compound 10. As outlined in Scheme 10, acid 42 was

prepared by reacting 25 with tert-butyl bromoacetate followed by

a deprotection of the tert-butyl group. Coupling of acid 42 with 3-

amino-1-propanol using EDCI and HOBT produced compound

43. Mitsunobu coupling of 43 with phenol 40 gave diester 44.

Diester 44 was hydrolyzed with K₂CO₃in MeOH/H₂O to give

the desired product 10.

Compound 11. Starting from 1,2:5,6-bis-O-(1-methylethyli-

dene)-D-chiroinositol, compound 11 was synthesized by the

same procedure as described for compound 8 (Scheme 11).

Compound 12. Treatment of compound 36 with chloroacetyl

chloride at 0 °C followed by mercaptopyridine at 100 °C led to

Compound 40. Synthesis of compound 40 is outlined in

Scheme 9. 1-Chloro-2-ﬂuoro-4-methoxy-5-nitrobenzene (19)

H

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 6. Synthesis of Compound 7

a

Reagents and conditions: (a) NaH, THF, 60 °C; (b) DIAD, TPP, DCM, 0−60 °C; (c) H₂, Pd−C, MeOH, rt; (d) Et₃N, THF, 0 °C; (e) Et₃N,

THF, 100 °C; (f) K₂CO₃, MeOH/H₂O, 80 °C.

a

Scheme 7. Synthesis of Compound 8

a

Reagents and conditions: (a) (1) NaIO₄, acetone/H₂O; (2) NaBH₄; (b) NaH, THF, 60 °C; (c) DIAD, TPP, DCM, 0−60 °C; (d) Zn, NH₄Cl,

acetone/water, rt; (e) HCl, MeOH; (f) Et₃N, THF, 0 °C; (g) Et₃N, THF, 100 °C; (h) K₂CO₃, MeOH/H₂O, 80 °C.

ester 50. Hydrolysis of 50 with K₂CO₃in MeOH/H₂O at 80 °C

gave compound 12 as shown in Scheme 12.

assessed whether they could reduce lactate levels in cancer cells.

We assessed extracellular lactate levels in a Burkitt’s lymphoma

cell line (Ramos) treated with compounds 9, 12, 13, an inactive

compound 14, or oxamate as a positive control (Figure 9).

Compound 14 produced virtually no inhibition of lactate

secretion, while oxamate reduced lactate levels to 48% and

56% of vehicle by 1 and 4 h, respectively. Like oxamate,

compound 9 produced an initial inhibition to 56% by 1 h and to

68% by the later time point, reﬂecting a sustained eﬀect.

Compounds 12 and 13 more dramatically blocked lactate levels

to 21% and 17%, respectively; however, both had begun to

recover by 4 h to 61% and 35%. While the extent of cellular

lactate inhibition does not directly correlate with LDH-A IC₅₀,

we cannot rule out factors such as cell permeability, import and

Compound 13. As outlined in Scheme 13, compound 52 was

synthesized via Suzuki coupling of 3-(Cbz-amino)-5-ﬂuorophe-

nylboronic acid and 2-bromopyridne followed by deprotection of

the Cbz group under hydrogenation conditions. With aniline 52

in hand, compound 13 was synthesized by the same procedure as

described for the synthesis of compound 9.

Compound 56. Compound 56 was synthesized using the

conditions reported in literature.²⁰

Compounds 57 and 58. Compounds 57 and 58 were

synthesized using published literature protocol.⁴²

Biology. After determining that the compounds described

here can directly inhibit LDH-A enzymatic activity, we next

I

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 8. Synthesis of Compound 9

a

Reagents and conditions: (a) PDC, MS, DCM, rt; (b) MS, DCM, NaHB(OAc)₃, rt; (c) Zn, NH₄Cl, acetone/water, rt; (d) HCl, MeOH; (e) Et₃N,

THF, 0 °C; (f) Et₃N, THF, 100 °C; (g) K₂CO₃, MeOH/H₂O.

Scheme 9. Synthesis of Compound 40

a

Scheme 10. Synthesis of Compound 10

a

Reagents and conditions: (a) tert-butyl 2-bromoacetate, K₂CO₃, THF, rt; (b) TFA, rt; (c) 3-amino-1-propanol, EDCI, HOBt, TEA, DMF, rt; (d)

DIAD, TPP, THF 0 °C to rt; (e) K₂CO₃, 90 °C.

export processes, and degradation kinetics in shaping the cellular

eﬀects of these compounds. Taken together, all compounds

except 14, which lacks LDH-A activity, reduced cellular lactate

levels reﬂecting activity against the LDH-A enzyme.

Evaluation of Enzyme Activity for Previously Pub-

lished Inhibitors of LDH-A. During the course of the project

we evaluated several previously published LDH-A inhibitors

using enzymatic assays and biophysical methods. For the

majority of compounds we were unable to reproduce key

literature results under our experimental conditions. As an

example, Supporting Information Figure 2SA shows the enzyme

inhibition of compound 56 (FX11). While no enzyme inhibition

J

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 11. Synthesis of Compound 11

a

Reagents and conditions: (a) (1) NaIO₄, acetone/H₂O; (2) NaBH₄; (b) NaH, THF, 60 °C; (c) DIAD, TPP, DCM, 0 to 60 °C; (d) Zn, NH₄Cl,

acetone/water, rt; (e) HCl, MeOH; (f) Et₃N, THF, 0 °C; (g) Et₃N, THF, 100 °C; (h) K₂CO₃, MeOH/H₂O, 80 °C.

a

Scheme 12. Synthesis of Compound 12

a

Reagents and conditions: (a) Et₃N, THF, 0 °C; (f) Et₃N, THF, 100 °C; (g) K₂CO₃, MeOH/H₂O.

a

Scheme 13. Synthesis of Compound 13

a

Reagents and conditions: (a) Pd(PPh₃)₄, K₂CO₃, dioxane/H₂O; (b) H₂, Pd−C; (c) molecular sieves, DCM, NaHB(OAc)₃, rt; (d) Zn, NH₄Cl,

acetone/water, rt; (e) HCl, MeOH; (f) Et₃N, THF, 0 °C; (g) B, Et₃N, THF, 100 °C; (h) K₂CO₃, MeOH/H₂O.

K

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

the potency-enhancing eﬀect of linking fragments. Although

fragment growing itself failed to gain suﬃcient potency, ulti-

mately a linking strategy proved to be successful as a means to

enhance potency, resulting in potent and nonaggregating

inhibitors of LDH-A. While we obtained compounds that dem-

onstrated high potency in enzymatic assays, the physicochemical

properties of the compounds are not optimal for cellular per-

meability. Further work remains to be done in making potent

compounds with desirable ADME properties for this valuable

target.

EXPERIMENTAL SECTION

■

NMR Spectroscopy. NMR experiments were performed at a tem-

perature of 300 K on a Bruker Avance DRX 600 MHz spectrometer,

equipped with a z-gradient 1.0 mm TXI microprobe and a Bruker

SampleJet autosampler. STD-NMR experiments were recorded with 2 s

saturation time, applying 50 ms Gaussian shaped pulses with a nominal

ﬂip angle of 720°. The frequency of the on-resonance irradiation was set

to 0.8 ppm, and the oﬀ-resonance pulse was at 30 ppm. The residual

water signal was suppressed using excitation sculpting. For the primary

screen, rabbit LDH-A (Aldrich) was dissolved at 20 μM in D₂O based

buﬀer containing 50 mM sodium phosphate, 100 mM NaCl, pH 7.4, and

0.33 mM of TSP as chemical shift reference. After addition of fragment

from concentrated DMSO-d₆stock solution at a ﬁnal concentration of

2−3 mM per sample, about 10 μL of solution was transferred to 1.0 mm

NMR tubes using a Gilson liquid handler. Samples contained up to 7%

DMSO and were kept at room temperature until measured. Data sets

shown were the average of 320 scans. All data processing was performed

using the Bruker Topspin 1.3 software.

SPR. SPR studies were performed at room temperature on a BIAcore

T100 or a Pioneer SensiQ instrument (as indicated). Human LDH-A

was immobilized onto a CM5 chip via standard amine coupling which

yielded ∼13000 RU of immobilized protein. Carbonic anhydrase II

(Sigma) was used as reference protein. Binding of NADH to LDH-A

was tested using 250 μM as the highest concentration in a 2-fold dilution

series with nine concentrations in total. The running buﬀer contained

10 mM HEPES, 150 mM NaCl, and 3% DMSO. Each concentration was

tested three times. An overlay of all the binding responses is shown in

the Supporting Information (Figure 1S). The responses were well

reproducible indicating that the LDH-A surface was active. The ﬁt of the

equilibrium response data to a 1:1 interaction model yielded a binding

aﬃnity of NADH of 13 μM at room temperature.

Figure 9. Inhibition of cellular lactate levels by compounds 9, 12, 13,

and 14. Relative lactate levels were determined at 1 and 4 h treatments

with the four compounds (200 μM) or the positive control oxamate

(10 mM). A representative experiment is shown.

is observed in the presence of 0.1% Triton, the compound

shows strong enzyme inhibition in the absence of detergent, with

measured IC₅₀depending on experimental parameters, such as

protein−compound incubation time, and ranging from ∼40 μM

to >200 μM. These observations, together with the Hill co-

eﬃcient, are indicative of enzyme inhibition by aggregation.⁴³

This conclusion was further corroborated by SPR studies

(Supporting Information Figure 2SB) which indicate super-

stoichiometic binding and slow on and oﬀ kinetics for this

compound, implying unspeciﬁc binding due to compound

aggregation.⁴⁴Other previously reported compounds, such as

compounds 57 and 58, known as common ligand mimics

(CLM)⁴²showed similar behavior, as illustrated in Supporting

Information Figure 3S. In the presence of detergent in the buﬀer,

no enzyme inhibition was observed for those compounds when

tested at up to 1 mM. Not surprisingly, attempts to produce

cocrystal structures with those compounds were unsuccessful.

Crystallography. LDH from several organisms including human,

rabbit, and dogﬁsh were screened for a soaking suitable crystal system.

Although we crystallized all the LDH proteins, only rabbit LDH gave

well-diﬀracting crystals when cocrystallized with AMP.

Rabbit LDH, purchased from Sigma-Aldrich (catalog no. L2500), was

puriﬁed from rabbit muscle containing predominantly LDH-5.The

sequence of rabbit LDH-5 form is identical to its human counterpart (an

all LDH-A tetramer) in the active site loop and near identical in the

entire active site.³⁴The purchased protein sample was then polished in

a size-exclusion column in a buﬀer containing 50 mM Hepes, pH 7.5,

50 mM NaCl and 1 mM DTT. Cocrystals of rabbit LDH (25 mg/mL)

with AMP (10 mM) were obtained at 4 °C in a condition containing

0.1 M Tris, pH 8.0, and 28% PEG 550 MME.

Soaking of the fragment hits and lead compounds was performed

following a back-soaking procedure in which the LDH/AMP cocrystals

were transferred to a fresh mother liquor for 2−15 h. Such a procedure is

designed to allow the prebound AMP to dissociate from the enzyme.

Then the apo crystals were soaked with inhibitors at 1−20 mM for up to

3 days and cryo-protected by 38% PEG 550 MME before ﬂash-freezing

with liquid nitrogen. Diﬀraction data were collected at 100 K on a Rigaku

rotating anode and R-AXIS IV++ imaging system. The data were

indexed and scaled using the HKL2000 package. The rabbit LDH

crystals belong to P21 space group with cell dimensions of a = 83.9 Å, b =

139.7 Å, c = 138.9 Å, and β = 94.5°.

CONCLUSION

■

LDH-A, an enzyme is the glycolysis pathway, is considered a

signiﬁcant small-molecule oncology target in cancer metabolism.

We demonstrate here the successful application of a fragment-

based drug discovery strategy to ﬁnd novel inhibitors of LDH-A.

LDH-A proved to be a diﬃcult target. It has a small substrate

binding site formed in part by cofactor binding, leading to a need

for inhibitors capable displacing the cofactor, and furthermore

contains a polar and largely solvent-exposed binding site. Despite

these challenges, a series of potent compounds was identiﬁed

through growing and linking fragments. Binding of these potent

linked fragment-based ligands extended through both cofactor

and substrate binding sites, utilizing hydroxylated linkers to bind

in the NADH pyrophosphate binding site. We demonstrated

experimentally a lack of aggregator artifacts on binding of these

compounds to LDH-A and conﬁrmed their enzymatic activity

with SPR binding assays where the ligands showed stoichio-

metric binding to the target. Additionally, we obtained several

crystal structures of linked compounds bound to LDH-A. Finally,

we demonstrated the ability of representative compounds to

inhibit lactate levels in cells.

Our work highlights the power of fragment-based approaches

in locating binding hot spots even in challenging targets and of

The structures were determined by molecular replacement with

AMoRe using a tetramer from the structure of human LDH-A bound

L

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

with NADH and oxamate (PDB code 1i10) as a search model. There are

two LDH-5 tetramers in the asymmetric unit. The structures were

reﬁned with CNX combined with model building in Quanta (Accelrys

Inc.). The inhibitor molecules were built into the density after several

cycles of reﬁnement and model building. As a control, the structure of

rabbit LDH determined after back-soaking showed AMP soaked out

readily thus permitting subsequent inhibitor binding.

¹H NMR (400 MHz, CDCl₃) δ 9.18 (dd, J = 2.0, 0.8 Hz, 1H), 9.26

(dd, J = 8.4, 2.0 Hz, 1H), 7.66 (dd, J = 8.4, 0.8 Hz, 1H), 7.19 (s, 1H), 7.13

(t, J = 1.6 Hz, 1H), 7.01 (ddd, J = 10.0, 2.0, 1.6 Hz, 1H), 6.41 (ddd, J =

10.0, 2.4, 2.0 Hz, 1H), 3.90 (s, 3H).

Compound 3. A suspension of compound 15 (40 mg, 0.16 mmol) in

MeOH (1.0 mL) was treated with LiOH (2 equiv) and drops of water at

room temperature for 1 h. Formic acid (drops) was added, and then

solvent was evaporated. The residue was puriﬁed by preparative HPLC

to give the title compound (25 mg, 66%) as a light yellow solid.

Compound 16. The 4,6-dichloronicotinic acid (9.5 g, 49.5 mmol)

was reﬂuxed in 100 mL of thionyl chloride for 2 h and evaporated. The

residue was dissolved in 100 mL of anhydrous ethanol, and sodium

ethoxide (3.70 g, 54 mmol) was added. The reaction was stirred at room

temperature overnight and ﬁltered. The ﬁltrate was evaporated and

chromatographed (heptane/ethyl acetate 10:1) to give an oil, 8.8 g, in

Molecular Modeling. The X-ray crystal structure of rabbit LDH-A

with soaked-in AMP and the published ternary complex structure (PDB

code 1i10) were used in modeling studies. To explore the binding cavity,

SiteMap³⁵and WaterMap were applied to the crystal structure. SiteMap

was applied with default settings; cofactor and substrate were used to

deﬁne binding site regions. WaterMap⁴⁵was performed with default

options on the crystal structure. An initial minimization and a molecular

dynamics simulation was run for 9 ns on a solvated protein system in the

absence of the ligand. The preparation of the protein for docking was

performed with PrepWizard using the standard protocol, including the

addition of hydrogens and the assignment of bond order and of correct

protonation states. The hydrogen bonding network of the protein was

optimized. Docking experiments were performed using Glide XP.⁴⁶For

substrate-site binders, a hydrogen-bond constraint to the Arg168 was

used. For design of the linked molecules, the initial crystallographic

poses of the two fragments were used as templates. Then a linker was

built in and the resulting molecule minimized to obtain the optimal

positioning of the linker. The fragments and ligands were treated with

LigPrep⁴⁷to obtain 3D structures, and protonation states were assigned

using Epik. MM-GB/SA in Prime⁴⁸was performed on top scoring ligand

poses, with default settings.

Biology. The human Burkitt’s lymphoma cell line Ramos (American

Type Cell Culture, Manassus, VA, catalog no. CRL-1596) was plated at

5 × 10⁶cells/mL and treated with 200 μM 9, 12, 13, and 14, 10 mM

oxamate, or vehicle. Then 5 μL of supernatant was harvested, and lactate

levels were determined after 1 and 4 h of treatment using a two-step

lactate assay (Trinity Biotech, Wicklow, Ireland, catalog no. 735-10)

(Supporting Information Figure 4S). Both time points were determined

to be within the linear range of the reaction. A control reaction using

puriﬁed lactate instead of supernatant was performed to conﬁrm that the

compounds do not aﬀect lactate oxidase or peroxidase activity.

Chemistry. General Remarks. Commercially available chemicals

and solvents were purchased and used without further puriﬁcation. ¹H

NMR spectra were recorded on a Bruker Avance 400 spectrometer.

Chemical shifts (δ) are reported relative to CDCl₃at 7.26 ppm or

DMSO-d₆at 2.50 ppm as an internal standard. Mass spectra were re-

corded on a Waters Micromass ZQ spectrometer. HPLC was performed

on an Agilent 1100 HPLC system. All tested compounds were puriﬁed

to >95% purity as determined by ¹H NMR and by HPLC (C-18 column,

MeCN/H₂O with 0.1% TFA as the mobile phase).

Compound 2. To a microwaveable test tube was loaded a mixture of

6-bromonicotinic acid (101 mg, 0.5 mmol), 3-(carboxymethoxy)-5-

ﬂuorophenylboronic acid (105 mg, 0.5 mmol), and Pd(PPh₃)₄(60 mg,

0.05 mmol). This septum-capped tube was degassed via three cycles of

vacuum-reﬁlling of N₂. To this tube was added dioxane (3 mL, Aldrich

Sureseal bottle) and 2 M aqueous K₂CO₃(2 mL, predegassed). The

resulting suspension was heated under microwave at 120 °C for 20 min.

The reaction was quenched with addition of 2 N aqueous HCl, and the

solvents were removed on rotavap. The crude solid was dissolved in

DMSO, and the ﬁltered solution was subjected to preparative HPLC

puriﬁcation to furnish 2 as white solid (20 mg, 14%). ¹H NMR (DMSO-

d₆, 400 MHz) δ 9.14 (dd, 1H, J = 2.0, 0.8 Hz), 8.33 (dd, 1H, J = 8.0, 2.0

Hz), 8.16 (d, 1H, J = 8.8 Hz), 7.60 (s, 1H), 7.57 (d, 1H, J = 2.4 Hz), 6.96

(d, 1H, J = 10.8 Hz), 4.80 (s, 2H). MS (ES+) m/z 292 (M + 1).

Compound 15. A mixture of methyl 6-chloronicotinate (1.0 g,

5.83 mmol), 3-amino-5-ﬂuorophenylboronic acid (0.9 g, 1.0 mmol),

Pd(PPh₃)₄(360 mg), dioxane (10 mL), and water (5 mL) was degassed

and then heated at 100 °C for 15 min. The reaction mixture was diluted

with ethyl acetate (10 mL). Organic layer was separated and evaporated.

The residue was puriﬁed by column chromatography (elution, heptane/

ethyl acetate = 1:1) to give the title compound 15 (1.1 g, 77%) as an oﬀ

white solid.

1

81% yield. H NMR (ppm) (CDCl₃) 8.77 (s, 1H), 7.39 (s, 1H), 4.36

(q, J = 7.2 Hz, 2H), 1.34 (t, J = 7.2 Hz, 3H).

Compound 17. Compound 16 (3.3 g, 15.0 mmol) and Pd(PPh₃)₄

(303.6 mg) were stirred in 13 mL of dimethoxyethane at room tem-

perature for 15 min. To this suspension was then added phenylboronic

acid (1.74 g, 14.2 mmol) in 15 mL of isopropanol and 13.2 mL of 2.0 M

K₂CO₃. The mixture was stirred at 95 °C for 1 h. After evaporation, the

mixture was worked up with saturated NaHCO₃and DCM and chro-

matographed (EtOAc/heptane = 1:10) to give a white solid, 2.4 g, in

61% yield. ¹H NMR (ppm) (CDCl₃) 9.04 (s, 1H), 7.94 (m, 2H), 7.74 (s,

1H), 7.44 (m, 3H), 4.38 (q, J = 7.2 Hz, 2H), 1.34 (t, J = 7.2 Hz, 3H) .

Compound 18. Compound 17 (127 mg, 0.485 mmol) was dissolved

in 10 mL of THF, and Pd(PPh₃)₄(30 mg, 0.025 mmol) was added. The

mixture was stirred under N₂for 5 min to get a clear solution. Methylzinc

chloride (2.0 M in THF, 0.5 mL, 1.0 mmol) was added at room tem-

perature, and the mixture was stirred at 70 °C overnight. Another 1.0 mL

of CH₃ZnCl and Pd(PPh₃)₄(40 mg) were added, and the stirring was

continued at 80 °C overnight. The mixture was evaporated and worked

up with saturated NH₄Cl and DCM and puriﬁed with ﬂash chro-

matography (EtOAc/heptane = 1:10) to give an oil, 106 mg, in 91%

1

yield. H NMR (ppm) (CDCl₃) 9.10 (s, 1H), 7.95 (m, 2H), 7.52 (s,

1H), 7.42 (m, 3H), 4.34 (q, J = 7.2 Hz, 2H), 2.62 (s, 3H), 1.34 (t, J = 7.2

Hz, 3H).

Compound 4. Compound 18 (103 mg, 0.43 mmol) was dissolved in

10 mL of 2.0 M Na₂CO₃and 10 mL of EtOH. The mixture was stirred at

room temperature for 15 min, and LiOH (30 mg) was added. The

mixture was stirred at room temperature for 30 min and evaporated

to half of the volume. Then 10 mL of water was added, and the solution

was washed with ether (10 mL) and DCM (5 mL). The solution was

adjusted to pH 5 with concentrated HCl to get a precipitate. The

precipitate was ﬁltered and the solid was washed with water and ether

and dried to give a white solid, 57.9 mg, in 62% yield. ¹H NMR (ppm)

(DMSO) 13.20 (s, 1H), 9.02 (s, 1H), 8.15 (m, 2H), 7.96 (s, 1H), 7.47−

7.55 (m, 3H), 2.64 (s, 3H).

Compound 19. Compound 19 was made from 3-ﬂuoro-4-

chlorophenol as reported.⁴⁰

Compound 20. A mixture of compound 19 (205 mg, 1.0 mmol) and

1,2-isopropylidenglycerin (145 mg, 1.1 mmol) in DMF (1.0 mL) was

treated with potassium carbonate (1.0 g) at 60 °C for 10 min. The

mixture was diluted with DCM (3.0 mL), and then the solid was ﬁltered

oﬀ. The solution was evaporated, and the residue was used for the next

step.

Compound 21. The residue 20 was dissolved in acetic acid (2.0 mL)

and then treated with iron (0.5 g, powder). The mixture was heated to

reﬂux and then diluted with ethyl acetate (4.0 mL). After ﬁltration,

solvent was evaporated and the residue was puriﬁed with a preparative

TLC plate (elution, DCM/MeOH = 100:3) to give the title compound

product 21 (120 mg, 42% two steps) as an oﬀ white solid.

Compound 6. To a solution of 21 (57 mg, 0.20 mmol) in THF

(1.0 mL) and triethylamine (0.01 mL) was added chloroacetyl chloride

(2.0 M solution in THF) at 0 °C, monitoring the reaction with HPLC

until the starting material disappeared. 6-Mercaptonicotinic acid

(120 mg, 0.77 mmol) was introduced. The resulting mixture was heated

in a sealed tube at 100 °C for 5 min. Solvent was evaporated, and the

residue was treated with HCl in dioxane (1.0 mL) at room temperature

M

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

for 6 h. Solvent was evaporated and the residue was puriﬁed with pre-

parative HPLC chromatography to give the title compound 6 (31 mg,

35%) as an oﬀ white solid. MS (ES⁺) m/z (M + 1) 443; ¹H NMR (400

MHz, DMSO) δ 9.67 (s, 1H), 8.95 (d, J = 1.6 Hz, 1H), 8.17 (s, 1H), 8.12

(dd, J = 8.4, 2.4 Hz, 1H), 8.05 (s, 1H), 7.54 (d, J = 8.4 Hz, 1H), 6.86 (s,

1H), 4.13 (s, 2H), 4.07 (dd, J = 10.0, 4.4 Hz, 1H), 4.00 (dd, J = 10.0, 6.4

Hz, 1H), 3.85 (s, 3H), 3.81(quint, J = 5.6 Hz, 1H), 3.49 (dd, J = 10.8, 5.6

Hz, 1H), 4.07 (dd, J = 10.8, 5.6 Hz, 1H).

4.07 (t, J = 6.0 Hz, 2H), 4.00 (s, 3H), 3.99 (s, 2H), 3.97 (s, 3H), 3.74 (s,

3H), 3.68 (t, J = 6.0 Hz, 2H), 3.66 (t, J = 6.4 Hz, 2H).

Compound 7. A solution of 28 (125 mg, 0.17 mmol) in methanol

(0.5 mL) was treated with potassium carbonate (250 mg) and water

(drops) in a capped vial at 80 °C for 5 min. The organic layer was

transferred to a new vial and diluted with water (2.0 mL). The mixture

was then treated with formic acid to pH 3−4. The precipitate was

collected by ﬁltration and washed with methanol (1.5 mL) to give the

title compound 7 (58 mg, 49%) as a yellow solid. MS (ES⁺) m/z (M + 1)

700; ¹H NMR (400 MHz, DMSO) δ 13.45 (br, 1H), 9.61 (s, 1H), 9.12

(dd, J = 2.4, 0.8 Hz, 1H), 8.93 (d, J = 1.2 Hz, 1H), 8.31 (dd, J = 8.4, 2.0

Hz, 1H), 8.12 (dd, J = 8.4, 0.8 Hz, 1H), 8.10 (dd, J = 8.4, 2.4 Hz, 1H),

8.05 (s, 1H), 7.55−7.49 (m, 3H), 6.91 (dt, J = 9.2, 2.4 Hz, 1H), 6.78 (s,

1H), 4.14−4.09 (m, 6H), 3.78 (s, 3H), 3.60−3.55 (m, 4H), 2.02−1.93

(m, 4H).

Compound 29. To a solution of 1,2:5,6-bis-O-(1-methylethyli-

dene)-^D-chiroinositol (2.0 g, 7.7 mmol) in acetone (10 mL) was added

CaCO₃(1.0 g), and the resulting mixture was treated with NaIO₄(2.0 g)

in water (20 mL) at room temperature for 2 h. The mixture was ex-

tracted with EtOAc (2 × 10 mL), and the combined organic solution

was dried and evaporated for the next step. The residue was dissolved in

MeOH (20 mL) and treated with NaBH4 (1.0 g, excess) at room

temperature for 30 min. After aqueous workup, the product was puriﬁed

by a ﬂash column (DCM/MeOH = 100:8) to give a white solid (1.3 g,

65%).

Compound 30. A solution of 2,3:4,5-bis-O-(1-methylethylidene)-^D-

mannitol 29 (5.5 g, 26.8 mmol) in THF (100 mL) was treated with

sodium hydride (0.8 g, 20 mmol, 60% in mineral oil) at room tem-

perature and stirring for 1 h. 2-Nitro-4-chloro-5-ﬂuoroanisole (4.0 g,

19.5 mmol) was added, and the resulting mixture was warmed to 60 °C

for 15 min. Solvent was evaporated and the residue puriﬁed with a ﬂash

column chromatography (elution, DCM/MeOH = 100:0 to 5) to give

the title compound (4.4 g, 50%) as a yellow solid. ¹H NMR(400 MHz,

CDCl₃) δ 8.10 (s, 1H), 6.64 (s, J = 8.4, 1H), 6.70 (ddd, J = 10.8, 6.4, 4.8

Hz, 1H), 4.56 (dd, J = 6.4, 4.4 Hz, 1H), 4.45−4.39 (m, 2H), 4.33 (dd, J =

11.6, 5.2 Hz, 1H), 4.20 (dd, J = 9.2, 4.4 Hz), 4.01 (s, 3H), 3.84−3.76 (m,

2H), 2.39 (dd, J = 7.2, 5.6 Hz, 1H), 1.57 (s, 3H), 1.51 (s, 3H), 1.46 (s,

3H), 1.27 (s, 3H).

Compound 31. To a solution of triphenylphosphine (150 mg,

0.57 mmol) in DCM (3.9 mL) was added DIAD (100 uL, 0.51 mmol)

at −20 °C, and the mixture was warmed to 0 °C for 10 min. A mixture of

compounds 30 (168 mg, 0.37 mmol) and 25 (90 mg, 0.36 mmol) was

introduced. After 10 min at room temperature, the reaction vial was

capped and heated at 60 °C for 10 min. Solvent was evaporated and the

residue was puriﬁed with two preparative TLC plates (elution, DCM/

AcOEt = 100:10) to give the title compound (152 mg, 61%) as a white

solid. ¹H NMR(400 MHz, CDCl₃) δ 9.29 (dd, J = 2.4, 0.8 Hz, 1H), 8.39

(dd, J = 8.4, 2.0 Hz, 1H), 8.08 (s, 1H), 7.78 (dd, J = 8.4, 0.8 Hz, 1H), 7.48

(dd, J = 2.0, 1.6 Hz, 1H), 7.39 (ddd, J = 9.6, 2.4, 1.6 Hz, 1H), 6.75 (ddd,

J = 10.0, 2.4, 2.0 Hz, 1H), 6.63 (s, 1H), 4.65−4.70 (m, 2H), 4.63 (dd, J =

6.0, 4.4 Hz, 1H), 4.53 (dd, J = 6.0, 4.4 Hz, 1H), 4.45 (dd, J = 9.6, 7.6 Hz,

1H), 3.34 (dd, J = 10.0, 7.6 Hz, 1H), 4.20 (dt, J = 9.2, 4.4 Hz, 2H), 4.01

(s, 3H), 4.00 (s, 3H), 1.54 (s, 3H), 1.53 (s, 3H), 1.36 (s, 3H), 1.33

(s, 3H).

Compound 23. Compound 23 was synthesized according to the

procedure described.⁴¹

Compound 24. A solution of 3,3′-dihydroxydipropyl ether 15

(500 mg, 3.73 mmol) in THF (10 mL) was treated with NaH (77 mg,

1.9 mmol, 60% in mineral oil) at room temperature and stirring for 1 h.

2-Nitro-4-chloro-5-ﬂuoroanisole (400 mg, 1.62 mmol) was added, and

the resulting mixture was warmed to 60 °C for 15 min. Solvent was

evaporated and the residue was puriﬁed by a ﬂash column chro-

matography (elution, DCM/MeOH = 100:0 to 5) to give the title

1

compound (330 mg, 64%) as a yellow solid. H NMR (400 MHz,

CDCl₃) δ 8.09 (s, 1H), 6.57 (s, 1H), 4.22 (t, J = 6.0 Hz, 2H), 4.00 (s,

3H), 3.77 (t, J = 6.0 Hz, 2H), 3.69 (t, J = 6.0 Hz, 2H), 3.66 (t, J = 6.0 Hz,

2H), 2.16 (quint, J = 6.0 Hz, 2H), 1.85 (quint, J = 6.0 Hz, 2H).

Compound 25. 2-Chloro 6-methylpyridinylcarboxylate (344 mg,

2.0 mmol), 3-ﬂuoro-5-hydroxyphenylboronic acid (406 mg, 2.6 mmol),

and Pd(PPh₃)₄(110 mg, 0.095 mmol) were weighed into a 10 mL

round-bottom ﬂask under Ar. Dioxane (12 mL) was added to the

mixture, followed by 2.5 mL of K₂CO₃solution (1.6 M, 4 mmol). The

content was stirred at 80 °C overnight. To the mixture were added

EtOAc and 5% citric acid. The organic layer was separated, and the

aqueous layer was extracted with more EtOAc. Combined organic layers

were dried over Na₂SO₄, concentrated. Crude product was puriﬁed by

ﬂash chromatography (EtOAc/hexanes) to give 350 mg product as a

white solid (yield, 54%).

Compound 26. To a solution of triphenylphosphine (613 mg,

2.34 mmol) in THF (3.0 mL) was added DIAD (400 uL, 2.0 mmol)

at −20 °C, and the mixture was warmed to 0 °C for 10 min. A mixture of

compound 16 (500 mg, 1.56 mmol) and compound 35 (386 mg, 1.56

mmol) was introduced. After 10 min at room temperature, the reaction

vial was capped and heated at 60 °C for 10 min. Solvent was evaporated

and the residue was puriﬁed by ﬂash column chromatography (elution,

DCM/MeOH = 100:3) to give the title compound (560 mg, 65%) as a

1

white solid. H NMR (400 MHz, CDCl₃) δ 9.30 (dd, J = 2.0, 0.8 Hz,

1H), 8.40 (dd, J = 8.0, 2.0 Hz, 1H), 8.07 (s, 1H), 7.80 (dd, J = 8.4, 0.8 Hz,

1H), 7.49 (t, J = 1.6 Hz, 1H), 7.35 (ddd, J = 9.6, 2.0, 1.2 Hz, 1H), 6.72

(dt, J = 10.4, 2.4 Hz, 1H), 6.54 (s, 1H), 4.24 (t, J = 6.0 Hz, 2H), 4.18 (t,

J = 6.0 Hz, 2H), 4.04 (s, 3H), 3.95 (s, 3H), 3.74 (t, J = 6.0 Hz, 2H), 2.20

(quint, J = 6.0 Hz, 2H), 2.13 (quint, J = 6.0 Hz, 2H).

Compound 27. A solution of 26 (279 mg, 0.51 mmol) in methanol

(8 mL) was charged with Pd−C (0.5 g). The mixture was hydrogenated

under a hydrogen balloon for 15 min. The catalyst was ﬁltered oﬀ and

solvent was evaporated to give the title compound (230 mg, 87%) as a

yellow solid. ¹H NMR (400 MHz, CDCl₃) δ 9.25 (dd, J = 2.0, 0.8 Hz,

1H), 8.33 (dd, J = 8.4, 2.4 Hz, 1H), 7.74 (dd, J = 8.4, 0.8 Hz, 1H), 7.41 (t,

J = 1.6 Hz, 1H), 7.33 (ddd, J = 9.6, 2.0, 1.6 Hz, 1H), 6.69 (dt, J = 10.0, 2.4

Hz, 1H), 6.68 (s, 1H), 6.48 (s, 1H), 4.10 (t, J = 6.0 Hz, 2H), 4.04 (t, J =

6.0 Hz, 2H), 3.97 (s, 3H), 3.78 (s, 3H), 3.67 (t, J = 6.0 Hz, 2H), 3.65 (t,

J = 6.4 Hz, 2H), 2.10−2.02 (m, 4H).

Compound 28. To a solution of 27 (230 mg, 0.44 mmol) in THF

(4.0 mL) and triethylamine (0.3 mL) was added chloroacetyl chloride

(2.0 M in THF) at −20 °C, monitoring the reaction with HPLC until

the starting material disappeared. 6-Mercaptonicotinic acid methyl ester

(105 mg, 0.62 mmol) was introduced. The resulting mixture was heated

in a sealed tube at 100 °C for 5 min. Solvent was evaporated and the

residue was puriﬁed by using two preparative TLC plates (elution,

DCM/THF = 100:5) to give the title compound (180 mg, 56%) as a

yellow solid. ¹H NMR (400 MHz, CDCl₃) δ 9.57 (s, 1H), 9.27 (dd, J =

2.4, 0.8 Hz, 1H), 9.11 (dd, J = 2.4, 0.8 Hz, 1H), 8.39 (s, 1H), 8.37 (dd, J =

8.4, 2.4 Hz, 1H), 8.13 (dd, J = 4.4, 2.0 Hz, 1H), 7.76 (dd, J = 8.4, 0.8 Hz,

1H), 7.42 (t, J = 1.2 Hz, 1H), 7.35 (dd, J = 8.4, 0.8 Hz, 1H), 7.33 (m,

1H), 6.68 (dt, J = 8.4, 2.0 Hz, 1H), 6.44 (s, 1H), 4.12 (t, J = 6.4 Hz, 2H),

Compound 32. A suspension of compound 31 (550 mg, 0.81 mmol)

in acetone (5.0 mL) and water (0.5 mL) was treated with zinc (1.0 g,

excess) and ammonium chloride (0.5 g) at room temperature for 15 min.

Organic layer was transferred to a new vial, and the residue was extracted

with ethyl acetate (3.0 mL). Combined organic solution was dried and

evaporated. The residue was puriﬁed using ﬂash column chromatography

(elution, DCM/MeOH = 100:3) to give the title compound (410 mg,

78%) as an oﬀ white solid. ¹H NMR(400 MHz, CDCl₃) δ 9.27 (dd, J =

2.4, 0.8 Hz, 1H), 8.36 (dd, J = 8.0, 2.0 Hz, 1H), 7.60 (dd, J = 8.4, 0.8 Hz,

1H), 7.45 (t, J = 1.6 Hz, 1H), 7.40 (ddd, J = 9.2, 2.0, 1.2 Hz, 1H), 6.74

(ddd, J = 10.0, 2.4, 2.0 Hz, 1H), 6.70 (s, 1H), 6.54 (s, 1H), 4.72 (dd, J =

11.6, 6.0 Hz, 1H), 4.63 (t, J = 6.4 Hz, 1H), 4.62−4.59 (m, 1H), 4.59 (dd,

J = 6.4, 6.0 Hz, 1H), 4.29 (dd, J = 9.6, 6.4 Hz, 1H), 4.25 (dd, J = 9.6, 7.6 Hz,

1H), 4.15 (dd, J = 9.6, 5.2 Hz, 1H), 4.01 (dd, J = 9.6, 4.4 Hz, 1H), 3.99 (s,

3H), 3.83 (s, 3H), 1.55 (s, 3H), 1.54 (s, 3H), 1.40 (s, 3H), 1.39 (s, 3H).

N

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Compound 33. A solution of 32 (220 mg, 0.39 mmol) in THF

(3.0 mL) and triethylamine (60 μL) was treated with chloroacetyl

chloride (250 uL, 2.0 M solution in THF). The mixture was warmed up

to room temperature for 5 min, and then 6-mercaptonicotinic acid

methyl ester (130 mg, 0.77 mmol) was added. The mixture was heated

in a sealed tube at 100 °C for 5 min. Solvent was evaporated and the

residue puriﬁed with a preparative TLC plate (DCM/MeOH = 100:8)

to give an oﬀ white solid (140 mg, 46%). ¹H NMR (400 MHz, DMSO) δ

9.63 (s, 1H), 9.18 (dd, J = 2.0, 0.8 Hz, 1H), 8.96 (d, J = 1.2 Hz, 1H), 8.37

(dd, J = 8.4, 2.0 Hz, 1H), 8.21 (dd, J = 8.4, 0.8 Hz, 1H), 8.14 (dd, J = 8.4,

2.0 Hz, 1H), 8.04 (s, 1H), 7.63 (t, J = 2.0 Hz, 1H), 7.58 (d, J = 8.4 Hz,

1H), 7.57 (dm, J = 9.2 Hz, 1H), 6.98 (dt, J = 9.2, 2.4 Hz, 1H), 6.90 (s,

1H), 4.95 (d, J = 6.0 Hz, 1H), 4.9 (d, J = 6.0 Hz, 1H), 4.45 (d, J = 6.4 Hz,

1H), 4.43 (d, J = 6.0 Hz, 1H), 4.36−4.31 (m, 2H), 4.16 (s, 2H), 4.14 (dd,

J = 9.2, 5.2 Hz, 1H), 4.14−4.11 (m, 2H), 3.93 (s, 3H), 3.88 (s, 3H), 3.86

(s, 3H), 3.87−3.77 (m, 2H).

Compound 8. A solution of 33 (80 mg, 0.10 mmol) in methanol

(0.5 mL) was treated with potassium carbonate (300 mg) and water

(0.2 mL) in a capped vial at 80 °C for 5 min. Organic layer was

transferred to a new vial, and then drops of formic acid were added. The

pH was checked to ensure that the mixture is mildly acidic. The mixture

was puriﬁed using preparative HPLC to give the title compound (31.5

mg, 41%) as an oﬀ white solid. MS (ES⁺) m/z (M + 1) 748; ¹H NMR

(400 MHz, DMSO) δ 13.36 (br, 1H), 9.67 (s, 1H), 9.14 (dd, J = 2.0, 0.8

Hz, 1H), 8.94 (d, J = 1.6 Hz, 1H), 8.32 (dd, J = 8.4, 2.0 Hz, 1H), 8.15 (d,

J = 8.0 Hz, 1H), 8.11 (dd, J = 8.4, 2.0 Hz, 1H), 8.05 (s, 1H), 7.61 (d, J =

1.6 Hz, 1H), 7.54 (dt, J = 9.6, 2.0 Hz, 1H), 6.95 (dt, J = 9.6, 2.4 Hz, 1H),

6.89 (s, 1H), 4.93−4.90 (m, 2H), 4.43 (m, 2H), 4.32 (t, J = 8.0 Hz, 2H),

4.14−4.10 (m, 2H), 4.13 (s, 2H), 3.88−3.77 (m, 2H), 3.85 (s, 3H).

Compound 34. To a solution of 30 (2.0 g, 4.5 mmol) in DCM

(80 mL) was added molecular sieves (3.0 g, 4 Å) and PDC (3.0 g,

8.0 mmol). The mixture was stirred at room temperature for 2 h and

then diluted with ether (100 mL). The mixture was ﬁltered through a

Celite pad, and solvent was evaporated. The residue was puriﬁed with

silica gel column chromatography (elution, DCM/MeOH = 100:0 to 2)

to give the title compound (1.4 g, 70%) as a light yellow solid. ¹H NMR

(400 MHz, CDCl₃) δ 8.08 (s, 1H), 6.62 (s, 1H), 4.83 (dd, J = 6.8, 2.4 Hz,

1H), 4.72 (dd, J = 11.2, 4.4 Hz, 1H), 4.69−4.65 (m, 1H), 4.49 (dd, J =

6.8, 2.4 Hz, 1H), 4.41 (dd, J = 9.2, 7.6 Hz, 1H), 4.24−4.19 (m, 1H), 4.18

(s, 1H), 4.15 (s, 1H), 3.99 (s, 3H), 3.47 (s, 3H), 1.57 (s, 3H), 1.50 (s,

3H), 1.45 (s, 3H), 1.20 (s, 3H),

Compound 35. To a solution of 34 (450 mg, 1.0 mmol) in DCM

(5.0 mL) were added molecular sieves (0.5 g) and compound 15

(250 mg, 1.0 mmol). The mixture was stirred for 5 min, and NaBH(OAc)₃

(0.5 g, 2.3 mmol) was added. The resulting mixture was stirred at room

temperature for 2 h and diluted with aqueous sodium bicarbonate.

Organic layer was dried, evaporated and the residue was puriﬁed with

silica gel column chromatography (elution, DCM/MeOH = 100:3 to 5)

to give the title compound 35 (470 mg, 70%) as a yellow solid. ¹H NMR

(400 MHz, CDCl₃) δ 9.27 (dd, J = 2.4, 0.8 Hz, 1H), 8.35 (dd, J = 8.4, 2.0

Hz, 1H), 8.08 (s, 1H), 7.75 (dd, J = 7.6, 0.8 Hz, 1H), 7.29 (s, 1H), 7.20

(t, J = 2.0 Hz, 1H), 7.06 (dd, J = 9.6, 2.0 Hz, 1H), 6.63 (s, 1H), 6.44 (dt,

J = 10.8), 6.68 (dt, J = 10.0, 4.4 Hz, 1H), 4.52−4.46 (m, 4H), 4.43 (dd,

J = 9.2, 8.4 Hz, 1H), 4.26 (dt, J = 9.2, 1.6 Hz, 1H), 4.19 (dd, J = 9.6, 4.4

4.00 (s, 6H), 3.56−3.52 (m, 1H), 3.42 (dd, J = 12.4, 6.4 Hz, 1H), 1.58 (s,

3H), 1.53 (s, 3H), 1.47 (s, 3H), 1.29 (s, 3H)

Compound 36. A suspension of 35 (100 mg, 0.15 mmol) in acetone

(1.0 mL) and water (0.3 mL) was treated with zinc (0.3 g) and ammo-

nium chloride (0.3 g) at room temperature for 10 min. Organic layer was

transferred to a new vial. The residue was extracted with acetone

(1.0 mL), and the combined organic solution was evaporated. The residue

was dissolved in methanol (1.0 mL) and then treated with HCl (2.0 mL,

2.0 N). The mixture was warmed to 60 °C for 10 min, and solvent was

evaporated. The residue was treated with wet sodium bicarbonate and

extracted with methanol (3 mL). Organic solution was evaporated, and

the residue was used for next step.

0.59 mmol) was introduced, and the resulting mixture was heated in a

sealed tube at 100 °C for 5 min. Solvent was evaporated and the residue

was puriﬁed using preparative HPLC to give the title compound 14 (45

mg, 39% two steps) as a yellow solid. MS (ES⁺) m/z (M + 1) 775; ¹H

NMR (400 MHz, DMSO) δ 9.62 (s, 1H), 9.13 (dd, J = 2.4, 0.8 Hz, 1H),

8.95 (d, J = 1.6 Hz, 1H), 8.33 (dd, J = 8.4, 2.4 Hz, 1H), 8.13 (dd, J = 8.4,

2.4 Hz, 1H), 8.03 (d, J = 8.4, Hz, 1H), 8.02 (s, 1H), 7.57 (d, J = 8.4 Hz,

1H), 7.30 (t, J = 1.6 Hz, 1H), 7.05 (dt, J = 10.0, 2.4 Hz, 1H), 6.88 (s, 1H),

6.56 (dt, J = 10.0, 2.4 Hz, 1H), 6.01 (t, J = 2.0 Hz, 1H), 4.89 (d, J = 6.0

Hz, 1H), 4.76 (d, J = 7.2 Hz, 1H), 4.42 (d, J = 7.2 Hz, 1H), 4.31 (d, J =

8.8 Hz, 1H), 4.15 (s, 2H), 4.12 (dd, J = 10.4, 6.0 Hz, 1H), 3.91 (s, 3H),

3.87 (s, 3H), 3.85 (s, 3H), 3.77 (t, J = 8.4 Hz, 3H), 3.72−3.67 (m, 2H),

3.52−3.47 (m, 1H), 3.10−3.04 (m, 1H).

Compound 9. A solution of 14 (95 mg, 12 mmol) in MeOH

(0.5 mL) was treated with potassium carbonate (300 mg) and water

(0.2 mL) in a capped vial at 80 °C for 5 min. Organic layer was

transferred to a new vial, and then drops of formic acid were added. The

pH was checked to ensure that the mixture is mildly acidic. The mixture

was puriﬁed using preparative HPLC to give the title compound (38 mg,

42%) as a yellow solid. MS (ES⁺) m/z (M + 1) 747; ¹H NMR(400 MHz,

DMSO) δ 13.19 (br, 1H), 11.66 (s, 1H), 9.65 (s, 1H), 9.11 (dd, J = 2.4,

0.8 Hz, 1H), 8.94 (d, J = 1.2 Hz, 1H), 8.30 (dd, J = 6.0, 2.4 Hz, 1H), 8.10

(dd, J = 8.4, 2.4 Hz, 1H), 8.04 (s, 1H), 8.00 (dd, J = 8.4, 0.8 Hz, 1H), 7.54

(d, J = 8.4 Hz, 1H), 7.29 (t, 1.6 Hz, 1H), 7.04 (ddd, J = 9.6, 2.0, 1.2 Hz),

6.88 (s, 1H), 6.54 (ddd, J = 12.0, 2.4, 2.0 Hz, 1H), 5.99 (t, J = 4.8 Hz,

1H), 6.89 (d, J = 6.0 Hz, 1H), 4.76 (br, 1H), 4.41−4.25 (m, 3H), 4.13 (s,

2H), 4.11−4.09 (m, 1H), 3.84 (s, 3H), 3.82−3.65 (m, 4H), 3.52−3.47

(m, 1H), 3.10−3.03 (m, 1H).

Compound 37. A solution of 1-chloro-2-ﬂuoro-4-methoxy-5-

nitrobenzene (4.1 g, 20 mmol) (19) in DMSO (4.0 mL) was treated

with potassium hydroxide (5.0 g) in water (5.0 mL). The mixture was

heated slowly to 100 °C and then kept at this temperature for 1 h.

The mixture was cooled to room temperature and diluted with DCM

(50 mL) and then treated with KHSO₄(excess). The organic solution

was transferred, and the residue was washed with DCM (50 mL). The

combined organic solution was dried and evaporated. The resulting

solid was washed with water and a small amount DCM to give a yellow

solid (3.5 g, 86%). ¹H NMR (400 MHz, CD₃Cl) δ 8.11 (s, 1H), 6.76 (s,

1H), 3.98 (s, 3H).

Compound 38. A mixture of 37 (5.89 g, 28.9 mmol), Zn (7.2 g),

NH₄Cl (10 g), and water (13 mL) in 64 mL of acetone was stirred at

room temperature for 30 min. After another 2 g of zinc was added, the

mixture was stirred for another 15 min. The mixture was evaporated, and

250 mL of DCM was added. The mixture was stirred for 15 min and

ﬁltered. The residue was washed with EtOAc. The combined organics

were dried, ﬁltered, and evaporated to a dark brown solid (2.6 g, 52%).

Compound 39. A mixture of ethyl 6-chloro-3-pyridinecarboxylate

(1.85 g, 10 mmol) and mercaptoacetic acid (1.2 g, 10 mmol) was heated

without solvent at 100 °C for 10 min. The mixture was cooled to room

temperature and washed with methanol (∼10 mL) to give a white solid

(2.1 g, 87%). ¹H NMR (400 MHz, CDCl₃) δ 13.06 (br, 1H), 8.24 (d, J =

2.0 Hz, 1H), 8.04 (dd, J = 9.6, 2.4 Hz, 1H), 6.60 (d, J = 9.6 Hz, 1H), 4.25

(q, J = 7.2 Hz, 2H), 1.38 (t, J = 7.2 Hz, 3H).

Compound 40. To a solution of the acid 39 (2.4 g, 9.95 mmol) and

38 (1.83 g, 10.5 mmol) in 70 mL of DMF were added DCC (2.63 g, 12.7

mmol) and HOBt (1.86 g, 13.8 mmol) at room temperature. The

mixture was stirred at room temperature overnight and evaporated. The

residue was worked up with DCM/saturated NaHCO₃and chromato-

graphed (EtOAc/heptane 1:3 to 1:1) to give a light brown solid (1.82 g,

46%). ¹H NMR (CDCl₃) (ppm) 9.66 (1H, s), 9.05 (1H, s), 8.35 (1H, s),

8.06 (dd, J = 6.2, 2.2 Hz, 1H), 7.27 (dd, J = 7.6, 0.8 Hz, 1H), 6.46 (1H, s),

4.36 (q, J = 7.1 Hz, 2H), 3.91 (2H, s), 3.71 (3H, s), 1.34 (t, J = 7.1 Hz,

3H).

Compound 41. Compound 25 (405 mg, 1.64 mmol) and K₂CO₃

(678 mg, 4.9 mmol) were suspended in 5 mL of anhydrous THF in a

25 mL round-bottom ﬂask under Ar. The content was cooled to 0 °C, and

then tert-butyl bromoacetate (955 mg, 2.0 mmol) was added in one

portion. The reaction mixture was stirred at room temperature

overnight. The solvent was removed on a rotavapor, and the residue

was partitioned with EtOAc/water. The organic layer was separated, and

Compound 14. To a solution of 36 in THF (2.0 mL) and triethyl

amine (0.1 mL) was added chloroacetyl chloride (2.0 M solution in

THF) at 0 °C, monitoring the reaction with HPLC until the starting

material disappeared. 6-Mercaptonicotinic acid methyl ester (100 mg,

O

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

the aqueous layer was extracted with EtOAc once more. The combined

organic layers were dried over anhydrous Na₂SO₄. The ﬁnal product was

puriﬁed by ﬂash chromatography (EtOAc/hexanes) to give the ﬁnal

product as a thick oil (546 mg, yield 92%).

¹H NMR (400 MHz, DMSO) δ 9.14 (dd, J = 2.4, 0.8 Hz, 1H), 8.36 (dd,

J = 8.0, 2.0 Hz, 1H), 8.18 (dd, J = 8.4, 0.8 Hz, 1H), 8.05 (s, 1H), 7.60−

7.56 (m, 2H), 7.05 (ddd, J = 10.8, 2.4, 2.0 Hz, 1H), 7.02 (s, 1H), 4.66−

4.61 (m, 2H), 4.67−4.46 (m, 3H), 4.34−4.27 (m, 2H), 4.15 (dd, J =

10.0, 6.4 Hz, 1H), 4.00 (s, 3H), 3.92 (s, 3H), 1.46 (s, 3H), 1.44 (s, 3H),

1.32 (s, 3H), 1.30 (s, 3H).

Compound 48. A suspension of 47 (450 mg, 0.66 mmol) in ace-

tone (3.0 mL) and water (0.5 mL) was treated with zinc (3.0 g) and

ammonium chloride (1.5 g) at room temperature for 10 min. Organic

layer was transferred to a new vial. The residue was extracted with

AcOEt (6.0 mL), and the combined organic solution was evaporated.

The residue was dissolved in methanol (2.0 mL) and treated with HCl

(4.0 mL, 2N) at 60 °C for 15 min. The solvent was evaporated to dry and

the residue was dissolved in a mixture of THF (3.0 mL) and methanol

(1.5 mL). The mixture was treated with wet sodium bicarbonate. After

ﬁltration, the organic solution was evaporated and the residue was used

for next step.

Compound 42. To compound 41 (314 mg, 0.87 mmol) in a 5 mL

round-bottom ﬂask was added TFA (1.3 mL). The solution was stirred

at room temperature overnight. The solvent was removed in vacuo and

the crude product was obtained (263 mg, 100%) as a white solid.

Compound 43. To a stirred solution of acid 42 (270 mg, 0.88

mmol), 3-amino 1-propanol (63 mg, 0.88 mmol), and HOBt (150 mg,

1.1 mmol) in DMF (9 mL) was added EDCI (300 mg, 1.49 mmol) in

one portion at room temperature followed by addition of TEA (0.3 mL,

2.2 mmol), and the resultant mixture was stirred for overnight. Solvent

was stripped under vacuum, and the residue was extracted with EtOAc

(3 × 15 mL) and washed with 5% NaHCO₃solution. The combined

organic extracts were dried over Na₂SO₄and concentrated in vacuo. The

crude material was then ﬂash-chromatographed over silica gel, eluting

with hexane−EtOAc (2:8) to furnish the desired product (220 mg,

69%). ¹H NMR (DMSO-d₆, 400 MHz) δ 1.60 (m, 2H), 3.57 (m, 2H),

3.70 (s, 3H), 3.90 (m, 2H), 4.70 (s, 2H), 7.00 (m, 1H), 7.61 (m, 1H,),

7.75 (s, 1H) 8.15 (d, 1H, J = 8.8 Hz) 8.25 (m, 1H,) 8.40 (dd, 1H, J = 8.1,

2.0 Hz), 9.15 (dd, 1H, J = 2.0, 0.8 Hz). MS (ES+) m/z 362 (M + 1).

Compound 44. To a solution of triphenylphosphine (53 mg, 0.57

mmol) in THF (2.5 mL) was added DIAD (41 uL, 0.2 mmol) at 0 °C,

and the mixture was stirred for 5 min. A mixture of the above compound

43 (73 mg, 0.2 mmol) and phenol 40 (80 mg, 0.2 mmol) were added

together. The mixture was then stirred at room temperature overnight.

Solvent was evaporated and the crude product was subjected to ﬂash

chromatography over silica gel, eluting with EtOAc to aﬀord the product

(50 mg, 33%). ¹H NMR (MeOD-d₄, 400 MHz) δ 1.40 (t, 3H), 1.91 (m,

2H), 3.55 (m, 2H), 3.70 (s, 3H), 3.90 (s, 2H), 3.91 (m, 2H), 4.33 (q,

2H), 4.70 (s, 2H), 6.80 (m, 1H), 7.33 (m, 1H,), 7.44 (m, 2H) 7.79 (d,

1H J = 9.2 Hz,) 8.00 (s, 1H,) 8.10 (dd, 1H, J = 8.1, 2.0 Hz), 8.25 (dd, 1H,

J = 8.1, 2.0 Hz), 8.50 (s, 1H), 9.10 (dd, 2H, J = 1.6, 9.6 Hz). MS (ES+)

m/z 740.17 (M + 1).

Compound 10. A solution of diester 44 (30 mg, 0.04 mmol) in

MeOH (1.5 mL) was treated with potassium carbonate (200 mg) and

water (0.6 mL) in a sealed microwave vial heated at 90 °C for 15 min.

The mixture was sequestered with formic acid and directly injected on a

preparative HPLC column, eluting with a mixture of ACN−H₂O with

0.1% HCO₂H to yield an oﬀ white solid (6 mg, 21%). ¹H NMR (MeOD-

d₄, 400 MHz) δ 1.91 (m, 2H), 3.55 (m, 2H), 3.90 (s, 2H), 3.91 (m, 2H),

4.70 (s, 2H), 6.80 (m, 1H), 7.33 (m, 3H), 7.70 (d, 1H, J = 8.4 Hz) 8.15

(s, 1H), 8.20 (dd, 1H, J = 8.1, 2.0 Hz), 8.30 (m, 3H), 9.10 (dd, 2H, J =

1.2, 10.6 Hz). MS (ES+) m/z 698.12 (M + 1).

Compound 45. 2,3:4,5-di-O-Isopropylidenemannitol (45) was

synthesized from 1,2:5,6-bis-O-(1-methylethylidene)-^D-chiroinositol

(6.5 g, 25 mmol) by utilizing the procedure as described for the

synthesis of compound 29 in 70% yield as white solid.

Compound 46. A solution of 2,3:4,5-di-O-isopropylidenemannitol

45 (1.5 g, 5.7 mmol) in THF (30 mL) was treated with NaH (0.2 g, 5.0

mmol, 60% in mineral oil) at room temperature for 0.5 h. 2-Nitro-4-

chloro-5-ﬂuoroanisole (1.0 g, 4.9 mmol) was added, and the resulting

mixture was warmed to 60 °C for 15 min. Solvent was evaporated and

the residue was puriﬁed with a ﬂash column chromatography (elution,

DCM/MeOH = 100:0 to 5) to give the title compound (1.4 g, 64%) as a

yellow solid.

Compound 49. To a solution of 48 in THF (3.0 mL) and tri-

ethylamine (0.3 mL) was added chloroacetyl chloride (2.0 M solution in

THF) at −20 °C, monitoring the reaction with HPLC until the starting

material disappeared. 6-Mercaptonicotinic acid methyl ester (200 mg,

1.2 mmol) was introduced. The resulting mixture was heated in a sealed

tube at 100 °C for 5 min. Solvent was evaporated and the residue was

puriﬁed with two preparative TLC plates (elution, DCM/MeOH/

THF = 100:7:10) to give the title compound (160 mg, 31% three steps)

1

as a white solid. MS (ES⁺) m/z (M + 1) 776; H NMR (400 MHz,

DMSO) δ 9.62 (s, 1H), 9.17 (dd, J = 2.4, 0.8 Hz, 1H), 8.95 (d, J = 1.6 Hz,

1H), 8.36 (dd, J = 8.4, 2.4 Hz, 1H), 8.20 (dd, J = 8.4, 0.8 Hz, 1H), 8.13

(dd, J = 8.4, 2.4 Hz, 1H), 8.03 (s, 1H), 7.62 (d, J = 1.6 Hz, 1H), 7.58−

7.54 (m, 2H), 6.97 (ddd, J = 10.8, 2.4, 2.0 Hz, 1H), 6.89 (s, 1H), 4.94 (d,

J = 6.0 Hz, 2H), 4.91 (d, J = 6.0 Hz, 2H), 4.43 (d, J = 6.8 Hz, 2H), 4.42

(d, J = 6.8 Hz, 2H), 4.32 (m, 2H), 4.15 (s, 2H), 4.14−4.11 (m, 2H), 3.92

(s, 3H), 3.88 (s, 3H), 3.85 (s, 3H), 3.83−3.76 (m, 2H), 3.17 (d, J =

5.6, 1H).

Compound 50. To a solution of 36 in THF (2.0 mL) and tri-

ethylamine (0.1 mL) was added chloroacetyl chloride (2.0 M solution in

THF) at 0 °C, monitoring the reaction with HPLC until the starting

material disappeared. Mercaptopyridine (35 mg, 0.31 mmol) was

introduced. The resulting mixture was heated in a sealed tube at 100 °C

for 5 min. Solvent was evaporated. The residue was treated with aqueous

sodium bicarbonate, and extraction was with DCM. Organic solution

was evaporated, and the residue was used for next step.

Compound 12. A solution of 50 in methanol (0.5 mL) was treated

with potassium carbonate (300 mg) and water (0.2 mL) in a capped vial

at 80 °C for 5 min. Organic layer was transferred to a new vial, and then

drops of formic acid were added. The pH was checked to ensure that

the mixture is mildly acidic. The mixture was puriﬁed using preparative

HPLC to give the title compound (11 mg, 10% from 36) as a yellow

solid. MS (ES⁺) m/z (M + 1) 703; ¹H NMR (400 MHz, DMSO) δ 13.35

(br, 1H), 9.73 (s, 1H), 9.10 (d, J = 1.6 Hz, 1H), 8.52 (d, J = 4.0 Hz, 1H),

8.28 (dd, J = 8.4, 2.0 Hz, 1H), 8.10 (s, 1H), 7.97 (d, J = 8.0 Hz, 1H), 7.72

(td, J = 7.6, 2.0 Hz, 1H), 7.43 (d, J = 8.0 Hz, 1H), 7.29 (s, 1H), 7.22 (ddd,

J = 7.2, 4.8, 0.8 Hz, 1H), 7.04 (dm, J = 10.0, 1H), 6.88 (s, 1H), 6.50 (dm,

J = 11.6 Hz), 5.99 (t, J = 5.6 Hz, 1H), 4.90 (d, J = 5.6 Hz, 1H), 4.90 (d, J =

5.6 Hz, 1H), 4.77 (d, J = 4.4 Hz, 1H), 4.44 (d, J = 7.2 Hz, 1H), 4.37 (d,

J = 7.2 Hz, 1H), 4.31 (d, J = 4.4 Hz, 1H), 4.12 (dd, J = 10.4, 6.4 Hz, 1H),

4.03 (s, 2H), 3.88 (m, 1H), 3.84 (s, 3H), 3.77 (t, J = 8.4 Hz, 1H), 3.69

(m, 1H), 3.55−3.47 (m, 1H), 3.10−3.04 (m, 1H).

Compound 53. To a solution of 30 (220 mg, 0.49 mmol) in DCM

(3.0 mL) were added molecular sieves (0.5 g) and compound 52 (96

mg, 0.51 mmol). The resulting mixture was treated with NaHB(OAc)₃

(250 mg, 1.2 mmol) at room temperature for 2 h. Drops of water were

added, and organic solution was transferred to a new vial. Solvent was

evaporated and the residue was puriﬁed by using two preparative TLC

plates (elution, DCM/MeOH = 100:3) to give the title compound (250

mg, 83%) as a yellow solid. ¹H NMR (400 MHz, CDCl₃) δ 8.71 (dt, J =

4.8, 0.8 Hz, 1H), 8.07 (s, 1H), 7.82 (dd, J = 8.0, 1.2 Hz, 1H), 7.71 (d, J =

8.0 Hz, 1H), 7.31 (td, J = 7.2, 0.8 Hz, 1H), 7.19 (s, 1H), 6.99 (ddd, J =

9.6, 2.0, 1.6 Hz, 1H), 6.63 (s, 1H), 6.42 (ddd, J = 10.8, 2.4, 2.0 Hz, 1H),

4.71−4.68 (m, 1H), 4.54−4.47 (m, 3H), 4.43 (dd, J = 9.6, 8.0 Hz, 1H),

¹H NMR (400 MHz, CDCl₃) δ 8.10 (s, 1H), 6.64 (s, 1H), 4.70 (ddd,

J = 10.8, 6.4, 4.8 Hz, 1H), 4.56 (dd, J = 6.4, 4.4 Hz, 1H), 4.45−4.39 (m,

2H), 4.33 (dd, J = 11.6, 5.2 Hz, 1H), 4.21 (dd, J = 9.2, 4.4 Hz, 1H), 4.01

(s, 3H), 3.84−3.76 (m, 2H), 2.40 (br, 1H), 1.57 (s, 3H), 1.51 (s, 3H),

1.46 (s, 3H), 1.27 (s, 3H).

Compound 47. To a solution of triphenylphosphine (280 mg, 1.1

mmol) in DCM (3.0 mL) was added DIAD (135 uL, 0.69 mmol) at

−20 °C, and the mixture was warmed to 0 °C for 10 min. A mixture of 46

(225 mg, 0.50 mmol) and 25 (125 mg, 0.50 mmol) was introduced.

After 10 min at room temperature, the reaction vial was capped and

heated at 60 °C for 10 min. Solvent was evaporated and the residue was

puriﬁed with two preparative TLC plates (elution, DCM/AcOEt =

100:10) to give the title compound (237 mg, 70%) as a light yellow solid.

P

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

4.20 (dd, J = 9.6, 4.4 Hz, 1H), 4.01 (m, 1H), 3.99 (s, 2H), 3.55 (dd,

J = 12.4, 4.8 Hz, 1H), 3.42 (dd, J = 12.4, 7.2 Hz, 1H), 1.58 (s, 3H), 1.52

(s, 3H), 1.47 (s, 3H), 1.28 (s, 3H).

4.85 (s, 2H). Methanol (10 mL) was charged with Pd−C (0.4 g) and

hydrogenated under a hydrogen balloon at room temperature.

Compound 54. A mixture of 53 (350 mg, 0.57 mmol) in acetone

(1.5 mL) was treated with zinc (0.5 g), ammonium chloride (0.3 g), and

water (drops) at room temperature for 10 min. Organic solution was

transferred to a new vial, and the residue was extracted with DCM

(3.0 mL). The combined organic solution was dried and evaporated for

the next step. The residue was dissolved in methanol (1.0 mL) and

treated with HCl (2.0 mL, 2 N) at 60 °C for 15 min. The mixture was

evaporated to dry, and the residue was dissolved in a mixture of THF

(1.0 mL) and methanol (0.5 mL). The resulting solution was treated

with wet sodium bicarbonate. After ﬁltration, the organic solution was

evaporated and the residue was for nest step.

Compound 55. To a solution of 54 in THF (2.0 mL) and tri-

ethylamine (0.05 mL) was added chloroacetyl chloride (2.0 M solution

in THF) at 0 °C, monitoring the reaction with HPLC until the starting

material disappeared. 6-Mercaptonicotinic acid methyl ester (120 mg,

0.71 mmol) was introduced. The resulting mixture was heated in a

sealed tube at 100 °C for 5 min. Solvent was evaporated, and the residue

was puriﬁed with a preparative TLC plate (elution, DCM/MeOH =

100:12) to give the title compound (110 mg, 27% from 53) as an oﬀ

white solid . ¹H NMR (400 MHz, DMSO) δ 9.63 (s, 1H), 8.96 (d, J = 1.6

Hz, 1H), 8.65 (dt, J = 4.4, 1.2 Hz, 1H), 8.14 (dd, J = 8.4, 2.4 Hz, 1H),

7.87 (dd, J = 4.4, 1.2 Hz, 1H), 7.58 (d, J = 8.4 Hz, 1H), 7.53 (m, 2H),

7.23 (t, J = 1.2 Hz, 1H), 6.98 (ddd, J = 10.4, 2.0, 1.2 Hz, 1H), 6.89 (s,

1H), 6.50 (dt, J = 12.0, 2.0 Hz, 1H), 5.91 (t, J = 5.6 Hz, 1H), 4.90 (d, J =

6.0 Hz, 1H), 4.76 (d, J = 6.0 Hz, 1H), 4.43 (d, J = 7.2 Hz, 1H), 4.37 (d,

J = 7.6 Hz, 1H), 4.32 (dd, J = 10.4, 1.6 Hz, 1H), 4.16 (s, 2H), 4.13 (dd,

J = 10.0, 2.0 Hz, 1H), 3.88 (s, 3H), 3.86 (s, 3H), 3.78 (t, J = 8.4 Hz, 1H),

3.74−3.65 (m, 2H), 3.50 (ddd, J = 12.8, 6.4, 2.4 Hz), 3.07 (ddd, J = 12.8,

7.2, 1.2 Hz, 1H).

Compound 13. A solution of 55 (90 mg, 0.13 mmol) in methanol

(0.5 mL) was treated with potassium carbonate (300 mg) and water

(0.2 mL) in a capped vial at 80 °C for 5 min. Organic layer was transferred

to a new vial and diluted with DCM (2.0 mL) and then treated with drops

of formic acid. The pH was checked to be pH 4−5. The yellow solid was

collected by ﬁltration and wash with water (1.5 mL), then DCM (1.0mL).

The dried solid was washed with a small amount of methanol to give the

title compound (48 mg, 53%) as a yellow solid. MS (ES⁺) m/z (M + 1)

703; ¹H NMR (400 MHz, DMSO) δ 9.66 (s, 1H), 8.94 (d, J = 1.2 Hz,

1H), 8.64 (ddd, J = 4.4, 1.6, 1.2 Hz, 1H), 8.11 (dd, J = 8.4, 2.0 Hz, 1H),

8.05 (s, 1H), 7.86 (d, J = 4.4 Hz, 1H), 7.85 (d, J = 4.8 Hz, 1H), 7.54 (d, J =

8.4 Hz, 1H), 7.34 (q, J = 4.4 Hz, 1H), 7.21 (t, J = 1.6 Hz, 1H), 6.97 (ddd,

J = 9.6, 2.0, 1.6 Hz, 1H), 6.48 (ddd, J = 8.0, 2.4, 2.0 Hz, 1H), 5.90 (m, 1H),

4.89 (d, J = 2.0 Hz, 1H), 4.75 (d, J = 4.0 Hz, 1H), 4.42 (d, J = 6.4 Hz, 1H),

3.76 (d, J = 7.2 Hz, 1H), 4.31 (dd, J = 10.4, 2.0 Hz, 1H), 4.15 (s, 2H), 4.10

(d, J = 6.4 Hz, 1H), 3.87 (m, 1H), 3.85 (s, 3H), 3.79−3.04 (m, 4H), 3.51−

3.47 (m, 1H), 3.07−3.03 (m, 1H).

Compound 51. A mixture of 2-bromopyridine (158 mg, 1.0 mmol),

3-(Cbz-amino)-5-ﬂuorophenylboronic acid (289 mg, 1.0 mmol), K₃PO₄

(300 mg), Pd₂(dba)₃(50 mg), 2-(dicyclohexylphosphino)biphenyl

(38 mg), dioxane (2.0 mL), and water (1.0 mL) was degassed and then

heated at 100 °C for 5 min. The reaction mixture was diluted with ethyl

acetate (10 mL). Organic layer was separated and evaporated. The

residue was puriﬁed by using two preparative TLC plates (elution,

heptane/ethyl acetate = 3:1) to give the title compound (263 mg, 82%)

as an oﬀ white solid. ¹H NMR (400 MHz, CDCl₃) δ 8.57 (ddd, J = 4.2,

2.0, 1.2 Hz, 1H), 7.65 (td, J = 8.0, 2.0 Hz, 1H), 7.60 (t, J = 1.6 Hz, 1H),

7.56 (d, J = 8.0 Hz, 1H), 7.36−7.23 (m, 7H), 7.15 (ddd, J = 7.2, 4.8, 0.8

Hz, 1H), 7.30 (br, 1H), 5.12 (s, 2H).

ASSOCIATED CONTENT

* Supporting Information

SPR sensograms, evaluation of reference compounds, and lactate

assay reaction scheme. This material is available free of charge via

the Internet at http://pubs.acs.org.

Accession Codes

The following X-ray crystal structures have been deposited in the

RCSB Protein Data Bank: codes 4I8X, 4I9U, 4I9N, and 4I9H.

■

S

AUTHOR INFORMATION

Corresponding Author

*Phone: 617-621-2349. E-mail: anna.kohlmann@ariad.com.

■

Present Addresses

^†Biogen Idec, 14 Cambridge Center, Cambridge, Massachusetts

02142, United States.

^‡Belfer Institute, Dana Farber Cancer Institute, 77 Louis Pasteur

Avenue, Boston, Massachusetts 02115, United States.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

The authors thank R. Anjum and V. Fantin for their work on

target validation, M. Weigele for helpful discussions, L. Xue for

assistance with analogue searches, and D. Myszka (Biosensor

Tools, LLC) and A. Vinitsky (Aﬃna Biotechnologies, Inc.) for

providing SPR data and helpful discussions.

ABBREVIATIONS USED

■

LDH, lactate dehydrogenase; LDH-A, lactate dehydrogenase A;

LDH-B, lactate dehydrogenase B; STD-NMR, saturation transfer

diﬀerence NMR; SPR, surface plasmon resonance; CA-II,

carbonic anhydrase II; CLM, common ligand mimic

REFERENCES

■

(1) Warburg, O. On respiratory impairment in cancer cells. Science

1956, 124, 269−270.

(2) Vander Heiden, M. G.; Cantley, L. C.; Thompson, C. B.

Understanding the Warburg effect: the metabolic requirements of cell

proliferation. Science 2009, 324, 1029−1033.

(3) Tennant, D. A.; Duran, R. V.; Gottlieb, E. Targeting metabolic

transformation for cancer therapy. Nat. Rev. Cancer 2010, 10, 267−277.

(4) Warburg, O. On the origin of cancer cells. Science 1956, 123, 309−

314.

(5) Ward, P. S.; Thompson, C. B. Metabolic reprogramming: a cancer

hallmark even Warburg did not anticipate. Cancer Cell 2012, 21, 297−

308.

(6) Hanahan, D.; Weinberg, R. A. Hallmarks of cancer: the next

generation. Cell 2011, 144, 646−674.

(7) Eszes, C. M.; Sessions, R. B.; Clarke, A. R.; Moreton, K. M.;

Holbrook, J. J. Removal of substrate inhibition in a lactate dehydrogen-

ase from human muscle by a single residue change. FEBS Lett. 1996, 399,

193−197.

(8) Hewitt, C. O.; Eszes, C. M.; Sessions, R. B.; Moreton, K. M.;

Dafforn, T. R.; Takei, J.; Dempsey, C. E.; Clarke, A. R.; Holbrook, J. J. A

general method for relieving substrate inhibition in lactate dehydro-

genases. Protein Eng. 1999, 12, 491−496.

(9) Read, J. A.; Winter, V. J.; Eszes, C. M.; Sessions, R. B.; Brady, R. L.

Structural basis for altered activity of M- and H-isozyme forms of human

lactate dehydrogenase. Proteins 2001, 43, 175−185.

(10) Goldman, R. D.; Kaplan, N. O.; Hall, T. C. Lactic dehydrogenase

in human neoplastic tissues. Cancer Res. 1964, 24, 389−399.

Compound 52. A solution of 51 (400 mg, 1.24 mmol) in methanol

(10 mL) was charged with Pd−C (0.4 g) and hydrogenated under a

hydrogen balloon at room temperature for 30 min. The catalyst was

ﬁltered oﬀ and the solvent evaporated to give the title compound

(200 mg, 86%) as a brown solid. ¹H NMR (400 MHz, MeOD) δ 8.56

(ddd, J = 5.2, 2.0, 1.2 Hz, 1H), 8.12 (td, J = 8.0, 1.2 Hz, 1H), 7.72 (dt, J =

8.0, 1.2 Hz, 1H), 7.31 (ddd, J = 7.6, 4.8, 1.2 Hz, 1H), 7.05 (t, J = 2.0 Hz,

1H), 6.92 (ddd, J = 10.8, 2.4, 1.2 Hz, 1H), 6.50 (dt, J = 10.8, 2.0 Hz, 1H),

Q

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(11) Giatromanolaki, A.; Sivridis, E.; Gatter, K. C.; Turley, H.; Harris,

A. L.; Koukourakis, M. I. Lactate dehydrogenase 5 (LDH-5) expression

in endometrial cancer relates to the activated VEGF/VEGFR2(KDR)

pathway and prognosis. Gynecol. Oncol. 2006, 103, 912−918.

(12) Koukourakis, M. I.; Giatromanolaki, A.; Sivridis, E.; Gatter, K. C.;

Harris, A. L. Lactate dehydrogenase 5 expression in operable colorectal

cancer: strong association with survival and activated vascular

endothelial growth factor pathwaya report of the tumour angio-

genesis research group. J. Clin. Oncol. 2006, 24, 4301−4308.

(13) Lu, Y.; Liu, P.; Wen, W.; Grubbs, C. J.; Townsend, R. R.; Malone,

J. P.; Lubet, R. A.; You, M. Cross-species comparison of orthologous

gene expression in human bladder cancer and carcinogen-induced

rodent models. Am. J. Transl. Res. 2011, 3, 8−27.

(30) Irwin, J. J.; Sterling, T.; Mysinger, M. M.; Bolstad, E. S.; Coleman,

R. G. ZINC: a free tool to discover chemistry for biology. J. Chem. Inf.

Model. 2012, 52, 1757−1768.

(31) Congreve, M.; Carr, R.; Murray, C.; Jhoti, H. A “rule of three” for

fragment-based lead discovery? Drug Discovery Today 2003, 8, 876−877.

(32) Mayer, M.; Meyer, B. Characterization of ligand binding by

saturation transfer difference NMR spectroscopy. Angew. Chem., Int. Ed.

1999, 38, 1784−1788.

(33) Neumann, T.; Junker, H. D.; Schmidt, K.; Sekul, R. SPR-based

fragment screening: advantages and applications. Curr. Top. Med. Chem.

2007, 7, 1630−1642.

(34) Sass, C.; Briand, M.; Benslimane, S.; Renaud, M.; Briand, Y.

Characterization of rabbit lactate dehydrogenase-M and lactate

dehydrogenase-H cDNAs. Control of lactate dehydrogenase expression

in rabbit muscle. J. Biol. Chem. 1989, 264, 4076−4081.

(35) Halgren, T. New method for fast and accurate binding-site

identification and analysis. Chem. Biol. Drug Des. 2007, 69, 146−148.

(36) Halgren, T. A. Identifying and characterizing binding sites and

assessing druggability. J. Chem. Inf. Model. 2009, 49, 377−389.

(37) Abel, R.; Young, T.; Farid, R.; Berne, B. J.; Friesner, R. A. Role of

the active-site solvent in the thermodynamics of factor Xa ligand

binding. J. Am. Chem. Soc. 2008, 130, 2817−2831.

(38) 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.; Garcia-Ochoa, S.; Gamo, F. J.;

Sanz, L.; Leon, L.; Ruiz, J. R.; Gabarro, R.; Mallo, A.; Gomez de las

Heras, F. Identification and activity of a series of azole-based compounds

with lactate dehydrogenase-directed anti-malarial activity. J. Biol. Chem.

2004, 279, 31429−31439.

(39) Conners, R.; Schambach, F.; Read, J.; Cameron, A.; Sessions, R.

B.; Vivas, L.; Easton, A.; Croft, S. L.; Brady, R. L. Mapping the binding

site for gossypol-like inhibitors of Plasmodium falciparum lactate

dehydrogenase. Mol. Biochem. Parasitol. 2005, 142, 137−148.

(40) Kuntz, K.; Uehling, D. E.; Waterson, A. G.; Emmitte, K. A.;

Stevens, K.; Shotwell, J. B.; Smith, S. C.; Nailor, K. E.; Salovich, J. M.;

Wilson, B. J.; Cheung, M.; Mook, R. A.; Baum, E. W.; Moorthy, G. US

2008300242 A1, 2008; 249 pp; Glaxosmithkline LLC.

(41) Mueller, R.; Yang, J.; Duan, C.; Pop, E.; Zhang, L. H.; Huang, T.

B.; Denisenko, A.; Denisko, O. V.; Oniciu, D. C.; Bisgaier, C. L.; Pape,

M. E.; Freiman, C. D.; Goetz, B.; Cramer, C. T.; Hopson, K. L.; Dasseux,

J. L. Long hydrocarbon chain ether diols and ether diacids that favorably

alter lipid disorders in vivo. J. Med. Chem. 2004, 47, 5183−5197.

(42) Sem, D. S.; Bertolaet, B.; Baker, B.; Chang, E.; Costache, A. D.;

Coutts, S.; Dong, Q.; Hansen, M.; Hong, V.; Huang, X.; Jack, R. M.;

Kho, R.; Lang, H.; Ma, C. T.; Meininger, D.; Pellecchia, M.; Pierre, F.;

Villar, H.; Yu, L. Systems-based design of bi-ligand inhibitors of

oxidoreductases: filling the chemical proteomic toolbox. Chem. Biol.

2004, 11, 185−194.

(14) Kolev, Y.; Uetake, H.; Takagi, Y.; Sugihara, K. Lactate

dehydrogenase-5 (LDH-5) expression in human gastric cancer:

association with hypoxia-inducible factor (HIF-1alpha) pathway,

angiogenic factors production and poor prognosis. Ann. Surg. Oncol.

2008, 15, 2336−2344.

(15) Koukourakis, M. I.; Giatromanolaki, A.; Sivridis, E.; Bougioukas,

G.; Didilis, V.; Gatter, K. C.; Harris, A. L. Lactate dehydrogenase-5

(LDH-5) overexpression in non-small-cell lung cancer tissues is linked

to tumour hypoxia, angiogenic factor production and poor prognosis. Br.

J. Cancer 2003, 89, 877−885.

(16) Deberardinis, R. J.; Sayed, N.; Ditsworth, D.; Thompson, C. B.

Brick by brick: metabolism and tumor cell growth. Curr. Opin. Genet.

Dev. 2008, 18, 54−61.

(17) Hsu, P. P.; Sabatini, D. M. Cancer cell metabolism: Warburg and

beyond. Cell 2008, 134, 703−707.

(18) Kroemer, G.; Pouyssegur, J. Tumor cell metabolism: cancer’s

Achilles’ heel. Cancer Cell 2008, 13, 472−482.

(19) Dang, C. V.; Semenza, G. L. Oncogenic alterations of metabolism.

Trends Biochem. Sci. 1999, 24, 68−72.

(20) Le, A.; Cooper, C. R.; Gouw, A. M.; Dinavahi, R.; Maitra, A.;

Deck, L. M.; Royer, R. E.; Vander Jagt, D. L.; Semenza, G. L.; Dang, C. V.

Inhibition of lactate dehydrogenase A induces oxidative stress and

inhibits tumor progression. Proc. Natl. Acad. Sci. U.S.A. 2010, 107,

2037−2042.

(21) Fantin, V. R.; St-Pierre, J.; Leder, P. Attenuation of LDH-A

expression uncovers a link between glycolysis, mitochondrial physiol-

ogy, and tumor maintenance. Cancer Cell 2006, 9, 425−434.

(22) Kanno, T.; Sudo, K.; Maekawa, M.; Nishimura, Y.; Ukita, M.;

Fukutake, K. Lactate dehydrogenase M-subunit deficiency: a new type of

hereditary exertional myopathy. Clin. Chim. Acta 1988, 173, 89−98.

(23) Rao, S. T.; Rossmann, M. G. Comparison of super-secondary

structures in proteins. J. Mol. Biol. 1973, 76, 241−256.

(24) McClendon, S.; Vu, D. M.; Clinch, K.; Callender, R.; Dyer, R. B.

Structural transformations in the dynamics of Michaelis complex

formation in lactate dehydrogenase. Biophys. J. 2005, 89, L07−09.

(25) McClendon, S.; Zhadin, N.; Callender, R. The approach to the

Michaelis complex in lactate dehydrogenase: the substrate binding

pathway. Biophys. J. 2005, 89, 2024−2032.

(26) Choi, S. R.; Beeler, A. B.; Pradhan, A.; Watkins, E. B.; Rimoldi, J.

M.; Tekwani, B.; Avery, M. A. Generation of oxamic acid libraries:

antimalarials and inhibitors of Plasmodium falciparum lactate dehydro-

genase. J. Comb. Chem. 2007, 9, 292−300.

(27) Choi, S. R.; Pradhan, A.; Hammond, N. L.; Chittiboyina, A. G.;

Tekwani, B. L.; Avery, M. A. Design, synthesis, and biological evaluation

of Plasmodium falciparum lactate dehydrogenase inhibitors. J. Med.

Chem. 2007, 50, 3841−3850.

(28) Granchi, C.; Bertini, S.; Macchia, M.; Minutolo, F. Inhibitors of

lactate dehydrogenase isoforms and their therapeutic potentials. Curr.

Med. Chem. 2010, 17, 672−697.

(43) Ferreira, R. S.; Bryant, C.; Ang, K. K.; McKerrow, J. H.; Shoichet,

B. K.; Renslo, A. R. Divergent modes of enzyme inhibition in a

homologous structure−activity series. J. Med. Chem. 2009, 52, 5005−

5008.

(44) Giannetti, A. M.; Koch, B. D.; Browner, M. F. Surface plasmon

resonance based assay for the detection and characterization of

promiscuous inhibitors. J. Med. Chem. 2008, 51, 574−580.

(45) Abel, R.; Salam, N.; Beuming, T.; Sherman, W.; Farid, R.

Presented at the National Meeting of the American Chemical Society,

2010; COMP-79.

(46) Glide, version 5.5; Schrodinger, LLC.: New York, NY, 2009.

(47) Ligprep, version 2.3; Schrodinger, LLC.: New York, NY, 2009.

(48) Prime, version 2.1; Schrodinger, LLC.: New York, NY, 2009.

(29) 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. Design

and synthesis of novel lactate dehydrogenase A inhibitors by fragment-

based lead generation. J. Med. Chem. 2012, 55, 3285−3306.

R

dx.doi.org/10.1021/jm3014844 | J. Med. Chem. XXXX, XXX, XXX−XXX

Article Doi

Fragment growing and linking lead to novel nanomolar lactate dehydrogenase inhibitors

DOI: 10.1021/jm3014844

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm3014844

Full text of DOI:10.1021/jm3014844

Products guided by the article

R&D Labs maybe for 1417792-31-4

Relevant to this article

Hot Product