Structural basis of the promiscuous inhibitor susceptibility of Escherichia coli LpxC

Article abstract of DOI:10.1021/cb400067g

The LpxC enzyme in the lipid A biosynthetic pathway is one of the most promising and clinically unexploited antibiotic targets for treatment of multidrug-resistant Gram-negative infections. Progress in medicinal chemistry has led to the discovery of potent LpxC inhibitors with a variety of chemical scaffolds and distinct antibiotic profiles. The vast majority of these compounds, including the nanomolar inhibitors L-161,240 and BB-78485, are highly effective in suppressing the activity of Escherichia coli LpxC (EcLpxC) but not divergent orthologs such as Pseudomonas aeruginosa LpxC (PaLpxC) in vitro. The molecular basis for such promiscuous inhibition of EcLpxC has remained poorly understood. Here, we report the crystal structure of EcLpxC bound to L-161,240, providing the first molecular insight into L-161,240 inhibition. Additionally, structural analysis of the EcLpxC/L-161,240 complex together with the EcLpxC/BB-78485 complex reveals an unexpected backbone flipping of the Insert I βa?βb loop in EcLpxC in comparison with previously reported crystal structures of EcLpxC complexes with L-threonyl-hydroxamate-based broad-spectrum inhibitors. Such a conformational switch, which has only been observed in EcLpxC but not in divergent orthologs such as PaLpxC, results in expansion of the active site of EcLpxC, enabling it to accommodate LpxC inhibitors with a variety of head groups, including compounds containing single (Ror S-enantiomers) or double substitutions at the neighboring C? atom of the hydroxamate warhead group. These results highlight the importance of understanding inherent conformational plasticity of target proteins in lead optimization.

Full text of DOI:10.1021/cb400067g

Articles
pubs.acs.org/acschemicalbiology
Structural Basis of the Promiscuous Inhibitor Susceptibility of
Escherichia coli LpxC
Chul-Jin Lee,†,∥ Xiaofei Liang,‡ Ramesh Gopalaswamy,‡ Javaria Najeeb,†,∥ Eugene D. Ark,§
Eric J. Toone,†,∥,‡ and Pei Zhou*,†,∥,‡
†Department of Biochemistry, Duke University Medical Center, Durham, North Carolina 27710, United States
∥Structural Biology & Biophysics Program, Duke University, Durham, North Carolina 27710, United States
‡Department of Chemistry, Duke University, Durham, North Carolina 27708, United States
§Trinity College of Arts & Sciences, Duke University, Durham, North Carolina 27708, United States
S
* Supporting Information
ABSTRACT: The LpxC enzyme in the lipid A biosynthetic pathway is one
of the most promising and clinically unexploited antibiotic targets for
treatment of multidrug-resistant Gram-negative infections. Progress in
medicinal chemistry has led to the discovery of potent LpxC inhibitors
with a variety of chemical scaffolds and distinct antibiotic profiles. The vast
majority of these compounds, including the nanomolar inhibitors L-161,240
and BB-78485, are highly effective in suppressing the activity of Escherichia
coli LpxC (EcLpxC) but not divergent orthologs such as Pseudomonas
aeruginosa LpxC (PaLpxC) in vitro. The molecular basis for such promiscuous
inhibition of EcLpxC has remained poorly understood. Here, we report the
crystal structure of EcLpxC bound to L-161,240, providing the first molecular
insight into L-161,240 inhibition. Additionally, structural analysis of the
EcLpxC/L-161,240 complex together with the EcLpxC/BB-78485 complex
reveals an unexpected backbone flipping of the Insert I βa−βb loop in EcLpxC in comparison with previously reported crystal
structures of EcLpxC complexes with L-threonyl-hydroxamate-based broad-spectrum inhibitors. Such a conformational switch,
which has only been observed in EcLpxC but not in divergent orthologs such as PaLpxC, results in expansion of the active site of
EcLpxC, enabling it to accommodate LpxC inhibitors with a variety of head groups, including compounds containing single (R-
or S-enantiomers) or double substitutions at the neighboring Cα atom of the hydroxamate warhead group. These results
highlight the importance of understanding inherent conformational plasticity of target proteins in lead optimization.
ical companies and academic laboratories, and potent LpxC
inhibitors with distinct chemical scaffolds have emerged in the
last two decades (Figure 1B) (earlier compounds reviewed by
refs 3 and 4 and newer inhibitors described in refs 5−9).
Small molecule LpxC inhibitors were initially discovered in
the late 1990s using random compound screenings that
measured 3H-galactose uptake in E. coli cells.10 Lead
optimization led to the discovery of L-161,240, a hydrox-
amate-containing phenyloxazoline compound (Figure 1B) that
inhibited the growth of E. coli with potency comparable to that
of ampicillin.10 Administration of L-161,240 rescued mice
challenged with a lethal dose of E. coli, demonstrating its
therapeutic potential. Despite the impressive antibiotic activity
of L-161,240 against E. coli in vitro and in vivo, the
ineffectiveness of L-161,240 in controlling the growth of
Pseudomonas aeruginosa was noted early on.10 Additional
experimentation has shown that the differential antibiotic
INTRODUCTION
■
LpxC is an essential zinc-dependent enzyme in Gram-negative
bacteria that catalyzes the deacetylation of UDP-3-O-(acyl)-N-
acetylglucosamine in the biosynthesis of lipid A, the membrane
anchor of lipopolysaccharide.1 As the major lipid component of
the outer monoleaflet of the Gram-negative bacterial outer
membrane, lipid A protects bacterial cells from the damage of
external agents, such as detergents and antibiotics. The
biosynthesis of Kdo2-lipid A, the active component of
endotoxin, consists of nine steps in Escherichia coli (Figure
1A).1 With the exception of few Gram-negative bacteria, the
first step of lipid A biosynthesis catalyzed by LpxA is
thermodynamically unfavorable and reversible.2 Therefore, the
second reaction catalyzed by LpxC is generally considered the
committed step of lipid A biosynthesis. Since inhibition of
LpxC is bactericidal for most Gram-negative bacteria and since
LpxC has never been exploited by existing antibiotics, it has
become one of the most attractive novel antibiotic targets for
treatment of multidrug-resistant Gram-negative infections.3
Consequently, significant efforts have been devoted to
developing effective LpxC-targeting antibiotics by pharmaceut-
Received: January 26, 2013
Accepted: October 11, 2013
Published: October 11, 2013
© 2013 American Chemical Society
237
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
Figure 1. Antibiotics targeting LpxC in the lipid A biosynthetic pathway. (A) LpxC is an essential enzyme that catalyzes the first irreversible step of
lipid A biosynthesis. (B) Representative LpxC inhibitors show distinct stereochemistry at the Cα position next to the hydroxamate group.
Figure 2. Structure of EcLpxC in complex with L-161,240. (A) Ribbon diagram of the crystallographic dimer of the EcLpxC/L-161,240 complex.
Monomers are colored distinctly. The L-161,240 molecules are shown with a space-filling model. (B) Comparison of the EcLpxC/L-161,240
structure with EcLpxC/LPC-009. The EcLpxC/L-161,240 complex has an unstructured C-terminal tail, whereas the C-terminus of the EcLpxC/
LPC-009 complex folds into the gap between domains I and II. Insert I, Insert II, and the C-terminus (residues 288−300, modeled) are highlighted
in orange, blue, and red, respectively. Inhibitors are shown with a stick model, and the active site zinc ion is shown with a space-filling model.
activities of L-161,240 toward E. coli and P. aeruginosa are due
to the greater binding affinity of L-161,240 toward E. coli LpxC
(EcLpxC) than toward P. aeruginosa LpxC (PaLpxC) but not
due to differences in intrinsic resistance mechanisms of these
two bacterial strains, such as membrane permeability or efflux.11
A similar result was observed for the sulfonamide−hydroxamate
compound BB-78485 containing two naphthalene rings (Figure
1B).12 BB-78485 prevented E. coli growth with an antibiotic
activity slightly better than that of L-161,240, but it is still
ineffective against P. aeruginosa.
Recent medicinal chemistry efforts have led to the discovery
of more potent LpxC-targeting antibiotics that are also effective
for a wide range of Gram-negative pathogens, including P.
aeruginosa.5,6,9,13−15 In parallel, structural studies of LpxC and
its inhibitor-bound complexes have started to unveil the
architecture of this essential enzyme and the molecular basis
of compound selectivity.5,15−23 Notably, structural elucidation
of different LpxC orthologs in complex with the same
diacetylene-based inhibitor LPC-009 has revealed large,
inherent conformational differences of the hydrophobic,
substrate-binding passage encompassed by Insert II of different
LpxC orthologs, which plays a predominant role in
accommodating the hydrophobic “tail” of LpxC-targeting
antibiotics and in determining their activity profiles.22
In contrast to many structurally characterized LpxC
inhibitors containing a rigid hydrophobic tail inserted into
the hydrophobic substrate-binding passage, L-161,240 has a
flexible propyl chain and two methoxy groups attached to a
phenyloxazoline moiety. Although previous NMR studies have
shown that the propyl chain of L-161,240 similarly inserts into
the hydrophobic passage as other LpxC-targeting antibiotics,19
the flexible propyl chain of L-161,240 is unlikely to be affected
by conformational variation of the substrate-binding passage
accompanied by large changes of the Insert II helix orientation
in distinct LpxC orthologs, suggesting that the molecular
determinant rendering EcLpxC exquisitely sensitive to the
inhibition of L-161,240 arises from structural features other
than the Insert II region of EcLpxC.
In order to understand the molecular basis underlying the
promiscuous susceptibility of E. coli LpxC to a variety of LpxC
inhibitors, including L-161,240 and BB-78485, we determined
the structure of EcLpxC in complex with L-161,240 at 2.1 Å
resolution. This structure, together with a 1.8 Å resolution
structure of EcLpxC bound to BB-78485, reveals a previously
unanticipated molecular determinant of the inhibitor specificity
238
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
Figure 3. Binding mode of L-161,240 in EcLpxC. (A) Interactions between L-161,240 and EcLpxC in the active site. EcLpxC is shown as a ribbon
diagram with Insert I and Insert II highlighted in orange and blue, respectively. L-161,240 and interacting LpxC residues are shown with a stick
model. Purple mesh represents the mFo − DFc omit map (contoured at 3.4σ) surrounding the inhibitor. The active site zinc ion is shown as a space-
filling model. Hydrogen bonds are denoted by dashed lines. (B) The zinc coordination geometry of L-161,240 (trigonal-bipyramidal) differs
significantly from those of BB-78485, TU-514, and LPC-009 (square pyramidal). In the right panel, PaLpxC/BB-78485 (magenta), AaLpxC/TU-514
(blue), and EcLpxC/LPC-009 (orange) complexes are superimposed on the EcLpxC/L-161,240 structure (green). Inhibitors and side chains of
protein involved in zinc coordination are shown as sticks, and zinc ions are shown as spheres. The equatorial plane of the trigonal-bipyramid
geometry and the square base of square-pyramidal geometry are colored in yellow. Arrows from zinc ions indicate the apical points of coordination
geometry. (C) External view of the hydrophobic substrate-binding passage in the EcLpxC/L-161,240 complex. (D) Binding of L-161,240 to EcLpxC
causes backbone flipping of the βa−βb loop in Insert I, with the side chain of C63 swinging out of the active site to generate more space for the bulky
oxazoline moiety of L-161,240. Complexes of EcLpxC/L-161,240 and EcLpxC/LPC-009 are shown in green and orange, respectively. The side chain
of residue C63, the backbone of residues in the Insert I loop, and inhibitors are shown as sticks. Zinc ions are shown as spheres.
and sheds structural insights into further development of LpxC-
targeting antibiotics.
structures.16−18 EcLpxC consists of two domains with a
β−α−α−β sandwich fold, and each domain contains a unique
insert region: Insert I of Domain I and Insert II of Domain II.
Interestingly, the EcLpxC/L-161,240 complex has a fully
extended C-terminal tail that interacts with the neighboring
protomer in the unit cell in a domain-swapped fashion
(Supplemental Figure S1, Supporting Information). Such an
extended tail conformation differs from other EcLpxC−
inhibitor structures,15,22 in which the C-terminus folds into
the gap between Domains I and II (Figure 2B). Given the
solution NMR observation of a flexible and unprotected C-
terminal tail of the EcLpxC/L-161,240 complex, which allows
for its recognition and degradation by the FtsH protease,24 the
domain swapped loop conformation of the C-terminal tail in
the EcLpxC/L-161,240 complex likely reflects a crystal packing
artifact.
The molecular scaffold of L-161,240 can be divided into
three distinct regions: a hydroxamate group, an oxazoline ring,
and a phenyl ring substituted with two methoxy groups and a
propyl chain (Figure 1B). Each of these three regions exhibits
unique interactions with LpxC to form a tight-binding complex.
The hydroxamate headgroup of L-161,240, a primary affinity
determinant shared by many potent LpxC inhibitors,
participates in coordination of the catalytic zinc with its two
RESULTS AND DISCUSSION
■
Structure of EcLpxC in complex with L-161,240. The
crystal structure of EcLpxC in complex with L-161,240 was
determined at 2.1 Å resolution using molecular replacement
with the previously reported EcLpxC/LPC-009 complex
structure as the search model.22 Because of disorder at the C-
terminus, the last six amino acids of EcLpxC were not modeled.
The EcLpxC/L-161,240 complex crystals were grown in
significantly different crystallization conditions from previously
reported EcLpxC−inhibitor crystals, and they belong to the
trigonal R32 space group rather than the hexagonal P61 space
group observed for other EcLpxC structures.15,22 In addition,
previously reported crystals contained a single molecule in each
asymmetric unit, whereas two EcLpxC/L-161,240 complexes
arrange as a parallel homodimer to form the crystallographic
asymmetric unit (Figure 2A). Since EcLpxC behaves as a
monomer in solution, this crystallographic dimer is unrelated to
its biological function.
The EcLpxC/L-161,240 complex shows a similar overall
topology as previously reported Aquifex aeolicus LpxC
239
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
Figure 4. Inhibitor-induced conformational change in Insert I of LpxC. (A) Active site interactions between BB-78485 (purple) and EcLpxC. Purple
mesh represents the mFo−DFc omit map (contoured at 3.4σ) of BB-78485. The color scheme and model representation of EcLpxC are identical to
those in Figure 3A. (B) Insert I of the EcLpxC/BB-78485 complex (purple) adopts the same conformation as observed for the EcLpxC/L-161,240
complex (green). (C) Isotropic temperature factor distribution in EcLpxC/BB-78485, PaLpxC/BB-78485, and PaLpxC/LPC-009. Ribbons are
colored by B factor, with dark blue to red indicating low to high values, respectively.
oxygen atoms (Figure 3A). The hydroxamate oxygen atoms of
L-161,240 form a typical penta-coordination together with
three conserved LpxC residues, H79, H238, and D242;
however, the coordination geometry in the EcLpxC/L-
161,240 complex is vastly different from other LpxC/
hydroxamate inhibitor complexes, which coordinate the
catalytic zinc with an approximate square-pyramidal geometry.
Instead, L-161,240 chelates the catalytic zinc ion in a trigonal-
bipyramidal geometry (Figure 3B), with H79, H238, and the
hydroxyl oxygen of the hydroxamate group forming the central
plane and with D242 and the carbonyl oxygen of the
hydroxamate group forming two apical vertices. Compared
with other LpxC−inhibitor complexes, such as the EcLpxC/
LPC-009 complex,22 the PaLpxC/BB-78485 complex,21 and
the AaLpxC/TU-514 complex,25 the plane formed by the
hydroxamate atoms in the EcLpxC/L-161,240 complex rotates
∼25° around the hydroxamate hydroxyl oxygen. The nitrogen
atom of the hydroxamate group of L-161,240 does not appear
to form direct hydrogen bonds with the side chains of the
catalytically important residues E78 and H265 as in other
LpxC/inhibitor complexes (Figure 3A). The hydroxamate
carbonyl oxygen, however, still forms a hydrogen bond with
the hydroxyl group of the conserved T191 side chain.
The second region of L-161,240, the oxazoline group, is
connected to the hydroxamate group by a chiral carbon. The
vast majority of known LpxC inhibitors can be categorized
loosely as amino acid derivatives (Figure 1B), with the
substituted nitrogen group and the hydroxamate group being
R1 and R2 and the Cα substitutions being R3 and R3′. In L-
161,240, the nitrogen and alkoxy sides of the bulky cyclic
moiety stem from the R1 and R3′ positions of the chiral Cα
carbon, respectively, rendering L-161, 240 as an R-enantiomer,
as opposed to the S-enantiomer inhibitors, such as L-threonine-
based CHIR-090 and LPC-009 containing a proton at R3′ and
the threonine side chain at R3 (Figure 1B). While the threonyl
moiety of the S-enantiomer compounds CHIR-090 or LPC-009
forms extensive interactions with highly conserved residues on
the Insert II side of the LpxC active site, the oxazoline moiety
of L-161,240, due to its R-configuration at the Cα-position next
to the hydroxamate group, tilts toward the βa−βb loop of Insert
240
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
Table 1. MIC Values
a
b
LpxC replacement strain (E. coli lpxC gene replaced by P. aeruginosa lpxC gene).15 Values for CHIR-090 and LPC-009 were adapted from refs 15
and 22.
I on the opposite side of the active site of EcLpxC, with its
nitrogen atom forming a hydrogen bond with the backbone
carbonyl group of L62 to hold the oxazoline ring (Figure 3A).
The third moiety of L-161,240, the substituted phenyl
moiety, interacts extensively with the hydrophobic substrate-
binding passage. The phenyl ring is located at the “mouth” of
the hydrophobic passage, and its propyl group inserts into the
hydrophobic passage (Figure 3A,C), with the terminal methyl
group of the propyl chain reaching to the close proximity of the
backbone amides of G210 and S211 of the Insert II at the exit
of the hydrophobic passage, a configuration consistent with
prior solution NMR studies of the EcLpxC/L-161,240
complex.19 Although the penetration of the L-161,240 is
shallower than that of CHIR-090 or LPC-009, the EcLpxC/L-
161,240 complex shows an equivalent opening of the external
exit of the hydrophobic passage (Figure 3C). Within the active
site, the phenyl ring of L-161,240 is supported by favorable van
der Waals (vdW) contacts with the methyl groups of L62.
Additionally, water-mediated hydrogen bonds are formed
between the two methoxy groups of L-161,240 and the side
chain of D197 and may serve to fixate the orientation of the
Insert II helix (Figure 3A).
Backbone Flipping of Insert I Residues in EcLpxC.
Despite the overall structural similarity of the EcLpxC/L-
161,240 complex with other EcLpxC/inhibitor complexes,
detailed structural comparison of the EcLpxC/L-161,240 and
EcLpxC/LPC-009 (PDB entry 3P3G) complexes has revealed
an unexpected conformational change of Insert I (Figure 3D).
In particular, we note that the βa−βb loop of Insert I in the
EcLpxC/L-161,240 complex undergoes complete backbone
flipping for L62 and C63 located in the loop connecting βa and
βb strands of Insert I. As a result of such backbone flipping, the
side chain of C63 rotates by almost 180° relative to its
orientation in the EcLpxC/LPC-009 complex, pulling away
from the hydrophobic passage to avoid a potential steric clash
with the oxazoline moiety of L-161,240. In contrast, in the
LPC-009 complex that contains a proton at R3′ instead of a
bulky substituent, the side chain of C63 flips down and points
toward the hydrophobic passage of EcLpxC. Such a notable
conformational change of the backbone flipping and side chain
rotation in the EcLpxC/L-161,240 complex causes the βa−βb
loop to shift away from the active site by ∼2.0 Å, creating a
more spacious active site cavity than that observed for the
EcLpxC/LPC-009 complex to accommodate the R-stereo-
specific substitution at the Cα position next to the hydroxamate
group of L-161,240.
Having shown that the EcLpxC/LPC-009 and EcLpxC/L-
161,240 complexes have relatively minor global conformational
differences, but very distinct configurations of the βa−βb loop
of Insert I, we suspected that the ability of the βa−βb loop of
EcLpxC to adopt two distinct conformations directly underlies
its extraordinary ability to accommodate a variety of diverse
inhibitors, including BB-78485, which also is an R-enantiomer
at the Cα position next to the hydroxamate group, with one
naphthalene ring connected through a sulfonamide group to
the R1 position of the chiral carbon and the other naphthalene
ring connected to the R3′ position (Figure 1B). To test this
hypothesis, we additionally determined the crystal structure of
the EcLpxC/BB-78485 complex, which reveals an overall
similar topology for LpxC and an identical binding mode of BB-
78485 as the previously reported PaLpxC/BB-78485 struc-
ture,21 with both naphthalene groups pointing toward the
hydrophobic passage (Figure 4A).
Strikingly, in the EcLpxC/BB-78485 complex, we observed
the same Insert I conformation found in the EcLpxC/L-
161,240 complex, where the βa−βb loop of Insert I flips out to
create a more spacious active site to accommodate BB-78485
(Figure 4B). The magnitude of the Insert I conformational
241
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
Figure 5. Conformational flexibility of Insert I in EcLpxC accommodates the Cα-methyl substitution of LPC-138. (A) The binding mode of LPC-
138 (slate) is shown in the stick model, with the same color scheme and model representation of EcLpxC as those in Figure 3A. Favorable protein−
inhibitor interactions are denoted as dashed lines. Purple mesh represents the mFo − DFc omit map (contoured at 3.4σ) surrounding the inhibitor.
(B) The EcLpxC/LPC-138 complex displays both unflipped and flipped conformations of the Insert I loop. Green mesh represents the 2mFo − DFc
electron density for C63 and its flanking residues. The contouring level of the electron map is 0.7σ. (C) Superimposition of EcLpxC/LPC-009
(orange) and EcLpxC/LPC-138 (slate). Introduction of the Cα-methyl group in LPC-138 results in rotation of the neighboring amide bond and the
proximal phenyl ring to avoid internal steric clashes. The amide plane of LPC-138 is rotated ∼36° away from the corresponding plane of LPC-009 in
EcLpxC.
rearrangement induced by BB-78485 binding is even more
pronounced than that induced by L-161,240, due to the
presence of the bulkier naphthalene-sulfonamide moiety. Thus,
the conformational plasticity of the βa−βb loop of Insert I of
EcLpxC may be a unique structural feature that enables
EcLpxC to accommodate inhibitors with divergent chemical
scaffolds in the active site. In support of this notion, PaLpxC
appears to have a more rigid βa−βb loop in Insert I, which does
not undergo any significant conformational change upon BB-
78485 binding, resulting in severe penalty of the binding
energy. Accordingly, PaLpxC is inhibited poorly by BB-78485.
Its MIC value is at least 16-fold higher against an E. coli−
PaLpxC knock-in strain than against WT E. coli (Table 1),
indicating a similar reduction of binding affinity of BB-78485
toward PaLpxC compared with EcLpxC. Additionally, binding
of BB-78485 significantly increases the isotropic temperature
factor of the βa−βb loop of Insert I in the PaLpxC/BB-78485
complex,21 whereas the corresponding region has a much
smaller B-factor in the PaLpxC/LPC-009 complex22 and in the
EcLpxC/BB-78485 complex (this work) (Figure 4C), suggest-
ing that although the unflipped Insert I conformation of
PaLpxC can easily accommodate S-enantiomer compounds
such as L-threonine-based LPC-009,22 such a rigid conforma-
tion is far less capable of accommodating the bulky naphthalene
moiety of BB-78485.
In order to further investigate the role of the βa−βb loop in
inhibitor selectivity, we attempted to generate chimeric LpxCs
in which the βa−βb loop was swapped between EcLpxC and
PaLpxC and vice versa. Unfortunately, expression of the
PaLpxC chimera with the EcLpxC βa−βb loop (58-
RDTMLCTC-65) formed an inclusion body, whereas the
EcLpxC chimera with the PaLpxC βa−βb loop (58-
GETTMSTT-65) is unstable in solution and does not allow
for reliable measurements for inhibitor binding.
Conformational Flexibility of Insert I of E. coli LpxC
Accommodates Cα Doubly Substituted LpxC Inhibitors.
One limitation of hydromaxate compounds is their relatively
high plasma clearance rates due to the metabolic turnover of
the hydroxamate group through glucuronide conjugation and
aldehyde oxidase-mediated metabolism.26 Substituting the
hydroxamate group with the carboxylate group significantly
improves the half-life of parent compounds, but such a
substitution invariably results in a complete loss of antibiotic
activity of LpxC inhibitors.5,27 Another approach to suppress
the metabolic turnover of the hydroxamate group is through
introduction of sterically hindered functional groups (e.g., an
additional methyl group) at the neighboring Cα position of the
hydroxamate group.28 Although such substitution generates
vdW clashes in our structural model of the EcLpxC/LPC-009
complex, the observation that Insert I of EcLpxC is capable of
undergoing backbone flipping to generate an expanded active
site to accommodate R-enantiomer compounds L-161,240 and
BB-78485 suggests that with a flipped Insert I, EcLpxC may be
able to tolerate inhibitors containing such modification. To test
this hypothesis, we synthesized a derivative of LPC-009 and
LPC-011, which contains a Cα-methyl substituted threonyl-
hydroxamate headgroup and an amino-substituted biphenyl
diacetylene scaffold. Indeed, such a methyl substituted
compound (LPC-138) is well tolerated by EcLpxC, with a
MIC value of ∼0.24 μg/mL, which is similar to that of CHIR-
090 and slightly worse than LPC-009 and LPC-011 (∼5−8-
fold; Table 1). In contrast, although CHIR-090, LPC-009, and
LPC-011 display broad-spectrum antibiotic activities against a
variety of Gram-negative bacteria, the Cα-methyl substituted
compound LPC-138 has become a specific antibiotic similar to
L-161,240 and BB-78485 (Table 1), and it is effective only for
E. coli and closely related Gram-negative bacteria, whose LpxC
enzymes share a high degree of sequence identity with EcLpxC
and presumably a flexible Insert I capable of conformational
flipping to interact with Cα doubly substituted inhibitors
(Table 1).
In order to verify this predicted mode of interaction of LPC-
138 with LpxC, we carried out structural analysis of EcLpxC in
complex with LPC-138 (Figure 5). As expected, LPC-138 binds
in the hydrophobic passage, with an overall binding mode very
similar to that of LPC-009 reported previously.22 The
introduction of the Cα-methyl substitution however causes
rotation of the neighboring amide bond and the proximal
phenyl ring to avoid internal steric clashes, and these changes
generate a flipped Insert I conformation of EcLpxC that
coexists with an unflipped Insert I conformation. The
coexistence of two conformations of the EcLpxC Insert I is
242
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
a
Scheme 1. Synthesis of LPC-138
a
Reagents and conditions: (a) NH2OBn·HCl, 1 M LiHMDS, THF, −78 °C, 1 h; (b) 2 M HCl, MeOH, RT (8 h) then reflux (4 h); (c)
pentafluorophenol, DCC, DMAP, THF, RT, 12 h; (d) NEt3, DMF, RT, 12 h; (e) 1 M BCl3, DCM, −10 °C, 1 h.
consistent with the smaller spatial footprint of the Cα-methyl
group and its neighboring amide group compared with the R-
substituted oxazoline ring in L-161,240 and the naphthalene
group of BB-78485, and such an observation reinforces the
notion of conformational flexibility of the Insert I loop of E. coli
LpxC.
passage of distinct LpxC orthologs and is essential for its broad-
spectrum antibiotic activity.22 In this study, we reveal another
key motif, the βa−βb loop of Insert I, that confers resistance to
L-161,240 and BB-78485, suggesting that more flexible
inhibitors would be necessary to overcome the conformational
heterogeneity of both Insert I and Insert II of LpxC.
Conversely, subtle changes of broad-spectrum inhibitors such
as LPC-009 and LPC-011 that exploit the unique conforma-
tional plasticity of the βa−βb loop in EcLpxC and closely
related orthologs can provide a strategy to design highly potent
inhibitors for use against a specific set of Gram-negative
pathogens, such as Shigella, Salmonella, Yersinia, Vibrio, and
Klebsiella.
Another important insight of this work is the trigonal-
bipyramidal geometry of the catalytic zinc coordination in the
EcLpxC/L-161,240 complex. Because square pyramidal geom-
etry has been observed for all other hydroxamate-containing
inhibitors, this zinc coordination state has generally been
regarded as a primary restraint for structure-based design of
LpxC inhibitors. However, it is well-known that many zinc-
dependent enzymes utilize the Berry pseudorotation process,
the isomerization between two penta-coordination geometries,
for transition-state stabilization in catalysis. Thus, our
observation of a trigonal-bipyramidal zinc coordination may
provide important insights into LpxC catalysis and new metal
constraints for computational design of novel inhibitors.
Furthermore, this alternative zinc coordination may affect the
overall position of an inhibitor in the active site and enable the
incorporation of new scaffolds into LpxC-targeting antibiotics.
Implication for Inhibitor Design. Since the discovery of
L-161,240, LpxC inhibitors with a variety of chemical scaffolds
have been reported, each with distinct antibiotic profiles. A vast
majority of these compounds efficiently inhibit the growth of E.
coli and are selective inhibitors for EcLpxC, but not for PaLpxC.
How EcLpxC binds these compounds, including compounds
with two distinct stereospecific configurations at the Cα-
position of the hydroxamate group has remained a mystery.
Here, we shed light on this conundrum by determining the
structures of EcLpxC in complex with L-161,240, BB-78485,
and a newly designed compound LPC-138. Together, these
structures reveal the conformational flexibility of the Insert I
loop of EcLpxC as the molecular basis underlying the
extraordinary susceptibility of EcLpxC to diverse inhibitors.
While LpxC orthologs with a rigid Insert I region are only
capable of interacting with S-enantiomer inhibitors such as
CHIR-090, LpxC orthologs with a flexible Insert I region,
represented by EcLpxC, are able to alter the size and the shape
of the active site in response to inhibitor binding by adopting
different conformations of Insert I. Such conformational
flexibility allows EcLpxC to accommodate a variety of LpxC
inhibitors with distinct scaffolds, including compounds that
contain R-substituted bulky moieties at the Cα-position next to
the hydroxamate group and compounds with double
substitutions at the Cα position.
METHODS
■
Chemical Synthesis. CHIR-090, BB-78485, L-161,240, LPC-009,
and LPC-011 were synthesized as described.12,13,15,29 Synthesis of
LPC-138 (Scheme 1) commenced from methyl (2R,3S,7R,7aS)-7-
methoxy-2,3,7,7a-tetramethyl-5-oxotetrahydro-2H-oxazolo[4,3-b]-
oxazole-3-carboxylate (1), which was generated as reported.30
Treatment of the bicyclic derivative 1 with O-benzylhydroxylamine
hydrochloride in the presence of LiHMDS provided benzyl protected
Our study suggests that the successful design of broad-
spectrum LpxC inhibitors will require navigating differences in
conformations and flexibilities of distinct LpxC orthologs. The
previously reported diacetylene-based inhibitor LPC-009 is an
excellent example of a compound whose elastic scaffold allows
it to overcome structural variations in the Insert II hydrophobic
243
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
Table 2. Data Collection and Refinement Statistics
EcLpxC/L-161,240
R32
EcLpxC/BB-78485
P61
EcLpxC/LPC-138
P61
space group
cell dimensions
a, b, c (Å)
90.0, 90.0, 120.0
117.4, 117.4, 253.7
37998/506964
35.92−2.13 (2.18−2.13)
99.1 (100.0)
90.0, 90.0, 120.0
106.8, 106.8, 52.5
31218/172488
20.98−1.80 (1.86−1.80)
98.1 (98.0)
90.0, 90.0, 120.0
107.5, 107.5, 53.6
23784/245347
29.42−2.01 (2.07−2.01)
99.6 (99.0)
α, β, γ (deg)
reflections (unique/total)
resolution range (Å)
completeness (%)
I/σ
a
39.5 (3.9)
32.8 (3.4)
36.3 (4.3)
R-merge (%)
no. of atoms
protein
10.7 (56.5)
5.8 (44.4)
6.8 (55.6)
4735
44
2358
30
2422
29
inhibitor
water
276
53
310
119
18.3
21.1
166
48
other molecules
R-factor (%)
R-free (%)
21.6
24.7
18.4
21.1
avg B-factor (Å2)
protein
59.74
54.45
36.99
51.36
33.19
24.17
22.81
43.86
53.85
53.61
31.85
55.94
inhibitor
catalytic zinc
water
rmsd from ideal geometry
bond lengths (Å)
bond angles (deg)
Ramachandran plot
favored (%)
0.007
0.705
0.008
1.345
0.011
1.253
95.46
97.65
96.67
allowed (%)
100.00
100.00
100.00
MolProbity
all-atom clashscore
clashscore percentile
11.01
83rd
9.11
79th
5.57
96th
b
a
b
Values in parentheses are for highest-resolution shell. 100th percentile is the best among structures of comparable resolution; 0th percentile is the
worst.
hydroxamate 231 in 81% yield. Acidic hydrolysis of the substrate 2
afforded amino acid hydroxamate 3 in 55% yield. Reaction of the
amino acid derivative 3 with pentafluorophenyl (PFP) ester 5, which
was prepared from corresponding carboxylic acid 4,15 provided amide
6 in 81% yield. Upon subsequent deprotection of the benzyl group in
the substrate 6, using BCl3, the final compound LPC-138 was obtained
in 56% yield.
Crystallization and X-ray Data Collection. Protein samples
were prepared as described previously.15,22 For the L-161,240 complex,
purified EcLpxC was diluted to a final concentration of 6 mg mL−1,
and a 4-fold molar excess of L-161,240 dissolved in DMSO was added
to the diluted protein. The protein and inhibitor were incubated on ice
for 1 h to obtain a homogeneous sample before setting up
crystallization screening. EcLpxC/L-161,240 complex crystals were
produced with the sitting drop vapor diffusion method at 4 °C, in
drops containing 2 μL of the EcLpxC/L-161,240 mixture and 1 μL of
well solution consisting of 0.1 M sodium citrate tribasic dehydrate, pH
5.6, 20% 2-propanol, 20% w/v PEG 4000, and 10 mM DTT.
Diffraction data were collected at the Southeast Regional Collaborative
Access Team (SER-CAT) 22-BM beamline at the Advanced Photon
Source (APS) at Argonne National Laboratory.
collected in-house at 100 K using a Rigaku MicroMax-007 HF rotating
anode generator and R-Axis IV++ detector.
For the LPC-138 complex, initial screening failed to yield any
crystals. After reviewing the previously published crystallization
structure of the EcLpxC/LPC-009,22 we were able to identify an
impurity molecule that contains a threonyl-hydroxamate group
coupled to a phenyl-acetylene group (4-ethynyl-N-((2S,3R)-3-
hydroxy-1-(hydroxyamino)-1-oxobutan-2-yl)benzamide) lying at the
interface of two EcLpxC molecules. Such a compound does not bind
EcLpxC with an appreciable affinity based on ITC studies (data now
shown). Inclusion of such a molecule (1.2 mM) similarly stabilizes the
crystal packing of the EcLpxC/LPC-138 complex, and we were able to
obtain diffraction quality crystals by mixing 10 mg mL−1 protein
solution with a 4-fold molar excess of LPC-138 dissolved in DMSO.
The EcLpxC/LPC-138 crystals were obtained with the sitting drop
vapor diffusion method at 4 °C, in drops containing 1.2 μL of the
complex sample and 0.9 μL of well solution consisting of 0.1 M
HEPES, pH 7.5, 1.7 M lithium sulfate, and 16 mM DTT. The crystals
were cryoprotected with perfluoropolyether (PFO-X175/08) before
flash-freezing. Data set of the EcLpxC/LPC-138 complex cocrystal was
collected at the SER-CAT 22-ID beamline at the Advanced Photon
Source (APS) at Argonne National Laboratory. All the X-ray
diffraction data were processed with HKL2000.32
For the BB-78485 complex, EcLpxC protein was diluted to a final
concentration of 10 mg mL−1, mixed with 4-fold molar excess of BB-
78485 dissolved in DMSO, and incubated on ice for 1 h. The complex
crystals were obtained using the hanging drop vapor diffusion method
at 4 °C, with a reservoir solution containing 3.6 M sodium formate
(Hampton Research), 10% glycerol, and 10 mM DTT. Two
microliters of the complex sample was mixed with an equal amount
of reservoir solution prior to equilibration. Diffraction data were
Model Building and Refinement. Molecular replacement with
the program PHASER33 was used to obtain initial phases for LpxC-
inhibitor structures using the EcLpxC/LPC-009 complex (PDB entry
3P3E) as the search model. Water molecules were added using
PHENIX and verified with COOT.34,35 An additional molecule (ligand
or additive) was found at the packing interface of EcLpxC molecules in
the EcLpxC/BB-78485 complex and in the EcLpxC/LPC-138
244
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
complex, which likely stabilizes the crystal packing of the complex. The
statistics for these structures are shown in Table 2.
(5) Brown, M. F., Reilly, U., Abramite, J. A., Arcari, J. T., Oliver, R.,
Barham, R. A., Che, Y., Chen, J. M., Collantes, E. M., Chung, S. W.,
Desbonnet, C., Doty, J., Doroski, M., Engtrakul, J. J., Harris, T. M.,
Huband, M., Knafels, J. D., Leach, K. L., Liu, S., Marfat, A., Marra, A.,
McElroy, E., Melnick, M., Menard, C. A., Montgomery, J. I., Mullins,
L., Noe, M. C., O’Donnell, J., Penzien, J., Plummer, M. S., Price, L. M.,
Shanmugasundaram, V., Thoma, C., Uccello, D. P., Warmus, J. S., and
Wishka, D. G. (2012) Potent inhibitors of LpxC for the treatment of
Gram-negative infections. J. Med. Chem. 55, 914−923.
(6) Montgomery, J. I., Brown, M. F., Reilly, U., Price, L. M.,
Abramite, J. A., Arcari, J., Barham, R., Che, Y., Chen, J. M., Chung, S.
W., Collantes, E. M., Desbonnet, C., Doroski, M., Doty, J., Engtrakul, J.
J., Harris, T. M., Huband, M., Knafels, J. D., Leach, K. L., Liu, S.,
Marfat, A., McAllister, L., McElroy, E., Menard, C. A., Mitton-Fry, M.,
Mullins, L., Noe, M. C., O’Donnell, J., Oliver, R., Penzien, J., Plummer,
M., Shanmugasundaram, V., Thoma, C., Tomaras, A. P., Uccello, D. P.,
Vaz, A., and Wishka, D. G. (2012) Pyridone methylsulfone
hydroxamate LpxC inhibitors for the treatment of serious gram-
negative infections. J. Med. Chem. 55, 1662−1670.
(7) McAllister, L. A., Montgomery, J. I., Abramite, J. A., Reilly, U.,
Brown, M. F., Chen, J. M., Barham, R. A., Che, Y., Chung, S. W.,
Menard, C. A., Mitton-Fry, M., Mullins, L. M., Noe, M. C., O’Donnell,
J. P., Oliver, R. M., 3rd, Penzien, J. B., Plummer, M., Price, L. M.,
Shanmugasundaram, V., Tomaras, A. P., and Uccello, D. P. (2012)
Heterocyclic methylsulfone hydroxamic acid LpxC inhibitors as Gram-
negative antibacterial agents. Bioorg. Med. Chem. Lett. 22, 6832−6838.
(8) Warmus, J. S., Quinn, C. L., Taylor, C., Murphy, S. T., Johnson,
T. A., Limberakis, C., Ortwine, D., Bronstein, J., Pagano, P., Knafels, J.
D., Lightle, S., Mochalkin, I., Brideau, R., and Podoll, T. (2012)
Structure based design of an in vivo active hydroxamic acid inhibitor of
P. aeruginosa LpxC. Bioorg. Med. Chem. Lett. 22, 2536−2543.
(9) Liang, X., Lee, C. J., Zhao, J., Toone, E. J., and Zhou, P. (2013)
Synthesis, Structure, and Antibiotic Activity of Aryl-Substituted LpxC
Inhibitors. J. Med. Chem. 56, 6954−6966.
(10) Onishi, H. R., Pelak, B. A., Gerckens, L. S., Silver, L. L., Kahan,
F. M., Chen, M. H., Patchett, A. A., Galloway, S. M., Hyland, S. A.,
Anderson, M. S., and Raetz, C. R. H. (1996) Antibacterial agents that
inhibit lipid A biosynthesis. Science 274, 980−982.
(11) Mdluli, K. E., Witte, P. R., Kline, T., Barb, A. W., Erwin, A. L.,
Mansfield, B. E., McClerren, A. L., Pirrung, M. C., Tumey, L. N.,
Warrener, P., Raetz, C. R., and Stover, C. K. (2006) Molecular
validation of LpxC as an antibacterial drug target in Pseudomonas
aeruginosa. Antimicrob. Agents Chemother. 50, 2178−2184.
(12) Clements, J. M., Coignard, F., Johnson, I., Chandler, S., Palan,
S., Waller, A., Wijkmans, J., and Hunter, M. G. (2002) Antibacterial
Activities and Characterization of Novel Inhibitors of LpxC.
Antimicrob. Agents Chemother. 46, 1793−1799.
(13) Anderson, N. H., Bowan, J., Erwin, A., Harvood, E., Kline, T.,
Mdluli, K., Pfister, K. B., Shawar, R., Wagman, A., and Yabannavar, A.
(2004) Antibacterial Agents. International Patent WO 2004/02601 A2
(Chiron, Emeryville, CA).
MIC Tests. MIC tests were carried out as described previously
using a modified NCCLS protocol, which is adapted to 96-well plates
and LB media in the presence of 5% DMSO.22,36 Briefly, 100 μL of
various concentrations of compounds were prepared on a standard 96-
well plate (Corning Costar 3596, flat bottomed with lid, polystyrene
wells), in 2-fold dilution series with a range of 0.0005 to 100 μg/mL.
Bacterial cells, grown to OD600 = 0.6, were diluted 1:100 into LB
medium, and 100 μL of the diluted cells was added into each well and
incubated at 37 °C for 22 h. After the incubation, 50 μL of 1 mg mL−1
[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide solution
(MTT) was added and incubated at 37 °C for another 3 h. The
MIC was determined as the lowest concentration of an antibiotic that
prevented color change (yellow to black).
ASSOCIATED CONTENT
* Supporting Information
■
S
Supplemental Figure S1 and synthesis and characterization of
LPC-138 and intermediates. This material is available free of
charge via the Internet at http://pubs.acs.org.
Accession Codes
Structure factors and coordinates for the EcLpxC/L-161,240,
EcLpxC/BB-78485, and EcLpxC/LPC-138 complexes have
been deposited to the RCSB Protein Data Bank with the
accession codes 4IS9, 4ISA and 4MQY, respectively.
AUTHOR INFORMATION
Corresponding Author
*E-mail: peizhou@biochem.duke.edu. Phone: 919-668-6409.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by National Institutes of Health
Grants AI055588 and GM-51310. Diffraction data for the
EcLpxC/BB-78485 complex were collected and processed at
the Duke University X-ray Crystallography Shared Resource.
Diffraction data for the EcLpxC/L-161,240 and EcLpxC/LPC-
138 complexes were collected at the Southeast Regional
Collaborative Access Team (SER-CAT) 22-ID and 22-BM
beamlines at the Advanced Photon Source, Argonne National
Laboratory, supported by Duke University. Use of the
Advanced Photon Source was supported by the U.S. Depart-
ment of Energy, Office of Science, Office of Basic Energy
Sciences, under Contract No. W-31-109-Eng-38. We would like
to thank M. J. Kuehn for providing Klebsiella pneumoniae and
Vibrio cholerae strains.
(14) McClerren, A. L., Endsley, S., Bowman, J. L., Andersen, N. H.,
Guan, Z., Rudolph, J., and Raetz, C. R. (2005) A slow, tight-binding
inhibitor of the zinc-dependent deacetylase LpxC of lipid A
biosynthesis with antibiotic activity comparable to ciprofloxacin.
Biochemistry 44, 16574−16583.
(15) Liang, X., Lee, C. J., Chen, X., Chung, H. S., Zeng, D., Raetz, C.
R., Li, Y., Zhou, P., and Toone, E. J. (2011) Syntheses, structures and
antibiotic activities of LpxC inhibitors based on the diacetylene
scaffold. Bioorg. Med. Chem. 19, 852−860.
(16) Coggins, B. E., Li, X., McClerren, A. L., Hindsgaul, O., Raetz, C.
R. H., and Zhou, P. (2003) Structure of the LpxC deacetylase with a
bound substrate-analog inhibitor. Nat. Struct. Biol. 10, 645−651.
(17) Coggins, B. E., McClerren, A. L., Jiang, L., Li, X., Rudolph, J.,
Hindsgaul, O., Raetz, C. R. H., and Zhou, P. (2005) Refined solution
structure of the LpxC-TU-514 complex and pKa analysis of an active
site histidine: insights into the mechanism and inhibitor design.
Biochemistry 44, 1114−1126.
REFERENCES
■
(1) Raetz, C. R. H., and Whitfield, C. (2002) Lipopolysaccharide
endotoxins. Annu. Rev. Biochem. 71, 635−700.
(2) Anderson, M. S., Bull, H. G., Galloway, S. M., Kelly, T. M.,
Mohan, S., Radika, K., and Raetz, C. R. H. (1993) UDP-N-
acetylglucosamine acyltransferase of Escherichia coli. The first step of
endotoxin biosynthesis is thermodynamically unfavorable. J. Biol.
Chem. 268, 19858−19865.
(3) Barb, A. W., and Zhou, P. (2008) Mechanism and inhibition of
LpxC: An essential zinc-dependent deacetylase of bacterial lipid A
synthesis. Curr. Pharm. Biotechnol. 9, 9−15.
(4) Zhang, J., Zhang, L., Li, X., and Xu, W. (2012) UDP-3-O-(R-3-
hydroxymyristoyl)-N-acetylglucosamine deacetylase (LpxC) inhibi-
tors: A new class of antibacterial agents. Curr. Med. Chem. 19,
2038−2050.
245
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246

ACS Chemical Biology
Articles
(18) Whittington, D. A., Rusche, K. M., Shin, H., Fierke, C. A., and
Christianson, D. W. (2003) Crystal structure of LpxC, a zinc-
dependent deacetylase essential for endotoxin biosynthesis. Proc. Natl.
Acad. Sci. U.S.A. 100, 8146−8150.
(19) Barb, A. W., Jiang, L., Raetz, C. R., and Zhou, P. (2007)
Structure of the deacetylase LpxC bound to the antibiotic CHIR-090:
Time-dependent inhibition and specificity in ligand binding. Proc. Natl.
Acad. Sci. U.S.A. 104, 18433−18438.
(35) Emsley, P., and Cowtan, K. (2004) Coot: model-building tools
for molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 60,
2126−2132.
(36) Wikler, M., Low, D., Cockerill, F., Sheehan, D., Craig, W.,
Tenover, F., Dudley, M. (2007) Methods for dilution antimicrobial
susceptibility tests for bacteria that grow aerobically:approved
standard-seventh editionClinical Laboratory Standards Institute
(formerly NCCLS), M7-A7.
(20) Cole, K. E., Gattis, S. G., Angell, H. D., Fierke, C. A., and
Christianson, D. W. (2011) Structure of the metal-dependent
deacetylase LpxC from Yersinia enterocolitica complexed with the
potent inhibitor CHIR-090. Biochemistry 50, 258−265.
(21) Mochalkin, I., Knafels, J. D., and Lightle, S. (2008) Crystal
structure of LpxC from Pseudomonas aeruginosa complexed with the
potent BB-78485 inhibitor. Protein Sci. 17, 450−457.
(22) Lee, C. J., Liang, X., Chen, X., Zeng, D., Joo, S. H., Chung, H. S.,
Barb, A. W., Swanson, S. M., Nicholas, R. A., Li, Y., Toone, E. J., Raetz,
C. R., and Zhou, P. (2011) Species-specific and inhibitor-dependent
conformations of LpxC: Implications for antibiotic design. Chem. Biol.
18, 38−47.
(23) Mansoor, U. F., Vitharana, D., Reddy, P. A., Daubaras, D. L.,
McNicholas, P., Orth, P., Black, T., and Siddiqui, M. A. (2011) Design
and synthesis of potent Gram-negative specific LpxC inhibitors. Bioorg.
Med. Chem. Lett. 21, 1155−1161.
(24) Barb, A. W., Jiang, L., Raetz, C. R., and Zhou, P. (2010)
Assignment of 1H, 13C and 15N backbone resonances of Escherichia coli
LpxC bound to L-161,240. Biomol. NMR Assign 4, 37−40.
(25) Gennadios, H. A., Whittington, D. A., Li, X., Fierke, C. A., and
Christianson, D. W. (2006) Mechanistic Inferences from the Binding
of Ligands to LpxC, a Metal-Dependent Deacetylase. Biochemistry 45,
7940−7948.
(26) Dalvie, D., Cosker, T., Boyden, T., Zhou, S., Schroeder, C., and
Potchoiba, M. J. (2008) Metabolism distribution and excretion of a
matrix metalloproteinase-13 inhibitor, 4-[4-(4-fluorophenoxy)-
benzenesulfonylamino]tetrahydropyran-4-carboxylic acid hydroxya-
mide (CP-544439), in rats and dogs: Assessment of the metabolic
profile of CP-544439 in plasma and urine of humans. Drug Metab.
Dispos. 36, 1869−1883.
(27) Jackman, J. E., Fierke, C. A., Tumey, L. N., Pirrung, M.,
Uchiyama, T., Tahir, S. H., Hindsgaul, O., and Raetz, C. R. H. (2000)
Antibacterial agents that target lipid A biosynthesis in gram-negative
bacteria. Inhibition of diverse UDP-3-O-(r-3-hydroxymyristoyl)-n-
acetylglucosamine deacetylases by substrate analogs containing zinc
binding motifs. J. Biol. Chem. 275, 11002−11009.
(28) Fisher, J. F., and Mobashery, S. (2006) Recent advances in
MMP inhibitor design. Cancer Metastasis Rev. 25, 115−136.
(29) Chen, M. H., Steiner, M. G., de Laszlo, S. E., Patchett, A. A.,
Anderson, M. S., Hyland, S. A., Onishi, H. R., Silver, L. L., and Raetz,
C. R. (1999) Carbohydroxamido-oxazolidines: antibacterial agents that
target lipid A biosynthesis. Bioorg. Med. Chem. Lett. 9, 313−318.
́ ́
(30) Aydillo, C., Avenoza, A., Busto, J. H., Jimenez-Oses, G.,
Peregrina, J. M., and Zurbano, M. M. (2008) α-Alkylation versus retro-
O-Michael/γ-alkylation of bicyclic N,O-acetals: An entry to α-
methylthreonine. Tetrahedron: Asymmetry 19, 2829−2834.
(31) Gissot, A., Volonterio, A., and Zanda, M. (2005) One-Step
Synthesis of O-Benzyl Hydroxamates from Unactivated Aliphatic and
Aromatic Esters. J. Org. Chem. 70, 6925−6928.
(32) Otwinowski, Z., and Minor, W. (1997) Processing of X-ray
diffraction data collected in oscillation mode. Methods Enzymol. 276,
307−326.
(33) McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M.
D., Storoni, L. C., and Read, R. J. (2007) Phaser crystallographic
software. J. Appl. Crystallogr. 40, 658−674.
(34) Adams, P. D., Grosse-Kunstleve, R. W., Hung, L. W., Ioerger, T.
R., McCoy, A. J., Moriarty, N. W., Read, R. J., Sacchettini, J. C., Sauter,
N. K., and Terwilliger, T. C. (2002) PHENIX: Building new software
for automated crystallographic structure determination. Acta Crystal-
logr., Sect. D: Biol. Crystallogr. 58, 1948−1954.
246
dx.doi.org/10.1021/cb400067g | ACS Chem. Biol. 2014, 9, 237−246