Thioamide hydroxypyrothiones supersede amide hydroxypyrothiones in potency against anthrax lethal factor

Article abstract of DOI:10.1021/jm8013212

Anthrax lethal factor (LF) is a critical virulence factor in the pathogenesis of anthrax. A structure-activity relationship (SAR) of potential lethal factor inhibitors (LFi) is presented in which the zinc-binding group (ZBG), linker, and backbone moieties

Full text of DOI:10.1021/jm8013212

J. Med. Chem. 2009, 52, 1063–1074
1063
Thioamide Hydroxypyrothiones Supersede Amide Hydroxypyrothiones in Potency against
Anthrax Lethal Factor
Arpita Agrawal,† Ce´sar Augusto F. de Oliveira,†,‡ Yuhui Cheng,†,‡ Jennifer A. Jacobsen,† J. Andrew McCammon,†,‡ and
Seth M. Cohen*,†
Department of Chemistry and Biochemistry, Howard Hughes Medical Institute, Center for Theoretical Biological Physics, and Department of
Pharmacology, UniVersity of California, San Diego, 9500 Gilman DriVe, La Jolla, California 92093
ReceiVed October 18, 2008
Anthrax lethal factor (LF) is a critical virulence factor in the pathogenesis of anthrax. A structure-activity
relationship (SAR) of potential lethal factor inhibitors (LFi) is presented in which the zinc-binding group
(ZBG), linker, and backbone moieties for a series of hydroxypyrone-based compounds were systematically
varied. It was found that hydroxypyrothione ZBGs generate more potent inhibitors than hydroxypyrone
ZBGs. Furthermore, coupling the hydroxypyrothione to a backbone group via a thioamide bond improves
potency when compared to an amide linker. QM/MM studies show that the thioamide bond in these inhibitors
allows for the formation of two additional hydrogen bonds with the protein active site. In both types of
hydroxypyrothione compounds, ligand efficiencies of 0.29-0.54 kcal mol-1 per heavy atom were achieved.
The results highlight the need for a better understanding to optimize the interplay between the ZBG, linker,
and backbone to get improved LFi.
Introduction
three molecules of LF and/or EF can bind.8 The complex then
undergoes receptor mediated endocytosis, and the low pH in
the endosome triggers a conformational change that converts
the prepore to a mature cation-specific pore. LF and EF are
translocated across the mature pore to the cytosol of the cell
where they exert their toxicity.9-14 EF is a calcium and
calmodulin dependent adenylate cyclase that causes elevated
levels of cAMP in the cytosol of infected cells and also plays
a role in impairment of the immune system. Together with PA,
EF forms the edema toxin (ETx).15,16 LF is a zinc-dependent
hydrolytic metalloenzyme that cleaves the N-terminus of
mitogen activated protein kinase kinases (MAPKKs) to disrupt
downstream signaling pathways and cause macrophage apop-
tosis. In combination with PA, LF forms the lethal toxin
(LeTx).15,17-19 There are several published reviews describing
the pathogenesis of anthrax via its toxins, and despite extensive
research in the field, the exact pathway via which LF imparts
toxicity is still somewhat unclear; nevertheless, this protein is
an important target for inhibition.19-22 Current therapies against
B. anthracis include FDA approved antibiotics such as ciproflax-
in that target the bacteria but are ineffective toward the toxins
secreted by the bacterium. Inactivation of the LF gene in B.
anthracis leads to a 1000-fold reduction in its virulence, which
suggests that anthrax pathology is largely dictated by LF.22
Anthrax is one of the oldest documented diseases on record
known to infect animals and to this day poses a serious threat
to both animals and humans.1 Anthrax is caused by the Gram-
positive, rod-shaped bacterium Bacillus anthracis that is notori-
ous for its ability to form endospores. The bacteria adopt a
dormant spore structure when threatened by external factors and
can survive for decades in this state before entering a host. B.
anthracis spores are mostly soil-borne, and their dormant
longevity in the soil significantly contributes to their lethality.
Anthrax spores are hence among the most worrisome biological
weapons used, with recent attacks in the U.S. in 2001 sparking
significant concern.2-4
Anthrax infection can occur via three routes: inhalational,
gastrointestinal, and subcutaneous, with inhalational being the
most fatal. When B. anthracis spores are inhaled, they bind to
alveolar macrophages, which phagocytose the spores and traffic
them to regional lymph nodes. En route, the spores germinate
to pathogenic bacteria that release a potent anthrax toxin.5
Anthrax toxin is composed of three proteins: protective antigen
(PA, 83 kDa), edema factor (EF, 89 kDa), and lethal factor (LF,
90 kDa).a Independently the proteins are nontoxic but in concert
can induce cell death. PA first binds to one of two ubiquitous
receptors, ANTXR1 (tumor endothelium marker 8) or ANTXR2
(capillary morphogenesis protein 2).6,7 Once bound, PA is
activated by the cleavage of a 20 kDa N-terminal fragment by
membrane bound furin-like proteases. Upon activation, the 63
kDa PA oligomerizes to form a heptameric prepore to which
Several groups have been successful in developing potent
lethal factor inhibitors (LFi), some of which include known
matrix metalloproteinase inhibitors (MMPi).23-34 To date, some
of the most potent LFi carry a chelating hydroxamic acid zinc-
binding group (ZBG) similar to other zinc metalloprotease
inhibitors. Hydroxamic acids are known to be limited by poor
oral availability, limited zinc(II) ion selectivity, and poor
pharmacokinetics.35,36 To overcome the limitations of hydrox-
amic acids, the design of LFi that incorporate alternative ZBGs
merits investigation.
* To whom correspondence should be addressed. Telephone: (858) 822-
5596. Fax: (858) 822-5598. E-mail: scohen@ucsd.edu.
†
Department of Chemistry and Biochemistry.
Howard Hughes Medical Institute, Center for Theoretical Biological
‡
Physics, and Department of Pharmacology.
a Abbreviations: LF, lethal factor; LeTx, lethal toxin; EF, edema factor;
ETx, edema toxin; PA, protective antigen; AHA, acetohydroxamic acid;
LFi, lethal factor inhibitor(s); ZBG, zinc binding group; QM/MM, quantum
mechanics/molecular mechanics; LE, ligand efficiency; BPCA, 3-(benzy-
loxy)-4-oxo-4H-pyran-2-carboxylic acid; BDCA, 1-(benzyloxy)-6-oxo-1,6-
dihydropyridine-2-carboxylic acid.
In this study we have focused on a three component strategy
to the design of an LFi: (i) a ZBG to chelate and inactivate the
catalytic Zn2+ ion, (ii) a backbone to interact noncovalently with
the LF active site, and (iii) a linker to connect the backbone to
10.1021/jm8013212 CCC: $40.75
2009 American Chemical Society
Published on Web 01/26/2009

1064 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Agrawal et al.
Figure 1. ZBGs and linkers used in the LFi reported here. Abbreviations for each subset of compounds are shown in parentheses. The heteroatom
terminology used throughout is highlighted in the top three structures by the use of bold and italics atom labels. Shown in the bottom right is the
complete LFi 11b.
the ZBG. A similar overall scheme has been used in the
development of MMP and histone deacetylase (HDAC)
inhibitors.37,38 In the work presented here, the specific ZBGs
employed are derivatives of 3-hydroxy-2-methyl-4-pyrone (mal-
tol), 3-hydroxy-2-methyl-4-pyrothione (thiomaltol), 1-hydroxy-
pyridin-2(1H)-one (1,2-HOPO), and 1-hydroxypyridine-2(1H)-
thione (1,2-HOPTO). We have shown that these compounds
are effective ZBGs for inhibiting LF and other metalloprotein-
ases.39-44 The study presented here is focused on the synthesis
and structure-activity relationship (SAR) of the hydroxypyrone-
and hydroxypyrothione-based LFi. Different aromatic backbones
were attached to the ZBGs via an amide or a thioamide linkage.
The results presented elucidate the contribution of each com-
ponent to the potency of the different LFi. The potency of each
inhibitor was tested in vitro by using a fluorescence-based assay
that has been previously reported.40,45 The potency of each LFi
is reflected in calculated kinetic parameters such as IC50, Ki,
and ligand efficiency. In order to better understand the impor-
tance of the linker and backbone in protein-inhibitor interac-
tions, computational studies were performed to build three-
dimensional models of the inhibitors bound to the LF active
site. Our computational models are in agreement with the
experimental results and suggest a key role of the thioamide
group in stabilizing the protein-inhibitor complex.
ZBG, compounds 1a-10a were similarly converted to their O,S
donor hydroxypyrothione analogues. Reaction of the hydroxy-
pyrone compounds 1a-10a with phosphorus pentasulfide
(P4S10) and hexamethyldisiloxane (HMDO) in benzene generates
two products as shown in Scheme 2. One major product contains
a single sulfur atom, designated (O,S,O) compounds, at the
pyrone carbonyl group. The second major product contains two
sulfur atoms, designated (O,S,S) compounds, with one sulfur
atom at the pyrone carbonyl group and the other at the amide
carbonyl linker group. This structural assignment was based on
the spectroscopic characterization of the compounds (1H NMR
and 13C NMR) and was unambiguously confirmed by an X-ray
crystal structure of compound 5c (vide infra).
Upon thionation of the hydroxypyrone (O,O,O) compounds,
the (O,S,O) compounds are produced first followed by produc-
tion of the (O,S,S) compounds. The pyrone carbonyl oxygen
of each compound reacts prior to conversion of the amide to a
thioamide linker, and therefore, the percentage of each major
product can be modulated by the reaction time. However, the
reactivity of each compound was found to also depend on the
backbone substituent. When 1a, with a biphenyl backbone, was
treated with P4S10 and HMDO for 3 h, the reaction yielded 1b
(O,S,O) in 16% and 1c (O,S,S) in 43% yield, respectively. When
1a was treated for about 1 h, 1b and 1c were obtained in 69%
and 2% yields, respectively. When 8a, with a simple ethyl
backbone was treated for only 30 min, the isolated products
consisted of 55% 8b (O,S,O) and 20% 8c (O,S,S). Similarly,
when the reaction time was extended to 1 h, 8b was obtained
in 25% yield, whereas 8c was isolated in a 43% yield. The
results obtained show that optimal reaction times have to be
considered for each compound of interest. An additional point
of interest is that the Rf values for the (O,S,O) and (O,S,S)
compounds tend to be very similar and chromatographic
separation can be challenging (hence, the poor to moderate
yields reported here).
Results
Synthesis of LFi. Our group has previously identified
alternative ZBGs that are more potent than simple hydroxamic
acids against LF.39,40 On the basis of these earlier findings, a
novel hydroxypyrothione-based (O,S metal-binding donor atom
set) LFi (AM-2S, 11b, Figure 1) was designed, synthesized,
and found to be moderately potent against LF (IC50 ) 13.9
µM).40 11b thereby inspired the design of additional hydroxy-
pyrone, hydroxypyrothione, N-hydroxypyridinone, and N-hy-
droxypyridinethione LFi shown in Figures 1 and 2. Compounds
1a (PY-2) and 9a (1,2-HOPO-2) were previously synthesized
and identified as potent nanomolar inhibitors against deep pocket
MMPs.46 Several similar compounds comprising hydroxypyrone
ZBGs (O,O metal-binding donor atom set), amide linking
groups, and a series of different backbone substituents were
synthesized (Scheme 1, Figure 2, 1a-10a).46 Compounds
1a-10a were universally found to be ineffective as LFi (vide
supra).
As for the N-hydroxypyridinone-based (1,2-HOPO) inhibitors,
synthesis of 9a was previously reported;46 the synthesis of 10a
parallels the synthetic procedure for 9a (Scheme 3). Unlike the
hydroxypyrothiones described above, the syntheses of the
N-hydroxypyridinethione-based (1,2-HOPTO) inhibitors are not
derived from their O,O chelate analogues. The synthesis of
1-hydroxy-6-thioxo-1,6-dihydropyridine-2-carboxylic acid (HT-
DCA) in Scheme 4 is based on literature procedures with slight
modifications (see Experimental Section).47,48 Syntheses of 9b
and 10b from HTDCA are outlined in Scheme 4. The 4-phe-
Because compounds 1a-10a were not potent LFi and our
previously reported inhibitor (11b) used a hydroxypyrothione

Thioamide Hydroxypyrothiones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1065
Figure 2. Structures of the inhibitors tested in this study. The heteroatom terminology used throughout is highlighted in the top three structures
by the use of bold and italics atom labels. Shown in the box is the complete LFi 1a (for clarity). Other inhibitors are hydroxypyrones with an amide
linker (1a-8a), hydroxypyrothiones with an amide linker (1b-8b), hydroxypyrothiones with a thioamide linker (1c-8c), N-hydroxypyridinones
with an amide linker (9a, 10a, below thick line), and N-hydroxypyridinethiones with an amide linker (9b, 10b below thick line).
nylbenzylamine and 4-fluorobenzylamine backbones are ap-
pended to HTDCA under standard amide coupling conditions.
EDCI (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide) and
NHS (N-hydroxysuccinimide) were used as the coupling
reagents for 9b, whereas EDCI and HOBt (N-hydroxybenzot-
riazole) were the coupling reagents used in the production of
10b.
is directly proportional to LF activity.45 The IC50 values (Table
1) were calculated for each synthesized inhibitor as described
under Experimental Section. Kinetic parameters (Vmax, Km, Ki)
were also calculated using the same fluorogenic assay. Ad-
ditionally, ligand efficiencies of most inhibitors were calculated
from the respective Ki values (vide infra).49-51 Compounds
1a-8a (Table 1, column “a”), which consist of a hydroxypyrone
ZBG and amide linker (O,O,O), are ineffective against LF with
IC50 values greater than 100 µM. Compounds 1b-8b (Table 1,
column “b”), which consist of a hydroxypyrothione ZBG and
amide linker (O,S,O), show moderate potency against LF with
IC50 values ranging from 13 to >100 µM. Surprisingly,
compounds 1c-8c (Table 1, column “c”), which consist of a
hydroxypyrothione ZBG and thioamide linker (O,S,S), are
uniformly the most potent compounds with IC50 values in the
range 5-11 µM. Together these data show that (i) a hydroxy-
pyrone ZBG when coupled via an amide linker (O,O,O) is
ineffective against LF regardless of the backbone used, (ii) a
hydroxypyrothione ZBG when coupled via an amide linker
(O,S,O) can have moderate potency against LF depending on
the backbone substituent, and (iii) a hydroxypyrothione ZBG
when coupled via a thioamide linker (O,S,S) provides consistent
potency against LF with little influence from the backbone
group. The minimal effect of the backbone in compounds 1c-8c
is highlighted by inhibitor 8c, which utilizes only a simple ethyl
Single-Crystal X-ray Structure of 5c. To provide evidence
for the conversion of both carbonyl oxygen atoms to sulfur, an
X-ray diffraction study was performed on inhibitor 5c. Large
red plates of 5c were obtained by slow evaporation from a
concentrated solution of the compound in chloroform. The
crystal structure (Figure S1 and Table S2 of Supporting
Information) confirms the presence of the two expected sulfur
atoms. In addition to the measured electron density, the C-S
bond lengths confirm the correct identity of the sulfur atoms.
The thionyl C(4)-S(4) bond distance of 1.66 Å lies within the
range of C-S bonds from other hydroxypyrothiones (1.68 Å).42
The thioamide C(1)-S(1) bond distance of 1.68 Å is similar to
the average thioamide bond distance (1.67 Å) predicted from
the Cambridge Crystallographic Data Center (CCDC) and is
too long to be an amide C-O bond (∼1.22 Å).
In Vitro Potency. The in vitro potency of each inhibitor was
evaluated against LF using an established procedure with a
quenched fluorogenic substrate, where observed fluorescence

1066 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Agrawal et al.
Scheme 1. Synthesis of Hydroxypyrone (O,O,O) Inhibitors (1a-8a)a
a Reagents and conditions: (a) RNH2, EDCI, HOBt, CH2Cl2, N2, room temp; (b) 1:1 HCl/CH3COOH, room temp.
Scheme 2. Synthesis of Hydroxypyrothione Analogues 1b (O,S,O) and 1c (O,S,S) Compoundsa
a Reagents and conditions: (a) P4S10, HMDO, benzene, 110 °C.
Scheme 3. Synthesis of N-Hydroxypyridinone 10aa
a Reagents and conditions: (a) EDCI, HOBt, CH2Cl2, N2, room temp; (b) 1:1 HCl/CH3COOH, room temp, 32%.
backbone and yet has a similar IC50 value compared to the rest
of the series that employ a variety of aromatic backbones. These
findings thus underline the importance of all three components,
the ZBG, linker, and backbone, and that each of these
components must be considered when optimizing LFi.
differences are seen between the 1,2-HOPO (9a, 10a) and 1,2-
HOPTO (9b, 10b) compounds, with the latter inhibitors showing
very poor potency (>100 µM). The lack of potency of the 1,2-
HOPO and 1,2-HOPTO chelators may be due to (a) a signifi-
cantly different binding mode of these chelators to the active
site zinc(II) ion or (b) the significantly greater acidity of these
chelators relative to the hydroxypyrones and hydroxypy-
rothiones. In either case, this data show that the nature of the
ZBG is critical and that even small changes in structure can
result in large differences in efficacy.
Computational Modeling. In order to verify the structural
and energetic factors that are relevant for the differences in
observed potency, we built QM/MM optimized structures for
the bound complex between the LF active site and several of
the hydroxypyrothione inhibitors. Figure 3 displays the models
for the bound structures of inhibitors 1b, 1c, 3b, 3c, 4b, and 4c
in the active site of LF. Interestingly, optimized structures of
molecules 1b and 1c in the active site of LF presented very
similar configurations and no significant difference was observed
in the orientation of the backbone fragment or the amide and
thioamide linker groups. This is consistent with the observations
that 1b and 1c have similar IC50 values against LF. In contrast,
In order to confirm the importance of the ZBG and linker in
generating effective LFi, we synthesized four additional inhibi-
tors, based on N-hydroxypyridinone (9a, 10a) and N-hydroxy-
pyridinethione (9b, 10b) chelators. Compounds 9a and 9b
contain the same biphenyl backbone and amide linker found in
1a. Compounds 10a and 10b have the same fluorobenzyl
backbone and amide linker as compound 2a. Therefore, on the
basis of our earlier classification, 9a and 10a are (O,O,O)
inhibitors like 1a and 2a; 9b and 10b are (O,S,O) inhibitors
similar to 1b and 2b. Notably, the only difference between these
inhibitors lies in the use of a N-hydroxypyridinone (1,2-HOPO)
versus hydroxypyrone, or N-hydroxypyridinethione (1,2-
HOPTO) versus hydroxypyrothione ZBG. These heterocyclic
ZBGs are essentially isosteric and very similar in overall
composition. With that in mind, it was surprising to find that
whereas the (O,O,O) hydroxypyrones (1a, 2a) are less effective
than the (O,S,O) hydroxypyrothiones (1b, 2b), no substantial

Thioamide Hydroxypyrothiones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1067
Scheme 4. Synthesis of N-Hydroxypyridinethione Inhibitors 9b and 10ba
a Reagents and conditions: (a) TFA, 30% H2O2, 80 °C, 94%; (b) Na+HS-, H+Cl-, 95 °C, 89%; (c) EDCI, NHS, CH2Cl2, 46%; (d) EDCI, HOBt, CH2Cl2,
95%.
Table 1. IC50 Values of LFi Using a Fluorescence-Based Assaya
IC50 (µM)
compd
a (O,O,O)
b (O,S,O)
c (O,S,S)
1
2
3
4
5
6
7
8
9
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
13.2 ((0.9)
29.2 ((1.4)
50.4 ((2.4)
14.1 ((2.0)
33.5 ((2.3)
37.4 ((3.1)
>100
8.3 ((0.1)
9.2 ((0.2)
10.6 ((0.6)
5.0 ((0.2)
8.7 ((0.8)
6.6 ((0.8)
7.9 ((0.6)
11.4 ((0.7)
72.5 ((5.4)
>50b
>100
10
a All values represent the average and standard deviation from at least
three independent experiments. b Determination of a more accurate IC50
value was precluded by the limited solubility of this compound.
even though the backbone fragments of 3b and 3c adopt similar
configurations, the ZBG relative to the amide/thioamide linker
groups shows distinct orientations. By comparing the structures
of 1c and 3c, we observed that the sulfur atom of the thioamide
group acts as a hydrogen bond acceptor, interacting with the
backbone amide N-H group of residues Lys392 and Gly393.
Similar hydrogen bonding was also observed in the optimized
structures of thioamide containing LFi: 2c, 7c, and 8c (Figure
S3). Figure 3 shows that although the two hydrogen bonds were
observed in the complexes of LF with 1b, 1c, and 3c, they were
not found with compound 3b. The distance between the amide
oxygen of 3b and the backbone N-H groups was greater than
4.0 Å (Figure 3). These observations are consistent with the
IC50 values obtained for these compounds (Table 1), where 3b
is clearly the least effective of these inhibitors (50.4 µM). Similar
interactions involving either Lys392 or Gly393 have already
been characterized in crystal structures of anthrax lethal factor
bound to small molecule inhibitors, which indicate that these
residues may play an important role in the inhibitor binding.32,33
Previous work suggested that the plasticity and flexibility of
the S1′ binding pocket allow the receptor to accommodate small
and large hydrophobic residues at the P1′ position.33 These
studies also indicate that the design of compounds with greater
complementarity to the S1′ pocket would lead to potent and
selective inhibition of LF. In order to explore interactions with
Figure 3. QM/MM optimized structures of the complex between
anthrax LF (ribbons) and inhibitors 1b, 1c, 3b, 3c, 4b, and 4c (ball-
and-stick). Some active site residues (ball-and-stick) and catalytic
zinc(II) ion (gray sphere) are also shown.
the S1′ binding pocket, inhibitors 4b and 4c were designed. In
fact, 4c shows a modest, ∼2-fold improvement in LF potency
when compared to 1c.
To further investigate the quality of our models, all optimized
structures were submitted to energy calculation procedures by

1068 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Agrawal et al.
using the program Autodock 4.0.52 Because there is a good
agreement between the calculated structures and the experi-
mentally determined IC50 values, the calculated models were
used to obtain a rough estimation of the binding free energy of
molecules 1b, 1c, 2b, 2c, 3b, 3c, 4b, and 4c. Figure S2 of
Supporting Information shows the plot of log(IC50) value against
the estimated free energy of binding (∆G), which demonstrates
that the trend in potency is captured reasonably well by the
calculated models.
4a-c. The “bifurcated” biphenyl backbone of 4c was designed
to interact with the same pocket as found for 11b while also
targeting the S1′ pocket. Compound 4c showed a slight but not
pronounced improvement in inhibition compared to the other
LFi reported here. To confirm the dominance of the ZBG and
linker over the backbone in this series of LFi, compounds with
a simple ethyl backbone were prepared (8a-c). Indeed, the
potency of 8c showed that the backbone groups play a minimal
role in these compounds, and further design is required to fully
exploit additional protein-specific interactions.
To probe the specific interactions of the ZBG, linker, and
backbone in the active site, computational models for the binding
mode of several of the inhibitors were calculated using the
previously calculated 11b model as a starting point.40 QM/MM
minimization procedures were employed on each model in order
to take into account electronic effects in the interaction between
zinc(II), the active site residues, and the ZBG. The models show
a good agreement, at least qualitatively, with the experimentally
observed IC50 values. The optimized structures reveal that
different backbone fragments have little effect on the orientation
of the ZBG in the active site. The main structural difference
found in our models was in the orientation of the amide/
thioamide linker relative to the ZBG. The models show that
twisting about the bond between the ZBG and the amide group
(e.g., breaking planarity of the amide group and aromatic pyrone
rings) allows for formation of specific interactions between
residues Lys392, Gly393, and the oxygen/sulfur amide atoms
of the inhibitors. Dihedral angles between the amide/thioamide
groups and pyrone rings in 1b, 1c, 3b, 3c, 4b, and 4c were
found to be 71°, 62°, 28°, 98°, 61°, and 57°, respectively. The
hydrogen bonding is most pronounced in the inhibitor with bulky
backbones (1b,c and 4b,c), where the hydrogen-bonded amide/
thioamide linkers sit in essentially the same configuration (Figure
3). The smaller angle in 3b is consistent with the lack of
hydrogen bonding (Figure 3) with the protein active site
compared to the other LFi. It is important to note that QM
minimizations of the inhibitors, in the absence of protein and
solvation, converge to a planar thioamide-pyrone ring conju-
gated system. This was validated by minimizing the conforma-
tion of 5c, which converged to the structure found by crystal-
lographic analysis (Figure S1 of Supporting Information). The
hydrogen bond formed by the thioamide LFi was somewhat
surprising, as thioamides are known to be weaker hydrogen bond
acceptors compared to their amide analogues.54,55 However,
Allen and co-workers have shown that thioamides possess a
resonance induced hydrogen bonding potential that arises from
the lone pair of electrons on the neighboring nitrogen atom of
the amino group.54 This leads to a net increase in electronega-
tivity of the thioamide sulfur and a lengthening of the CdS
bond, which likely facilitates hydrogen bonding in these
molecules.
Discussion
Anthrax lethal factor is a zinc dependent endopeptidase that
cleaves MAPKKs 1-4, 6, and 7 to inhibit downstream pathways
that culminate in macrophage apoptosis. The Zn2+ ion in the
lethal factor active site is tetrahedrally coordinated via two
histidine residues (His686 and His690), one glutamic acid
residue (Glu735), and one water molecule. The bound water
molecule in this motif, which occurs in several different zinc
metalloproteases, is responsible for substrate cleavage.53 Our
goal is to exploit the presence of a metal ion and develop LFi
that can chelate and inactivate the catalytic zinc(II) ion while
optimizing the noncovalent interactions of the backbone com-
ponent with the protein active site. Most LFi and metalloprotein
inhibitors in the literature utilize a hydroxamic acid as the ZBG;
however, the poor pharmacokinetics of hydroxamates has made
the search for alternative chelators a prudent topic of interest.39,43
Of the previously reported ZBGs, hydroxypyrothiones were
determined to be the most potent against LF and thus were
incorporated into the LFi 11b as shown in Figure 1.40 11b
inspired the design of additional hydroxypyrothione inhibitors
as shown in Figure 2, along with their hydroxypyrone precur-
sors, to develop a structure-activity-relationship (SAR). The
overall goal was to design and synthesize different inhibitors
with variations in three key components (the ZBG, linker, and
backbone) and to analyze the contribution of each component
toward potency against LF.
Figure 2 shows the compounds that were synthesized for this
study which also include MMPi (1a and 9a) that were previously
reported.46 As shown in Figure 2, four different ZBGs (hy-
droxypyrone, hydroxypyrothione, N-hydroxypyridinone, N-hy-
droxypyridinethione), two different linkers (amide and thioa-
mide), and eight different backbones were assembled in different
combinations to generate 28 compounds. Table 1 lists the IC50
values of each compound against LF. As previously noted, the
N-hydroxypyridinone- (both O,O,O and O,S,O) and the hy-
droxypyrone-based (O,O,O) compounds are poor inhibitors of
LF. In contrast, the (O,S,O) hydroxypyrothione ZBG clearly
improves inhibition of LF, consistent with the earlier studies
on 11b.40
A three-dimensional structure of the bound complex formed
between 11b and anthrax LF showed the biphenyl backbone of
11b targeting a specific pocket in the LF active site.40 To
optimize the backbone interactions in LF, different substituted
and unsubstituted aromatic backbones were appended to the
ZBGs via an amide or thioamide linker. The hydroxypyrothiones
coupled via an amide bond (O,S,O) showed moderate potency
(consistent with 11b) with some variations depending on the
backbone group. However, the use of a thioamide linker with
a hydroxypyrothione ZBG (O,S,S) showed consistently im-
proved potency against LF, regardless of the backbone used.
These results suggested that the potency of the (O,S,S)
compounds was primarily a result of the ZBG and linker alone.
To improve potency by enhancing backbone interactions, we
targeted the S1′ pocket of LF with the design of inhibitor series
Despite the difference in potency among the (O,S,O) and
(O,S,S) compounds, the ligand efficiencies (LE) of both sets of
hydroxypyrothiones fall in a reasonably efficient, small window
of 0.28-0.54 kcal mol-1 per heavy atom.49-51 LE is the ratio
of the binding free energy of a ligand to the number of heavy
atoms in the ligand. LE is used to normalize the potency of
various ligands with respect to their molecular weight.49-51,56
In this way, LE is an important metric in controlling the size of
a druggable lead while maintaining potency. The LE of the
hydroxypyrothione ZBG (i.e., thiomaltol) is 0.58 kcal mol-1
per heavy atom and with the addition of a thioamide ethyl
linkage (8c) is 0.54 kcal mol-1 per heavy atom. The remaining
hydroxypyrothione LFi with aromatic backbones have an

Thioamide Hydroxypyrothiones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1069
[M - H]-. Anal. (C19H15NO3S·H2O). C, H, N, S, respectively: calcd
64.21, found 64.51; calcd 4.82, found 4.94; calcd 3.94, found 3.89;
calcd 10.80, found 11.26.
average LE of 0.33 kcal mol-1 per heavy atom. The potency
contributed by the thioamide bond is further reflected in the
similar LE values obtained for thiomaltol and 8c and remains
relatively consistent throughout the LFi series compared to their
amide analogues. As discussed above, the potency of the
thioamide-linked compounds may be attributed to the longer
CdS bond and twisting of the ZBG, which helps to introduce
hydrogen bonding with the protein active site.54,57
There is some concern about the clinical use of thioamides,
as in some cases they have been associated with hepatotoxic
effects.58,59 However, other studies have shown thioamides to
improve the stability of compounds against certain proteases.57,60,61
In this study, in addition to being more potent, the thioamide
LFi reported here meet certain drug discovery criteria, with
molecular weights of less than 500 Da, possessing less than 5
hydrogen-bond donors and less than 10 hydrogen-bond acceptors
with adequate ligand efficiencies.56 To further develop the
hydroxypyrothione ZBGs with a thioamide linker, we plan to
utilize our computational models to probe the active site and
further elaborate the (O,S,S) scaffold with more specific, protein-
interacting backbones.
N-(Biphenyl-4-ylmethyl)-3-hydroxy-4-thioxo-4H-pyran-2-car-
bothioamide (1c). 1c was obtained from 1a from the same reaction
as that of 1b (vide supra) as a dark-brown solid (94 mg, 0.27 mmol).
Yield ) 43%. 1H NMR (400 MHz, CDCl3, 25 °C): δ ) 5.01 (d, J
) 4.4 Hz, NHCH2), 7.35 (t, J ) 7.4 Hz, 1H; ArH), 7.42 (m, 5H;
ArH), 7.58 (m, 4H; ArH), 7.75 (d, J ) 5.2 Hz, 1H; ArH), 9.53
(brt, 1H; CONHCH2), 10.30 (brs, 1H; ArOH). 13C NMR (100 MHz,
CDCl3, 25 °C): δ ) 50.0 (CH2), 125.3 (ArC), 127.0 (ArC), 127.5
(ArC), 127.7 (ArC), 128.6 (ArC), 128.8 (ArC), 133.9 (ArC), 134.2
(ArC), 140.2 (ArC), 141.2 (ArC), 147.4 (ArC), 149.3 (ArC), 184.8
(CdS), 191.2 (ArCdS). APCI-MS(+) m/z 353.98 [M + H]+. Anal.
(C19H15NO2S2 · 0.65H2O). C, H, N, S, respectively: calcd 62.49,
found 662.29; calcd 4.50, found 4.45; calcd 3.84, found 3.66; calcd
17.56, found 17.40.
N-(4-Fluorobenzyl)-3-hydroxy-4-oxo-4H-pyran-2-carboxam-
ide (2a). To a solution of 3-(benzyloxy)-4-oxo-4H-pyran-2-car-
boxylic acid (BPCA)46 (500 mg, 2.0 mmol) in dichloromethane
was added NHS (1.0 equiv, 230 mg, 2.0 mmol), EDCI (1.0 equiv,
383 mg, 2.0 mmol), and 4-fluorobenzylamine (1.0 equiv, 250 mg,
227 µL, 2.0 mmol). The mixture was stirred overnight at room
temperature under N2 and then poured into a separatory funnel and
extracted with water and dichloromethane. The layers were
separated, and the organic layers were combined, dried over
anhydrous magnesium sulfate, filtered, and concentrated to an
orange oil. The crude oil was purified via silica column chroma-
tography (0-2% MeOH/CH2Cl2) to yield a yellow solid of the
benzyl protected product. To the yellow product was added 10 mL
of a 1:1 solution of concentrated HCl and glacial acetic acid. The
reaction mixture was stirred for 5 d and then concentrated in vacuo
to yield an off-white residue which was coevaporated with MeOH
and dried to yield an off-white solid (269 mg, 1.02 mmol). Yield
Conclusion
In this report we have successfully shown the effect of the
ZBG and linker on the potency of a series of metal-chelating
LFi. Some of the compounds include potent, isoform-specific
MMPi (1a and 9a) that showed negligible activity against LF.
The active inhibitors identified here elicit the effect by tight
binding of the ZBG and strong hydrogen bonding to the protein
via a thioamide linking group. Even though at present we have
not identified an optimal backbone component, we have
succeeded in identifying a potent ZBG-linker scaffold from
which an optimized backbone can be grafted to obtain a new
class of highly potent LFi.
1
) 93%. H NMR (400 MHz, CDCl3, 25 °C): δ ) 4.46 (d, J ) 6
Hz, NHCH2), 6.32 (d, J ) 5.6 Hz, 1H; ArH), 6.90 (t, J ) 8.4 Hz,
2H; ArH), 7.20 (dd, J ) 3.6, 14, 2H; ArH), 7.67 (d, J ) 5.2 Hz,
1H; ArH), 8.11 (brt, 1H; CONHCH2). 13C NMR (100 MHz, CDCl3,
25 °C): δ ) 42.1 (CH2), 114.8 (ArC), 115.0 (ArC), 115.2 (ArC),
129.1 (ArC), 129.2 (ArC), 132.9 (ArC), 148.9 (ArC), 153.2 (ArC),
160.5 (ArC), 162.4 (ArC), 162.9 (CdO), 173.6 (ArCdO). ESI-
MS(+) m/z 264.02 [M + H]+. Anal. (C13H10FNO4) C, H, N.
N-(4-Fluorobenzyl)-3-hydroxy-4-thioxo-4H-pyran-2-carbox-
amide (2b). This compound was prepared from 2a according to
the procedure outlined for 1b. 2a (150 mg, 0.57 mmol) was treated
with P4S10 and HMDO for 1 h to yield an orange-red solid of 2b
Experimental Section
Chemistry. Unless otherwise noted, all chemicals were pur-
chased from commercial suppliers (Aldrich and Fisher) and used
as received. Flash silica gel chromatography was performed using
Merck silica gel 40-63 µm mesh. Inert reactions were carried out
under a dinitrogen atmosphere. 1H/13C NMR spectra were recorded
at ambient temperature on a 300 or 400 MHz Varian FT-NMR
instrument, property of the Department of Chemistry and Biochem-
istry, University of California, San Diego. Mass spectra were
obtained at the Small Molecule Mass Spectrometry Facility in the
Department of Chemistry and Biochemistry at the University of
California, San Diego. Elemental analysis was performed by
NuMega Resonance Laboratories, San Diego, CA.
1
(48 mg, 0.17 mmol). Yield ) 30%. H NMR (400 MHz, CDCl3,
25 °C): δ ) 4.63 (d, J ) 6 Hz, NHCH2), 7.02 (t, J ) 9.4 Hz, 2H;
ArH), 7.31 (m, 2H; ArH), 7.42 (dd, J ) 3.6, 6.4 Hz, 1H; ArH),
7.72 (dd, J ) 3.6, 6.4 Hz, 1H; ArH), 7.90 (brt, 1H; CONHCH2),
9.65 (brs, 1H; ArOH). 13C NMR (100 MHz, CDCl3, 25 °C): δ )
42.5 (CH2), 115.0 (ArC), 115.2 (ArC), 126.4 (ArC), 129.1 (ArC),
129.2 (ArC), 131.3 (ArC), 146.2 (ArC), 152.7 (ArC), 160.7 (ArC),
162.2 (ArC), 163.1 (CdO), 191.5 (ArCdS). APCI-MS(+) m/z
280.06 [M + H]+. Anal. (C13H10FNO3S) C, H, N, S.
N-(Biphenyl-4-ylmethyl)-3-hydroxy-4-oxo-4H-pyran-2-car-
boxamide (1a). This compound was prepared as previously
reported.46
N-(4-Fluorobenzyl)-3-hydroxy-4-thioxo-4H-pyran-2-carboth-
ioamide (2c). 2c was obtained from 2a from the same reaction as
that of 2b as a brown solid (47 mg, 0.16 mmol). Yield ) 28%. 1H
NMR (400 MHz, CDCl3, 25 °C): δ ) 4.94 (d, J ) 4.8 Hz, NHCH2),
7.06 (t, J ) 8.8 Hz, 2H; ArH), 7.35 (dd, J ) 2.8, 13.6 Hz, 2H;
ArH), 7.41 (d, J ) 5.2 Hz, 1H; ArH), 7.76 (d, J ) 4.8 Hz, 1H;
ArH), 9.50 (brt, 1H; CONHCH2), 10.18 (brs, 1H; ArOH). 13C NMR
(100 MHz, CDCl3, 25 °C): δ ) 49.4 (CH2), 115.8 (ArC), 116.0 (ArC),
125.2 (ArC), 129.8 (ArC), 130.0 (ArC), 133.8 (ArC), 147.4 (ArC),
149.1 (ArC), 161.2 (ArC), 163.7 (ArC), 184.8 (CdS), 191.1 (ArCdS).
APCI-MS(+) m/z 296.03 [M + H]+. Anal. (C13H10FNO2S2). C, H,
N, S, respectively: calcd 52.87, found 52.43; calcd 3.41, found 3.50;
calcd 4.74, found 4.55; calcd 21.71, found 21.32.
N-(Biphenyl-4-ylmethyl)-3-hydroxy-4-thioxo-4H-pyran-2-car-
boxamide (1b). To a solution of 1a (200 mg, 0.62 mmol) in
benzene (20 mL) was added P4S10 (0.366 equiv, 102 mg, 0.23
mmol) and HMDO (3.339 equiv, 336 mg, 442 µL, 2.07mmol), and
the mixture was heated to reflux at 105 °C for 3 h. Solid byproducts
were removed by vacuum filtration, and the filtrate was evaporated
in vacuo to a crude brown solid. The crude product was purified
via column chromatography (50% EtOAc/hexanes) to afford 1b
1
(33 mg, 0.10 mmol). Yield ) 16%. H NMR (400 MHz, CDCl3,
25 °C): δ ) 4.71 (d, J ) 6 Hz, NHCH2), 7.34 (t, J ) 7.2 Hz, 1H;
ArH), 7.42 (m, 5H; ArH), 7.57 (m, 4H; ArH), 7.72 (d, J ) 5.2 Hz,
1H; ArH), 7.92 (brt, 1H; CONHCH2), 9.73 (brs, 1H; ArOH). 13C
NMR (100 MHz, CDCl3, 25 °C): δ ) 43.6 (CH2), 125.7 (ArC),
127.1 (ArC), 127.4 (ArC), 127.6 (ArC), 128.3 (ArC), 128.8 (ArC),
132.4 (ArC), 136.0 (ArC), 140.5 (ArC), 140.9 (ArC), 147.4 (ArC),
151.6 (ArC), 160.9 (CdO), 190.5 (ArCdS). APCI-MS(-) m/z 336.11
3-Hydroxy-N-(4-methoxybenzyl)-4-oxo-4H-pyran-2-carbox-
amide (3a). 3a was prepared according to the same procedure
outlined for 2a starting from BPCA (1.0 g, 4.1 mmol), EDCI (786

1070 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Agrawal et al.
mg, 4.1 mmol), HOBt (554 mg, 4.1 mmol), and 4-methoxybenzy-
lamine (562 mg, 536 µL, 4.1 mmol) to yield an off-white solid of
3a (600 mg, 2.2 mmol). Yield ) 54%. 1H NMR (400 MHz, CDCl3,
25 °C): δ ) 3.81 (s, 3H; OCH3), 4.55 (d, J ) 4.8 Hz, NHCH2),
6.44 (d, J ) 6 Hz, 1H; ArH), 6.89 (d, J ) 8.8 Hz, 2H; ArH), 7.02
(brt, 1H; CONHCH2), 7.26 (d, J ) 9.6 Hz, 2H; ArH), 7.68 (d, J )
5.6 Hz, 1H; ArH). 13C NMR (100 MHz, CDCl3, 25 °C): δ ) 43.1
(CH2), 55.3 (OCH3), 114.3 (ArC), 115.5 (ArC), 128.5 (ArC), 129.4
(ArC), 135.9 (ArC), 149.6 (ArC), 153.0 (ArC), 159.4 (ArC), 163.1
(CdO), 173.6 (ArCdO). ESI-MS(+) m/z 275.98 [M + H]+. Anal.
(C14H13NO5) C, H, N.
7.76 (d, J ) 5.2 Hz, 1H; ArH), 9.92 (brd, CONHCH2), 10.15 (brs,
1H; ArOH). 13C NMR (100 MHz, DMSO-d6, 25 °C): δ ) 61.7
(CH2), 125.5 (ArC), 127.6 (ArC), 127.7 (ArC), 128.6 (ArC), 139.2
(ArC), 139.7 (ArC), 147.7 (ArC), 184.4 (CdS), 191.1 (ArCdS).
APCI-MS(+) m/z 353.85 [M + H]+. Anal. (C19H15NO2S2 ·0.4H2O).
C, H, N, S, respectively: calcd 63.27, found 63.47; calcd 4.42, found
4.86; calcd 3.88, found 3.90; calcd 17.78, found 18.16.
3-Hydroxy-N-(2-methoxybenzyl)-4-oxo-4H-pyran-2-carbox-
amide (5a). This compound was prepared according to the same
procedure outlined for 3a starting from BPCA (500 mg, 2.0 mmol),
EDCI (1.2 equiv, 713 mg, 2.4 mmol), HOBt (1.2 equiv, 324 mg,
2.4 mmol), and 2-methoxybenzylamine (1.2 equiv, 329 mg, 310
µL, 2.4 mmol), to yield an off-white solid of 5a (355 mg, 1.3 mmol).
3-Hydroxy-N-(4-methoxybenzyl)-4-thioxo-4H-pyran-2-car-
boxamide (3b). This compound was prepared from 3a according
to the procedure outlined for 1b. 3a (150 mg, 0.54 mmol) was
treated with P4S10 and HMDO for 1 h to yield an orange-red solid
1
Yield ) 65%. H NMR (400 MHz, CDCl3, 25 °C): δ ) 3.86 (s,
3H; OCH3), 4.57 (d, J ) 3.6 Hz, NHCH2), 6.39 (d, J ) 5.2 Hz,
1H; ArH), 6.87 (q, J ) 7.2 Hz, 2H; ArH), 7.25 (m, 2H; ArH), 7.57
(brt, 1H; CONHCH2), 7.71 (d, J ) 6 Hz, 1H; ArH). 13C NMR
(100 MHz, CDCl3, 25 °C): δ ) 39.6 (CH2), 55.4 (OCH3), 110.5
(ArC), 115.5 (ArC), 120.8 (ArC), 124.5 (ArC), 129.6 (ArC), 130.1
(ArC), 136.1 (ArC), 149.7 (ArC), 152.8 (ArC), 157.6 (ArC), 163.0
(CdO), 173.7 (ArCdO). APCI-MS(+) m/z 275.88 [M + H]+. Anal.
(C14H13NO5) C, H, N.
1
of 3b (65 mg, 0.22 mmol). Yield ) 41%. H NMR (400 MHz,
CDCl3, 25 °C): δ ) 3.77 (s, 3H; OCH3), 4.56 (d, J ) 5.2 Hz,
NHCH2), 6.85 (d, J ) 8 Hz, 2H; ArH), 7.24 (d, J ) 8.4 Hz, 2H;
ArH), 7.36 (d, J ) 4.8 Hz, 1H; ArH), 7.64 (d, J ) 4.8 Hz, 1H;
ArH), 7.80 (brt, 1H; CONHCH2), 9.95 (brs, 1H; ArOH). 13C NMR
(100 MHz, CDCl3, 25 °C): δ ) 43.3 (CH2), 55.2 (OCH3), 114.1
(ArC), 125.9 (ArC), 128.8 (ArC), 129.1 (ArC), 132.0 (ArC), 146.9
(ArC), 151.6 (ArC), 159.1 (ArC), 160.8 (CdO), 190.6 (ArC)S).
ESI-MS(-) m/z 290.09 [M - H]-. Anal. (C14H13NO4S·0.35H2O).
C, H, N, S, respectively: calcd 56.50, found 56.90; calcd 4.64, found
5.04; calcd 4.71, found 4.49; calcd 10.77, found 10.38.
3-Hydroxy-N-(2-methoxybenzyl)-4-thioxo-4H-pyran-2-car-
boxamide (5b). This compound was prepared from 5a according
to the procedure outlined for 1b. 5a (400 mg, 1.4 mmol) was treated
with P4S10 and HMDO for 2.5 h to yield a reddish brown solid of
5b (290 mg, 1.0 mmol). Yield ) 76%. 1H NMR (300 MHz, CDCl3,
25 °C): δ ) 3.90 (s, 3H; OCH3), 4.65 (d, J ) 5.7 Hz, NHCH2),
6.90 (q, J ) 7.8 Hz, 2H; ArH), 7.28-7.35 (m, 2H; ArH), 7.40 (d,
J ) 4.8 Hz, 1H; ArH), 7.68 (d, J ) 5.7 Hz, 1H; ArH), 8.08 (brt,
1H; CONHCH2), 9.88 (brs, 1H; ArOH). 13C NMR (75 MHz, CDCl3,
25 °C): δ ) 39.9 (CH2), 55.4 (OCH3), 110.4 (ArC), 120.7 (ArC),
125.0 (ArC), 125.8 (ArC), 129.4 (ArC), 129.9 (ArC), 133.4 (ArC),
147.0 (ArC), 152.1 (ArC), 157.6 (ArC), 161.4 (CdO), 191.7
(ArCdO). ESI-MS(+) m/z 292.00 [M + H]+. HRMS calcd for
C14H13NO4S: 291.0560. Found: 291.0563.
3-Hydroxy-N-(2-methoxybenzyl)-4-thioxo-4H-pyran-2-car-
bothioamide (5c). 5c was obtained from 5a from the same reaction
as that of 5b as a dark-brown solid (90 mg, 0.29 mmol). Yield )
21%. 1H NMR (400 MHz, CDCl3, 25 °C): δ ) 3.90 (s, 3H; OCH3),
4.99 (d, J ) 5.2 Hz, NHCH2), 6.93 (q, J ) 7.2 Hz, 2H; ArH), 7.33
(q, J ) 8 Hz, 2H; ArH), 7.40 (d, J ) 5.2 Hz, 1H; ArH), 7.73 (d,
J ) 5.2 Hz, 1H; ArH), 9.70 (brt, 1H; CONHCH2), 10.43 (brs, 1H;
ArOH). 13C NMR (100 MHz, CDCl3, 25 °C): δ ) 45.4 (CH2),
55.1 (OCH3), 110.3 (ArC), 120.4 (ArC), 122.9 (ArC), 125.7 (ArC),
129.5 (ArC), 130.2 (ArC), 133.6 (ArC), 146.4 (ArC), 150.1 (ArC),
157.4 (ArC), 183.6 (CdS), 191.8 (ArCdS). APCI-MS(+) m/z
307.91 [M + H]+. Anal. (C14H13NO3S2 ·0.5H2O). C, H, N, S,
respectively: calcd 53.15, found 52.90; calcd 4.46, found 4.49; calcd
4.43, found 4.12; calcd 20.27, found 20.44.
3-Hydroxy-N-(4-methoxybenzyl)-4-thioxo-4H-pyran-2-car-
bothioamide (3c). 3c was obtained from 3a from the same reaction
as that of 3b as a dark-brown solid (20 mg, 0.07 mmol). Yield )
7%. 1H NMR (400 MHz, CDCl3, 25 °C): δ ) 3.82 (s, 3H; OCH3),
4.87 (d, J ) 5.6 Hz, NHCH2), 6.91 (d, J ) 8.8 Hz, 2H; ArH), 7.31
(d, J ) 8.4 Hz, 2H; ArH), 7.40 (d, J ) 4.8 Hz, 1H; ArH), 7.73 (d,
J ) 5.2 Hz, 1H; ArH), 9.38 (brt, 1H; CONHCH2), 10.33 (brs, 1H;
ArOH). 13C NMR (100 MHz, CDCl3, 25 °C): δ ) 49.8 (CH2),
55.2 (OCH3), 114.2 (ArC), 125.3 (ArC), 127.1 (ArC), 129.6 (ArC),
133.6 (ArC), 147.0 (ArC), 149.4 (ArC), 159.4 (ArC), 184.2 (CdS),
191.2 (ArCdS). APCI-MS(-) m/z 306.06 [M - H]-. Anal.
(C14H13NO3S2) C, H, N, S.
N-Benzhydryl-3-hydroxy-4-oxo-4H-pyran-2-carboxamide (4a).
This compound was prepared according to the same procedure
outlined for 3a starting from BPCA (1.0 g, 4.1 mmol), EDCI (1.2
equiv, 939 mg, 4.9 mmol), HOBt (1.2 equiv, 662 mg, 4.9 mmol),
and diphenylmethanamine (1.2 equiv, 898 mg, 844 µL, 4.9 mmol)
1
to yield a tan solid of 4a (732 mg, 2.3 mmol). Yield ) 56%. H
NMR (400 MHz, CDCl3, 25 °C): δ ) 6.37 (d, J ) 8 Hz, NHCH),
6.45 (d, J ) 5.2 Hz, 1H; ArH), 7.05 (brd, J ) 7.6 Hz, CONHCH2),
7.28-7.43 (m, 10H; ArH), 7.72 (d, J ) 6 Hz, 1H; ArH). 13C NMR
(100 MHz, DMSO-d6, 25 °C): δ ) 56.4 (CH2), 114.4 (ArC), 127.3
(ArC), 128.5 (ArC), 137.7 (ArC), 141.2 (ArC), 147.9 (ArC), 155.2
(ArC), 160.8 (CdO), 173.5 (ArCdO). ESI-MS(+) m/z 321.85 [M
+ H]+. Anal. (C19H15NO4 ·0.25H2O). C, H, N: calcd 70.04, found
70.09; calcd 4.79, found 5.16; calcd 4.30, found 4.47.
3-Hydroxy-N-(3-methoxybenzyl)-4-oxo-4H-pyran-2-carbox-
amide (6a). This compound was prepared according to the same
procedure outlined for 3a starting from BPCA (500 mg, 2.0 mmol),
EDCI (1.2 equiv, 713 mg, 2.4 mmol), HOBt (1.2 equiv, 324 mg,
2.4 mmol), and 3-methoxybenzylamine (1.2 equiv, 329 mg, 310
µL, 2.4 mmol) to yield a tan solid of 6a (330 mg, 1.2 mmol). Yield
N-Benzhydryl-3-hydroxy-4-thioxo-4H-pyran-2-carboxam-
ide (4b). This compound was prepared from 4a according to the
procedure outlined for 1b. 4a (600 mg, 1.9 mmol) was treated with
P4S10 and HMDO for 3 h to yield a light-brown solid of 4b (90
1
1
mg, 0.27 mmol). Yield ) 14%. H NMR (400 MHz, CDCl3, 25
) 60%. H NMR (400 MHz, CDCl3, 25 °C): δ ) 3.79 (s, 3H;
°C): δ ) 6.44 (d, J ) 8 Hz, NHCH), 7.30-7.38 (m, 10H; ArH),
7.41 (d, J ) 5.2 Hz, 1H; ArH), 7.70 (d, J ) 4.8 Hz, 1H; ArH),
8.27 (brd, J ) 7.2 Hz, CONHCH2), 9.72 (brs, 1H; ArOH). 13C
NMR (100 MHz, DMSO-d6, 25 °C): δ ) 56.5 (ArC), 127.4 (ArC),
127.5 (ArC), 128.5 (ArC), 132.5 (ArC), 140.9 (ArC), 147.0 (ArC),
153.5 (ArC), 161.9 (CdO), 191.9 (ArCdS). APCI-MS(+) m/z
338.18 [M + H]+. Anal. (C19H15NO3S·0.6H2O). C, H, N, S,
respectively: calcd 65.54, found 65.60; calcd 4.69, found 5.05; calcd
4.02, found 4.02; calcd 9.21, found 10.33.
N-Benzhydryl-3-hydroxy-4-thioxo-4H-pyran-2-carbothioam-
ide (4c). 4c was obtained from 4a from the same reaction as that
of 4b as a dark-brown solid (216 mg, 0.61 mmol). Yield ) 32%.
1H NMR (400 MHz, CDCl3, 25 °C): δ ) 6.93 (d, J ) 7.2 Hz,
NHCH), 7.30-7.40 (m, 10H; ArH), 7.42 (d, J ) 5.2 Hz, 1H; ArH),
OCH3), 4.57 (d, J ) 4.8 Hz, NHCH2), 6.52 (d, J ) 6.0 Hz, 1H;
ArH), 6.82-6.92 (m, 3H; ArH), 7.24 (t, J ) 7.8 Hz, 1H; ArH),
7.60 (brt, 1H; CONHCH2), 7.81 (d, J ) 5.2 Hz, 1H; ArH). 13C
NMR (100 MHz, CDCl3, 25 °C): δ ) 43.5 (CH2), 55.3 (OCH3),
113.2 (ArC), 113.4 (ArC), 113.7 (ArC), 120.1 (ArC), 130.0 (ArC),
136.0 (ArC), 138.1 (ArC), 149.6 (ArC), 153.1 (ArC), 159.9 (ArC),
163.1 (CdO), 173.6 (ArCdO). APCI-MS(+) m/z 275.96 [M +
H]+. Anal. (C14H13NO5) C, H, N.
3-Hydroxy-N-(3-methoxybenzyl)-4-thioxo-4H-pyran-2-car-
boxamide (6b). This compound was prepared from 6a according
to the procedure outlined for 1b. 6a (40 mg, 0.15 mmol) was treated
with P4S10 and HMDO for 40 min to yield a brown solid of 6b (40
mg, 0.14 mmol). Yield ) 95%. 1H NMR (300 MHz, CDCl3,
25 °C): δ ) 3.81 (s, 3H; OCH3), 4.64 (d, J ) 6.0 Hz, NHCH2),

Thioamide Hydroxypyrothiones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1071
6.83-6.89 (m, 2H; ArH), 6.92 (d, J ) 7.8 Hz, 1H; ArH), 7.28 (d,
J ) 7.8 Hz, 1H; ArH), 7.41 (d, J ) 5.4 Hz, 1H; ArH), 7.70 (d, J
) 4.8 Hz, 1H; ArH), 7.86 (brt, 1H; CONHCH2), 9.74 (brs, 1H;
ArOH). 13C NMR (75 MHz, DMSO-d6, 25 °C): δ ) 42.4 (CH2),
55.0 (OCH3), 112.5 (ArC), 113.2 (ArC), 119.5 (ArC), 127.8 (ArC),
129.5 (ArC), 131.7 (ArC), 139.6 (ArC), 154.1 (ArC), 159.3 (ArC),
161.9 (ArC), 163.4 (CdO), 192.4 (ArCdS). ESI-MS(+) m/z 292.05
[M + H]+. HRMS calcd for C14H13NO3S: 291.0560. Found:
291.0556.
off-white solid of 8a (210 mg, 1.2 mmol). Yield ) 64%. 1H NMR
(400 MHz, CDCl3, 25 °C): δ ) 1.26 (t, J )7.4 Hz, 2H; CH2CH3),
3.47 (p, J ) 7.2 Hz, 2H; NHCH2CH3), 6.46 (d, J ) 5.2 Hz, 1H;
ArH), 6.75 (brt, 1H; CONHCH2), 7.71 (d, J ) 5.6 Hz, 1H; ArH).
13C NMR (100 MHz, CDCl3, 25 °C): δ ) 14.6 (CH2CH3), 34.6
(CH2CH3), 115.5 (ArC), 136.0 (ArC), 149.5 (ArC), 152.9 (ArC),
163.2 (CdO), 173.7 (ArCdO). ESI-MS(+) m/z 184.09 [M + H]+.
HRMS calcd for C8H9NO4: 183.0526. Found: 183.0525.
N-Ethyl-3-hydroxy-4-thioxo-4H-pyran-2-carboxamide (8b).
This compound was prepared from 8a according to the procedure
outlined for 1b. 8a (107 mg, 0.6 mmol) was treated with P4S10 and
HMDO for 25 min to yield a reddish-brown solid of 8b (30 mg,
0.15 mmol). Yield ) 25%. 1H NMR (300 MHz, CDCl3, 25 °C): δ
) 1.26 (t, J ) 7.4 Hz, 2H; CH2CH3), 3.50 (p, J ) 7.0 Hz, 2H;
CH2CH3), 7.42 (d, J ) 5.2 Hz, 1H; ArH), 7.71 (brt, 1H;
CONHCH2), 7.70 (d, J ) 5.2 Hz, 1H; ArH), 9.86 (brs, 1H; ArOH).
ESI-MS(+) m/z 222.02 [M + Na]+. HRMS calcd for C8H9NO3SNa:
222.0195. Found: 222.0193.
N-Ethyl-3-hydroxy-4-thioxo-4H-pyran-2-carbothioamide (8c).
This compound was prepared from 8a according to the procedure
outlined for 1c. 8a (165 mg, 0.90 mmol) was treated for 2.5 h to
yield a dark-brown solid of 8c (77 mg, 0.36 mmol). Yield ) 40%.
1H NMR (400 MHz, CDCl3, 25 °C): δ ) 1.38 (t, J ) 7.4 Hz, 2H;
CH2CH3), 3.81-3.88 (m, 2H; CH2CH3), 7.41 (d, J ) 5.2 Hz, 1H;
ArH), 7.74 (d, J ) 4.8 Hz, 1H; ArH), 9.21 (brt, 1H; CONHCH2),
10.44 (brs, 1H; ArOH). 13C NMR (100 MHz, CDCl3, 25 °C): δ )
12.9 (CH2CH3), 41.1 (CH2CH3), 125.3 (ArC), 133.7 (ArC), 147.1
(ArC), 149.3 (ArC), 184.6 (CdS), 191.4 (ArCdS). APCI-MS(+)
m/z 216.01 [M + H]+. Anal. (C8H9NO2S2). C, H, N, S, respectively:
calcd 44.63, found 44.89; calcd 4.21, found 4.70; calcd 6.51, found
6.51; calcd 29.79, found 30.34.
3-Hydroxy-N-(3-methoxybenzyl)-4-thioxo-4H-pyran-2-car-
bothioamide (6c). This compound was prepared from 6a according
to the procedure outlined for 1c. 6a (450 mg, 1.6 mmol) was treated
for 2 h to yield a dark-greenish-brown solid of 6c (126 mg, 0.43
1
mmol). Yield ) 27%. H NMR (400 MHz, CDCl3, 25 °C): δ )
3.83 (s, 3H; OCH3), 4.93 (d, J ) 5.2 Hz, NHCH2), 6.89-6.92 (m,
2H; ArH), 6.96 (d, J ) 8.0 Hz, 1H; ArH), 7.30 (t, J ) 8.0 Hz, 1H;
ArH), 7.42 (d, J ) 5.2 Hz, 1H; ArH), 7.76 (d, J ) 5.2 Hz, 1H;
ArH), 9.47 (brt, 1H; CONHCH2), 10.26 (brs, 1H; ArOH). 13C NMR
(75 MHz, CDCl3, 25 °C): δ ) 50.4 (CH2), 55.3 (OCH3), 113.6
(ArC), 113.9 (ArC), 120.4 (ArC), 125.2 (ArC), 129.2 (ArC), 130.2
(ArC), 133.9 (ArC), 136.8 (ArC), 147.5 (ArC), 160.0 (ArC), 185.3
(CdS), 197.0 (ArCdS). APCI-MS(+) m/z 307.96 [M + H]+. Anal.
(C14H13NO3S2 · 0.55H2O). C, H, N, S, respectively: calcd 52.99,
found 52.83; calcd 4.48, found 4.57; calcd 4.41, found 4.40; calcd
20.21, found 21.52.
N-Benzyl-3-hydroxy-4-oxo-4H-pyran-2-carboxamide (7a). This
compound was prepared according to the same procedure outlined
for 3a starting from BPCA (575 mg, 2.3 mmol), EDCI (1.0 equiv,
441 mg, 2.3 mmol), HOBt (1.0 equiv, 311 mg, 2.3 mmol), and
benzylamine (1.0 equiv, 246 mg, 251 µL, 2.3 mmol) to yield a tan
solid of 7a (294 mg, 1.2 mmol). Yield ) 52%. 1H NMR (400 MHz,
CDCl3, 25 °C): δ ) 4.63 (d, J ) 6 Hz, NHCH2), 6.46 (d, J ) 5.2
Hz, 1H; ArH), 7.04 (brt, 1H; CONHCH2), 7.34-7.41 (m, 5H; ArH),
7.69 (d, J ) 6 Hz, 1H; ArH). 13C NMR (100 MHz, CDCl3, 25 °C):
δ ) 43.6 (CH2), 115.5 (ArC), 128.0 (ArC), 128.1 (ArC), 129.0
(ArC), 135.9 (ArC), 136.5 (ArC), 149.6 (ArC), 153.0 (ArC), 163.1
(CdO), 173.5 (ArCdO). APCI-MS(+) m/z 246.00 [M + H]+. Anal.
(C13H11NO4) C, H, N.
N-Benzyl-3-hydroxy-4-thioxo-4H-pyran-2-carboxamide (7b).
This compound was prepared from 7a according to the procedure
outlined for 1b. 7a (250 mg, 1.0 mmol) was treated for 1.5 h to
yield a mustard-brown solid of 7b (85 mg, 0.32 mmol). Yield )
32%. 1H NMR (300 MHz, CDCl3, 25 °C): δ ) 4.51 (d, J ) 6 Hz,
NHCH2), 7.15-7.20 (m, 1H; ArH), 7.23 (s, 3H; ArH), 7.24 (s, 1H;
ArH), 7.27 (d, J ) 4.4 Hz, 1H; ArH), 7.58 (d, 7.04, J ) 4.4 Hz,
1H; ArH), 8.29 (brt, 1H; CONHCH2), 9.80 (brs, 1H; ArOH). 13C
NMR (75 MHz, DMSO-d6, 25 °C): δ ) 43.3 (CH2), 126.3 (ArC),
127.4 (ArC), 127.5 (ArC), 128.5 (ArC), 131.5 (ArC), 136.7 (ArC),
146.4 (ArC), 152.2 (ArC), 161.7 (CdS), 191.1 (ArCdS). APCI-
MS(+) m/z 262.05 [M + H]+. HRMS calcd for C13H11NO3S:
261.0454. Found: 291.0455.
N-Benzyl-3-hydroxy-4-thioxo-4H-pyran-2-carbothioamide (7c).
This compound was prepared from 7a according to the procedure
outlined for 1c. 7a (180 mg, 0.73 mmol) was treated with P4S10
and HMDO for 2.5 h to yield a dark-red solid of 7c (24 mg,
0.09 mmol). Yield ) 12%. 1H NMR (400 MHz, CDCl3, 25 °C):
δ ) 4.961 (d, J ) 5.2 Hz, NHCH2), 7.347-7.393 (m, 5H; ArH),
7.409 (d, J ) 5.2 Hz, 1H; ArH), 7.744 (d, J ) 5.2 Hz, 1H; ArH),
9.473 (brt, 1H; CONHCH2), 10.305 (brs, 1H; ArOH). 13C NMR
(100 MHz, CDCl3, 25 °C): δ ) 50.021 (CH2), 125.403 (ArC),
128.042 (ArC), 128.149 (ArC), 128.855 (ArC), 133.563 (ArC),
135.079 (ArC), 146.847 (ArC), 149.639 (ArC), 184.477 (CdS),
191.316 (ArCdS). APCI-MS(+) m/z 278.08 [M + H]+. Anal.
(C13H11NO2S2 ·0.5H2O). C, H, N, S, respectively: calcd 54.52, found
54.69; calcd 4.22, found 4.62; calcd 5.71, found 4.89; calcd 22.39,
found 21.89.
N-(Biphenyl-4-ylmethyl)-1-hydroxy-6-oxo-1,6-dihydropyridine-
2-carboxamide (9a). This compound was prepared as previously
reported.46
N-(4-Fluorobenzyl)-1-hydroxy-6-oxo-1,6-dihydropyridine-2-
carboxamide (10a). To a solution of 1-(benzyloxy)-6-oxo-1,6-
dihydropyridine-2-carboxylic acid (BDCA)46 (200 mg, 0.82 mmol)
in dichloromethane was added EDCI (1.0 equiv, 157 mg, 0.82
mmol), HOBt (1.0 equiv, 111 mg, 0.82 mmol), and 4-fluoroben-
zylamine (1.0 equiv, 93 µL, 0.82 mmol). The mixture was stirred
overnight at room temperature under N2 and then poured into a
separatory funnel and extracted with water and dichloromethane.
The organic layers were combined, dried over anhydrous magne-
sium sulfate, filtered, and concentrated to give the benzyl-protected
intermediate. The intermediate was dissolved in 10 mL of a 1:1
solution of concentrated HCl and glacial acetic acid. The reaction
mixture was stirred for 5 d and then concentrated in vacuo,
coevaporated with MeOH, and dried to yield an off-white solid
1
(55 mg, 0.21 mmol). Yield ) 26%. H NMR (400 MHz, CDCl3,
25 °C): δ ) 4.51 (d, J ) 5.6 Hz, 2H; NHCH2), 6.88 (q, J ) 9.0
Hz, 2H; ArH), 7.23 (m, 4H; ArH), 7.41 (t, J ) 8.0 Hz, 1H; ArH),
9.66 (br t, 1H; CONHCH2). APCI-MS(+) m/z 263.01 [M + H]+.
2-Bromopyridine-6-carboxylic Acid 1-Oxide (BCAO). The
synthesis is based on a modified literature procedure.48 To
6-bromopicolinic acid (6-BPA) (2.0 g, 9.9 mmol) was added a
solution of 26 mL of trifluoroacetic acid and 4 mL of 30% hydrogen
peroxide. The reaction mixture was heated to reflux at 80 °C for
7 h under N2. The solution was concentrated in vacuo and the
concentrate was added to 200 mL of water, which resulted in the
immediate formation of a precipitate. The precipitate was vacuum-
filtered and dried to afford an off-white solid (2.03 g, 9.3 mmol).
1
Yield ) 94%. H NMR (400 MHz, DMSO-d6, 25 °C): δ 7.69
(t, J ) 8.0 Hz, 1H; ArH), 8.24 (dd, J ) 6.0, 10.4 Hz, 1H; ArH),
8.30 (dd, J ) 5.6, 10.4 Hz, 1H; ArH). ESI-MS(-): m/z 215.86 [M
- H]-.
1-Hydroxy-6-thioxo-1,6-dihydropyridine-2-carboxylic Acid
(HTDCA). The synthesis is based on a modified literature proce-
dure.47 BCAO (3.0 g, 14.0 mmol) was dissolved in a 50 mL solution
of Na2S·9H2O (3.3 g, 14 mmol) at 70 °C. To this yellow solution
was added 14 mmol of NaHS in 50 mL of water, and the pH was
adjusted to 10. The resulting reaction mixture was heated to reflux
N-Ethyl-3-hydroxy-4-oxo-4H-pyran-2-carboxamide (8a). This
compound was prepared according to the same procedure outlined
for 3a starting from BPCA (500 mg, 2.0 mmol), EDCI (1.2 equiv,
713 mg, 2.4 mmol), HOBt (1.2 equiv, 324 mg, 2.4 mmol), and 2
M ethylamine in THF (1.2 equiv, 1.2 mL, 2.4 mmol) to yield an

1072 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Agrawal et al.
at 95 °C for 6 h under N2 and acidified while still hot with
concentrated HCl to pH 2. The solution was concentrated in vacuo
and the resulting precipitate was vacuum-filtered to yield a crude
yellow crystalline solid. The crude acid was partially dissolved in
dichloromethane and extracted 2× with 5 M NaOH. The aqueous
phase was acidified to pH 1 with concentrated HCl and was further
extracted 3× with dichloromethane. The organic phase was dried
over anhydrous magnesium sulfate that was then vacuum-filtered
off. The filtrate was concentrated to yield a pure bright-yellow solid
(2.1 g, 12.3 mmol). Yield ) 88%. 1H NMR (400 MHz, CDCl3, 25
°C): δ 7.48 (t, J ) 8.0 Hz, 1H; ArH), 7.67 (dd, J ) 5.6, 10.4 Hz,
1H; ArH), 8.18 (dd, J ) 6.0, 9.6 Hz, 1H; ArH). APCI-MS(+): m/z
172.12 [M + H]+. HR-EI-MS calcd for C6H5NO3S 170.9984.
Expected 170.9985. Anal. (C6H5NO3S) C, H, N.
Construction of an Inhibitor/Anthrax LF Complex Model.
Our starting structure was the complex from Lewis et al.,40 in which
11b is in the anthrax lethal factor (LF) active site. The 6-methyl
group of 11b was replaced with a hydrogen atom to construct
inhibitor 1b. The inhibitor 1c was obtained by replacing the carbonyl
oxygen atom of 1b with a sulfur atom. The atomic charges for
both 1b and 1c were derived by fitting to HF/6-31G* electrostatic
potentials (ESP) using the RESP module. Other force parameters
for the two inhibitors were adapted from the standard force field
by following the general parametrization procedures outlined in
AMBER manual.63,64 After the added atoms in the gas phase were
relaxed, each structure was immersed in a cubic TIP3P water box
and neutralized by addition of Na+ counterions using the AMBER
Leap module. This led to the [1b·LF] (or [1c·LF]) simulation
system of 55 268 atoms. MD simulations were conducted in the
NPT ensemble at 300 K and 1 atm. The SHAKE algorithm65 was
used to constrain all bond lengths involving hydrogens. A 10.0 Å
cutoff was used for nonbonded interactions, and the neighbor pair
list was updated every 10 steps. The long-range electrostatic
interactions were treated with the particle mesh Ewald method.66
The two prepared systems were first equilibrated with a series of
minimizations interspersed by short MD simulations, and then two
snapshots from two models were selected for the following QM/
MM investigations. These selected structures were first minimized
using the MM method and then optimized with the B3LYP(6-31G*)
QM/MM calculations using an iterative minimization approach.67
QM/MM Minimization Calculations. The pseudobond ab initio
QM/MM approach68,69 has achieved great success in accurate
modeling of the chemistry at an enzyme active site while properly
including the effects of the enzymatic environment. It has been
applied to various enzyme reactions, including enolase,70 acetyl-
cholinesterase,71 4-oxalocrotonate tautomerase,72 kinase,73 and
methyl transferase.74 The QM/MM approach divides the whole
enzyme-substrate system into a QM and a MM subsystem. The
active site of the enzyme was described by a QM Hamiltonian,
and the influence of the remainder of the protein and the solvent
was included via a coupled MM potential. The code combining
the modified Gaussian 9874 and Tinker 3.675 was utilized for all
the calculations. An efficient iterative optimization procedure67 was
repeatedly applied to minimize the enzyme-substrate system into
a local minimum. For each minimization cycle, the large MM
subsystem was relaxed with the truncated Newton method in
Cartesian coordinates while the small QM subsystem was treated
using the quasi-Newton minimizer at the B3LYP/6-31G* level in
redundant internal coordinates. For the QM/MM calculations on
the [1b·LF] and [1c·LF] models, the QM subsystem consists of
the side chain of His422, His426, Glu471, zinc(II), and 1b (or 1c),
resulting in a total of 70 QM atoms. Geometry optimizations for
both the QM subgroups were at the B3LYP(6-31G*) level, with
the boundary between the QM and MM subsystems treated using
the pseudobond approach.69 The scheme used in our QM/MM
calculations corresponds to a good compromise between accuracy
and computational cost. While being economical for the large
system size, B3LYP is quite reliable for geometry optimizations
including those of Zn(II) complexes, which is the primary issue in
this work. More accurate schemes are available where computational
requirements may be more demanding.76 All other atoms are
described by the classical MM force fields. Similarly, we con-
structed the [2b·LF] and [2c·LF], [3b·LF] and [3c·LF], and
[7b·LF] and [7c·LF] QM/MM models.
N-(Biphenyl-4-ylmethyl)-1-hydroxy-6-thioxo-1,6-dihydropy-
ridine-2-carboxamide (9b). To a solution of HTDCA (105 mg,
0.6 mmol) in dichloromethane was added NHS (1.0 equiv, 70 mg,
0.6 mmol), EDCI (1.0 equiv, 117 mg, 0.6 mmol), and 4-phenyl-
benzylamine (1.0 equiv, 112 mg, 0.6 mmol). The mixture was
stirred overnight at room temperature under N2 and then poured
into a separatory funnel and extracted with saturated NaHCO3 and
dichloromethane. The organic layers were combined, dried over
anhydrous magnesium sulfate, filtered, and concentrated to a yellow
residue. The crude residue was purified via silica column chroma-
tography (1% MeOH/CHCl3). Maltol (3-hydroxy-2-methyl-4H-
pyran-4-one) was first run through the column to remove any metal
impurities in the silica. The compound 9b was isolated as a yellow
solid after removal of solvent (94 mg, 0.3 mmol). Yield ) 46%.
1H NMR (400 MHz, CDCl3, 25 °C): δ 4.73 (d, J ) 6.0 Hz, 2H;
NHCH2,), 7.33 (t, J ) 6.4 Hz, 1H; ArH), 7.42-7.59 (m, 9H; ArH),
7.67 (dd, J ) 6.4, 10.0 Hz, 1H; ArH), 8.32 (dd, J ) 6.0, 10.0 Hz,
1H; ArH), 11.17 (t, J ) 5.8 Hz, 1H; NHCH2). APCI-MS(-): m/z
335.04 [M - H]-. Anal. (C19H16N2O2). C, H, N, S, respectively:
calcd 67.84, found 67.36; calcd 4.79, found 5.06; calcd 8.33, found
8.13; calcd 9.53, found 9.32.
N-(4-Fluorobenzyl)-1-hydroxy-6-thioxo-1,6-dihydropyridine-
2-carboxamide (10b). The synthesis of 10b was performed as
described for 9b, starting from HTDCA (50 mg, 0.3 mmol), EDCI
(1.0 equiv, 56 mg, 0.3 mmol), HOBt (1.0 equiv, 39 mg, 0.3 mmol),
and 4-fluorobenzylamine (1.0 equiv, 36 mg, 33 µL, 0.3 mmol).
Yield ) 95%. 1H NMR (400 MHz, CDCl3, 25 °C): δ 4.65 (d, J )
6.0 Hz, 2H; NHCH2,), 7.01 (t, J ) 8.8 Hz, 2H; ArH), 7.34 (qd, J
) 3.2, 4.8 Hz, 2H; ArH), 7.43 (t, J ) 8.4 Hz, 1H; ArH), 7.66 (dd,
J ) 6.4, 10.4 Hz, 1H; ArH), 8.30 (dd, J ) 6.8, 9.6 Hz, 1H; ArH),
11.13 (t, J ) 5.6 Hz, 1H; NHCH2). APCI-MS(-): m/z 276.96 [M
- H]-.
Single-Crystal X-ray Diffraction. A single crystal of 5c suitable
for X-ray diffraction structural determination was mounted on a
nylon loop with Paratone oil and placed under a nitrogen cold
stream (100 K). Data were collected on a Bruker P4 diffractometer
using Mo KR radiation (λ ) 0.710 73 Å) controlled using the APEX
2.0 software package. A semiempirical method utilizing equivalents
was employed to correct for absorption.62 The data collection was
solved and refined using SHELXTL suite. All hydrogen atoms were
fixed at calculated positions with isotropic thermal parameters, while
all non-hydrogen atoms were refined anisotropically.
Inhibition Assays. The compounds were dissolved in dimethyl
sulfoxide (DMSO) at a concentration of 50 mM. Serial dilutions
of the inhibitor were made from this stock and were diluted 500×
in LF buffer (20 mM HEPES, 1 mM CaCl2, 0.1 mg/mL BSA,
0.01% Tween-20, pH 7). The assay was carried out in 96-well
plates. Each well contained a total volume of 100 µL ) 25 µL
buffer, 10 nM (30 µL) recombinant LF (List Biological Labora-
tories), 20 µL LFi, and the reaction was initiated by the addition
of 3 µM (25 µL) fluorogenic LF substrate (Cou)-N-Nle-Lys-Lys-
Lys-Lys-Val-Leu-Pro-Ile-Gln-Leu-Asn-Ala-Ala-Thr-Asp-Lys-(QSY-
35)-Gly-Gly-NH2 (Calbiochem) after a 35 min incubation period
at 25 °C. Kinetic measurements were recorded every minute for
15 min with excitation and emission wavelengths at 380 and 450
nm, respectively.
Acknowledgment. The authors thank Kristine K. Tanabe
for solving the crystal structure of 5c and Dr. Yongxuan Su
(University of California, San Diego) for obtaining the mass
spectrometry data. This work has been supported by a grant
from the National Institutes of Health (S.M.C., Grant R03
AI070651-01) and in part by grants from NSF, NIH, the Center
for Theoretical Biological Physics, the National Biomedical
Computation Resource, NSF Supercomputing Centers, and
Accelrys, Inc. (J.A.M).

Thioamide Hydroxypyrothiones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1073
anthrax lethal factor protease. Bioorg. Med. Chem. Lett. 2006, 16,
5183–5189.
(26) Johnson, S. L.; Chen, L. H.; Pellecchia, M. A high-througput screening
approach to anthrax lethal factor inhibition. Bioorg. Chem. 2007, 35,
306–312.
(27) Johnson, S. L.; Jung, D.; Forino, M.; Chen, Y.; Satterhwait, A.;
Rozanov, D. V.; Strongin, A. Y.; Pellecchia, M. Anthrax lethal factor
protease inhibitors: synthesis, SAR, and structure-based 3D QSAR
studies. J. Med. Chem. 2006, 49, 27–30.
(28) Johnson, S. L.; Pellecchia, M. Structure- and fragment-based ap-
proaches to protease inhibition. Curr. Top. Med. Chem. 2006, 6, 317–
329.
(29) Numa, M. M. D.; Lee, L. V.; Hsu, C. C.; Bower, K. E.; Wong, C. H.
Identification of novel anthrax lethal factor inhibitors generated by
combinatorial Pictet-Spengler reaction followed by screening in situ.
ChemBioChem 2005, 2005, 1002–1006.
(30) Panchal, R. G.; Hermone, A. R.; Nguyen, T. L.; Wong, T. Y.;
Schwarzenbacher, R.; Schmidt, J.; Lane, D.; McGrath, C.; Turk, B. E.;
Burnett, J.; Aman, M. J.; Little, S.; Sausville, E. A.; Zaharevitz, D. W.;
Cantley, L. C.; Liddington, R. C.; Gussio, R.; Bavari, S. Identification
of small molecule inhibitors of anthrax lethal factor. Nat. Struct. Mol.
Biol. 2004, 11, 67–72.
Supporting Information Available: Crystallographic informa-
tion for compound 5c in CIF format, results from combustion
analysis, table of crystallographic data collection and structural
parameters for 5c, and structural diagram of 5c. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Pile, J. C.; Malone, J. D.; Eitzen, E. M.; Friedlander, A. M. Anthrax
as a potential biological warfare agent. Arch. Intern. Med. 1998, 158,
429–434.
(2) Collier, R. J.; Young, J. A. T. Anthrax toxin. Annu. ReV. Cell DeV.
Biol. 2003, 19, 45–70.
(3) Smith, H.; Keppie, J. Observations on experimental anthrax: demon-
stration of a specific lethal factor produced in vivo by Bacillus
anthracis. Nature 1954, 173, 869–870.
(4) Tournier, J. N.; Quesnel-Hellmann, A.; Cleret, A.; Vidal, D. R.
Contribution of toxins to the pathogenesis of inhalational anthrax. Cell.
Microbiol. 2007, 9, 555–565.
(5) Bradley, K. A.; Mogridge, J.; Mourez, M.; Collier, R. J.; Young,
J. A. T. Identification of the cellular receptor for anthrax toxin. Nature
2001, 414, 225–229.
(31) Schepetkin, I. A.; Khlebnikov, A. I.; Kirpotina, L. N.; Quinn, M. T.
Novel small-molecule inhibitors of anthrax lethal factor identified by
high-throughput screening. J. Med. Chem. 2006, 49, 5232–5244.
(32) Shoop, W. L.; Xiong, Y.; Wiltsie, J.; Woods, A.; Guo, J.; Pivnichny,
J. V.; Felcetto, T.; Michael, B. F.; Bansal, A. C., R. T.; Cunningham,
B. R.; Friedlander, A. M.; Douglas, C. M.; Patel, S. B.; Wisniewski,
D.; Scapin, G.; Salowe, S. P.; Zaller, D. M.; Chapman, K. T.; Scolnick,
E. M.; Schmatz, D. M.; Bartizal, K.; MacCoss, M. Anthrax lethal factor
inhibition. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 7958–7963.
(33) Turk, B. E.; Wong, T. Y.; Schwarzenbacher, R.; Jarrell, E. T.; Leppla,
S. H.; Collier, R. J.; Liddington, R. C.; Cantley, L. C. The structural
basis for substrate and inhibitor selectivity of the anthrax lethal factor.
Nat. Struct. Mol. Biol. 2004, 11, 60–66.
(34) Xiong, Y.; Wiltsie, J.; Woods, A. G., J.; Pivnichny, J. V.; Tang, W.;
Bansal, A.; Cummings, R. T.; Cunningham, B. R.; Friedlander, A. M.;
Douglas, C. M.; Salowe, S. P.; Zaller, D. M.; Scolnick, E. M.; Schmatz,
D. M.; Bartizal, K.; Hermes, J. D.; MacCoss, M.; Chapman, K. T.
The discovery of a potent and selective lethal factor inhibitor for
adjunct therapy of anthrax infection. Bioorg. Med. Chem. Lett. 2006,
16, 964–968.
(35) Konstantinopoulos, P. A.; Karamouzis, M. V.; Papatsoris, A. G.;
Papavassiliou, A. G. Matrix metalloproteinase inhibitors as anticancer
agents. Int. J. Biochem. Cell Biol. 2008, 40, 1156–1168.
(36) Levitt, N. C.; Eskens, F. A. L. M.; O’Byrne, K. J.; Propper, D. J.;
Denis, L. J.; Owen, S. J.; Choi, L.; Foekens, J. A.; Wilner, S.; Wood,
J. M.; Nakajima, M.; Talbot, D. C.; Steward, W. P.; Harris, A. L.;
Verweij, J. Phase I and pharmacological study of the oral matrix
metalloproteinase inhibitor, MMI1270 (CGS27023A), in patients with
advanced solid cancer. Clin. Cancer Res. 2001, 7, 1912–1922.
(37) Andrianov, V.; Gailite, V.; Lola, D.; Loza, E.; Semenikhina, V.;
Kalvinsh, I.; Finn, P.; Petersen, K. D.; Ritchie, J. W. A.; Khan, N.;
Tumber, A.; Collins, L. S.; Vadlamudi, S. M.; Bjorkling, F.; Sehested,
M. Novel amide derivatives as inhibitors of histone deacetylase: design,
synthesis and SAR. Eur. J. Med. Chem. 2008, doi:10.1016/j.ej-
mech.2008.06.020.
(38) Hajduk, P. J.; Sheppard, G.; Nettesheim, D. G.; Olejniczak, E. T.;
Shuker, S. B.; Meadows, R. P.; Steinman, D. H.; Carrera, G. M., Jr.;
Marcotte, P. A.; Severin, J.; Walter, K.; Smith, H.; Gubbins, E.;
Simmer, R.; Holzman, T. F.; Morgan, D. W.; Davidsen, S. K.;
Summers, J. B.; Fesik, S. W. Discovery of potent nonpeptide inhibitors
of stromelysin using SAR by NMR. J. Am. Chem. Soc. 1997, 119,
5818–5827.
(39) Jacobsen, F. E.; Lewis, J. A.; Cohen, S. M. A new role for old ligands:
discerning chelators for zinc metalloproteinases. J. Am. Chem. Soc.
2006, 10, 3156–3157.
(40) Lewis, J. A.; Mongan, J.; McCammon, J. A.; Cohen, S. M. Evaluation
and binding-mode prediction of thiopyrone-based inhibitors of anthrax
lethal factor. ChemMedChem 2006, 7, 694–697.
(41) Lewis, J. A.; Puerta, D. T.; Cohen, S. M. Metal complexes of the
trans-influencing ligand thiomaltol. Inorg. Chem. 2003, 42, 7455–7459.
(42) Lewis, J. A.; Tran, B. L.; Puerta, D. T.; Rumberger, E. M.;
Hendrickson, D. N.; Cohen, S. M. Synthesis, structure and spectros-
copy of new thiopyrones and hydroxythiopyrone transition-metal
complexes. Dalton Trans. 2005, 15, 2588–2596.
(43) Puerta, D. T.; Griffin, M. O.; Lewis, J. A.; Romero-Perez, D.; Garcia,
R.; Villarreal, F. J.; Cohen, S. M. Heterocyclic zinc-binding groups
for use in next generation matrix metalloproteinase inhibitors: potency,
toxicity, and reactivity. J. Biol. Inorg. Chem. 2006, 11, 131–138.
(44) Puerta, D. T.; Lewis, J. A.; Cohen, S. M. New beginnings for matrix
metalloproteinase inhibitors: identification of high-affinity zinc-binding
groups. J. Am. Chem. Soc. 2004, 126, 8388–8389.
(6) Miller, C. J.; Elliot, J. L.; Collier, R. J. Anthrax protective antigen:
prepore-to-pore conversion. Biochemistry 1999, 38, 10432–10441.
(7) Scobie, H. M.; Rainey, J. A.; Bradley, K. A.; Young, J. A. T. Human
capillary morphogenesis protein 2 functions as an anthrax toxin
receptor. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 5170–5174.
(8) Young, J. A.; Collier, R. J. Anthrax toxin: receptor binding, internal-
ization, pore formation, and translocation. Annu. ReV. Biochem. 2007,
76, 243–265.
(9) Krantz, B. A.; Finkelstein, A.; Collier, R. J. Protein translocation
through the anthrax toxin transmembrane pore is driven by a proton
gradient. J. Mol. Biol. 2006, 355, 968–979.
(10) Leppla, S. H. Anthrax toxin edema factor: a bacterial adenylate cyclase
that increases cyclic AMP concentrations in eukaryotic cells. Proc.
Natl. Acad. Sci. U.S.A. 1982, 79, 3162–3166.
(11) Paccani, S. R.; Tonello, F.; Patrussi, L.; Capitani, N.; Simonato, M.;
Montecucco, C.; Baldari, C. T. Anthrax toxins inhibit immune cell
chemotaxis by perturbing chemokine receptor signalling. Cell. Mi-
crobiol. 2007, 9, 924–929.
(12) Scobie, H. M.; Marlett, J. M.; Rainey, G. J. A.; Lacy, D. B.; Collier,
R. J.; Young, J. A. T. Anthrax toxin receptor 2 determinants that dictate
the pH threshold of toxin pore formation. PLoS One 2007, 2, e329.
(13) Sun, J.; Lang, A. E.; Aktories, K.; Collier, R. J. Phenylalanine-427 of
anthrax protective antigen functions in both pore formation and protein
translocation. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 4346–4351.
(14) Wimalasena, D. S.; Cramer, J. C.; Janowiak, B. E.; Jurius, S. J.;
Melnyk, R. A.; Anderson, D. E.; Kirk, K. L.; Collier, R. J.; Bann,
J. G. Effect of 2-fluorohistidine labeling of the anthrax protective
antigen on stability, pore formation, and translocation. Biochemistry
2007, 46, 14928–14936.
(15) Agrawal, A.; Pulendran, B. Anthrax lethal toxin: a weapon of
multisystem destruction. Cell. Mol. Life Sci. 2004, 61, 2859–2865.
(16) Park, J. M.; Greten, F. R.; Li, Z. W.; Karin, M. Macrophage apoptosis
by anthrax lethal factor through p38 MAP kinase inhibition. Science
2002, 297, 2048–2051.
(17) Chopra, A. P.; Boone, S. A.; Liang, X.; Duesbery, N. S. Anthrax lethal
factor proteolysis and inactivation of MAPK kinase. J. Biol. Chem.
2003, 278, 9402–9406.
(18) Puhar, A.; Montecucco, C. Where and how do anthrax toxins exit
endosomes to intoxicate host cells? Trends Microbiol. 2007, 15, 477–
482.
(19) Tonello, F.; Ascenzi, P.; Montecucco, C. The metalloproteolytic
activity of the anthrax lethal factor is substrate-inhibited. J. Biol. Chem.
2003, 278, 40075–40078.
(20) Ascenzi, P.; Visca, P.; Ippolito, G.; Spallarossa, A.; Bolognesi, M.;
Montecucco, C. Anthrax toxin: a tripartite lethal combination. FEBS
Lett. 2002, 531, 384–388.
(21) Hanna, P. C.; Acosta, D.; Collier, R. J. On the role of macrophages
in anthrax. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 10198–10201.
(22) Pezard, C.; Berche, P.; Mock, M. Contribution of individual toxin
components to virulence of Bacillus anthracis. Infect. Immun. 1991,
59, 3472–3477.
(23) Forino, M.; Johnson, S. L.; Wong, T. Y.; Rozanov, D. V.; Savinov,
A. Y.; Li, W.; Fattorusso, R.; Becattini, B.; Orry, A. J.; Jung, D.;
Abagyan, R. A.; Smith, J. W.; Alibek, K.; Liddington, R. C.; Strongin,
A. Y.; Pellecchia, M. Efficient synthetic inhibitors of anthrax lethal
factor. Proc. Natl. Acad. Sci. U.S.A. 2005, 2005, 9499–9504.
(24) Gaddis, B. D.; Avramova, L. V.; Chmielewski, J. Inhibitors of anthrax
lethal factor. Bioorg. Med. Chem. Lett. 2007, 17, 4575–4578.
(25) Jiao, G. S.; Simo, O.; Nagata, M.; O’Malley, S.; Hemscheidt, T.;
Cregar, L.; Millis, S. Z.; Goldman, M. E.; Tang, C. Selectively
guanidinylated derivatives of neamine. Synthesis and inhibition of

1074 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Agrawal et al.
(45) Cummings, R. T.; Salowe, S. P.; Cunningham, B. R.; Wiltsie, J.; Park,
Y. W.; Sonatore, L. M.; Wisniewski, D.; Douglas, C. M.; Hermes,
J. D.; Scolnick, E. M. A peptide-based fluorescence resonance energy
transfer assay for Bacillus anthracis lethal factor protease. Proc. Natl.
Acad. Sci. U.S.A. 2002, 99, 6603–6606.
(46) Agrawal, A.; Romero-Perez, D.; Jacobsen, J. A.; Villarreal, F. J.;
Cohen, S. M. Zinc-binding groups modulate selective inhibition of
MMPs. ChemMedChem 2008, 3, 812–820.
(47) Abu-Dari, K.; Raymond, K. N. Ferric ion sequestering agents. 23.
Synthesis of tris(hydroxypyridinethione) ligands and their ferric
complexes; X-ray structure analysis of N,N′,N′′-tris((1,2-didehydro-
1-hydroxy-2-thioxopyrid-6-yl)carbonyl-2,2′,2′-triamino-triethylamina-
to)iron(III). Inorg. Chem. 1991, 30, 519–524.
(48) Scarrow, R. C.; Riley, P. E.; Abu-Dari, K.; White, D. L.; Raymond,
K. N. Ferric ion sequestering agents. 13. Synthesis, structures, and
thermodynamics of complexation of cobalt(III) and iron(III) tris
complexes of several chelating hydroxypyridinones. Inorg. Chem.
1985, 24, 954–967.
(49) Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand efficiency: a useful
metric for lead selection. Drug DiscoVery Today 2004, 9, 430–431.
(50) Reynolds, C. H.; Bembenek, S. D.; Tounge, B. A. The role of
molecular size in ligand efficiency. Bioorg. Med. Chem. Lett. 2007,
17, 4258–4261.
(51) Reynolds, C. H.; Tounge, B. A.; Bembenek, S. D. Ligand binding
efficiency: trends, physical basis, and implications. J. Med. Chem.
2008, 51, 2432–2438.
(52) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.;
Belew, R. K.; Olson, A. J. Automated docking using a Lamarckian
genetic algorithm and and empirical binding free energy function.
J. Comput. Chem. 1998, 19, 1639–1662.
(53) Parkin, G. The bioinorganic chemistry of zinc: synthetic analogues
of zinc enzymes that feature tripodal ligands. Chem. Commun. 2000,
2000, 1971–1985.
(54) Allen, F. H.; Bird, C. M.; Rowland, R. S.; Raithby, P. R. Resonance-
induced hydrogen bonding at sulfur acceptors in R1R2CdS and R1CS2-
systems. Acta Crystallogr. 1997, B53, 680–695.
(55) Hori, T.; Otani, Y.; Kawahata, M.; Yamaguchi, K.; Ohwada, T.
Nonplanarstructuresofthioamidesderivedfrom7-azabicyclo[2.2.1]heptane.
Electronically tunable planarity of thioamides. J. Org. Chem. 2008,
73, 9102–9108.
(56) Pellecchia, M.; Bertini, I.; Cowburn, D.; Dalvit, C.; Giralt, E.; Jahnke,
W.; James, T. L.; Homans, S. W.; Kessler, H.; Luchinat, C.; Meyer,
B.; Oschkinat, H.; Peng, J.; Schwalbe, H.; Siegal, G. Perspectives on
NMR in drug discovery: a technique comes of age. Nat. ReV. Drug
DiscoVery 2008, 7, 738–745.
(57) Venkatesan, N.; Kim, B. H. Synthesis and enzyme inhibitory activities
of novel peptide isosteres. Curr. Med. Chem. 2002, 9, 2243–2270.
(58) Hanzlik, R. P.; Vyas, K. P.; Traiger, G. J. Substituent effects on the
hepatotoxicity of thiobenzamide derivatives in the rat. Toxicol. Appl.
Pharmacol. 1978, 46, 685–694.
(59) Ikehata, K.; Duzhak, T. G.; Galeva, N. A.; Ji, T.; Koen, Y. M.; Hanzlik,
R. P. Protein targets of reactive metabolites of thiobenzamide in rat
liver in vivo. Chem. Res. Toxicol. 2008, 21, 1432–1442.
(60) Jeschke, P.; Harder, A.; Etzel, W.; Gau, W.; Thielking, G.; Bonse,
G.; Iinuma, K. Synthesis and anthelmintic activity of thioamide
analogues of cyclic octadepsipeptides such as PF1022A. Pest Manage.
Sci. 2001, 57, 1000–1006.
(61) Pfund, E.; Lequeux, T.; Masson, S.; Vazeux, M.; Cordi, A.; Pierre,
A.; Serre, V.; Herve, G. Efficient synthesis of fluorothiosparfosic acid
analogues with potential antitumoral activity. Bioorg. Med. Chem.
2005, 13, 4921–4928.
(63) Case, D. A.; Perlman, D. A.; Caldwell, J. W.; Chetham, T. E., III;
Ross, W. S.; Simmerling, C. L.; Darden, T. A.; Merz, K. M.; Stanton,
R. V.; Cheng, A. L.; Vincent, J. J.; Crowley, M.; Tsui, V.; Gohlke,
H.; Radmer, R. J.; Duan, Y.; Pitera, J.; Massova, I.; Seibel, G. L.;
Singh, U. C.; Weiner, P. K.; Kollman, P. A. AMBER, version 7;
University of California: San Francisco, CA, 2002.
(64) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M.;
Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman,
P. A. A 2nd generation force-field for the simulation of proteins,
nucleic-acids, and organic-molecules. J. Am. Chem. Soc. 1995, 117,
5179–5197.
(65) Ryckaert, J. P.; Ciccotti, G.; Berendsen, H. J. C. Numerical-intergration
of Cartesian equations of motion of a system with constraints,
Molecular-dynamics of N-alkanes. J. Comput. Phys. 1977, 23, 327–
341.
(66) Darden, T.; York, D.; Pedersen, L. Particle mesh Ewald, an N·log(N)
method for Ewald sums in large systems. J. Chem. Phys. 1993, 98,
10089–10092.
(67) Zhang, Y. K.; Liu, H. Y.; Yang, W. T. Free energy calculation on
enzyme reactions with an efficient iterative procedure to determine
minimum energy paths on a combined ab intio QM/MM potential
energy surface. J. Chem. Phys. 2000, 112, 3483–3492.
(68) Liu, H. Y.; Zhang, Y. K.; Yang, W. T. How is the active site of enolase
organized to catalyze two different reaction steps? J. Am. Chem. Soc.
2000, 122, 6560–6570.
(69) Zhang, Y. K.; Lee, T. S.; Yang, W. T. A pseudobond approach to
combining quantum mechanical and molecular mechanical methods.
J. Chem. Phys. 1999, 110, 46–54.
(70) Zhang, Y. K.; Kua, J.; McCammon, J. A. Role of the catalytic triad
and oxyanion hole in acetylcholinesterase catalysis: an ab initio QM/
MM study. J. Am. Chem. Soc. 2002, 124, 10572–10577.
(71) Cisneros, G. A.; Wang, M.; Silinski, P.; Fitzgerald, M. C.; Yang, W. T.
The protein backbone makes important contributions to 4-oxalocro-
tonate tautomerase enzyme catalysis: understanding from theory and
experiment. Biochemistry 2004, 43, 6885–6892.
(72) Cheng, Y. H.; Zhang, Y. K.; McCammon, J. A. How does the cAMP-
dependent protein kinase catalyze the phosphorylation reaction: an
ab intio QM/MM study. J. Am. Chem. Soc. 2005, 127, 1553–1562.
(73) Hu, P.; Zhang, Y. K. Catalytic mechanism and product specificity of
the histone lysine methyltransferase SET7/9: An ab initio QM/MM-
FE study with multiple initial structures. J. Am. Chem. Soc. 2006,
128, 1272–1278.
(74) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone,
V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts,
R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.5; Gaussian,
Inc.: Pittsburgh, PA, 1998.
(75) Ponder, J. W. TINKER, Software Tools for Molecular Design, version
3.6; 1998.
(76) Amin, E. A.; Truhlar, D. G. Zn coordination chemistry: development
of benchmark suites for geometries, dipole moments, and bond
dissociation energies and their use to test and validate density
functionals and molecular orbital theory. J. Chem. Theory Comput.
2008, 4, 75–85.
(62) Sheldrick, G. M. S. SHELXTL, version 2.10; Bruker AXS, Inc.:
Madison, WI, 2004. The semiempirical method used is based on a
method of Blessing, R. H. Acta Crystallogr. 1995, A51, 33.
JM8013212