Discovery of a cyclic boronic acid β-lactamase inhibitor (RPX7009) with utility vs class A serine carbapenemases

Article abstract of DOI:10.1021/acs.jmedchem.5b00127

The increasing dissemination of carbapenemases in Gram-negative bacteria has threatened the clinical usefulness of the β-lactam class of antimicrobials. A program was initiated to discover a new series of serine β-lactamase inhibitors containing a boronic acid pharmacophore, with the goal of finding a potent inhibitor of serine carbapenemase enzymes that are currently compromising the utility of the carbapenem class of antibacterials. Potential lead structures were screened in silico by modeling into the active sites of key serine β-lactamases. Promising candidate molecules were synthesized and evaluated in biochemical and whole-cell assays. Inhibitors were identified with potent inhibition of serine carbapenemases, particularly the Klebsiella pneumoniae carbapenemase (KPC), with no inhibition of mammalian serine proteases. Studies in vitro and in vivo show that RPX7009 (9f) is a broad-spectrum inhibitor, notably restoring the activity of carbapenems against KPC-producing strains. Combined with a carbapenem, 9f is a promising product for the treatment of multidrug resistant Gram-negative bacteria.

Full text of DOI:10.1021/acs.jmedchem.5b00127

Drug Annotation
pubs.acs.org/jmc
Discovery of a Cyclic Boronic Acid β‑Lactamase Inhibitor (RPX7009)
with Utility vs Class A Serine Carbapenemases
Scott J. Hecker,*,† K. Raja Reddy,† Maxim Totrov,‡ Gavin C. Hirst,† Olga Lomovskaya,†
David C. Griffith,† Paula King,† Ruslan Tsivkovski,† Dongxu Sun,† Mojgan Sabet,† Ziad Tarazi,†
Matthew C. Clifton,§ Kateri Atkins,§ Amy Raymond,§ Kristy T. Potts,§ Jan Abendroth,§ Serge H. Boyer,†
Jeffrey S. Loutit,† Elizabeth E. Morgan,† Stephanie Durso,† and Michael N. Dudley†
†Rempex Pharmaceuticals, Inc., A Subsidiary of The Medicines Company, 3033 Science Park Rd., Suite 200, San Diego, California
92121, United States
‡Molsoft L.L.C., 11199 Sorrento Valley Road, San Diego, California 92121, United States
§Beryllium, 3 Preston Court, Bedford, Massachusetts 01730, United States
ABSTRACT: The increasing dissemination of carbapenemases in Gram-negative bacteria has
threatened the clinical usefulness of the β-lactam class of antimicrobials. A program was initiated
to discover a new series of serine β-lactamase inhibitors containing a boronic acid pharmacophore,
with the goal of finding a potent inhibitor of serine carbapenemase enzymes that are currently
compromising the utility of the carbapenem class of antibacterials. Potential lead structures were
screened in silico by modeling into the active sites of key serine β-lactamases. Promising candidate molecules were synthesized
and evaluated in biochemical and whole-cell assays. Inhibitors were identified with potent inhibition of serine carbapenemases,
particularly the Klebsiella pneumoniae carbapenemase (KPC), with no inhibition of mammalian serine proteases. Studies in vitro
and in vivo show that RPX7009 (9f) is a broad-spectrum inhibitor, notably restoring the activity of carbapenems against KPC-
producing strains. Combined with a carbapenem, 9f is a promising product for the treatment of multidrug resistant Gram-
negative bacteria.
treatment of infections due to CRE. More recently, the
diazabicyclooctane inhibitors avibactam (NXL-104)5 and
relebactam (MK-7655)6 have entered clinical development, in
combination with ceftazidime and imipenem, respectively. Both
compounds display a broad spectrum of β-lactamase inhibition
that includes the KPC enzyme.
INTRODUCTION
■
Rapidly rising resistance to multiple antimicrobial agents in
Gram-negative bacteria, commonly related to healthcare-
associated infections, is an emerging public health concern in
U.S. hospitals. While the cephalosporin class of β-lactams was the
mainstay of treatment in the 1980s, the dissemination of
extended-spectrum β-lactamases (ESBLs) over the past 2
decades has dramatically weakened the utility of this class and
brought about a corresponding reliance on the carbapenems.1
Although carbapenems are widely recognized as a safe and
effective class of antimicrobials, carbapenem-resistant Enter-
obacteriaceae (CRE) due to the Klebsiella pneumoniae
carbapenemase (KPC) and other β-lactamases now threatens
the usefulness of all β-lactam antibiotics.2 The Centers for
Disease Control (CDC) considers CRE to be an urgent
antimicrobial resistance threat that now has been detected in
nearly every U.S. state, with an alarming increase in incidence
over the past 5 years.3 The failure to develop antimicrobial agents
to manage CRE threatens to have a catastrophic impact on the
healthcare system.4
A proven strategy to overcome resistance to β-lactam
antibiotics has been to restore their activity by combining them
with an inhibitor of the β-lactamase enzymes responsible for their
degradation. Examples of clinically important β-lactamase
inhibitors (Figure 1) include clavulanic acid (combined with
amoxicillin), sulbactam (with ampicillin), and tazobactam (with
piperacillin). The KPC β-lactamase is poorly inhibited by these
β-lactamase inhibitors, and thus, they have no usefulness in the
Boronic acids have long been explored as inhibitors of serine
proteases.7 Mechanistically, the affinity of boronates for serine
hydrolases is due to the formation of a covalent adduct between
the catalytic serine side chain and the boronate moiety, which
effectively mimics the tetrahedral transition state on the acylation
or deacylation reaction path. β-Lactamase inhibition by boronic
acids such as phenylboronic acid 1 (Figure 2) was first reported
from researchers at Oxford University,8 who noted an earlier
observation9 that borate ions inhibit β-lactamase I. Following
publication of the high-resolution X-ray crystal structure of the
class A RTEM-1 β-lactamase of E. coli,10 the group of Jones
(Toronto) described a rationally designed inhibitor 2 that
displayed an inhibition constant of 110 nM.11,12 Further
structure-guided design yielded two highly potent inhibitors 3
and 4 with inhibition constants of 6 and 13 nM, respectively.13
From the late 1990s to the present, several publications from the
laboratory of B. Shoichet detailed structure-based design efforts
in this class (e.g., compounds 5 and 6).14,15 More recently,
published patent applications16−18 describe analogs (e.g., 7)
Received: January 22, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Figure 1. Currently marketed and promising new β-lactamase inhibitors.
Figure 2. Previously reported boronic acid β-lactamase inhibitors.
Figure 3. Design of cyclic boronate 9.
related to earlier compounds 3 and 4, as well as heterocyclic
variants such as 819 (see Figure 2). Notably, despite a high level of
interest by multiple investigators, there is not a single report
related to determination of efficacy in an animal infection model
with any of the compounds of these classes. Believing that
boronate-containing β-lactamase inhibitors might find utility in
inhibiting the problematic serine carbapenemases of CRE, we set
out to identify a potent inhibitor of the KPC enzyme for use in
combination with a carbapenem.
were proposed and were docked in silico with β-lactamase
enzymes from classes A, C, and D. Modeling of the precovalent
(Michaelis) complex as well as covalent adduct was performed.
High affinity of the precovalent complex is expected to facilitate
rapid complex formation, while favorable inhibitor/enzyme
interactions within the covalent adduct are important to maintain
the complex, since the boronate/serine side chain reaction is
reversible. On the basis of the docking results, the highest ranking
of the proposed inhibitor types was structure 9, from which the
fused benzo ring of 4b has been excised.
DESIGN
Encouragingly, structure 9 exhibited similar favorable
precovalent (free docking) and covalently bound poses in
representative class A and C enzyme active sites. Comparison to
an available X-ray structure of the Michaelis substrate complex in
the active site of AmpC (S64G mutant, Protein Data Bank entry
1KVL)20 showed that the putative inhibitor could capture key
substrate−enzyme interactions (Figure 4). The carboxylate
moiety is extensively coordinated within the subpocket that
coordinates the carboxylate of the substrate β-lactam. The amide
carbonyl forms hydrogen bonds to two hydrogen bond donor
groups. The free hydroxyl of the boronic acid moiety enters the
oxyanion hole, and the lipophilic portion of the ring engages in
hydrophobic interactions with two leucine side chains, again
matching the corresponding interactions of the substrate. While
no similar experimental structure was available for a class A
enzyme/substrate Michaelis complex at the time, our modeling
suggested that the inhibitor could also mimic the substrate
interactions closely. In contrast to the boron-containing “core,”
■
Of the various leads explored by others, we were particularly
interested in the inhibitors reported by Ness et al.13 for their high
potency in inhibiting some class A β-lactamases. These authors
found that installation of a hydroxyl group on the aromatic ring of
compound 4a caused a considerable increase in affinity to the
TEM-1 enzyme. They considered the possibility of formation of
cyclized variant 4b; however, they concluded from examination
of the X-ray crystal structure of this compound bound to TEM-1
that it existed in the acyclic form (4a, Figure 3). We were
intrigued by the possibility of using cyclic boronate ester
formation to constrain potential lead molecules into the
preferred conformation for enzyme complexation. We also
anticipated that cyclic boronates may have better selectivity
toward β-lactamases versus other serine hydrolases that have
linear substrates. The latter enzymes are likely to have more
sterically restricted active sites that can accommodate a linear but
not a cyclic transition state mimetic inhibitor. Several structures
B
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Acidic removal of all protecting groups afforded the cyclic
boronic ester analogs 9a−r.
EVALUATION OF ANALOGS
■
Potentiation of Carbapenems against the KPC-2
Carbapenemase. Initial evaluation of early analogs of 9 quickly
established that in addition to submicromolar activity against
most serine β-lactamases (vide infra), members of this structure
class displayed exceptional activity in inhibition of KPC-type
carbapenemase enzymes. It was quickly recognized that
inhibition of just this one class of β-lactamase would address a
key resistance mechanism that threatens the utility of the
carbapenem class. From the outset, we envisioned that our
ultimate product would be a combination with either
meropenem (because of its excellent safety record and
registration status in most of the world) or biapenem (being
the least effluxed of the carbapenems in Pseudomonas aeruginosa
and having the best stability to the metallo β-lactamase NDM-1).
Medicinal chemistry efforts were focused on understanding
the structure−activity relationships (SARs) of the N-acyl
substituent with respect to potentiation of a carbapenem against
KPC-producing strains. Results of evaluating the ability of
selected inhibitor analogs 9 to potentiate biapenem against a
KPC-producing strain of Klebsiella pneumoniae (KP1004) are
shown in Table 1. Potency is expressed as MPC1, defined as the
minimum concentration of the β-lactamase inhibitor required to
reduce the biapenem MIC from 32 μg/mL (no inhibitor) to 1
μg/mL; in addition to KPC-2, KP1004 also produces the β-
lactamase enzymes SHV-11 and TEM-1. The simple N-acetyl
analog 9a afforded potent activity, with 0.3 μg/mL being
sufficient to reduce the biapenem MIC from 32 to 1 μg/mL. In
most cases, further substitution had only a modest impact on
potency. Whereas phenylacetyl analog 9c was only 2-fold
improved relative to 9a (MPC1 = 0.15 μg/mL), the 2-thienyl
acetyl analog 9f (reminiscent of the marketed cephalosporins
cephalothin and cefoxitin) was 16-fold more potent than 9a
(MPC1 = 0.02 μg/mL). Replacement of the thiophene of 9f by
other heterocycles such as aminothiazole (9g) or pyridyl (9i−k)
resulted in lesser activity, although potency was restored by
incorporation of an amino group in aminopyridyl analog 9l
Figure 4. Model of 9 bound at the active site of class C β-lactamase
(magenta). X-ray structure of cephalothin (core portion, green) bound
to AmpC (enzyme is superimposed) is shown in green. Blue balls
represent hydrogen bonds.
the region of the binding site associated with the N-acyl
substituent was variable across the different serine β-lactamases;
it appeared that this region would benefit from a hydrophobic
substituent but more specific insights could not be gleaned from
the initial model.
SYNTHESIS
■
The synthesis of cyclic boronates 9a−y was accomplished in six
steps in an overall yield of about 30% (Scheme 1). Enantiopure
β-hydroxy ester 10, prepared by lipase-mediated kinetic
resolution of the corresponding racemate,21 was protected as
its silyl ether 11 with TBSCl and imidazole. Regioselective
hydroboration with catalysis by [Ir(COD)Cl]2 gave pinacol
boronate 12 which was converted to the more stable pinanediol
boronate 13. Stereoselective chloromethylation following
Matteson’s protocol at −100 °C afforded the (S)-chloro
homologue 14 as an 85:15 mixture of diastereomers. Stereo-
specific displacement of the chloro group with hexamethyldisi-
lazide followed by in situ acylation gave acylamidoboronate 15.
Scheme 1. Synthesis of Cyclic Boronate Analogs 9a−r
C
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Table 1. Potentiation of Biapenem by Members of Inhibitor Analog Series 9
(MPC1 = 0.04 μg/mL). Various hydroxyl substitutions (9m−o)
and basic groups (9p−r) were tolerated but offered no
advantage. None of these compounds displayed any antibacterial
activity when tested in the absence of a β-lactam. At the
conclusion of this SAR study, compound 9f (RPX7009)22 was
selected for further evaluation as a potential development
candidate because of its high potency as well as precedent for the
thienylacetyl side chain in the marketed cephalosporins
cephalothin and cefoxitin.
cells in combination with cefepime (Table 3).24 Restoration of
cefepime activity against various clinical strains expressing serine
enzymes from classes A, C, and D is observed. In general, we
observed a reasonably good correlation between the biochemical
enzyme inhibition results and cellular activity. For example, 9f
has the lowest potency against the TEM-10 enzyme; likewise, the
lowest extent of potentiation is observed against strains
expressing TEM β-lactamases. Still, in addition to intrinsic
potency, other factors most certainly affect the observed extent of
potentiation, including rate of uptake and active efflux of the
inhibitor as well as the enzyme expression level.
Whole-Cell Activity in Combination with Carbape-
nems. The potent inhibition of KPC enzymes by our series of β-
lactamase inhibitors along with the intrinsic stability of
carbapenems to ESBLs and ampC invited combination of our
inhibitors with these antimicrobials. Table 4 illustrates the ability
of compound 9f to potentiate various carbapenem antibiotics
against clinical isolates of Enterobacteriaceae expressing class A
carbapenemases, including strains producing other class A β-
lactamases associated with resistance to cephalosporins. It should
be noted that these carbapenems are stable to the β-lactamases
represented in the strains in Table 3 and thus are highly active on
their own against these strains. However, when KPC is present,
carbapenem MICs are also affected by mutations in the genes
coding for major porins such as Omp35 and OmpK36, and the
highest MICs are associated with strains that have double porin
mutations (for example, strains KP1084 and KP1087). Even in
these strains, compound 9f is effective in reducing the MICs to
susceptible levels.
FURTHER EVALUATION OF COMPOUND 9f IN
VITRO
■
Inhibition of Nitrocefin Degradation. The spectrum of β-
lactamase inhibition of 9f was initially evaluated by measuring
inhibition of nitrocefin degradation by purified β-lactamase
enzymes.23 This compound features a broad spectrum of
inhibition of β-lactamases, with particularly potent activity
against KPC, CTX-M, SHV, and CMY enzymes (Table 2).
Notably, 9f is considerably more potent than the β-lactamase
inhibitors clavulanic acid and tazobactam against the class A
carbapenemase KPC-2 as well as the class C enzymes P99 and
CMY-2.
Whole-Cell Activity in Combination with Cefepime.
The spectrum of β-lactamase inhibition was evaluated in whole
Table 2. Inhibition of Nitrocefin Degradation (Ki, μM) by
Compound 9f
enzyme
KPC-2
class
9f
clavulanic acid
tazobactam
A
A
A
A
C
C
0.069
0.044
0.029
0.110
0.053
0.099
41.2
1.6
Selectivity of Protease Inhibition. One concern for
inhibitors that interact covalently with β-lactamases was that
they might prove to be inhibitors of mammalian serine proteases.
Therefore, 9f was evaluated against a panel of common
mammalian serine proteases. As shown in Table 5, IC50 values
for 11 different mammalian serine proteases were all ≥1000 μM.
CTX-M-15
SHV-12
TEM-10
P99
0.027
≤0.039
0.020
1106
845
0.001
0.0004
0.005
1.10
CMY-2
0.71
D
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Table 3. MICs to Cefepime Alone or in Combination with 9f
organism
strain
enzyme
cefepime MIC (μg/mL)
cefepime MIC with 4 μg/mL 9f (μg/mL)
Escherichia coli
EC1008
KP1005
KP1009
KP1011
KP1010
EC1009
EC1011
ECL1003
ECL1002
ECL1061
EC1010
KX1001
EA1028
CTX-M-3
CTX-M-14
CTX-M-15
SHV-5
>64
64
>64
64
2
4
Klebsiella pneumoniae
Klebsiella pneumoniae
Klebsiella pneumoniae
Klebsiella pneumoniae
Escherichia coli
4
2
0.25
0.25
4
SHV-12
TEM-10
8
Escherichia coli
TEM-26
8
2
Enterobacter cloacae
Enterobacter cloacae
Enterobacter cloacae
Enterobacter cloacae
Klebsiella oxytoca
TEM, SHV
32
16
>64
>64
4
8
Hyper AmpC expression
0.5
2
Hyper AmpC expression, KPC-3
CMY-6
OXA-2
OXA-30
4
0.5
0.5
Enterobacter aerogenes
>64
Table 4. MIC values (in μg/mL) of Carbapenems in the Presence of Compound 9f (at 4 μg/mL) against Clinical Strains Producing
Serine Carbapenemases
biapenem
meropenem
ertapenem
imipenem
alone with 9f
organism
strain
enzyme
alone
with 9f
alone
with 9f
alone
with 9f
Escherichia coli
EC1007
ECL1058
KX1019
KX1017
KP1004
KP1008
KP1082
KP1087
KP1083
KP1084
KP1088
KPC-3
8
≤0.06
≤0.06
0.25
4
≤0.06
≤0.06
≤0.06
≤0.06
≤0.06
≤0.06
≤0.06
1
8
≤0.06
0.25
8
0.13
Enterobacter cloacae
Klebsiella oxytoca
KPC-3, SHV-11, TEM-1
KPC-2, OXA-2
8
8
32
16
16
32
8
8
0.25
8
4
0.25
4
0.13
Klebsiella oxytoca
KPC-2, OXA-2, SHV-30
KPC-2, TEM-1, SHV-11
KPC-2
4
≤0.06
≤0.06
≤0.06
≤0.06
0.25
4
0.25
8
0.13
Klebsiella pneumoniae
Klebsiella pneumoniae
Klebsiella pneumoniae
Klebsiella pneumoniae
Klebsiella pneumoniae
Klebsiella pneumoniae
Klebsiella pneumoniae
8
8
≤0.06
≤0.06
≤0.06
2
8
≤0.06
≤0.06
0.13
8
4
4
KPC-2, SHV-1
4
4
4
4
KPC-2, CTX-M-15, SHV-11, TEM-1
KPC-3, SHV-1, TEM-1
KPC-3, SHV-11, TEM-1
KPC-3, SHV-11, TEM-1
16
16
64
32
64
16
>64
8
>64
32
>64
16
16
16
64
32
0.25
≤0.06
0.25
≤0.06
0.5
≤0.06
4
0.13
0.25
≤0.06
≤0.06
≤0.06
≤0.06
Table 5. IC50 Values (in μM) for Inhibition of Mammalian
Serine Proteases by 9f
in the presence of 9f. Crystallization conditions are described in
the Experimental Section.25
Consistent with the design of 9f, the structures of CTX-M-15
(Figure 5) and AmpC (Figure 6) complexes clearly show that the
catalytic serine residue of each enzyme is covalently bound to the
boron atom of the inhibitor. The two complexes share several
additional common features. In each, the amide moiety of the
inhibitor is extensively coordinated, donating a hydrogen bond to
the S237/S318 backbone carbonyl and accepting hydrogen
bonds from the side chains of N132/N152 and N104/Q120
(CTX-M-15/AmpC residue numbers, respectively). The
carboxylate of the ligand is also well-coordinated, by T235/
T316, S237/S318, and S130 (CTX-M-15 only) side chain
hydroxyls. In the structure of CTX-M-15, the thiophene moiety
of 9f is bound in a single fixed position, whereas in AmpC it has
an increased range of motion allowing two unique orientations to
be observed. A remarkable difference between the two complexes
is the inversion of the six-membered cyclic boronic acid ester
ring: both conformations are “chair”; however in the CTX-M-15
complex the amide substituent is in the axial orientation and
carboxymethyl is equatorial, while in the AmpC complex the
amide is equatorial and carboxymethyl is axial. The conformation
of the inhibitor in the AmpC complex allows for a more
enzyme
IC50 (μM)
trypsin
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
1000
chymotrypsin
plasmin
thrombin
elastase
urokinase
tissue plasminogen activator (TPA)
chymase
Ddipeptidyl peptidase 7 (DPP7)
neutrophil elastase
cathepsin A
CRYSTALLOGRAPHY
■
To fully understand the mechanism of action and gain insights
that might influence the design of new inhibitors, we sought to
determine the structures of representative members of class A
(CTX-M-15) and class C (AmpC) bound to 9f. The structures of
CTX-M-15 and AmpC were solved to 1.5 and 1.7 Å, respectively,
E
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
strain of the “chair” conformer and the strength of receptor−
ligand interactions likely defines the observed conformation.
EVALUATION OF COMPOUND 9f IN VIVO
■
Pharmacokinetics. The pharmacokinetics of 9f were
studied in the rat at multiple doses (Table 6).26 The systemic
Table 6. Mean Pharmacokinetic Parameters of Compound 9f
in Rats
parameter
100 mg/kg
300 mg/kg
1000 mg/kg
Cmax (mg/L)
231
64
735
239
0.26
0.48
1.26
2595
1037
0.40
0.55
0.97
AUC(0−∞) (h·mg/L)
t1/2 (h)
0.42
0.97
1.60
Vd (L/kg)
Cl (L h−1 kg−1
)
Figure 5. Compound 9f bound to CTX-M-15.
exposure, as shown by the Cmax and AUC values, generally
increased in a dose proportional manner. Overall the parameters
were similar to those of most β-lactam antibiotics, evidencing
high Cmax and AUC, short half-life, and low volume of
distribution.
Efficacy. The enhancement of biapenem and meropenem
activity by compound 9f against a KPC-producing strain of
Klebsiella pneumoniae was studied in a neutropenic mouse lung
infection model (Figure 7).27 In this model, biapenem at a dose
Figure 6. Compound 9f bound to AmpC.
substrate-like interaction with the enzyme, with the hydroxyl
oxygen of the boronate interacting in the oxyanion hole formed
by the backbone NH groups of S64 and S318, while the ring
oxygen forms a hydrogen bond to the hydroxyl of Y150. These
two oxygen atoms likely correspond to the carbonyl oxygen and
the nitrogen of the β-lactam substrate, respectively. On the other
hand, in the CTX-M-15 complex both oxygen atoms are
interacting with the oxyanion hole, which is significantly
expanded. The conformation of the ring is likely dependent on
the amount and configuration of space available for the two
substituents; in CTX-M-15 the amide subpocket is significantly
shallower, making axial configuration of the amide sterically
preferable, while in AmpC the carboxylate subpocket is oriented
more “upwards” from the ring, resulting in axial positioning of
the carboxymethyl group. Notably, the ability of the inhibitor to
switch between two conformations allows it to retain activity
against two different enzyme classes.
We compared the X-ray structures to our earlier models that
informed the design. The model of the covalent class C enzyme/
9f complex was remarkably accurate (rmsd for all heavy atoms,
0.8 Å), with all interactions essentially identical between the X-
ray and the model. Interactions in class A enzyme complex were
also largely correctly reproduced, although our covalent complex
model differed from the X-ray in that the 6-ring was predicted to
assume “boat” conformation upon covalent binding, resulting in
somewhat higher rmsd of 1.2 Å. In this case our precovalent state
model was closer to the experimental (covalent) complex, with
rmsd of 1.0 Å. A delicate balance between the lower internal
Figure 7. Potentiation of biapenem and meropenem activity by
compound 9f against a carbapenem-resistant strain of K. pneumoniae in
the neutropenic mouse lung infection model.
of 50 mg/kg and meropenem at a dose of 100 mg/kg are
ineffective at reducing viable counts in lung tissue. In
combination with biapenem and meropenem, compound 9f
achieved a substantial reduction in viable counts at a dose of 50
mg/kg, attaining over a 2 log colony forming unit (CFU)
reduction.
Safety. At doses up to 1000 mg kg−1 day−1, compound 9f
exhibited no discernible toxicity in the standard battery of safety
pharmacology, repeat-dose toxicology, genotoxicity, and repro-
ductive and developmental toxicity studies.28
Phase 1 Pharmacokinetics. The pharmacokinetics of
compound 9f administered as 3 h infusions at doses of 250,
500, 750, 1000, 1250, and 1500 mg were evaluated in a total of 36
normal volunteers.29 The compound was well-tolerated with no
drug-related adverse events. Plasma concentrations were fit to a
noncompartmental model; exposures increased proportionally
with dose (Figure 8). At the 1 g dose, 9f displayed a Cmax of 22.0
μg/mL and a half-life of 1.23 h, the area under the curve (AUC)
was 100 μg·h/mL, clearance (Cl) was 12.8 L/h, and the steady-
state volume of distribution (Vss) was 21.0 L. These parameters
F
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
system. The data reported in Table 1 are on 85:15 mixtures of
diastereomers; all subsequent data on compound 9f were obtained on
diastereomerically pure material. Other than the presence of the minor
diastereomer, the purity of all tested compounds was ≥95%, as
determined by HPLC with UV detection at 220 nm. Purity and low-
resolution mass spectrometry (MS) data were determined on a
ThermoFinnigan Surveyor HPLC and ThermoFinnigan LCQ Advant-
age instruments equipped with ESI, using an Agilent Eclipse Plus C18
(4.6 mm × 150 mm, 5 μm) column. Mobile phase A was 0.1%
trifluoroacetic acid in water, and mobile phase B was 0.1% trifluoroacetic
acid in acetonitrile. The following conditions were used: a gradient from
10% to 90% B in 6 min and held at 90% B for 1.5 min; UV detection at
220 and 254 nm; a flow rate of 2 mL/min; an injection volume of 10 μL;
full scan, mass range from 100 to 1000 amu. Masses are given as an m/z
Figure 8. Pharmacokinetic profile of 9f in normal volunteers following a
single 3 h intravenous infusion of 1000 mg.
1
ratio. H NMR spectra were acquired in the indicated solvents on a
1
Bruker NMR spectrometer (Avance TM DRX500, 500 MHz for H).
Data are reported as the following: chemical shift, multiplicity (s =
singlet, d = doublet, t = triplet, q = quartet, bs = broad singlet, dd =
doublet of doublet, td = triplet of doublet, m = multiplet), coupling
constants, and number of protons.
are similar to those of most β-lactam antibiotics, which are
typified by short half-lives and low volumes of distribution.
SUMMARY AND CONCLUSIONS
All experiments utilizing animals were conducted under protocols
approved by the Rempex Institutional Animal Care and Use Committee
(IACUC). The phase I clinical study was conducted according to Good
Clinical Practices (GCP) and met the regulations of the Declaration of
Helsinki, the International Conference on Harmonization Good
Clinical Practice (E6) guidelines, the United States Food and Drug
Administration, and other local regulations and legal requirements. An
independent review board and ethics committee approved the study.
(R)-tert-Butyl 3-(tert-Butyldimethylsilyloxy)-pent-4-enoate
(11). A 500 mL round-bottom flask equipped with a condenser under
argon was charged with (R)-tert-butyl 3-hydroxy-pent-4-enoate (38.33
g, 0.22 mol) and CH2Cl2 (220 mL). Imidazole (22.73 g, 0.33 mol) was
added to the solution of alcohol at room temperature. Upon complete
dissolution, TBDMSCl (40.25 g, 0.27 mol) was added at once. A white
precipitate formed and after a few minutes resulted in a thick slurry.
After stirring at room temperature for 1 h, the white slurry was poured
into a flask containing a saturated NaHCO3 solution (250 mL) and
hexanes (250 mL). After stirring at room temperature for 5 min, the
biphasic mixture was poured into a separatory funnel and the layers were
separated. The organic layer was washed with a saturated NaHCO3
solution (250 mL), dried (Na2SO4), filtered, and concentrated to
dryness. The residual tan oil was distilled under reduced pressure (bp
62−54 °C at 0.5 Torr) to give 11 as a colorless oil (59.82 g, 94%); 1H
NMR (DMSO-d6) δ ppm 5.84 (ddd, 1H, J = 17.2, 10.4, 6.0 Hz), 5.21 (dt,
1H, J = 17.2, 1.5 Hz), 5.06 (dt, 1H, J = 10.4, 1.4 Hz), 4.52−4.48 (m, 1H),
2.37 (dd, 1H, J = 14.8, 4.9 Hz), 2.30 (dd, 1H, J = 14.8, 8.0 Hz), 1.39 (s,
9H), 0.84 (s, 9H), 0.04 (s, 3H), 0.02 (s, 3H).
(S)-tert-Butyl 3-(tert-Butyldimethylsilyloxy)-5-(4,4,5,5-
tetramethyl[1,3,2]dioxaborolan-2-yl)pentanoate (12). In a
round-bottom flask containing a solution of olefin 11 (48 g, 0.17 mol)
in CH2Cl2 (560 mL) at room temperature were added bis-diphenyl-
phosphinobutane (4.29 g, 10 mmol) and [Ir(COD)Cl]2 (3.38 g, 5
mmol). Upon complete dissolution of the catalyst, pinacolborane (28.8
mL, 0.20 mol) was added to the orange solution. After stirring at room
temperature for 4 h, the yellow solution was carefully quenched with
MeOH (15 mL). The quenched solution was poured into a separatory
funnel and washed with water (250 mL), dried (Na2SO4), filtered, and
concentrated to dryness. The residual orange oil was purified by column
chromatography, eluting with CH2Cl2 then hexanes/ethyl acetate 7/3 to
give 12 as a colorless oil (66.65 g, 96%); 1H NMR (DMSO-d6) δ ppm
4.00−3.94−4.48 (m, 1H), 2.31 (dd, 1H, J = 14.8, 4.8 Hz), 2.21 (dd, 1H, J
= 14.8, 7.1 Hz), 1.39 (s, 9H), 1.17 (s, 12H), 0.84 (s, 9H), 0.05 (s, 3H),
0.02 (s, 3H).
■
A program was initiated to discover a potent new inhibitor of β-
lactamases for use in treatment of serious Gram-negative
infections. In view of the poor activity of currently marketed β-
lactamase inhibitors against serine carbapenemases, the program
focused on discovery of an inhibitor of these enzymes for
combination with a carbapenem antibiotic. On the basis of the
boronic acid pharmacophore, a new structure was proposed
containing a cyclic α-acylaminoboronic acid. We hypothesized
that intramolecular formation of a boronic ester ring would
constrain the inhibitor into the preferred conformation for
binding, resulting in enhanced potency. Methods for synthesis
were developed, relying on the Matteson protocol for access to
the key α-chloroboronate species with control of stereo-
chemistry.
Evaluation of initial structures showed achievement of broad-
spectrum activity, notably with high potency against the KPC
carbapenemase. Variation of the acyl side chain ensued, with a
focus on optimizing potency against this key enzyme.
Compound 9f emerged as the most potent analog of this series.
Profiling of 9f in combination with currently marketed
carbapenems showed consistent potentiation against strains
with high MICs to the carbapenems alone. In vivo, compound 9f
displays pharmacokinetics in the rat remarkably similar to β-
lactam antibiotics including carbapenems. Efficacy against a
KPC-producing strain in a neutropenic mouse lung infection
model was clearly established. Selectivity for the β-lactamase
target was established by screening against a panel of mammalian
proteases; IC50 values of ≥1000 μM were observed.
Furthermore, in IND-enabling toxicology studies of 9f at doses
up to 1000 mg kg−1 day−1, no adverse effects were observed. In
phase 1 studies in healthy volunteers, compound 9f was well-
tolerated, and human pharmacokinetic parameters were as
expected based on preclinical species. Patient trials in
combination with meropenem are now under way.
EXPERIMENTAL SECTION
■
General Methods. Unless otherwise noted, solvents and reagents
were obtained from commercial suppliers and were used without further
purification. All reactions involving air- or moisture-sensitive reagents
were performed under an argon atmosphere. Thin-layer chromatog-
raphy was performed on Merck silica gel 60 plates coated with 0.25 mm
layer with fluorescent indicator. Components were visualized by UV
light (λ = 254 and 366 nm) and iodine vapors. Column chromatography
was conducted either under medium pressure on silica gel (Merck silica
gel 40−63 μm) or on prepacked silica gel cartridges (Isco) on an Isco
(3S)-tert-Butyl 3-(tert-Butyldimethylsilyloxy)-5-[(2S,6R)-2,9,9-
trimethyl-3,5-dioxa-4-boratricyclo[6.1.1.02,6]decan-4-yl)-
pentanoate (13). In a round-bottom flask containing a solution of
pinacol boronate 12 (22.13 g, 53.4 mmol) in THF (50 mL) was added
(+)-pinanediol (10.0 g, 58,7 mmol) at room temperature. After stirring
at room temperature for 24 h, the clear colorless reaction mixture was
concentrated to dryness and the residual oil was purified by column
G
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Table 7. Enzyme Substrates and Buffers Used in the Study
enzyme
buffer
substrate
N-Bz-R-AMC
substrate concn (μM)
trypsin
50 mM Tris-HCl, pH 8.0, 10 mM CaCl2, 100 mM NaCl
20 mM Tris-HCl, pH 8, 150 mM NaCl, 2.5 mM CaCl2
100 mM Tris-HCl, pH 7.5, 100 mM NaCl
20 mM Tris-HCl, pH 8, 150 mM NaCl, 2.5 mM CaCl2
25 mM Tris-HCl, pH 8.0
200
10
chymotrypsin
plasmin
Suc-AAPF-AMC
H-D-VLK-pNA
Benz-FVR-AMC
Suc-AAPA-pNA
NGK-pNA
200
10
thrombin
elastase
50
urokinase
50 mM Tris HCl, pH 8.5, 38 mM NaCl
100
100
40
tissue plasminogen activator (TPA)
chymase
30 mM Tris-HCl, pH 8.5, 30 mM imidazole, 130 mM NaCl
100 mM Tris, pH 8.0, 2 M NaCl, 0.01% Triton X-100
50 mM Na-Ac, pH 5.8
GK-pNA
Suc-AAPF-AMC
H-Lys-Pro-AMC
MeOSuc-AAVP-AMC
MCA-RPPGFSAFK-Dnp
Suc-AAPF-AMC
Lys-Pro-AMC
dipeptidyl peptidase 7 (DPP7)
neutrophil elastase
100
30
50 mM Tris-HCl, pH 7.5; 1 M NaCl
cathepsin A
25 mM MES, 5 mM DTT, pH 5.5
10
cathepsin G
50 mM Na-Ac, pH 5.8, 2 mM EDTA, 1 mM DTT
10 mM Tris, pH 7.4, 10 mM MgCl2, 0.05% Tween-20
25 mM Tris, pH 7.5, 0.1% BSA
100
100
100
dipeptidyl peptidase 8 (DPP8)
dipeptidyl peptidase 9 (DPP9)
Lys-Pro-AMC
chromatography (0−15% ethyl acetate in hexanes) to give 13 as a
colorless oil (22.7 g, 91%); 1H NMR (DMSO-d6) δ ppm 4.29 (dd, 1H, J
= 8.6, 1.8 Hz), 4.01−3.98 (m, 1H), 2.33 (dd, 1H, J = 14.8, 4.7 Hz), 2.30−
2.25 (m, 1H), 2.22 (dd, 1H, J = 14.8, 7.3 Hz), 2.19−2.13 (m, 1H), 1.95
(t, 1H, J = 5.5 Hz), 1.87−1.83 (m, 1H), 1.7−1.66 (m, 1H), 1.51 (q, 2H, J
= 6.0 Hz), 1.39 (s, 9H), 1.31 (s, 3H), 1.24 (s, 3H), 0.99 (d, 1H, J = 10.7
Hz), 0.83 (s, 9H), 0.81 (s, 3H), 0.78−0.64 (m, 2H), 0.04 (s, 3H), 0.02 (s,
3H).
temperature. After stirring at room temperature overnight, the reaction
mixture was washed with water, then brine, dried (Na2SO4), filtered, and
concentrated under vacuum. The residue was purified by column
chromatography (gradient of 100% dichloromethane to 40% EtOAc/
dichloromethane) to afford amide 15f as a white solid (12.5 g, 70% yield
from 14); 1H NMR (DMSO-d6) δ ppm 9.27 (br s, 1H), 7.40 (dd, 1H, J =
5.1, 1.3 Hz), 6.98−6.93 (m, 2H), 4.05−3.98 (m, 2H), 3.79 (s, 2H), 2.4−
2.35 (m. 1H), 2.31 (dd, 1H, J = 15.1, 5.0 Hz), 2.25 (dd, 1H, J = 15.1, 7.1
Hz), 2.22−2.15 (m, 1H), 2.05−1.95 (m. 1H), 1.83−1.78 (m, 2H),
1.77−1.73 (m, 1H), 1.64 (br d, 1H, J = 13.9 Hz), 1.56−1.4 (m, 4H), 1.39
(s, 9H), 1.34 (d, 1H, J = 9.8 Hz), 1.23 (s, 3H), 1.21 (s, 3H), 0.83 (s, 9H),
0.80 (s, 3H), 0.03 (s, 3H), 0.02 (s, 3H).
(3S,6S)-tert-Butyl 3-(tert-Butyldimethylsilyloxy)-6-chloro-6-
[(2S,6R)-2,9,9-trimethyl-3,5-dioxa-4-boratricyclo[6.1.1.02,6]-
decan-4-yl)hexanoate (14). In a round-bottom flask containing a
solution of CH2Cl2 (6.9 mL, 107.4 mmol) in THF (150 mL) at −95 °C
was added a solution of n-BuLi (23.4 mL, 58.6 mmol) along the sides of
the flask and keeping the internal temperature below −90 °C. Upon
completion of the addition, the reaction mixture was stirred at −95 °C
for 30 min. A solution of pinanediol boronate 13 (22.7 g, 48.8 mmol) in
THF (40 mL) was added to the solution of dichloromethyllithium,
keeping the temperature below −90 °C. Upon completion of the
addition, a solution of ZnCl2 in THF (107 mL, 53.7 mmol) was added at
−95 °C. The reaction mixture was then allowed to warm to room
temperature. After stirring at room temperature overnight, the reaction
mixture was quenched with a saturated solution of NH4Cl, diluted with
ethyl acetate (250 mL), and the layers were separated. The organic layer
was washed with a saturated solution of NH4Cl, dried (Na2SO4),
filtered, and concentrated to dryness. The residual oil was purified by
(3R,6S)-[2-Hydroxy-3-(2-thiophen-2-ylacetylamino)[1,2]oxa-
borinan-6-yl]acetic Acid (9f). To a solution of amide 15f (7.5 g, 12.11
mmol) in 1,4-dioxane (25 mL) was added 25 mL of 3 N HCl. The
reaction mixture was heated at reflux for 90 min, after which the cooled
reaction mixture was diluted with water (25 mL) and extracted with
diethyl ether (2 × 75 mL). The aqueous layer was concentrated to afford
a sticky residue which was azeotroped with MeCN (3 × 100 mL),
dissolved in 20% dioxane−water, and lyophilized to afford a white
powder (3.5 g, 97%). This material consisted of an 85:15 mixture of the
1
desired (3R,6S)-isomer and the (3S,6S)-isomer (by HPLC and H
NMR). This material was suspended in ethyl acetate (60 mL). Water
(15 mL) was added, and most of the compound appeared to go into the
water layer. After sonicating for 5 min, a white precipitate formed. The
solid was collected by filtration, washed with ethyl acetate (2 × 16 mL),
and dried to give 2.28 g (64%) of 9f as a white solid; 1H NMR (CD3OD)
δ ppm 7.35 (dd, 1H, J = 5.3, 1.3 Hz), 7.05 (dd, 1H, J = 3.5, 1.3 Hz), 7.0
(dd, 1H, J = 5.2 3.5 Hz), 4.15−4.05 (m, 1H), 3.98 (s, 2H), 2.61 (br d,
1H, J = 3.5 Hz), 2.37 (dd, 1H, J = 14.9, 7.4 Hz), 2.24 (dd, 1H, J = 14.9,
5.8 Hz), 1.74 (br d, 1H, J = 12.1 Hz), 1.66−1.52 (m, 2H), 1.03 (br q, 1H,
J = 13.1 Hz); ESIMS found for C12H16BNO5S m/z 280 (100%) (M −
H2O)+.
Modeling. Structures of representative enzymes of classes A and C
were extracted from Protein Data Bank (class A, β-lactamase TOHO,
PDB entry 1IYS;30 class C, β-lactamase CMY-2, PDB entry 1RGY31)
and prepared for docking in ICM (Molsoft, San Diego, CA). Docking of
candidate compound structures was performed using ICM Docking
module.32 The lowest-scoring bound conformation was used for analysis
of putative interactions for each candidate structure.
Susceptibility Testing. Clinical strains of Enterobacteriaceae
expressing single or multiple β-lactamases from various classes
(corresponding genes confirmed by PCR and sequence analysis) were
used to evaluate the antibiotic potentiation activity of 9f. MICs were
determined using Clinical and Laboratory Standards Institute (CLSI)
broth microdilution methods as described in CLSI document M07-A9
(2012).33 Cefepime or carbapenem MICs were determined in
combination with 9f at a fixed concentration of 4 μg/mL.
1
column chromatography to give 14 as a colorless oil (22 g, 88%); H
NMR (DMSO-d6) δ ppm 4.41 (dd, 1H, J = 8.7, 1.6 Hz), 4.05 (br p, 1H, J
= 5.9 Hz), 3.56 (dd, 1H, J = 8.1, 5.9 Hz), 2.35−2.25 (m, 3H), 2.21−2.16
(m, 1H), 1.99 (t, 1H, J = 5.5 Hz), 1.95−1.82 (m, 2H), 1.65−1.63 (m,
3H), 1.57−1.49 (m, 1H), 1.39 (s, 9H), 1.34 (s, 3H), 1.25 (s, 3H), 1.07
(d, 1H, J = 10.8 Hz), 0.84 (s, 9H), 0.81 (s, 3H), 0.04 (s, 3H), 0.02 (s,
3H).
(3S,6R)-tert-Butyl 3-(tert-Butyldimethylsilyloxy)-6-(2-thio-
phen-2-yl-acetylamino)-6-[(2S,6R)-2,9,9-trimethyl-3,5-dioxa-4-
boratricyclo[6.1.1.02,6]decan-4-yl)hexanoate (15f). A solution of
chloro intermediate 14 (15 g, 29.12 mmol) in THF (145 mL) was
cooled to −78 °C under nitrogen. To this solution was slowly added a 1
M solution of LiHMDS in THF (29.1 mL, 29.1 mmol). Upon
completion of the addition, the reaction flask was allowed to warm to
room temperature. After stirring at room temperature for 16 h, the
reaction mixture was concentrated and hexanes (300 mL) was added.
The precipitated lithium salts were filtered off, rinsed with hexanes and
the combined filtrates were concentrated to give 18 g of crude
displacement product.
To a stirred solution of 2-thiopheneacetic acid (4.96 g, 34.94 mmol)
in dichloromethane (900 mL) at 0 °C under nitrogen were added EDCI
(8.34 g, 43.68 mmol) and HOBT (4.71 g, 34.94 mmol). After stirring at
0 °C for 30 min, a solution of the hexamethyldisilazane displacement
product (29.1 mmol) in dichloromethane (70 mL) was added, followed
by N-methylmorpholine (6.39 mL, 58.24 mmol). Upon completion of
the addition, the reaction flask was allowed to warm to room
Determination of Ki Values on Inhibition of β-Lactamases. Ki
values of inhibition of β-lactamases purified from overexpressing
H
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
recombinant E. coli strains were determined spectrophotometrically
using nitrocefin (NCF) as reporter substrate. Enzymes were mixed with
inhibitors at varying concentrations in reaction buffer (50 mM Na
phosphate, pH 7.0, 0.1 mg/mL BSA) and incubated for 10 min at room
temperature. 50 μM NCF was added, and substrate cleavage profiles
were recorded at 490 nm every 10 s for 10 min on SpectraMax plate
reader. Ki values were calculated by method of Waley SG.34
affinity tag was removed by applying the digested pool over two 5 mL
Ni2+ charged HiTrap chelating HP columns. The flow through
contained the cleaved protein of interest. The protein was concentrated
for crystallography via centrifugal concentration (Amicon Regenerated
Cellulose, 10 kDa MWCO, Millipore) to 30 mg/mL for both CTX-M-
15 and AmpC.
Effect of 9f on Serine Proteases. All enzymes and substrates were
from commercial sources, and testing was performed according to the
manufacturer’s protocols with some modifications. Briefly, 50 μL of the
diluted enzyme was mixed with 50 μL of an inhibitor at various
concentrations and 50 μL of a corresponding buffer (Table 7). Reaction
mixtures were incubated for 10 min at 37 °C. Subsequently, 50 μL of
corresponding substrate (Table 7) was added and absorbance or
fluorescence was monitored for 30 min on a SpectraMax M2 plate reader
(Molecular Devices). 4-(2-Aminoethyl)benzenesulfonyl fluoride hydro-
chloride (AEBSF) and leupeptin were used as positive controls. Rates of
reaction were calculated and presented relative to “no treatment”
control. IC50 values were calculated based on inhibitor concentration
producing 50% of enzyme inhibition.
Pharmacokinetic Studies in Rats. After acclimation, rats (n = 3/
dose level) were administered single intravenous infusions of 9f at 100,
300, or 1000 mg/kg. Doses were infused over 0.5 h via an indwelling
femoral vein cannula. Plasma (∼0.3 mL) samples were collected from
each rat at designated time points up to 24 h. Blood samples were
centrifuged within 5 min of collection at 12 000g for 5 min to obtain
plasma. The plasma samples were analyzed using an HPLC−MS
method. PK analysis was performed using WinNonlin.
Mouse Efficacy Studies. Swiss Webster mice were rendered
neutropenic by the administration of cyclophosphamide and were
infected under isoflurane anesthesia by intratracheal administration of
strain KP1074 (biapenem MIC = 64 μg/mL; biapenem MIC = 0.25 μg/
mL in the presence of 4 μg/mL 9f; meropenem MIC = 128 μg/mL;
meropenem MIC = 1.0 μg/mL in the presence of 4 μg/mL 9f).
Treatments were given every 2 h by the intraperitoneal route, starting 2
h postinfection. Animals were sacrificed 24 h after the start of treatment,
and the lungs were removed, homogenized, and plated to determine
bacterial counts.
Construct Design and Expression. AmpC from E. cloacae
(GenBank entry X07274.1, residues 21−381) and CMX-M-15 from
K. pneumoniae (GenBank entry AAY46940.1, residues 29−291) were
each cloned using standard restriction enzyme cloning techniques into a
pET28 vector modified with a N-terminal 6× histidine-Smt tag. Ligation
products were transformed into Top10 E. coli cells and plated on 2YT
agar in the presence of 50 μg/mL kanamycin. Single colonies of AmpC
and CMX-M-15 were grown overnight under antibiotic selection in 2YT
media. Cells were harvested, and the plasmid DNA was isolated using a
mini-prep kit. All final constructs were sequence verified. A starter
culture containing 50 μg/mL (final concentration) kanamycin
(Teknova) was inoculated with a single colony and grown for 16 h at
37 °C. This was then transferred to 8 L of terrific broth (Teknova)
containing 50 μg/mL (final concentration) kanamycin and grown to
OD600 = 0.8. Protein expression was induced by adding 1 mM IPTG
(VWR) and grown for 16 h at 37 °C. The cells were harvested by
centrifugation (Beckman) at 5000 rpm for 15 min, and the pellets were
collected and stored at −80 °C.
Table 8. X-ray Data Processing Statistics
protein:
ligand:
AmpC
CTX-M-15
9f
9f
Data Collection and Processing
radiation source
wavelength (Å)
space group
unit cell
ALS 5.0.3
0.9765
APS 21ID-G
0.9786
P22121
P212121
62.6 Å, 69.6 Å, 76.3 Å,
90°, 90°, 90°
44.8 Å, 45.8 Å, 116.9 Å,
90°, 90°, 90°
resolution (Å)
I/σ
50−1.75 (1.8−1.75)
20.0 (4.48)
100 (100)
0.10 (0.46)
34340 (2503)
7.9 (7.3)
Refinement
15.4
50−1.5 (1.54−1.5)
27.6 (7.64)
completeness (%)
Rmerge
98.7 (98.2)
0.03 (0.13)
unique reflections
multiplicity
38972 (2806)
3.9 (3.0)
R
cryst (%)
free (%)
13.2
14.7
0.01
1.34
9.42
R
18.6
rmsd bonds (Å)
0.01
rmsd angles (deg)
mean B-factor (Å2)
1.57
9.67
Validation
98.3
Ramachandran
favored (%)
98.5
0.4
Ramachandran
allowed (%)
0.0
MolProbity score
Clash score
0.83 (100th)
1.08 (99th)
0.84 (99th)
1.21 (100th)
Crystallization and Structure Determination (Table 8).
Compound 9f was dissolved in DMSO as a 10 mM stock solution
and stored at −20 °C. To load the protein prior to crystallization, 5 μL of
ligand was added to 50 μL of protein and mixed gently. These samples
were allowed to incubate at room temperature for approximately 5 min
before gently mixing again and setting up the crystallization experi-
ments. The final concentration of ligand was ∼0.9 mM. Crystal trays for
both AmpC and CTX-M-15 were set with 100 μL wells and crystal drops
containing 0.4 μL of protein (premixed with ligand) and 0.4 μL of
crystallant. Crystallization trays were stored at 18 °C, and crystals
appeared in 3−20 days. AmpC was crystallized at 30 mg/mL in sitting
drop vapor diffusion trays containing a solution comprising 30% PEG
3000, 200 mM NaCl, and 100 mM Tris HCl, pH 7.0. CTX-M-15 was
crystallized at at 30 mg/mL in sitting drop vapor diffusion trays
containing a solution comprising 2.4 M ammonium sulfate and 100 mM
bice NaOH, pH 9.0. Crystals for both proteins were cryoprotected using
20% ethylene glycol and flash frozen with liquid nitrogen. Data for
AmpC were collected at the Advanced Light Source in Berkely, CA
(ALS), and data for CTX-M-15 were collected at the Advanced Photon
Source in Argonne, IL (APS). The data were reduced using XDS35 and
prepared using the Collaborative Computational Project Number 4
(CCP4) suite of programs.36 Phases were determined by molecular
replacement37 using PDB entry 1XX2 as a search model for AmpC and
PDB entry 1IYS as a search model for CTX-M-15. The structures were
refined using Refmac,38 and model building was performed with Coot.39
The structures were validated using MolProbity.40
Protein Purification. Cells were resuspended 1g/4 mL in 25 mM
Tris-HCl, pH 8.0, 200 mM NaCl, 0.5% glycerol, 0.02% CHAPS, 10 mM
imidazole, 50 mM L-arginine, 1 mM TCEP, 125 U of benzonase, 100 mg
of lysozyme, and one complete EDTA free protease inhibitor tablet
(Roche). The cells were lysed via sonication (Misonix) and clarified via
centrifugation at 45 000 rpm for 30 min at 4 °C and filtered with a 0.2
μm bottle top filter. The supernatant was applied to two 5 mL Ni2+
charged HiTrap chelating HP (GE Healthcare) columns, and the
protein eluted with a 500 mM imidazole gradient over 24 column
volumes. The fractions of interest were pooled, and the His-Smt tag was
removed via cleavage with ubiquitin-like-specific protease 1 (Ulp-1)
while dialyzing against 2 L of 25 mM Tris, pH 8.0, 200 mM NaCl, 10
mM imidazole, 1 mM TCEP, 50 mM L-arginine, and 0.25% glycerol
overnight at 4 °C utilizing 10 000 MWCO snakeskin dialysis tubing. The
I
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
(9) Dobozy, O.; Mile, I.; Ferencz, I.; Csanyi, V. Effect of electrolytes on
the activity and iodine sensitivity of penicillinase from B. cereus. Acta
Biochim. Biophys. Acad. Sci. Hung. 1971, 6 (2), 97−105.
(10) Strynadka, N. C.; Adachi, H.; Jensen, S. E.; Johns, K.; Sielecki, A.;
Betzel, C.; Sutoh, K.; James, M. N. Molecular structure of the acyl-
enzyme intermediate in beta-lactam hydrolysis at 1.7 A resolution.
Nature 1992, 359 (6397), 700−705.
(11) Martin, R.; Jones, J. B. Rational design and synthesis of a highly
effective transition state analog inhibitor of the RTEM-1 β-lactamase.
Tetrahedron Lett. 1995, 36 (46), 8399−8402.
ASSOCIATED CONTENT
■
Accession Codes
Protein Data Bank codes are the following: AmpC/9f, 4XUX;
CTX-M-15/9f, 4XUZ.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 858-875-6678. E-mail: scott.hecker@themedco.com.
(12) Strynadka, N. C.; Martin, R.; Jensen, S. E.; Gold, M.; Jones, J. B.
Structure-based design of a potent transition state analogue for TEM-1
beta-lactamase. Nat. Struct. Biol. 1996, 3 (8), 688−695.
Notes
The authors declare no competing financial interest.
(13) Ness, S.; Martin, R.; Kindler, A. M.; Paetzel, M.; Gold, M.; Jensen,
S. E.; Jones, J. B.; Strynadka, N. C. Structure-based design guides the
improved efficacy of deacylation transition state analogue inhibitors of
TEM-1 beta-lactamase. Biochemistry 2000, 39 (18), 5312−5321.
(14) Morandi, F.; Caselli, E.; Morandi, S.; Focia, P. J.; Blazquez, J.;
Shoichet, B. K.; Prati, F. Nanomolar inhibitors of AmpC beta-lactamase.
J. Am. Chem. Soc. 2003, 125 (3), 685−695.
ACKNOWLEDGMENTS
■
Chemistry efforts were supported by Acme Bioscience, Palo Alto,
CA. The Berkeley Center for Structural Biology is supported in
part by the National Institutes of Health, National Institute of
General Medical Sciences, and the Howard Hughes Medical
Institute. The Advanced Light Source is supported by the
Director, Office of Science, Office of Basic Energy Sciences, of
the U.S. Department of Energy under Contract DE-AC02-
05CH11231. Use of the Advanced Photon Source, an Office of
Science User Facility operated for the U.S. Department of Energy
(DOE) Office of Science by Argonne National Laboratory, was
supported by the U.S. DOE under Contract DE-AC02-
06CH11357. Use of the LS-CAT Sector 21 was supported by
the Michigan Economic Development Corporation and the
Michigan Technology Tri-Corridor (Grant 085P1000817).
(15) Wang, X.; Minasov, G.; Blazquez, J.; Caselli, E.; Prati, F.; Shoichet,
B. K. Recognition and resistance in TEM beta-lactamase. Biochemistry
2003, 42 (28), 8434−8444.
(16) Burns, C. J.; Jackson, R. W. Beta-lactamase inhibitors.
WO2009064413A1, 2009.
(17) Burns, C. J.; Jackson, R. W.; Goswami, R.; Xu, H. Beta-lactamase
inhibitors. WO2009064414A1, 2009.
(18) Burns, C. J.; Goswami, R.; Jackson, R. W.; Lessen, T.; Li, W.;
Pevear, D.; Tirunahari, P. K.; Xu, H. Beta-lactamase inhibitors.
WO2010130708A1, 2010.
(19) Prati, F.; Caselli, E. Boronic acid inhibitors of beta-lactamases as
therapeutic agents in treatment of antibiotic-resistant infection diseases.
WO2013053372A1, 2013.
(20) Beadle, B. M.; Trehan, I.; Focia, P. J.; Shoichet, B. K. Structural
milestones in the reaction pathway of an amide hydrolase: substrate,
acyl, and product complexes of cephalothin with AmpC beta-lactamase.
Structure 2002, 10 (3), 413−424.
(21) Vrielynck, S.; Vandewalle, M.; Garcia, A. M.; Mascarenas, J. L.;
Mourino, A. A chemoenzymic synthesis of A-ring key-intermediates for
1α,25-dihydroxyvitamin D3 and analogs. Tetrahedron Lett. 1995, 36
(49), 9023−9026.
(22) (a) Hirst, G.; Reddy, R.; Hecker, S.; Totrov, M.; Griffith, D. C.;
Rodny, O.; Dudley, M. N.; Boyer, S. Cyclic boronic acid ester derivatives
and therapeutic uses thereof. WO2012021455A1, 2012. (b) Hecker, S.
J.; Reddy, K. R.; Totrov, M.; Hirst, G. C.; Lomovskaya, O.; Griffith, D.
C.; King, P.; Tsivkovski, R.; Sun, D.; Sabet, M.; Tarazi, Z.; Dudley, M. N.
Discovery of RPX7009, a broad-spectrum beta-lactamase inhibitor with
utility vs. class A serine carbapenemases. Presented at the 52nd
International Conference on Antimicrobial Agents and Chemotherapy,
San Francisco, CA, 2012; No. F-848.
(23) Tsivkovski, R.; Griffith, D. C.; Hecker, S. J.; Dudley, M. N.;
Lomovskaya, O. Biochemical characterization of the beta-lactamase
inhibitor RPX7009. Presented at the 52nd Interscience Conference on
Antimicrobial Agents and Chemotherapy, San Francisco, CA, 2012; No.
F-849.
(24) King, P.; Sun, D.; Griffith, D. C.; Hecker, S. J.; Dudley, M. N.;
Lomovskaya, O. Microbiological characterization of beta-lactamase
inhibitor RPX7009. Presented at the 52nd Interscience Conference on
Antimicrobial Agents and Chemotherapy, San Francisco, CA, 2012; No.
F-850.
(25) Clifton, M. C.; Raymond, A.; Atkins, K.; Abendroth, J.; Totrov,
M.; Tsivkovski, R.; Hecker, S. J.; Lomovskaya, O. The broad-spectrum
serine beta-lactamase inhibitor RPX7009 binds covalently to active site
serine residues in CTX-M-15 and AmpC. Presented at the 53rd
Interscience Conference on Antimicrobial Agents and Chemotherapy,
Denver, CO, 2013; No. F-1199.
ABBREVIATIONS USED
■
KPC, Klebsiella pneumoniae carbapenemase; ESBL, extended
spectrum β-lactamase; CRE, carbapenem-resistant Enterobacter-
iaceae; CDC, Centers for Disease Control; MIC, minimum
inhibitory concentration; MPC, minimum potentiating concen-
tration; SAR, structure−activity relationship; rmsd, root-mean-
square deviation; AUC, area under the curve; Cl, clearance; Vd,
volume of distribution; Vss, steady-state volume of distribution;
CFU, colony-forming unit; IND, investigational new drug
application
REFERENCES
■
(1) Lynch, J. P.; Clark, N. M.; Zhanel, G. G. Evolution of antimicrobial
resistance among Enterobacteriaceae (focus on extended spectrum β-
lactamases and carbapenemases). Expert Opin. Pharmacother. 2013, 14
(2), 199−210.
(2) Nordmann, P.; Dortet, L.; Poirel, L. Carbapenem resistance in
Enterobacteriaceae: here is the storm. Trends Mol. Med. 2012, 18 (5),
263−272.
(3) Centers for Disease Control (CDC) Web site, 2014.
(4) Watkins, R. R.; Bonomo, R. A. Increasing prevalence of
carbapenem-resistant Enterobacteriaceae and strategies to avert a
looming crisis. Expert Rev. Anti-Infect. Ther. 2013, 11 (6), 543−545.
(5) Zhanel, G. G.; Lawson, C. D.; Adam, H.; Schweizer, F.; Zelenitsky,
S.; Lagace-Wiens, P. R. S.; Denisuik, A.; Rubinstein, E.; Gin, A. S.;
Hoban, D. J.; Lynch, J. P., 3rd; Karlowsky, J. A. Ceftazidime-avibactam: a
novel cephalosporin/β-lactamase inhibitor combination. Drugs 2013, 73
(2), 159−177.
(6) Livermore, D. M.; Warner, M.; Mushtaq, S. Activity of MK-7655
combined with imipenem against Enterobacteriaceae and Pseudomonas
aeruginosa. J. Antimicrob. Chemother. 2013, 68 (10), 2286−2290.
(7) Smoum, R.; Rubinstein, A.; Dembitsky, V. M.; Srebnik, M. Boron
containing compounds as protease inhibitors. Chem. Rev. 2012, 112 (7),
4156−4220.
(26) Griffith, D. C.; Sabet, M.; Tarazi, Z.; Nolan, T.; Lomovskaya, O.;
Hecker, S. J.; Dudley, M. N. Pharmacokinetics of the new beta-lactamase
inhibitor RPX7009 in preclinical species. Presented at the 52nd
(8) Kiener, P. A.; Waley, S. G. Reversible inhibitors of penicillinases.
Biochem. J. 1978, 169 (1), 197−204.
J
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Interscience Conference on Antimicrobial Agents and Chemotherapy,
San Francisco, CA, 2012; No. F-851.
(27) Sabet, M.; Tarazi, Z.; Lomovskaya, O.; Hecker, S. J.; Dudley, M.
N.; Griffith, D. C. In vivo efficacy of the beta-lactamase inhibitor
RPX7009 combined with the carbapenem RPX2003 against KPC-
producing K. pneumoniae. Presented at the 52nd Interscience
Conference on Antimicrobial Agents and Chemotherapy, San
Francisco, CA, 2012; No. F-858.
(28) Griffith, D. C.; Sabet, M.; Tarazi, Z.; Tsivkovski, R.; Lomovskaya,
O.; Hecker, S. J.; Dudley, M. N. Nonclinical toxicology and safety profile
of the beta-lactamase inhibitor RPX7009. Presented at the 52nd
Interscience Conference on Antimicrobial Agents and Chemotherapy,
San Francisco, CA, 2012; No. F-852.
(29) Griffith, D. C.; Loutit, J. S.; Morgan, E. E.; Durso, S.; Dudley, M.
N. A phase 1 study of the safety, tolerability, and pharmacokinetics of the
beta-lactamase inhibitor RPX7009 in healthy adult subjects. Presented at
the 24th European Congress of Clinical Microbiology and Infectious
Diseases, Barcelona, Spain, 2014; No. eP454.
(30) Ibuka, A. S.; Ishii, Y.; Galleni, M.; Ishiguro, M.; Yamaguchi, K.;
Frere, J. M.; Matsuzawa, H.; Sakai, H. Crystal structure of extended-
spectrum beta-lactamase Toho-1: insights into the molecular mecha-
nism for catalytic reaction and substrate specificity expansion.
Biochemistry 2003, 42 (36), 10634−10643.
(31) Nukaga, M.; Kumar, S.; Nukaga, K.; Pratt, R. F.; Knox, J. R.
Hydrolysis of third-generation cephalosporins by class C beta-
lactamases. Structures of a transition state analog of cefotoxamine in
wild-type and extended spectrum enzymes. J. Biol. Chem. 2004, 279
(10), 9344−9352.
(32) Totrov, M.; Abagyan, R. Flexible protein-ligand docking by global
energy optimization in internal coordinates. Proteins 1997, Suppl 1,
215−220.
(33) Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria
That Grow Aerobically, 9th ed.; M07-A9, Vol. 32, No. 2; Clinical and
Laboratory Standards Institute: Wayne, PA, 2012; pp 1−68.
(34) Waley, S. G. A quick method for the determination of inhibition
constants. Biochem. J. 1982, 205 (3), 631−633.
(35) Kabsch, W. XDS. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010,
66 (Part 2), 125−132.
(36) Collaborative Computational Project Number 4. The CCP4
suite: programs for protein crystallography. Acta Crystallogr., Sect. D:
Biol. Crystallogr. 1994, 50 (Part 5), 760−763.
(37) McCoy, A. J. Solving structures of protein complexes by
molecular replacement with Phaser. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2007, 63 (Part 1), 32−41.
(38) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement of
macromolecular structures by the maximum-likelihood method. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53 (Part 3), 240−255.
(39) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features and
development of Coot. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010,
66 (Part 4), 486−501.
(40) Chen, V. B.; Arendall, W. B., 3rd; Headd, J. J.; Keedy, D. A.;
Immormino, R. M.; Kapral, G. J.; Murray, L. W.; Richardson, J. S.;
Richardson, D. C. MolProbity: all-atom structure validation for
macromolecular crystallography. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2010, 66 (Part 1), 12−21.
K
DOI: 10.1021/acs.jmedchem.5b00127
J. Med. Chem. XXXX, XXX, XXX−XXX