Derivatives of salicylic acid as inhibitors of YopH in yersinia pestis

Article abstract of DOI:10.1111/j.1747-0285.2010.00996.x

Yersinia pestis causes diseases ranging from gastrointestinal syndromes to bubonic plague and could be misused as a biological weapon. As its protein tyrosine phosphatase YopH has already been demonstrated as a potential drug target, we have developed two series of forty salicylic acid derivatives and found sixteen to have micromolar inhibitory activity. We designed these ligands to have two chemical moieties connected by a flexible hydrocarbon linker to target two pockets in the active site of the protein to achieve binding affinity and selectivity. One moiety possessed the salicylic acid core intending to target the phosphotyrosine-binding pocket. The other moiety contained different chemical fragments meant to target a nearby secondary pocket. The two series of compounds differed by having hydrocarbon linkers with different lengths. Before experimental co-crystal structures are available, we have performed molecular docking to predict how these compounds might bind to the protein and to generate structural models for performing binding affinity calculation to aid future optimization of these series of compounds.

Full text of DOI:10.1111/j.1747-0285.2010.00996.x

ª 2010 John Wiley & Sons A/S
doi: 10.1111/j.1747-0285.2010.00996.x
Chem Biol Drug Des 2010; 76: 85–99
Research Article
Derivatives of Salicylic Acid as Inhibitors of YopH
in Yersinia pestis
Zunnan Huang1, Yantao He2, Xian Zhang2,
protein tyrosine phosphatase YopH of this bacterium can be an
effective therapeutic strategy. For example, altering the gene of
YopH to a non-functional one removed the bacterium's pathogenicity
(5–7). Mutating the essential catalytic cysteine residue of YopH to
Andrea Gunawan3, Li Wu2,3, Zhong-Yin
Zhang2 and Chung F. Wong1
1Department of Chemistry and Biochemistry and Center for
alanine also abolished its protein tyrosine phosphatase activity and
dampened the pathogenic effects of the bacterium (8,9). Conse-
quently, potent and selective YopH inhibitors are expected to serve
as novel anti-plague agents.
Nanoscience, University of Missouri-Saint Louis, One University
Boulevard, St. Louis, MO 63121, USA
2Department of Biochemistry and Molecular Biology, Indiana
University School of Medicine, 635 Barnhill Drive, Indianapolis,
IN 46202, USA
Several YopH inhibitors have already been identified over the last
few years: Sun et al. (4) developed p-nitrocatechol sulfate (pNCS)
and determined its co-crystal structure with YopH. Phan et al. (10)
designed a hexapeptide mimic, Ac-DADE-F2Pmp-L-NH2, of the pro-
tein's natural substrate (F2Pmp stands for difluoro-substituted phos-
phonomethylphenylalanine, which is a phosphotyrosine analog.) and
determined its co-crystal structure with the protein. Liang et al. (11)
identified aurintricarboxylic acid as a potent inhibitor of YopH and it
displayed 6- to 120-fold selectivity in favor of YopH over a panel of
mammalian protein tyrosine phosphatases. Tautz et al. (12) screened
the DIVERSetꢀ library (ChemBridge, Inc. San Diego, CA, USA) of
drug-like compounds and identified furanyl salicylate compounds as
potent inhibitors of YopH. Hu and Stebbins (13) performed molecular
docking and 3D-QSAR studies to rationalize the binding of derivatives
of a-ketocarboxylic acids and squaric acid to YopH and to provide 3D-
QSAR models to guide future refinement of this class of compounds.
3Chemical Genomics Core Facility, Indiana University School of
Medicine, 635 Barnhill Drive, Indianapolis, IN 46202, USA
*Corresponding author: Chung F. Wong, wongch@umsl.edu
Yersinia pestis causes diseases ranging from gas-
trointestinal syndromes to bubonic plague and
could be misused as a biological weapon. As its
protein tyrosine phosphatase YopH has already
been demonstrated as a potential drug target, we
have developed two series of forty salicylic acid
derivatives and found sixteen to have micromolar
inhibitory activity. We designed these ligands to
have two chemical moieties connected by a flexi-
ble hydrocarbon linker to target two pockets in
the active site of the protein to achieve binding
affinity and selectivity. One moiety possessed the
salicylic acid core intending to target the phosp-
hotyrosine-binding pocket. The other moiety con-
tained different chemical fragments meant to
target a nearby secondary pocket. The two series
of compounds differed by having hydrocarbon
linkers with different lengths. Before experimental
co-crystal structures are available, we have per-
formed molecular docking to predict how these
compounds might bind to the protein and to gen-
erate structural models for performing binding
affinity calculation to aid future optimization of
these series of compounds.
In spite of these encouraging developments, the search for addi-
tional drug leads remains vital as many factors can prevent existing
drug leads from passing through a series of stringent preclinical
and clinical evaluations to become successful drugs. In this regard,
most YopH inhibitors reported in the literature display unfavorable
pharmacological properties and are not cell permeable. Moreover,
multidrug-resistant strains of Yersinia pestis can emerge (14,15). To
develop YopH inhibitors that carry sufficient polar and non-polar
interactions with the active site and yet possess favorable pharma-
cological properties, we decided to capitalize our previous findings
that the natural product salicylic acid can serve as a pTyr surrogate
(16) and that naphthyl and polyaromatic salicylic acid derivatives
exhibit enhanced affinity for protein tyrosine phosphatase relative
to the corresponding single ring compounds (11,16). Therefore, in
this work, we synthesized a new class of benzofuran salicylic acids
and found many of them to demonstrate lM activity.
Key words: click chemistry, distance-dependent dielectric model,
docking by simulated annealing, Generalized Born model, receiver operat-
ing characteristics curve, sensitivity and specificity of screening models
Received 10 March 2010, revised 6 May 2010 and accepted for publica-
tion 6 May 2010
Our initial design principle assumed the benzofuran salicylic acid
core to bind to the phosphotyrosine-binding pocket. By introducing
an additional chemical entity, linked to the core by a flexible hydro-
phobic linker, we hoped to target a neighboring pocket simulta-
Yersinia pestis can cause human diseases such as gastrointestinal
syndromes and bubonic plague (1–3) and could be misused as a
biological weapon (4). There is already evidence that blocking the
85

Huang et al.
neously to improve potency and selectivity. This article presents
two series of these compounds differing by having different length
of the linker connecting the two chemical moieties (B and D series
shown in Figure 1).
that, upon regioselective bromination, afforded 5-bromo-4-hydroxy-
salicylic acid (2). This compound was selectively protected in the
presence of acetone and trifluoroacetic anhydride ⁄ trifluoroacetic
acid to furnish dioxanone 3, which then reacted with CH3I to give
the methylation product 4. Compound 4 was coupled with phenyl-
acetylene in the presence of Pd(PPh3)4 to furnish 5, which was then
subjected to I2-induced cyclization. Coupling of the iodination prod-
uct 6 with ethynyltrimethylsilane gave compound 7. Desilylation
and deacetylation of 7 provided the core compound 1.
To investigate whether these compounds are likely to bind the way
that we expected, we performed molecular docking using a flexible
ligand-flexible protein model we developed recently. The method
improved docking by going beyond the rigid-protein approximation
to account for induced-fit effects so that it could dock a wider
range of ligands properly to a protein. The model used molecular
dynamics simulation as a sampling tool. However, instead of run-
ning simulations at a constant temperature, it employed a simulated
annealing cycling protocol to improve sampling efficiency. The pro-
tein was not completely flexible but with harmonic constraints
applied to the a carbons to keep its structure near a suitable refer-
ence structure such as one obtained from X-ray crystallography.
However, all other atoms, including all the side chains, were unre-
strained (17,18). Although not yet a completely flexible protein
model, this model avoided artifacts resulting from non-optimal
energy and solvation models by focusing on exploring the conforma-
tional space near a known experimental structure. Previously, we
showed that this model successfully docked several small organic
ligands to protein kinases and the protein tyrosine phosphatase
YopH (17,18); a completely rigid-protein model, on the other hand,
failed for some ligands studied (18).
To increase potency and selectivity, the strategically positioned
alkyne in the benzofuran salicylic acid core was tethered to 40
azide-containing diversity elements (20 discrete amines with 2 alkyl
linkers of 2 and 4 methylene length), using click chemistry or the
Cu(I)-catalyzed [3 + 2] azide-alkyne cycloaddition reaction (Figure 2).
The click chemistry offers an expedient way to connect two compo-
nents together with high yield and purity under extremely mild con-
ditions (19). More importantly, the cycloaddition reaction can be
conducted in aqueous solution in the absence of deleterious
reagents, thus allowing direct screening and identification of hits
from the library. The azide-containing building blocks were synthe-
sized in a one-pot procedure, in which alkyl or aryl amines were
reacted with the acyl chloride linkers in N, N-Dimethylformamide
(DMF), followed by SN2 reaction with sodium azide to generate the
corresponding azides. To construct the 40-member library, each
azide was coupled with the alkyne containing core 1 in a mixed
solvent of ethanol and water (7:3) and the click reaction was initi-
ated by catalytic amount of Cu(I), which was generated by reacting
CuSO4 with sodium ascorbate. After 48 h, the products were col-
lected by simple centrifugation. All products were assessed by LC-
MS and determined to be at least 70–100% pure and were used
directly for screening without further purification.
In this application, we further leveraged this approach by perform-
ing docking in two stages to improve speed without significantly
sacrificing reliability. As we were studying relatively similar com-
pounds within a chemical series, we assumed docking several rep-
resentative ligands in stage 1 could generate all the major docking
modes accessible by every compound in the series. Then, in stage
2, we allowed each compound in the series to refine its structure
around these major docking modes by performing less expensive
docking on a focused subset of configurational space. We then
applied the resulting docking poses to compute binding affinity to
check further whether they yielded results consistent with experi-
mental IC50. In the future, one can use the best resulting structural
models for performing binding affinity calculation on new deriva-
tives to suggest new compounds that are worthwhile to make and
test experimentally.
Screening of the salicylic acid library for YopH
inhibitors
To screen the salicylic acid library for YopH inhibitors, the effect of
each library member on the YopH-catalyzed p-nitrophenyl phosphate
(pNPP) hydrolysis was determined. The YopH-catalyzed hydrolysis of
pNPP in the presence of 10 lM compound was assayed at 30 ꢁC in
a 200 lL reaction system in a 96-well plate. Each reaction con-
tained 2 lL of 1 mM compound in DMSO (final concentration
10 lM) and 198 lL assay buffer (50 mM 3,3-dimethylglutarate, 1 mM
EDTA, 1 mM DTT, pH 7.0 with an ionic strength of 0.15 M adjusted
by addition of NaCl) containing 2 mM pNPP and 10 nM YopH. The
PTP-catalyzed reaction was started by addition of the enzyme. As a
control, 2 lL of DMSO was used. The YopH-catalyzed hydrolysis of
pNPP was measured by monitoring the absorbance at 405 nm of the
product p-nitrophenol continuously, with a SpectraMAX 340 micro-
plate spectrophotometer (Molecular Devices, Sunnyvale, CA, USA).
The initial rate was obtained by calculating the slope of the product
versus the time curve. Compounds that display significant inhibition
at 10 lM were subject to IC50 measurements.
Methods
Experimental
Library synthesis
Figure 2 depicts a focused library-based strategy for the acquisition
of potent and selective YopH inhibitors that are capable of bridging
both the active site and an adjacent peripheral site. The library con-
tains (i) a benzofuran salicylic acid core to engage the active site,
and (ii) 2 alkyl linkers of 2 and 4 methylene unit to tether the pTyr
surrogate to (iii) a structurally diverse set of 20 amines, aimed at
capturing additional interactions with adjacent pockets surrounding
the active site. The benzofuran salicylic acid core 1 was prepared
from a commercially available compound 4-hydroxysalicylic acid
IC50 measurement
The YopH-catalyzed hydrolysis of pNPP in the presence of inhibitor
was assayed at 30 ꢁC in a 200 ll reaction system in the same
86
Chem Biol Drug Des 2010; 76: 85–99

Derivatives of Salicylic Acid as Inhibitors of YopH
O
O
O
O
O
O
HO
HO
HO
HO
HO
N
N
H
N
N
N
H
H
N
N
N
N
HO
N
N
N
O
O
O
(21) D01; IC50 > 20 µM
(11) B11; IC50 = 1.2 µM
(1) B01; IC50 > 20 µM
O
O
O
O
O
O
HO
HO
HO
N
H
N
H
N
N
N
N
H
N
N
N
HO
HO
N
HO
N
N
O
O
O
(22) D02; IC50 > 20 µM
(12) B12; IC50 = 3 µM
(2) B02; IC50 > 20 µM
O
O
O
O
O
O
HO
HO
HO
S
N
N
N
N
H
N
N
H
N
H
N
N
N
HO
HO
N
HO
N
N
O
O
O
(3) B03; IC50 > 20 µM
(13) B13; IC50 > 20 µM
(23) D03; IC50 = 3 µM
O
O
O
O
O
S
O
HO
HO
HO
N
N
N
N
H
N
N
N
H
H
N
N
N
HO
HO
N
HO
N
N
O
O
O
(4) B04; IC50 > 20 µM
(14) B14; IC50 = 4 µM
(24) D04; IC50 = 5 µM
O
O
O
O
O
O
S
HO
HO
HO
N
N
H
N
N
N
H
N
H
N
N
N
N
HO
N
HO
N
HO
N
O
O
(5) B05; IC50 > 20 µM
(15)OB15; IC50 =7 µM
(25) D05; IC50 > 20 µM
O
O
O
O
O
O
HO
HO
HO
N
N
H
N
N
N
H
N
N
H
N
N
HO
N
HO
N
HO
O
F
N
O
F
O
O
(6) B06; IC50 > 20 µM
(16)OB16; IC50 = 2.4 uM
(26) D06; IC50 > 20 µM
O
O
O
O
O
O
HO
HO
HO
N
N
H
N
N
H
N
N
N
N
H
N
HO
N
HO
N
HO
N
O
O
O
(27) D07; IC50 > 20 µM
(7) B07; IC50 > 20 µM
(17) B17; IC50 = 2.4 µM
O
O
O
O
O
O
HO
HO
HO
N
H
N
N
N
H
N
N
H
N
N
N
HO
N
HO
N
HO
N
O
O
(8) B08; IC50 = 2.6 µM
(18)OB18; IC50 > 20 µM
(28) D08; IC50 > 20 µM
O
O
O
O
O
N
N
O
O
O
HO
HO
HO
N
H
N
N
N
H
N
H
N
N
N
N
HO
HO
N
HO
N
N
O
O
O
(9) B09; IC50 = 6 µM
(19) OB19; IC50 > 20 µM
(29) D09; IC50 = 1.25 µM
O
O
O
O
HO
N
O
N
O
HO
HO
N
N
N
N
H
N
N
N
HO
N
N
H
N
HO
HO
N
N
O
(10)OB10; IC50 = 4 µM
(20) B20; IC50 > 20 µM
(30) OD10; IC50 = 6 µM
Figure 1: Chemical structure of ligands in the B series and the D series.
Chem Biol Drug Des 2010; 76: 85–99
87

Huang et al.
O
O
O
O
O
S
O
HO
HO
HO
HO
HO
HO
N
H
N
N
H
N
N
N
N
H
N
N
N
N
N
N
O
O
O
(31) D11; IC50 > 20 µM
(35) D15; IC50 > 20 µM
(38) D18; IC50 > 20 µM
O
O
O
O
O
O
HO
HO
HO
N
N
H
N
H
N
N
N
H
N
N
N
HO
HO
HO
N
N
N
O
O
O
O
(32) D12; IC50 =4 µM
(36) D16; IC50 = 4 µM
(39) D19; IC50 > 20 µM
O
O
O
O
O
O
HO
HO
HO
S
N
N
N
H
N
N
H
N
N
N
N
N
HO
HO
N
N
HO
N
O
O
O
(33) D13; IC50 > 20 µM
(37) D17; IC50 > 20 µM
(40) D20; IC50 > 20 µM
O
O
S
HO
N
N
N
H
N
HO
N
O
(34) D14; IC50 = 1.8 µM
Figure 1: (Continued).
assay buffer described above. At various concentrations of the com-
pound, the initial rate at fixed pNPP concentrations (equal to the
corresponding Km value for YopH) was measured by continuously
following the production of p-nitrophenol as described above. The
IC50 value was determined by plotting the relative PTP activity
toward pNPP versus inhibitor concentration and fitting to equation
(1) using Kaleidagraph.
errors in the subsequent computation of binding affinity to help dis-
tinguish actives from non-actives.
• Used the docking poses from stage 2 docking to estimate the
binding affinity of each compound in the series employing either a
distance-dependent dielectric model or the Generalized Born model
termed GBMV (20–22).
Vi=V0 ¼ IC50=ðIC50 þ ½IꢀÞ
ð1Þ
• Checked which or which mixture of major docking modes gave
the best agreement with experiment in terms of classifying the
compounds into actives and non-actives – defined as compounds
having IC50 less than and greater than 20 lM, respectively – using
several quantitative measures described below.
In this case, Vi is the reaction velocity when the inhibitor concentra-
tion is [I], V0 is the reaction velocity with no inhibitor.
General Computational Strategy
• Use the best docking models in the future to aid optimization of
these series of compounds by attaching different functional groups
to the core chemical skeleton and examine which modifications
could lead to inhibitors with improved binding affinity.
The following steps summarize our strategy:
• Stage 1 docking: Performed extensive flexible ligand-flexible pro-
tein docking for a few representative compounds in a series to esti-
mate how the ligands bound to YopH. This was the most expensive
part of the simulation. Clustering the twenty lowest-energy struc-
tures afterward then generated several major docking modes for
stage 2 docking.
Protein–ligand docking by simulated annealing
cycling
We performed molecular docking by using a molecular dynamics-
based simulated annealing (SA) cycling protocol we published ear-
lier (17,23). The method performed many short SA cycles in a
molecular dynamics simulation to sample many energy minima of a
protein–ligand system thoroughly. The lowest-energy poses then
recommended how the ligand might bind to the protein.
• Stage 2 docking: Used each major docking mode from stage 1 to
construct the structures of all the other compounds in the series
and refined these structures by performing focused docking in
which we only allowed the ligands and their surrounding protein
residues to move; this significantly reduced computational cost in
comparison to the less restrictive docking in stage 1. The focusing
of the final docking of a similar series of compounds to the most
relevant regions could also reduce statistical noise and systematic
We conducted the SA cycling simulations using the CHARMM
param22 force field (24,25). In the docking simulations, we used a
simple but inexpensive distance-dependent dielectric model with
e(r) = 4r where r was the distance between two atoms. In calculat-
88
Chem Biol Drug Des 2010; 76: 85–99

Derivatives of Salicylic Acid as Inhibitors of YopH
GBMV1 parameters of ChocholouÐovꢁ and Feig (22) and we applied
the same cutoff distances as described above for the e(r ) = 4r
model.
A
Library construction
O
H
N
O
O
N3
R
N
R
N
n
O
n
H
N
Two-stage docking strategy
N
Click
HO
HO
Chemistry
Stage 1 initial docking of representative
ligands
HO
O
HO
O
Core 1
n = 2, 4
For each of the two chemical series, B and D, studied here, we
selected three compounds for thorough flexible ligand-flexible pro-
tein docking. We chose compounds B11, B16, and B17 for the B
series and compounds D03, D09, and D14 for the D series (struc-
tures shown in Figure 1). For docking each ligand to YopH, we
started the simulations from four different positions ⁄ structures. One
lied inside the same pocket that pNCS binds [PDB entry 1PA9(4)].
Another was located on the surface of the protein near the binding
pocket. At each location, we placed the ligands in two near
anti-parallel orientations, thus generating four different starting
structures to initiate docking runs. The structure in PDB entry 1QZ0
provided the starting structure for the YopH protein (10). The initial
3D structures of the ligands were prepared by using ChemSketch
(26). We performed ten independent SA cycling simulations for each
of the four starting structures, thus giving forty trajectories.
Because each trajectory lasted 2 ns, the aggregate simulation time
for each protein–ligand system covered 80 ns. In these simulations,
we used a time step of 2 fs.
B
Synthesis of Core 1
OH
OH
Br
OMe
Br
OH
HO
HO
HO
HO
O
O
O
O
Br
c
a
b
O
O
O
O
2
3
4
OMe
O
e
f
d
O
O
O
I
5
O
6
O
O
O
O
g
h
O
O
O
O
Core 1
Si
O
O
7
8
Conditions:
a) Br2, AcOH, r.t., 94%; b) TFAA, TFA, Acetone, r.t., 55%;
c) MeI, K2CO3, DMSO, r.t., 95%; d) Phenylacetylene, Pd(PPh3)4,
CuI, Et3N, DMF, 70°C, 72%; e) I2, NaHCO3, MeCN, 40°C, 40%;
f) Trimethylsilylacetylene, Pd(PPh3)2Cl2, CuI, Et3N, DMF, r.t., 85%;
g) TBAF, THF, r.t., 89%; h) LiOH, THF/H2O, 80°C, 71%.
We allowed the protein to move in the docking simulations but with
an appropriate restraint to prevent the protein from deviating too far
away from the crystal structure in PDB code 1QZ0. We achieved this
by applying the harmonic potential F · D2, where F was a force con-
stant and D was the root-mean-square deviation (RMSD) of a dynam-
ics snapshot from the crystal structure (an option in CHARMM) (27).
Only the a-carbons were used to calculate the RMSD so that the
side chains and the other backbone atoms were free to move. In our
previous work, (17,18), we set F = 1000 kcal ⁄ mol ⁄ ꢀ2. In this work,
we used two F's in two different parts of the protein. We applied a
smaller force constant of 100 kcal ⁄ mol ⁄ ꢀ2 to the flexible WPD-loop
containing nine residues spanning from Gly-352 to Val-360. The
larger force constant F = 1000 kcal ⁄ mol ⁄ ꢀ2 was applied to the rest
of the protein. We used a smaller force constant for the WPD-loop
because it was more flexible than the other parts of the protein sur-
rounding the phosphotyrosine-binding pocket (3,28). These restraints
prevented sampling unrealistic structures because of the limitation of
current force fields and solvation models but permitted larger confor-
mational change of the protein to allow a larger range of compounds
to dock properly to the protein. Prior to each molecular dynamics
simulation, we performed 500 steps of steepest descent energy mini-
mization on the protein–ligand complex to remove bad contacts.
C
Synthesis of azides
O
n
O
O
n
NaN3
RNH2
Br
R
Br
N3
R
N
Cl
n
50-90%
(two steps)
N
H
H
n = 2, 4
Chemical Structures of 20 amines
D
RNH2
=
H
N
O
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
N
NH2
NH2
NH2
NH2
N
O
NH2
S
NH2
H3CO
N
NH2
NH2 NH2
NH2
F
S
NH2
N
Figure 2: Synthetic scheme.
ing binding affinity, we further used the more sophisticated implicit-
solvent GBMV model (20–22). During the simulation, we used a
non-bonded cutoff distance of 14 ꢀ, a switching function for the
electrostatic interactions that began at 10 ꢀ and ended at 12 ꢀ,
and a shifting function for the Lennard–Jones potential. We also
used this e(r) = 4r model in the energy minimization preceding each
molecular dynamics simulation to relieve bad contacts. In recalculat-
ing the binding affinity with the GBMV (20,21) model, we used the
Stage 2 docking of all compounds in each
series
Using each selected major docking mode obtained from stage 1,
we used the VEGA ZZ program (29) to generate the structures of
all the compounds in a series by making the appropriate chemical
modifications. We then refined these structures by performing SA
Chem Biol Drug Des 2010; 76: 85–99
89

Huang et al.
cycling simulations that allowed the ligands to move freely but
restricted the protein to a larger extent than in stage 1 docking. In
this stage 2 docking, we restrained the a carbons of protein resi-
dues within 5.0 ꢀ of a docking ligand with F · D2 in which
F = 1000 kcal ⁄ mol ⁄ ꢀ2. Here, in calculating D, we used a low-
energy pose obtained from stage 1 rather than from the crystal
structure. The rest of the protein was held fixed so that stage 2
docking took significantly less computational time than stage 1.
Table S1 shows the movable protein residues for the four represen-
tative low-energy poses for each compound series. In this stage,
we ran only four trajectories for each ligand with each trajectory
lasting 2 ns. As in stage 1, we also performed 500 steps of steep-
est descent energy minimization of the protein–ligand complex to
remove bad contacts prior to running each molecular dynamics
trajectory.
Table 1: Number of clusters obtained from 40 2-ns protein–
ligand trajectories for six ligands selected for stage 1 docking
Ligand
B11
B16
B17
D03
D09
D14
Small cluster
Large cluster
729
4
792
4
626
4
811
4
671
5
683
7
est-energy structure of each cluster to represent the structure of
the cluster. After clustering, we selected the twenty lowest-energy
clusters for each ligand. When graphically examining their three-
dimensional structures, we found that some clusters gave visually
similar structures. We therefore regrouped these twenty clusters
into a smaller number of bigger clusters. This was performed by re-
clustering the twenty clusters using a larger cluster radius of 5 ꢀ.
We also used the lowest-energy structure of each resulting cluster
to represent the structure of each bigger cluster. Table 1 shows the
number of clusters obtained from the first and second round of
clustering for the six ligands selected in stage 1 docking.
On a dual core-dual processor cluster node with 2.8 GHz Intel Xeon
EM64T processors, it took ꢁ28 h for each trajectory and ꢁ1120 h
for 40 trajectories performed for each ligand in stage 1 docking. On
the other hand, stage 2 docking only took ꢁ4–6 h for each trajec-
tory and ꢁ16–24 h per ligand with four trajectories. The simulation
time varied slightly among different ligands.
Rigid-protein docking
For comparison, we also performed rigid-protein docking with Auto-
dock v4.0.1 (34,35). To check the sensitivity of the results to the
choice of protein structures, we used three different crystal struc-
tures [PDB entries 1QZ0 (10) (closed form), 1PA9 (4) (closed form),
and 1YPT (3) (open form)]. In addition, for each protein–ligand struc-
ture obtained from stage 2 docking, we re-docked the ligand and
recomputed its binding affinity using Autodock and carried out the
performance analysis described below to check whether the flexi-
ble-receptor model gave better results.
Scoring of docking poses
To identify the best docking poses, we used the sum of the energy
of the ligand and the interaction energy between the protein and
the ligand with the e(r) = 4r model. We did not use the total energy
because it was noisy to use from finite simulations (18). Removing
the noisy components arising from the protein identified docking
poses better in our previous docking of small organic ligands to
protein kinases and to YopH (17,18,23).
Target preparation
Autodock prepared several potential grids for each protein structure.
These grids contained the electrostatic potential and the van der
Waals energies for the atom types C (aliphatic carbon), A (aromatic
carbon), N (nitrogen), O (oxygen), SA (aromatic sulfur), H (polar
hydrogen), and F (fluorine) on the ligand. For calculating electrostat-
ics interactions, we used Kollman charges for the protein atoms
and Gasteiger charges for the ligand atoms. Each grid measured
70 · 70 · 70 with a grid spacing of 0.375 ꢀ and it centered at the
heart of four pockets comprising the phosphotyrosine-binding pocket
and three neighboring secondary pockets.
Selecting representative docking modes from
stage 1 for stage 2 docking
As shown in Results and Discussions below, stage 1 docking did
not produce an unambiguous best docking mode for each series of
compounds. We therefore selected several most likely docking
modes to use for stage 2 docking and computed binding affinity
using each docking mode to find out which mode was most consis-
tent with measured IC50.
To identify a small number of docking modes for stage 2 docking
and binding affinity calculation, we first clustered the structures
obtained from stage 1 by using a self-organizing neural net
approach (30–33) implemented in CHARMM and accessible from
the Multiscale Modeling Tools for Structural Biology (MMTSB) tool-
kit. The cluster algorithm optimizes cluster assignment by minimiz-
ing the distance between members and their centroid structure
within each cluster and by requiring this distance to be within a
user-predefined cluster radius. One does not need to specify the
number of clusters as the algorithm determines the optimal number
that satisfies the above criteria. In this work, we measured the dis-
tance between two ligand structures by calculating the RMSD
between the Cartesian coordinates of their heavy atoms after we
superimposed their cognate protein structures with the crystal
structure. We set the cluster radius to 2 ꢀ and we used the low-
Docking protocol
We selected the Lamarckian genetic algorithm of Autodock to
search for the best scoring conformation. In addition, we set the
docking parameters to: 100 docking runs, population size = 150,
random starting position and orientation, maximum translation step
size = 2 ꢀ, maximum rotation allowed for each step = 35ꢁ, elit-
ism = 1, mutation rate = 0.02, crossover rate = 0.8, local search
rate = 0.06, and 25 million energy evaluations. We clustered the
final structures using a 2.0-ꢀ cutoff.
On a dual core-dual processor cluster node with 2.8 GHz Intel Xeon
EM64T processors, it took ꢁ28–51 h to dock each ligand to a rigid
90
Chem Biol Drug Des 2010; 76: 85–99

Derivatives of Salicylic Acid as Inhibitors of YopH
protein model. This computation time nearly doubled that of stage
2 docking using the flexible-protein model.
Results and Discussions
Figure 3A–F plot RMSDheavy versus energy for the six YopH–ligand
systems selected in stage 1 docking, each docking covered a total
of 80 ns of simulation time. Here, RMSDheavy represented the
RMSD of all heavy atoms of the ligand between a structure near a
local minimum (a structure obtained below 5 K) and one that had
the lowest energy. The plot for ligand D09 gave one deep-energy
well significantly separated from the next lowest-energy well. On
the other hand, the other ligands yielded two or more deep-energy
wells with relatively similar energies, making it difficult to identify
a single best docking mode. We therefore performed the clustering
described above to identify several major docking modes for stage
2 docking that included all ligands within a series and used the
resulting structures to perform binding affinity calculation to exam-
ine which model produced results most consistent with measured
IC50.
Judging model performance
To help judge how well each model performed in classifying com-
pounds into actives and non-actives, we calculated the following
quantities:
Sensitivity and specificity
Drug designers (36,37) defined the sensitivity (Se) to be the ratio of
the number of active molecules correctly predicted to the total num-
ber of actives present. In terms of the number of true positive, TP,
and the number of false negatives, FN:
Ncorrectly predicted actives
TP
TP þ FN
Se ¼
¼
ð2Þ
Ntotal actives
The clustering described earlier produced four major docking modes
denoted by B-Model I to IV for the B series and D-Model I to IV for
the D series in Table 2. We labeled the structural models such that
B-Model I was most similar to D-Model I, B-Model II was most
similar to D-Model II, B-Model III was most similar to D-Model III,
and B-Model IV was most similar to D-Model IV. However, remem-
ber from Figure 1 that ligands in the two series differed by the
length of the linker connecting the two chemical moieties intended
to target two different pockets. Therefore, one would expect minor
structural differences between B-Model I and D-Model I, B-Model II
and D-Model II etc. Also, recall that we obtained these models in
two rounds of clustering. The first round clustered every structure
near local energy minima obtained from the SA docking using a
cutoff of 2 ꢀ. The second round merged the twenty lowest-energy
clusters obtained from round 1 into a smaller number of larger clus-
ters using a larger cutoff distance of 5 ꢀ. We found four clusters
that were adopted by five of the six ligands and we chose these
clusters as the four major docking mode used for stage 2 docking.
Specificity (Sp), on the other hand, is the ratio of the number of
non-active compounds correctly predicted to the total number of
non-active molecules. In terms of the number of true negatives, TN,
and the number of false positive, FP,
Ncorrectly predicted nonꢂactives
TN
TN þ FP
Sp ¼
¼
ð3Þ
Ntotal nonꢂactives
Accuracy (Acc)
It describes the percentage of molecules classified correctly into
actives and non-actives (37,38):
Ncorrectly predicted actives þ Ncorrectly predicted nonꢂactives
Acc ¼
Ntotal
ð4Þ
TP þ TN
¼
TP þ FP þ TN þ FN
Enrichment factor
Table 2 shows how the twenty lowest-energy clusters obtained
in round 1 were grouped to form the four clusters in round 2.
For example, the notation 1st ⁄ 11 in D-Model I for ligand D09
means that eleven of the twenty lowest-energy clusters obtained
from round 1 were merged into one cluster and the lowest-
energy structure came from the 1st small cluster obtained in
round 1. (We labeled the clusters from 1st to 20th in increasing
energy.) The structure of the 1st cluster was used to represent
D-Model I and as a template to construct all the ligands for
stage 2 docking.
The enrichment factor, EF, measures how well a model increases
the fraction of actives identified relative to the fraction of actives
in a database: (38).
TP/(TP + FP)
EF ¼
(number of actives)/(total number of compounds in a database)
ð5Þ
Receiver operating characteristic (ROC) curve
This curve plots Se versus 1-Sp obtained by using different cut-
off values of a suitable quantity – calculated binding affinity in
this application – to separate actives from non-actives (36,37).
Se represents the ability of the model to pick out true positives.
On the other hand, 1-Sp hints on the tendency of the model to
produce false positives. Good models yield Se near unity and 1-
Sp close to zero. If one plots Se versus 1-Sp for different cut-
offs and calculates the area under the ROC curve (AUC), good
models yield areas near unity, random models give an area of
0.5, and models performing worse than random produce areas
smaller than 0.5.
Five (B11, B16, B17, D03, and D14) of these six ligands all yielded
the above four docking modes. D09, on the other hand, did not
assume Model IV. Instead, it adopted another docking mode in
which the ligand was sharply bended at the linker with the chemi-
cal moieties on its two ends tightly packed against each other
(shown in ``Others'' in Table 2). We did not include this docking
mode in stage 2 docking because this docking mode occurred infre-
quently and at higher energies and were thus less likely to be the
correct docking mode. Ligand D14 also took on a similar sharply
Chem Biol Drug Des 2010; 76: 85–99
91

Huang et al.
20
A
20
16
12
8
B
16
12
8
4
4
0
0
20
40
60
80
60
85
–30
–10
10
30
80
70
Energy (kcal/mol)
Energy (kcal/mol)
B11
D03
20
C
20
16
12
8
D
16
12
8
4
4
0
0
0
20
40
20
40
60
Energy (kcal/mol)
Energy (kcal/mol)
B16
D09
20
20
16
12
8
E
F
16
12
8
Figure 3: Plots of ligand
RMSDheavy versus energy for
docking six ligands to YopH. (A)
ligand B11, (B) ligand B16,
(C) ligand B17, (D) ligand D03,
(E) ligand D09 system, (F) ligand
D14.
4
4
0
0
25
45
65
10
30
50
Energy (kcal/mol)
Energy (kcal/mol)
B17
D14
Table 2: Twenty best structural clusters grouped into four larger
clusters
Table 3: The major docking modes adopted by the lowest-
energy structures of the six ligands used in stage 1 docking
Ligand
Ligand B11
B16
B17
D03
D09
D14
Model B-Model II B-Model III B-Model I D-Model I D-Model I D-Model I
Model
B11
B16
B17
D09
D03
D14
B-Model Ia ⁄
D-Model I
B-Model II ⁄
D-Model II
B-Model III ⁄
D-Model III
B-Model IV ⁄
D-Model IV
Others
14th ⁄ 1b 2nd ⁄ 6
1st ⁄ 3
1st ⁄ 11 1st ⁄ 3
6thc ⁄ 5
bended docking mode but again was ignored for stage 2 docking
for the same reason. Table 3 summarizes the docking mode adopted
by the lowest-energy structure of each ligand. The three ligands in
the B series took on different docking modes as their lowest-energy
docking structures. On the other hand, the lowest-energy structures
for the three ligands in the D series all accepted Model I.
1st ⁄ 7
4th ⁄ 8
4th ⁄ 4
1st ⁄ 6
2nd ⁄ 9
4th ⁄ 7
14th ⁄ 2 2nd ⁄ 5 8th ⁄ 5
6th ⁄ 2
6th ⁄ 4
3rd ⁄ 8
14th ⁄ 2
3rd ⁄ 2
3d ⁄ 6
11thc ⁄ 4 11th ⁄ 4 19th ⁄ 1
2d ⁄ 5
aB-Model I, D-Model I, etc., denotes one of four major docking modes repre-
sented by a large cluster of structures.
Figure 4 displays the structures of these docking modes. In each
model, two (for D-Model IV only) or three different ligands that
docked similarly were overlaid. For example, B-model I in Figure 4A
shows the lowest-energy structures from the 14th cluster for ligand
B11, from the 2nd cluster for ligand B16, and from the 1st cluster
for ligand B17. On the other hand, D-model I in Figure 4E displays
the lowest-energy structures from the 1st cluster for ligand D09,
the 1st cluster for ligand D03, and the 6th cluster for ligand D14.
The figure shows that six binding modes (B ⁄ D-Model I, B ⁄ D-Model
II, and B ⁄ D Model IV) had the salicylic acid core docked to
the phosphotyrosine-binding pocket. They differed mainly in the
bThe first number identifies the smaller cluster, within a bigger cluster,
which contained the lowest-energy structure. The second number tells how
many small clusters (formed by using a 2-ꢀ cutoff) were grouped to form
the bigger cluster (formed by regrouping the twenty lowest-energy small
clusters using a clustering cutoff of 5 ꢀ).
cIndicates that the second, rather than the first, lowest-energy structures
were selected because they were closer to the lowest-energy structures
obtained for the other two ligands.
dRepresent sharply bent docking modes that was ignored in stage 2
docking.
92
Chem Biol Drug Des 2010; 76: 85–99

Derivatives of Salicylic Acid as Inhibitors of YopH
B-Model I
B-Model II
A
B
C
B-Model III
D
B-Model IV
Figure 4: Four major docking
modes identified from flexible
ligand-flexible protein docking.
(A) B-Model I. (B) B-Model II. (C)
B-Model III. (D) B-Model IV. (E)
D-Model I. (F) D-Model II. (G)
D-Model III. (H) D-Model IV.
E
D-Model I
F
D-Model II
Coloring scheme: protein in green,
oxygen in red, nitrogen in blue,
carbon in cyan, and sulfur in
yellow. Structural representation:
protein in cartoon, ligand in stick,
and side chains (from Table S1) in
line mode. The pictures were
generated by Visual Molecular
Dynamics (VMD) (39).
D-Model III
D-Model IV
G
H
non-salicylic portion of the compounds. On the other hand, B ⁄ D
Model III differed by having the non-salicylic acid core docked
inside the phosphotyrosine-binding pocket, with the salicylic acid
core exposed to the protein surface.
hotyrosine-binding pocket. On the other hand, the other end of the
ligand fitted into three different secondary pockets. A closer com-
parison of these structures revealed that these pockets changed
somewhat depending on which ligand was bound to them – a flexi-
ble-protein model, as done here, is essential to capture such
effects.
Figure 5 uses the surface representation to show the different ways
that ligand B11 might fit into the protein. The four lowest-energy
structures in column 2 of Table 2 are shown. These structures show
that one end of the bidentate ligand always occupied the phosp-
Stage 2 docking assumed that all the compounds within a series
bound in one of these four docking modes with minor adjustment
Chem Biol Drug Des 2010; 76: 85–99
93

Huang et al.
A B-Model I
B B-Model II
Figure 5: Four major docking
modes from flexible ligand-flexible
protein docking for compounds in
the B series using a surface repre-
sentation of the protein.
(A) B-Model I. (B) B-Model II.
(C) B-Model III. (D) B-Model IV.
Ligands are displaced in stick
mode. Coloring scheme: oxygen in
red, nitrogen in blue, carbon in
cyan, and sulfur in yellow. The
picture was generated by VMD
(39).
B-Model IV
D
B-Model III
C
of the protein structures at the protein–ligand interface. D-Model I
appeared to be the most likely docking mode for the D series, as it
was the lowest-energy docking pose for all the three ligands in the
D series on which extensive flexible ligand-flexible protein docking
was performed in stage 1 (Table 3). On the other hand, the three
ligands in the B series found different major docking modes as their
lowest-energy docking pose, suggesting the possibility that not all
compounds in the B series might bind with the same major docking
mode.
divide the compounds into actives and non-actives. Because there
were nine known actives for the B series, we first classified the
nine compounds with the most favorable predicted binding affinity
as actives and the rest as non-actives. The sensitivity, specificity,
accuracy, and enrichment factor in the table then indicate that B-
Model I performed the best for both solvation models (with the
best accuracy, for example). However, the mixed structural model
was not far behind for both solvation models. B-Model IV also
scored quite well for the e(r) = 4r solvation model but did not
score as well with the GBMV model. Based on these data, we
therefore concluded that B-Model I was most likely, although the
mixed structural model could be a possibility as well. For B-Model
I, the salicylic acid core bound to the phosphotyrosine-binding
pocket as intended.
To check this further, we used each structural model to perform
binding affinity calculation to find out which model gave results
most consistent with experimental IC50. Table 4 gives the calcu-
lated binding affinity using the e(r) = 4r or the GBMV solvation
model for the twenty ligands in the B series and for each of the
four different structural models: B-Model I-IV obtained in stage 1.
The results also included a mixed structural model in which not
all ligands within the series were required to dock to the same
major docking mode. Instead, each ligand selected the docking
mode that gave the most favorable binding affinity. In this preli-
minary evaluation of the performance of the different docking
modes, we simply used one cutoff value of the binding affinity to
Table 5 shows the corresponding results for the D series. Because
the experimental results only showed seven actives, we classified
the seven compounds predicted to have the most favorable binding
affinity as actives. The results suggest that D-Model I performed
the best for both solvation model. In contrast to the B series, the
mixed structural model only scored well for the e(r) = 4r solvation
model but not for GBMV model. If we only trust docking models
94
Chem Biol Drug Des 2010; 76: 85–99

Derivatives of Salicylic Acid as Inhibitors of YopH
Table 4: Binding affinity of ligands in the B series obtained from the e(r) = 4r and the GBMV solvation models for the four major docking
modes identified from stage 1 docking
e(r) = 4r model
GBMV model
I
Ligands
I
II
III
IV
Mixa
II
III
IV
Mixa
B01
B02
B03
B04
B05
B06
B07
B08
B09
B10
B11
B12
B13
B14
B15
B16
B17
B18
)66.08
)67.98
)69.95
)64.95
)69.88
)73.54
)70.89
)80.83
)86.87
)81.33
)75.68
)76.83
)69.81
)79.53
)78.79
)74.46
)77.91
)74.03
)69.75
)64.09
9 ⁄ 9
)68.36
)68.76
)73.18
)65.61
)72.59
)76.71
)74.17
)74.16
)73.29
)74.26
)78.52
)74.03
)74.14
)76.35
)77.50
)76.23
)79.59
)73.87
)73.28
)64.14
7 ⁄ 9
)67.93
)66.91
)72.38
)65.05
)70.78
)72.70
)70.89
)72.45
)78.64
)71.13
)74.27
)75.86
)75.09
)75.32
)73.73
)76.62
)76.44
)75.37
)71.39
)62.45
7 ⁄ 9
)69.30
)69.16
)71.60
)69.58
)72.73
)75.12
)73.38
)79.39
)74.86
)78.41
)77.50
)76.85
)75.29
)79.78
)75.70
)76.02
)78.71
)77.30
)73.61
)66.51
8 ⁄ 9
)69.30
)69.16
)73.18
)69.58
)72.73
)76.71
)74.17
)80.83
)86.87
)81.33
)78.52
)76.85
)75.29
)79.78
)78.79
)76.62
)79.59
)77.30
)73.61
)66.61
8 ⁄ 9
)49.75
)47.59
)53.30
)50.50
)50.39
)51.48
)50.95
)54.00
)54.98
)57.49
)53.81
)54.54
)53.37
)54.17
)56.11
)52.20
)54.66
)51.86
)53.76
)47.76
8 ⁄ 9
)41.72
)42.07
)44.88
)40.59
)42.57
)39.61
)44.44
)41.40
)44.79
)46.95
)45.54
)45.61
)45.20
)42.63
)48.59
)46.71
)43.51
)39.21
)42.13
)37.56
6 ⁄ 9
)42.66
)41.31
)46.78
)43.15
)48.43
)48.07
)53.45
)47.71
)45.07
)39.28
)53.01
)46.00
)48.36
)55.08
)48.14
)57.20
)55.56
)54.03
)42.17
)36.28
5 ⁄ 9
)53.80
)51.27
)48.23
)50.40
)50.70
)52.41
)52.49
)56.04
)56.20
)51.72
)50.95
)59.52
)51.17
)56.08
)50.97
)50.48
)51.19
)51.87
)50.77
)38.66
5 ⁄ 9
)53.80
)51.27
)53.30
)50.50
)50.70
)52.41
)53.45
)56.04
)56.20
)57.49
)53.81
)59.52
)53.37
)56.08
)56.11
)57.20
)55.56
)54.03
)53.76
)47.76
8 ⁄ 9
B19
B20
TP ⁄ (TP + TN)
Se (%)
Sp (%)
Acc (%)
EF
100
100
100
2.22
78
82
80
1.73
78
82
80
1.73
89
91
90
1.98
89
91
90
1.98
89
91
90
1.98
67
73
70
1.48
56
64
60
1.23
56
64
60
1.23
89
91
90
1.98
aIn the mixed model, ligands were allowed to select the docking mode that gave the most favorable binding affinity.
Bold: the nine ligands giving the most favorable binding affinity for each structural model.
TP, true positive; TN, true negative; Se, sensitivity; Sp, Specificity; Acc, accuracy; EF, enrichment factor.
that performed well with both solvation models, D-Model I was the
best choice for the D series.
Figure 8 shows how the ligand interacted with the protein in the
lowest-energy structures for B-Model I and D-Model I as Model I
was among the highest performance structural models for both ser-
ies of compounds. Table S2 gives key protein-ligand interactions for
these two predicted structures. The salicylic ring docked into the
phosphotyrosine-binding pocket for both docking structures. In
B-Model I (Figure 8A), the hydroxyl group formed hydrogen bonds
with the main-chain NH groups of three P-loop residues (Arg-404,
Ala-405 and Gly-406). One oxygen of the carboxylic group interacted
with the guanidinium side chain of Arg-409 and the other oxygen
hydrogen-bonded with the amino side chain of Gln-450 and the back-
bone NH group of Gln-357. The nitrogens in the triazole-ring inter-
acted with the side-chain oxygen of Gln-357 and formed two
hydrogen bonds with the terminal amino group of Lys-447. In addi-
tion, the oxygen and nitrogen from the amide hydrogen-bonded with
the side-chain NH group of Arg-205 and the main-chain carbonyl
group of Gln-446. One phenyl ring of the ligand situated in a pocket
formed inside a bundle of three a-helices (a1, a6, and a7) and the
other phenyl ring protruded into solution. The latter phenyl ring might
be removed in the future to reduce the size of the compounds so that
other more useful functional groups can be introduced elsewhere.
The above analysis relied on using one single cutoff value to clas-
sify compounds into actives and non-actives. To evaluate the mod-
els more thoroughly, we also generated ROC curves and calculated
AUC which required repeating the calculations of sensitivity and
specificity using different cutoff values for classifying compounds
into actives and non-actives. Figure 6 shows these curves for
different B Models. Table 6 gives the corresponding AUC. From
these ROC plots and AUC values, we further confirmed that B-
Model I performed the best for both solvation models. However,
the mixed structural model had AUC almost as good. Therefore, our
results did not significantly favor B-Model I over the mixed model.
In future lead optimization, one may want to use both models to
obtain consensus scoring and only suggests new derivatives that
score well with both B-Model I and the mixed model for experimen-
tal testing.
Likewise, Figure 7 shows the ROC curves and Table 6 presents the
corresponding AUC for compounds in the D series. This time, only
D-Model I performed best with both solvation models. Therefore,
this might be the most likely docking mode adopted by compounds
in the D series.
The structure of D-Model
I demonstrated with ligand D09
(Figure 8B) had the phenylmorpholine ring tightly bound inside the
Chem Biol Drug Des 2010; 76: 85–99
95

Huang et al.
Table 5: Binding affinity of ligands in the D series obtained by using the e(r) = 4r and the GBMV solvation models for the four major dock-
ing modes identified from stage 1 docking
e(r) = 4r implicit model
GBMV implicit model
D-Series
ligands
I
II
III
IV
Mixa
I
II
III
IV
Mixa
D01
D02
D03
D04
D05
D06
D07
D08
D09
D10
D11
D12
D13
D14
D15
D16
D17
D18
D19
D20
TP ⁄ (TP + TN)
Se (%)
Sp (%)
Acc (%)
EF
)74.16
)74.30
)79.29
)71.30
)75.46
)78.47
)77.10
)86.62
)93.19
)90.97
)81.01
)82.27
)74.28
)85.38
)85.38
)79.63
)82.93
)78.20
)81.75
)69.63
4 ⁄ 7
)73.98
)76.32
)75.19
)75.45
)77.03
)78.31
)77.45
)77.26
)79.02
)79.97
)80.60
)77.31
)79.96
)84.20
)84.42
)77.41
)82.49
)79.81
)75.44
)72.09
2 ⁄ 7
)71.58
)68.61
)75.34
)66.67
)77.65
)78.92
)73.22
)83.61
)82.65
)76.31
)80.25
)78.87
)73.60
)84.75
)83.60
)78.88
)83.42
)78.29
)77.40
)62.05
2 ⁄ 7
)76.34
)76.05
)79.26
)74.34
)77.26
)80.92
)78.66
)84.12
)79.48
)84.04
)81.79
)80.79
)77.40
)83.37
)87.25
)81.96
)85.34
)80.97
)79.79
)73.18
3 ⁄ 7
)76.34
)76.32
)79.29
)75.45
)77.65
)80.92
)78.66
)86.62
)93.19
)90.97
)81.79
)82.27
)79.96
)85.38
)87.25
)81.96
)85.34
)80.97
)81.75
)73.18
4 ⁄ 7
)55.71
)54.21
)54.90
)61.99
)54.97
)58.26
)61.00
)59.30
)64.11
)62.40
)63.64
)62.48
)59.33
)62.85
)59.40
)62.25
)59.82
)59.19
)59.16
)61.09
6 ⁄ 7
)55.87
)54.86
)54.52
)53.63
)52.22
)59.44
)51.38
)52.73
)53.73
)53.84
)56.57
)53.74
)56.48
)58.98
)58.84
)54.19
)54.61
)50.64
)51.03
)45.48
1 ⁄ 7
)44.90
)49.46
)50.88
)48.55
)53.53
)53.43
)49.14
)58.06
)50.25
)55.71
)52.17
)54.49
)48.11
)48.63
)52.85
)61.03
)60.20
)63.65
)55.37
)36.44
3 ⁄ 7
)65.22
)71.45
)71.94
)65.15
)65.88
)69.02
)69.16
)68.04
)62.47
)71.16
)68.55
)68.98
)66.77
)66.94
)71.53
)68.41
)71.98
)70.45
)67.62
)57.23
2 ⁄ 7
)65.22
)71.45
)71.94
)65.15
)65.88
)69.02
)69.16
)68.04
)64.11
)71.16
)68.55
)68.98
)66.77
)66.94
)71.53
)68.41
)71.98
)70.45
)67.62
)61.09
2 ⁄ 7
57
77
70
1.63
29
62
50
0.82
29
62
50
0.82
43
69
60
1.22
57
77
70
1.63
86
92
90
2.45
14
54
40
0.41
43
69
60
1.22
29
62
50
0.82
29
62
50
0.82
aIn the mixed model, not all ligands were assumed to adopt the same major docking mode. Instead, the binding affinity was obtained from the mode that
yielded the most favorable binding affinity.
Bold: the seven ligands with the most favorable binding affinity for each structural model.
TP, true positive; TN, true negative; Se, sensitivity; Sp, Specificity; Acc, accuracy; EF, enrichment factor.
1
0.8
0.6
0.4
0.2
0
1
0.8
0.6
0.4
0.2
0
A
B
I
II
III
IV
I
II
III
IV
0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
Figure 6: Receiver operating
characteristics curves obtained by
using four major docking modes
and two solvation models for com-
pounds in the B series. (A) B-
Model I, II, III, and IV with the
e = 4r solvation model, (B) B-
Model I, II, III, and IV with the
GBMV model, (C) mixed structural
model with the e = 4r solvation
model, (D) mixed structural model
with the GBMV solvation model.
1–Specificity
1–Specificity
1
0.8
0.6
0.4
0.2
0
1
0.8
0.6
0.4
0.2
0
D
C
V
1
V
0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1–Specificity
1–Specificity
bundle of three helices: a1, a6, and a7. The hydroxyl group and the
carboxylate groups of the salicylic acid core formed extensive
hydrogen bonds with the side chains of Arg-409, Lys-447 and the
backbone NH groups of Gln-357, Arg-404 and Ala-405. Similar to B-
Model I, the oxygen in the furan ring hydrogen bonded with NH2 of
Arg-404. On the other hand, the three nitrogens of the triazole-ring
96
Chem Biol Drug Des 2010; 76: 85–99

Derivatives of Salicylic Acid as Inhibitors of YopH
Table 6: Area under curve (AUC) for receiver operating charac-
teristics (ROC) plots
Table 7: Area under curve (AUC) for ROC plots from rigid-protein
docking using Autodock
B Series
D Series
B ⁄ D Model
B Series
D Series
B or D Model
e(r) = 4r
GBMV
e(r) = 4r
GBMV
Model I
Model II
Model III
Model IV
1QZ0
1PA9
1YPT
Mixed
0.94
0.90
0.72
0.57
0.75
0.89
0.66
0.77
0.60
0.62
0.42
0.51
0.37
0.48
0.52
0.49
Model I
Model II
Model III
Model IV
Mixed
1
0.97
0.85
0.67
0.71
0.99
0.71
0.52
0.55
0.55
0.69
0.81
0.53
0.52
0.45
0.45
0.88
0.87
0.94
0.97
ROC, receiver operating characteristics.
formed hydrogen bonds with the side chains of Gln-357 and Lys-
447. The amide interacted with Arg-205 and Gln-446 while the mor-
pholine ring interacted with the side chains of Asp-448 and Arg-
437.
from the flexible-receptor models, suggesting that the rigid-
protein model did not perform as well as the flexible-protein
model. It is therefore better to use the best docking modes
obtained from flexible-receptor docking to guide future lead opti-
mization.
Table 7 summarizes the AUC values obtained from the eight dif-
ferent rigid-protein docking models using Autodock described
above. The best AUC values were smaller than those obtained
1
1
A
B
0.8
0.6
0.4
0.2
0
0.8
0.6
I
II
III
IV
I
II
III
IV
0.4
0.2
0
0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
Figure 7: Receiver operating
characteristics curves obtained by
using four major docking modes
and two solvation models for com-
pounds in the D series. (A) D-
Model I, II, III, and IV with the
e = 4r solvation model, (B) D-
Model I, II, III, and IV with the
GBMV model, (C) mixed structural
model with the e = 4r solvation
model, (D) mixed structural model
with the GBMV solvation model.
1–Specificity
1–Specificity
1
0.8
0.6
0.4
0.2
0
1
0.8
0.6
0.4
0.2
0
C
D
V
V
0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
1–Specificity
1–Specificity
Figure 8: Key interactions
occurring at protein–ligand inter-
face. (A) B-Model I (YopH-B17 sys-
tem). (B) D-Model I (YopH-D09
system). Coloring scheme: protein
in green, oxygen in red, nitrogen
in blue, carbon in cyan, and sulfur
in yellow. The picture was gener-
ated by VMD (39).
B-Model I
D-Model I
B
A
Chem Biol Drug Des 2010; 76: 85–99
97

Huang et al.
Conclusions
References
1. Bulter T. (1985) Textbook of Medicine. Philadelphia, PA: W. B.
Saunders.
2. Guan K.L., Dixon J.E. (1990) Protein tyrosine phosphatase activ-
ity of an essential virulence determinant in Yersinia. Sci-
ence;249:553–556.
3. Stuckey J.A., Schubert H.L., Fauman E.B., Zhang Z.-Y., Dixon
J.E., Saper M.A. (1994) Crystal structure of Yersinia protein tyro-
sine phosphatase at 2.5 ꢀ and the complex with tungstate. Nat-
ure;370:571–575.
In this work, we have developed two series of forty compounds
derived from the salicylic acid core and found 16 to have micromo-
lar inhibition activity against YopH. The initial design strategy relied
on introducing two chemical moieties, linked together by a flexible
hydrocarbon chain, to target two pockets in the active site of the
protein. We used the salicylic acid moiety to target the pocket that
phosphotyrosine bind and tried different chemical entities on the
opposite end to target a nearby secondary pocket.
4. Sun J.-P., Wu L., Fedorov A.A., Almo S.C., Zhang Z.-Y. (2003)
Crystal structure of the Yersinia protein-tyrosine phosphatase
YopH complexed with a specific small molecule inhibitor. J Biol
Chem;278:33392–33399.
5. Bçlin I., Wolf-Watz H. (1988) The plasmid-encoded Yop2b pro-
tein of Yersinia pseudotuberculosis is a virulence determinant
regulated by calcium and temperature at the level of transcrip-
tion. Mol Microbiol;2:237–245.
6. Michiels T., Cornelis G.R. (1988) Nucleotide sequence and tran-
scription analysis of yop51 from Yersinia enterocolitica W22703.
Microb Pathog;5:449–459.
7. Straley S.C., Bowmer W.S. (1986) Virulence genes regulated at
the transcriptional level by Ca2+ in Yersinia pestis include struc-
tural genes for outer membrane proteins. Infect Immun;51:445–
454.
To predict how these compounds bind to YopH, we have started
with flexible ligand-flexible protein docking using two different
solvation models: e(r) = 4r and GBMV. The docking suggested four
possible docking modes, three had the salicylic acid core bound
to the phosphotyrosine-binding pocket and one had the other end
of the ligands bound instead. As the docking models yielded simi-
lar energy for these docking modes, it was difficult to single out
the best docking mode. We therefore used each one of these
docking modes to perform binding affinity calculations to examine
which docking modes gave results most consistent with experi-
mental IC50. We also considered a mixed structural model in
which not all ligands were required to bind to the same binding
mode. Instead, the docking mode – of four major docking modes
identified from flexible-receptor docking – that yielded the most
favorable binding affinity was taken. Different performance analy-
sis such as calculating accuracy and area under receiver operating
characteristics curve suggested that compounds in the B series
might prefer a binding mode (B-Model I) in which all ligands
bound with the salicylic acid core situated in the phosphotyrosine-
binding pocket. However, the mixed model that allowed different
ligands to take on different ones of the four possible major bind-
ing modes was also possible. On the other hand, compounds in
the D series preferred a binding mode similar to B-Model I – i.e.,
D-Model I – in which all ligands bound in roughly the same way
to the protein with minor adjustments at the protein–ligand
interface.
8. Bliska J.B., Clemens J.C., Dixon J.E., Falkow S. (1992) The Yer-
sinia tyrosine phosphatase: specificity of a bacterial virulence
determinant for phosphoproteins in the J774A.1 macrophage.
J Exp Med;176:1625–1630.
9. Bliska J.B., Guan K.L., Dixon J.E., Falkow S. (1991) Tyrosine
phosphate hydrolysis of host proteins by an essential Yersinia
virulence determinant. Proc Natl Acad Sci USA;88:1187–1191.
10. Phan J., Lee K., Cherry S., Tropea J.E., Burke, T.R. Jr., Waugh D.S.
(2003) High-resolution structure of the Yersinia pestis protein
tyrosine phosphatase YopH in complex with a phosphotyrosyl
mimetic-containing hexapeptide. Biochemistry;42:13113–13121.
11. Liang F., Huang Z.H., Lee S.-Y., Liang J., Ivanov M.I., Alonso A.,
Bliska J.B., Lawrence D.S., Mustelin T., Zhang Z.Y. (2003) Aurin-
tricarboxylic acid blocks in vitro and in vivo activity of YopH, an
essential virulent factor of Yersinia pestis, the agent of plague.
J Biol Chem;278:41734–41741.
12. Tautz L., Bruckner S., Sareth S., Alonso A., Bogetz J., Bottini N.,
Pellecchia M., Mustelin T. et al. (2005) Inhibition of Yersinia
tyrosine phosphatase by furanyl salicylate compounds. J Biol
Chem;280:9400–9408.
13. Hu X., Stebbins C.E. (2005) Molecular docking and 3D-QSAR
studies of Yersinia protein tyrosine phosphatase YopH inhibitors.
Bioorg Med Chem;13:1101–1109.
We had also performed rigid-receptor docking using Autodock but
the performance was not as good as the molecular dynamics-
based flexible-receptor docking. The two best docking models
found for the B series (B-Model I and the mixed model) and the
best model found for the D series found from flexible-receptor
docking might thus be the best to use in the immediate future
to guide future optimization of these two series of compound,
before experimental structures of protein–ligand complexes are
available.
14. Galimand M., Guiyoule A., Gerbaud G., Rasoamanana B., Chan-
teau S., Carniel E., Courvalin P. (1997) Multidrug resistance in
Yersinia pestis mediated by a transferable plasmid. N Engl J
Med;337:677–680.
15. Guiyoule A., Gerbaud G., Buchrieser C., Galimand M., Rahalison
L., Chanteau S., Courvalin P., Carniel E. (2001) Transferable plas-
mid-mediated resistance to streptomycin in a clinical isolate of
Yersinia pestis. Emerg Infect Dis;7:43–48.
Acknowledgments
This research was supported by a Research Board Award from the
University of Missouri System, and the National Institutes of Health
(Grants AI071991 and CA126937). We also thank the University of
Missouri Bioinformatics Consortium and the University of Missouri-
Saint Louis Information Technology Services for providing computa-
tional resources.
98
Chem Biol Drug Des 2010; 76: 85–99

Derivatives of Salicylic Acid as Inhibitors of YopH
16. Sarmiento M., Wu L., Keng Y.-F., Song L., Luo Z., Huang Z., Wu
G.Z., Yuan A.K., Zhang Z.Y. (2000) Structure-based discovery of
small molecule inhibitors targeted to protein tyrosine phospha-
tase 1B. J Med Chem;43:146–155.
17. Huang Z., Wong C.F. (2007) A mining-minima approach to explor-
ing the docking pathways of p-nitrocatechol sulfate to YopH.
Biophys J;93:4141–4150.
31. Feig M., Karanicolas J., Brooks C.L. III (2004) MMTSB tool set:
enhanced sampling and multiscale modeling methods for appli-
cations in structural biology. J Mol Graph Model;22:377–395.
32. Karpen M.E., Tobias D.J., Brooks C.L. III (1993) Statistical clus-
tering techniques for the analysis of long molecular dynamics
trajectories: analysis of 2.2-ns trajectories of YPGDV. Biochemis-
try;32:412–420.
18. Huang Z., Wong C.F., Wheeler R.A. (2008) Flexible Protein-Flexi-
ble Ligand Docking with Disrupted Velocity Simulated Annealing.
Proteins: Struct Funct Bioinform;71:440–454.
19. Kolb H.C., Finn M.G., Sharpless K.B. (2001) Click chemistry:
diverse chemical function from a few good reactions. Angew
Chem Int Ed Engl;40:2004–2021.
33. Pao Y.H. (1989) Adaptive Pattern Recognition and Neural Net-
works. New York: Addison Wesley.
34. Morris G.M., Goodsell D.S., Halliday R.S., Huey R., Hart W.E.,
Belew R.K., Olson A.J. (1998) Automated docking using a
lamarckian genetic algorithm and an empirical binding free
energy function. J Comput Chem;19:1639–1662.
20. Lee M.S., Feig M., Salsbury F.R. Jr, Brooks C.L. III (2003) New
analytic approximation to the standard molecular volume defini-
tion and its application to generalized Born calculations. J Com-
put Chem;24:1348–1356.
21. Lee M.S., Salsbury F.R. Jr, Brooks C.L. III (2002) Novel general-
ized Born methods. J Chem Phys;116:10606–10614.
22. ChocholouÐovꢁ J., Feig M. (2006) Balancing an accurate repre-
sentation of the molecular surface in generalized born formal-
isms with integrator stability in molecular dynamics simulations.
J Comput Chem;27:719–729.
23. Huang Z., Wong C.F. (2009) Conformational selection of protein
kinase A revealed by flexible-ligand flexible-protein docking. J
Comput Chem;30:631–644.
35. Goodsell D.S., Morris G.M., Olson A.J. (1996) Automated dock-
ing of flexible ligands: applications of AutoDock. J Mol Recog-
nit;9:1–5.
36. Triballeau N., Acher F., Brabet I., Pin J.-P., Bertrand H.-O. (2005)
Virtual screening workflow development guided by the ``receiver
operating characteristic'' curve approach. Application to high-
throughput docking on metabotropic glutamate receptor subtype
4. J Med Chem;48:2534–2547.
37. Kirchmair J., Markt P., Distinto S., Wolber G., Langer T. (2008)
Evaluation of the performance of 3D virtual screening protocols:
RMSD comparisons, enrichment assessments, and decoy selec-
tion-What can we learn from earlier mistakes. J Comput Aided
Mol Des;22:213–228.
24. MacKerell A.D. Jr, Bashford D., Bellott M., Dunbrack R.L. Jr,
Evanseck J.D., Field M.J., Fischer S. et al. (1998) All-atom
empirical potential for molecular modeling and dynamics studies
of proteins. J Phys Chem B;102:3586–3616.
25. MacKerell A.D. Jr, Feig M., Brooks C.L. III (2004) Extending the
treatment of backbone energetics in protein force fields: limita-
tions of gas-phase quantum mechanics in reproducing protein
conformational distributions in molecular dynamics simulations.
J Comput Chem;25:1400–1415.
38. Jacobsson M., Lidꢂn P., Stjernschantz E., Bostrçm H., Norinder
U. (2003) Improving structure-based virtual screening by multi-
variate analysis of scoring data. J Med Chem;46:5781–5789.
39. Humphrey W., Dalke A., Schulten K. (1996) VMD: visual molecu-
lar dynamics. J Mol Graph;14:33–38.
Supporting Information
Additional Supporting Information may be found in the online ver-
sion of this article:
26. Advanced Chemistry Development in (2006) ACD ⁄ ChemSketch
Freeware. Toronto, ON.
Table S1. Protein residues allowed to move in Stage 2 protein-
ligand docking.
27. Brooks B.R., Bruccoleri R.E., Olafson B.D., States D.J., Swamina-
than S., Karplus M. (1983) CHARMM: a program for macromo-
lecular energy, minimization, and dynamics calculations.
Comput Chem;4:187–217.
J
Table S2. Key protein-ligand interactions for one representative
compound in the B series and one in the D series.
28. Hu X., Stebbins C.E. (2006) Dynamics of the WPD Loop of the
Yersinia protein tyrosine phosphatase. Biophys J;91:948–956.
29. Pedretti A., Villa L., Vistoli G. (2004) VEGA – an open platform
to develop chemo-bio-informatics applications, using plug-in
architecture and script programming. J Comput Aided Mol
Des;18:167–173.
Please note: Wiley-Blackwell is not responsible for the content or
functionality of any supporting materials supplied by the authors.
Any queries (other than missing material) should be directed to the
corresponding author for the article.
30. Carpenter G.A., Grossberg S. (1987) ART 2: self-organization of
stable category recognition codes for analog input patterns.
Appl Opt;26:4919–4930.
Chem Biol Drug Des 2010; 76: 85–99
99