β-Lactamases and β-Lactamase Inhibitors in the 21st Century

Review
β-Lactamases and β-Lactamase Inhibitors in

the 21st Century
Catherine L. Tooke † , Philip Hinchliffe † , Eilis C. Bragginton, Charlotte K. Colenso,

Viivi H.A. Hirvonen, Yuiko Takebayashi and James Spencer
School of Cellular and Molecular Medicine, University of Bristol Biomedical Sciences Building, University Walk, Bristol BS8 1TD,
United Kingdom
Correspondence to James Spencer: [email protected].

https://doi.org/10.1016/j.jmb.2019.04.002
Edited by C.G. Dowson
Abstract
The β-lactams retain a central place in the antibacterial armamentarium. In Gram-negative bacteria, β-lactamase
enzymes that hydrolyze the amide bond of the four-membered β-lactam ring are the primary resistance
mechanism, with multiple enzymes disseminating on mobile genetic elements across opportunistic pathogens
such as Enterobacteriaceae (e.g., Escherichia coli) and non-fermenting organisms (e.g., Pseudomonas
aeruginosa). β-Lactamases divide into four classes; the active-site serine β-lactamases (classes A, C and D) and
the zinc-dependent or metallo-β-lactamases (MBLs; class B). Here we review recent advances in mechanistic
understanding of each class, focusing upon how growing numbers of crystal structures, in particular for β-lactam
complexes, and methods such as neutron diffraction and molecular simulations, have improved understanding
of the biochemistry of β-lactam breakdown. A second focus is β-lactamase interactions with carbapenems, as
carbapenem-resistant bacteria are of grave clinical concern and carbapenem-hydrolyzing enzymes such as
KPC (class A) NDM (class B) and OXA-48 (class D) are proliferating worldwide. An overview is provided of
the changing landscape of β-lactamase inhibitors, exemplified by the introduction to the clinic of combinations of
β-lactams with diazabicyclooctanone and cyclic boronate serine β-lactamase inhibitors, and of progress and
strategies toward clinically useful MBL inhibitors. Despite the long history of β-lactamase research, we contend
that issues including continuing unresolved questions around mechanism; opportunities afforded by new
technologies such as serial femtosecond crystallography; the need for new inhibitors, particularly for MBLs; the
likely impact of new β-lactam:inhibitor combinations and the continuing clinical importance of β-lactams mean
that this remains a rewarding research area.
© 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://
creativecommons.org/licenses/by/4.0/).
Introduction antibacterial arsenal. The fact that they remain both the
single most prescribed antibiotic class and the most
important in terms of sales [2] attests to their continuing
β-Lactam antibiotics central role in the treatment of bacterial infections.
β-Lactams, like other antimicrobial classes, have
The discovery of penicillin in 1929 [1] is rightly undergone continuous development since their
recognized as a milestone in the history of medicine, original introduction in order to improve properties
and its introduction to the clinic in the 1940s revolu- such as potency, spectrum of activity, pharmacoki-
tionized our ability to treat bacterial infections. Despite netic and safety profiles and to counter the emer-
enormous progress in the field of antimicrobial chemo- gence of resistance [3]. At present, four main classes
therapy in the more than 70 years after the first use of of β-lactam antimicrobials are in clinical use (Fig. 1).
penicillin, as the centenary of Fleming's work ap- These comprise three types of bicyclic structure: the
proaches, the β-lactams remain the cornerstones of the penicillins (1), in which the four-membered β-lactam
0022-2836/© 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://
creativecommons.org/licenses/by/4.0/). Journal of Molecular Biology (2019) 431, 3472–3500
Review: β-Lactamases and β-Lactamase Inhibitors 3473
Fig. 1. Structures of representative β-lactams and β-lactamase inhibitors. 1. Penicillin scaffold. 2. Cephalosporin scaffold.
3. (1-methyl) Carbapenem scaffold. 4. Hydrolyzed carbapenem (Δ2-pyrroline form). 5. Hydrolyzed carbapenem (Δ1-pyrroline
form). 6. Monobactam scaffold. 7. Clavulanic acid. 8. Avibactam. 9. Relebactam. 10. Vaborbactam. 11. Bicyclic boronate.
ring is fused to a thiazolidine ring; the cephalosporins (carbapenem) by Brown and co-workers [6] in 1976
(2), where the fusion partner is a six-membered and monobactams by Sykes and Imada and their
dihydrothiazine; and the carbapenems (3), where the respective co-workers [7,8] in 1981], each has since
bicyclic system is completed by a five-membered undergone extensive programs of modification to
pyrroline. A fourth class, the monobactams (6) are create arrays of semi-synthetic derivatives.
monocyclic systems. While each class was originally The antibacterial activity of β-lactams was identified
identified as a natural product (penicillin in 1929, by Tipper and Strominger [9] as based on their
cephalosporins by Newton and Abraham [5] (building resemblance to the terminal D-Ala–D-Ala moiety of the
on the work of Brotzu [4]) in 1954, olivanic acid peptidoglycan stem pentapeptide, with the β-lactam
3474 Review: β-Lactamases and β-Lactamase Inhibitors
amide and adjoining carboxylate (or, in the case of resistance [18]. In consequence, this review will largely
monobactams, sulfonic acid) groups serving to mimic focus on Gram-negative bacteria, where β-lactamases
the peptide bond and terminal carboxylate of D-Ala–D- remain the primary β-lactam resistance mechanism,
Ala. Activity then arises from reaction of the β-lactam although it is important to note that in the clinic
ring with the nucleophilic serine of target penicillin- resistance is very often multifactorial and phenotype is
binding proteins (PBPs), leading to opening of the ring driven by combinations of mechanisms that frequently
and irreversible PBP acylation [10] that prevents include permeability modifications and/or efflux pump
formation of peptidoglycan transpeptide cross-links. In upregulation in addition to β-lactamase production. A
consequence, modification that allows for retention of further complication is the increasing prevalence of
antibacterial activity is possible at several positions on strains expressing multiple β-lactamases, often
the β-lactam scaffold: C6 of penicillins, C7 and C3 of encoded upon single multiresistance plasmids (for
cephalosporins, C2 of carbapenems and C3 of recent reviews, see Refs. [19,20]).
monobactams. Key developments arising from such
modifications include introduction of aminopenicillins Classification of β-lactamases
(e.g., ampicillin) that extended antibacterial activity of
penicillins to include Gram-negative bacteria [11]; Identification of growing numbers of β-lactamases,
introduction of methicillin to counter penicillin-resistant coupled with availability of protein, and subsequently
strains of Staphylococcus aureus [12]; and introduction nucleotide, sequence information, established that
of oxyiminocephalosporins [13] (e.g., cefotaxime, these enzymes do not comprise a single homogeneous
ceftazidime) to counter emergence of β-lactamase- group but instead can be subdivided into multiple
mediated resistance in Gram-negative bacteria. classes. Furthermore, as enzyme activity against
different β-lactam substrates began to be reported, it
β-Lactamase mediated β-lactam resistance became apparent that β-lactamases encompass a
range of biochemical properties. With the explosion of
As with other antimicrobial classes, extensive use of sequence information, the number of identified β-
β-lactams has led to the emergence and dissemina- lactamases has undergone a near-exponential in-
tion of resistance. Resistance can occur by multiple crease; at the time of writing, the β-lactamase database
mechanisms, including modification of the target (www.bldb.eu [21]) contains over 4300 such enzymes
(mutation or expression of alternative PBPs), reduc- that have undergone varying degrees of
tion in cell permeability through downregulation of characterization.
porins required for β-lactam entry, over-expression Two systems of classifying this array of enzymes are
of efflux systems and production of modifying or in use: the Bush–Jacoby–Medeiros activity-based
degradative enzymes [14]. In the case of β-lactams, system [22] and the Ambler system [23] based on
enzyme-mediated resistance arises from the activity sequence information. The latter divides β-lactamases
of β-lactamases, enzymes produced by both Gram- into four distinct classes, termed A, B C and D (Fig. 2),
positive and Gram-negative bacteria that hydrolyze identified on the basis of specific sequence motifs but
the β-lactam amide [15]. It is not always appreciated also distinguished by fundamental differences in
that enzymes able to degrade penicillin were identified hydrolytic mechanism. A further fundamental division
in Escherichia coli even before its first use in man [16], is between the three classes (A, C and D) of active-site
although as initial penicillin use focused largely on serine enzymes (seine β-lactamases; SBLs) and class
infections by Gram-positive pathogens, the signifi- B that comprises a heterogeneous group of zinc
cance of this finding was not immediately apparent. metalloenzymes (metallo-β-lactamases, or MBLs).
β-Lactamases, however, quickly became clinically SBLs are distantly related to PBPs [24] (with which
important as resistance in S. aureus arising from they share an invariant Ser–Xaa–Xaa–Lys motif),
production of the PC1 enzyme [17] (encoded by the employ this serine as the reaction nucleophile and
blaZ gene) rapidly compromised penicillin effective- hydrolyze β-lactams via a covalent acylenzyme
ness. This and the successful countering of PC1- intermediate. MBLs instead utilize a metal-activated
producing organisms through introduction of methicillin water nucleophile to drive the hydrolytic reaction.
[12] heralded the beginning of an “arms race” between Although all four classes are widely distributed in
medicinal chemists and bacterial evolution that con- multiple species of clinically significant and environ-
tinues to this day and that has seen the introduction mental bacteria, within each class, a few enzyme
of new β-lactams lead to emergence of new β- families have been spectacularly successful and have
lactamases both by mutation of previously known disseminated widely across the most important bacte-
families and dissemination of genes encoding new rial pathogens. These are generally considered to be
enzymes. Despite the initial importance of PC1- Gram-negative bacteria responsible for opportunistic
mediated penicillin resistance in S. aureus, in Gram- healthcare-associated infections of immunocompro-
positive bacteria, the emergence of strains carrying mised patients, including Enterobacteriaceae such as
PBP alterations has largely superseded β-lactamase E. coli and Klebsiella pneumoniae and non-fermenting
production as the primary mechanism of β-lactam species such as Pseudomonas aeruginosa and
Fig. 2. Overall structure of representative β-lactamases from each class. Crystal structures of β-lactamases from classes
A, B, C and D. Catalytic important residues of serine-β-lactamases (serine 64/70 and lysine 67/73, labeled) are colored
orange, and the metallo-β-lactamase zinc ions are shown as gray spheres. (a) Class A KPC-2 (PDB 5ul8). (b) Class B NDM-1
(PDB 5zgy). (c) Class C AmpC (PDB 1ke4). (d) Class D OXA-48 (PDB 3hbr). Figure (and other structural Figures) generated
with Pymol (www.pymol.org).
Acinetobacter baumannii. Key enzyme families in- actions of β-lactamases with carbapenems, as
clude TEM, SHV, CTX-M and KPC (class A); NDM and these are increasingly “first-choice” agents for
VIM (class B); and CMY and ADC (class C). Class D empiric therapy of healthcare-associated Gram-
enzymes are all termed oxacillinase (OXA); particular negative infections [25]. In the light of the continuing
concern surrounds the OXA-23 and 24/40, and the lack of new antibiotics with anti-Gram-negative
OXA-48 groups, responsible for carbapenem resis- activity, carbapenem resistance is a cause of much
tance in A. baumannii and Enterobacteriaceae, clinical concern [26,27].
respectively. (It is apparent from the above that the
field is considerably complicated by historical but now Mechanistic overview of serine β-lactamases
well-established inconsistencies of nomenclature.)
This review will focus on recent advances in our Although the three classes of SBLs differ substan-
mechanistic understanding of these and other key tially in sequence, all employ an acylation–deacylation
enzymes from each of the four Ambler classes, as mechanism, reminiscent of that of serine proteases, in
derived primarily from the growing availability of which the nucleophilic serine is activated by a general
structural (crystallographic) information describing base, attacks the carbonyl carbon of the scissile β-
how β-lactamases interact with their substrates. lactam amide bond and generates the acylenzyme
We will then briefly consider how new develop- intermediate via a tetrahedral oxyanion transition state
ments in the area of β-lactamase inhibitors promise (Fig. 3). Resolution of this species occurs when a
to enable some β-lactamase-mediated resistance water molecule, the so-called deacylating water, is
to be overcome. A particular interest is the inter- activated by a general base for hydrolysis of the
Fig. 3. Mechanistic overview of serine β-lactamases. Figure shows hydrolysis of generic penicillin substrate. (a)
General base B1 activates Ser for nucleophilic attack on the amide carbonyl carbon (C7) generating covalent acylenzyme
(c) via tetrahedral oxyanionic acylation transition state (b). General base B2 activates incoming deacylating water molecule
DW for nucleophilic attack on the acylenzyme carbonyl liberating penicilloate product (e) via tetrahedral deacylation
transition state (d). For clarity, details of proton transfers to N4 and Ser are omitted. Note that the identities of bases B1 and
B2 vary between β-lactamase classes.
acylenzyme and liberation of the degraded product. spectrum of activity as new substrates are introduced to
The three Ambler classes, however, diverge mecha- the clinic. Thus, acquisition of point mutations that
nistically, in particular with respect to the identities of, enable TEM and SHV to hydrolyze oxyiminocephalos-
and precise interactions made by, the necessary porins such as cefotaxime and ceftazidime has
general base(s). In classes A and C, this represents a generated the so-called “extended-spectrum” pheno-
long-standing source of debate within the field. type (extended-spectrum beta-lactamases, or ESBLs);
CTX-M enzymes, which possess some inherent
activity against some such substrates, also accumulate
Class A β-lactamases mutations to extend their activity and provide resistance
to an enlarged range of β-lactams. ESBLs now
The class A enzymes comprise the most widely significantly threaten the continued effectiveness of
distributed and intensively studied of all β-lactamases. cephalosporins in a number of clinical contexts [32].
Notable class A enzymes include PC1, responsible The extensive literature on mutational adaptation in
for penicillin failure in S. aureus; TEM [28] (named class A enzymes has been recently reviewed by
for patient Temoneira), the first plasmid-borne β- Palzkill [33]; hence, this will not be considered further
lactamase identified in Gram-negative bacteria and here. We instead focus on two aspects of class A
active against aminopenicillins and early cephalospo- biochemistry—the continuing quest for understanding
rins; SHV (sulfhydryl variant, an enzyme with similar of the acylation mechanism and the basis for their wide
activity to TEM, originally identified on the chromosome variation in activity against carbapenem substrates.
of K. pneumoniae [29] and subsequently mobilized
onto plasmids); CTX-M [30] (cefotaximase, an enzyme The acylation mechanism of class A β-lactamases
inherently active against the oxyiminocephalosporin
cefotaxime and that has rapidly disseminated world- Despite the wealth of accumulated information, the
wide); and KPC [31] (K. pneumoniae carbapenemase). acylation mechanism of the class A enzymes remains
Key to the success of enzymes such as the TEM, SHV a source of debate and controversy. Two main
and CTX-M families has been their dissemination on proposals (Fig. 4) have been put forward, differing in
plasmids and other mobile genetic elements across a the mechanism by which the serine nucleophile is
range of Gram-negative pathogens, particularly the activated for attack upon the β-lactam carbonyl carbon
Enterobacteriaceae, and their ability to expand their and, specifically, the identity of the residue acting as
Fig. 4. (A) alternative acylation mechanisms for class A β-lactamases. (a) Apoprotein active site shows Lys73
protonated and Glu166 deprotonated. Ser70 may be activated for nucleophilic attack upon the β-lactam carbonyl by
Glu166 via a water molecule (b). Alternatively, substrate binding reorganizes protonation in the active site such that neutral
Lys73 activates Ser70 (c). The transition state (d) resolves to the acylenzyme intermediate (e) by protonation of the amide
nitrogen from Ser130.
the general base in this process. Two conserved sources that, when coupled with rapid mixing or other
residues, Lys73 (part of the SBL SXXK motif) and reaction initiation technologies, offer the prospect of
Glu166, have been proposed to fulfill this function, with applying serial femtosecond crystallography (SFX)
both proposals supported by experimental evidence techniques [34] to capture mechanistically relevant but
arising from crystallographic studies of enzyme transiently populated species. The inherently favorable
complexes, high-level computational simulations diffraction characteristics of many β-lactamase crys-
and biochemical investigations of enzyme mutants. tals make these appealing initial targets for application
Recent contributions in this area stem primarily from a of SFX to studies of enzyme mechanism, the first
growing availability of high-resolution crystal struc- results of such studies are starting to emerge [35–37].
tures of both free enzymes and covalent complexes, Recent evidence supporting involvement of Glu166
providing direct evidence of protonation states in the as the general base for acylation is provided by
β-lactamase active site. These include an increasing both ultrahigh-resolution X‐ray and neutron diffraction
number of neutron diffraction studies, made possible studies of the apo-enzymes of CTX-M-9 [38] and
by availability of more powerful neutron sources Toho-1 [39] (also known as CTX-M-44), which
capable of delivering improved diffraction from macro- consistently depict active-site hydrogen bonding
molecular crystals. An attractive prospect is presented networks connecting Ser70 and Glu166 via a con-
by the growing availability of X-ray free electron laser served water molecule and identify Lys73 as present
in the protonated form and thus unable to activate inhibition of PBPs, and the ability to effectively acylate,
Ser70. Moreover, quantum mechanics/molecular and form long-lived acylenzymes with, the majority of
mechanics (QM/MM) calculations on TEM-1 [40–42] serine β-lactamases. This inhibitory activity encom-
support Glu166 as the general base, on the basis that passes the most widely disseminated class A
calculated energetic barriers correlate with values enzymes of the TEM, SHV and CTX-M families.
derived from experiment and stable species with However, new β-lactamases able to hydrolyze carba-
Lys73 neutral and Glu166 protonated were not penems have emerged with increasing carbapenem
observed during the simulations. use; in the case of class A, the most important of these
The ability of Lys73 to act as general base is are the KPC enzymes [54], which now have worldwide
supported by the generation of stable acylenzyme distribution [31]. The mechanistic basis for the activity
complexes, albeit at reduced rates, by mutants at of these enzymes against carbapenems, which other
Glu166 [43,44], consistent with the widely accepted class A β-lactamases enzymes lack, is attracting
role of this residue as the general base in the continued attention.
deacylation reaction [45,46], but implying the exis- Carbapenems are distinguished from other β-lactams
tence of a viable alternative acylation mechanism. In by the ability of the pyrroline ring to tautomerize between
contrast, alanine mutants of Lys73 have been used to the Δ2 and Δ1 forms (Fig. 1, 4, 5), characterized by
trap non-covalent Michaelis complexes [47], indicative migration of the double bond from C2_C3 to C3_N.
of an essential role for Lys73 in acylation, and high- Knowles and co-workers [55,56] identified that, on
resolution X‐ray [48] and neutron [49] structures of reaction with TEM β-lactamase, bound carbapenem,
some acylenzyme complexes of Glu166 mutants initially in the Δ2 tautomer, could either deacylate or
show Lys73 to be deprotonated, suggesting that alternatively isomerize to the more inert Δ1 form that
binding of substrate can reduce its pKa sufficiently accumulated as a stable enzyme-bound species.
for this to occur. A combined X‐ray/neutron study of a Subsequent work [57] identified protonation at C3,
preacylation complex, generated using a CTX-M-44 catalyzed by the Arg244 residue present in TEM-1 and
(Toho-1) Ser70Gly mutant, also reaches this conclu- many other class A enzymes, as central to this process.
sion and is supported by QM/MM calculations of Crystal structures of acylenzyme complexes generated
energy barriers for proton transfer from Lys73 to by exposure of class A enzymes to carbapenems
Glu166 via Ser70, which reduce in models of the (Fig. 5) have in due course described binding modes of
substrate complex compared to the apoenzyme [50]. both the Δ2 and Δ1 tautomers, but less expectedly also
Similarly, ab initio QM/MM calculations [51] have also revealed an ability, in the TEM [58] and SHV [59]
supported existence of a Lys73-dependent acylation enzymes, of the acylenzyme carbonyl to migrate
pathway in addition to a Glu166-dependent mecha- between positions within and outside the oxyanion
nism. An alternative approach utilizing an analogue of hole formed by the backbone amides of residues 70 and
the acylation transition state [52] also indicates proton 237. In the case of SHV-1, both of these conformations
transfers induced by ligand binding and postulates a are observed in a single high-resolution structure
low-barrier hydrogen bond to Lys73 as an important (Fig. 5a). Together with Raman crystallographic studies
contributor to stabilizing the Ser70 nucleophile. of SHV-1:carbapenem complexes [60], that exploit the
Despite these and other advances, however, uncer- distinctive signature of the β-lactam acylenzyme
tainties remain: the necessary use of enzyme mutants carbonyl to resolve it from bulk protein background,
or substrate analogues in structural characterization of and establish that an initial dual (Δ1/Δ2) population
transient species may alter active-site hydrogen resolves to the Δ1 form, these provide a picture of
bonding networks, while QM/MM simulations remain carbapenem binding in which conformational flexibility
sensitive to factors including the starting structure/ of the acylenzyme prevents facile deacylation, enabling
initial model, basis set and QM method used, with isomerization to and accumulation of the stable Δ1
consequences for the accuracy of predicted values species. An additional key observation is that the 6α-
for energy barriers. Thus, despite substantial recent hydroxyethyl substituent of carbapenems adopts posi-
advances, the acylation mechanism for class A β- tions that enable interaction with elements of the active
lactamases remains to be fully resolved. site important to deacylation, Glu166 in the case of the
Mycobacterium tuberculosis BlaC enzyme [61] and the
Carbapenem hydrolysis by class A β-lactamases deacylating water molecule in SHV-1 complexes [59]. It
is then highly likely that such interactions are able to
An area of investigation of class A β-lactamases of change the pattern of hydrogen bonding in the active
much current interest and both mechanistic and site, deactivating the complex for deacylation and
clinical relevance concerns the interactions of these instead favoring the competing isomerization reaction.
enzymes with carbapenems. Carbapenems, intro- A central question is then how class A β-lactamases
duced to the clinic in 1985 in the form of imipenem [53] with carbapenemase activity, such as KPC, avoid
(in combination with the renal dipeptidase inhibitor inhibitory interactions with carbapenems. As structural
cilastatin required to avoid breakdown in vivo), information on such enzymes became available
possess both potent antimicrobial activity through [62–64], it was recognized that they are distinguished
Fig. 5. Carbapenem complexes of class A β-lactamases. Close-up views of class A: carbapenem complexes,
carbapenem acylenzymes are covalently attached to Ser70. Note hydrogen bonding interactions of the acylenzyme
carbonyl with the oxyanion hole formed by the backbone amides of residues 70 (nucleophilic Ser) and 237. (a) SHV-1:
meropenem acylenzyme (PDB:2ZD8). Meropenem is modeled in two conformers with the carbonyl oxygen either pointing
into the oxyanion hole (interacting with Ala237 and Ser70) or pointing away from the oxyanion hole (interacting with
Ser130). In both models, the meropenem R group is not modeled due to disorder. (b) KPC-2:faropenem non-covalent
product complex (5UJ4). (c) SFC-1:meropenem acylenzyme (4EV4). (d) GES-5:imipenem acylenzyme (4H8R). Important
residues for catalysis and substrate binding (labeled) are represented as sticks, and the deacylating water is displayed as
a red sphere. Distances (Å) are displayed as dashed lines.
from other class A β-lactamases by possession of a preclude deactivating contacts with the deacylating
disulfide bridge between residues 69 and 238, and water molecule. The expanded active site of class A
that their active sites are to some degree expanded carbapenemases may then facilitate this interaction by
compared to those of enzymes inhibited by carba- enabling the necessary rotation about the C6\C7
penems [64], although there are not otherwise gross bond, a conclusion that is supported by molecular
changes to the active site. Moreover, the importance dynamics (MD) simulations of SFC-1. In the GES
of the disulfide varies between enzyme families; it enzymes, where carbapenemase activity is restricted
appears essential to activity of SME-1 against any β- to specific point variants of what are otherwise
lactam substrate [65], and its disruption by mutation extended-spectrum enzymes and remains relatively
severely destabilizes the NMC-A and SFC-1 en- modest, but still clinically relevant, an additional
zymes, but a Cys69Gly mutant of GES-5 is destabi- consideration is the longevity of hydrogen-bonding
lized but remains catalytically competent [66]. These interactions involving Glu166. In GES variants
authors conclude that the disulfide is structurally (e.g., GES-5) that hydrolyze carbapenems, persis-
important, but is not an essential and defining tent H-bonds to Ser170 are proposed, based upon MD
requirement for carbapenemase activity. simulations, to maintain Glu166 in a position and
More recently, crystal structures have become orientation necessary for deacylation. GES enzymes
available for carbapenem acylenzymes of the SFC lacking Ser170, such as GES-1, are unable to
[67] and GES [68] enzymes (Fig. 5, c, d), providing hydrolyze carbapenems.
direct observation of carbapenem binding to enzymes KPC enzymes are by far the most clinically
with carbapenemase activity. Compared to carbapen- significant of the class A carbapenemases, due to
em acylenzymes of other class A β-lactamases, these their global distribution in species, principally K.
reveal bound carbapenem exclusively in the Δ2, rather pneumoniae, responsible for the majority of opportu-
than Δ1, form and show that the carbapenem 6α- nistic infections of compromised patients in the
hydroxyethyl group is oriented to make stable healthcare setting [31]. Although structural information
hydrogen-bonding interactions, with Asn132, that on interactions of KPC with β-lactams remains limited,
numerous studies have utilized both natural variants between two potentially conflicting requirements:
[69] and laboratory mutants [70–72] to investigate the the need to rotate the carbapenem 6α-hydroxyethyl
involvement of specific residues in KPC in activity substituent away from the deacylating water and to
against a range of substrates, including carbapenems. constrain the acylenzyme carbonyl within the
While a carbapenem acylenzyme complex for KPC oxyanion hole and prevent access to “non-permis-
remains elusive, the structure of a complex with the sive” conformations. It is likely that multiple struc-
hydrolysis product of the penem faropenem (Fig. 5b tural features, including the conformations of the
[73]) is consistent with many of these findings, Trp105 and the omega loops, and the Cys69–
implicating residues such as Pro104, Thr237 and Cys238 disulfide bridge, provide the necessary
Val240 in substrate binding in the context of an active constraints upon the orientation of bound sub-
site that is overall more hydrophobic than that of strates. In this context, it is notable that clinical
enzymes such as CTX-M. Nevertheless, high-level KPC point variants with increased activity against
computational approaches, typically involving many the (notably bulky) oxyiminocephalosporin ceftazi-
microseconds of accumulated simulations, have in- dime exhibit reduced activity against carbapenems
creasingly been applied to study dynamics and [69], indicating a possible association between
conformational transitions in KPC with the aim of active-site expansion and reduced carbapenemase
establishing a basis for carbapenemase activity. A activity.
common finding is that movement of Trp105, a residue
that lies on one side of the active site cleft and that
adopts multiple conformations in crystal structures of Class B β-lactamases
the unliganded enzyme [73], defines transitions be-
tween states that differ with respect to their compe- The class B, zinc-dependent, MBLs are unrelated to
tence for β-lactam turnover. In simulations of known PBPs and instead are members of a large,
carbapenem acylenzyme complexes, committor anal- ancient and widely distributed metallohydrolase super-
ysis identifies the conformation of the Trp105 loop as family. These enzymes are predominantly found in
one factor differentiating so-called “permissive” confor- prokaryotes and act upon a range of substrates
mations (in which the β-lactam acylenzyme carbonyl extending from various small-molecules (thiolesters,
occupies the oxyanion hole and the deacylating water phosphonates) though to nucleic acids [77,78], and
is present) from “non-permissive” conformations with in higher eukaryotes are involved in DNA repair and
the carbonyl outside the oxyanion hole and the RNA processing pathways [79]. MBLs are distin-
deacylating water absent [74]. In multiple-replica guished by a His-Xaa-His-Xaa-Asp motif that forms a
parallel tempering metadynamics simulations, that metal center located at the interface of the two β-sheets
access timescales not available through conventional that comprise the core of the protein. It is proposed that
methodologies, unliganded KPC is shown to undergo the fold originally arose by duplication of an ancestral
motions of active site regions, including Trp105 and the αβ domain. The identities of the residues that make up
omega-loop containing Glu166, that are related to this center and its stoichiometry and architecture define
hydrophobic networks in the individual α and α/β three distinct MBL subfamilies (termed B1, B2 and B3).
domains and that direct interactions with substrates B1 enzymes, the most clinically important, possess a
[75]. Experimental studies of site-directed mutants binuclear zinc center comprising tri-His (termed Zn1)
support these findings by identifying Trp105 as a and Cys-His-Asp (Zn2) metal sites; in B3 enzymes, the
determinant of KPC activity [71]. QM/MM investiga- Zn2-coordinating Cys is replaced by an additional His
tions of carbapenem hydrolysis by class A enzymes, residue; and in B2 enzymes, the first His of the defining
however, remain in their infancy, in part as a result of motif is replaced by Asn, resulting in a mononuclear
the relative paucity of crystal structures on which to enzyme in which only the Zn2 site is occupied [80,81].
base starting models. However, a comparison of In the binuclear enzymes (Fig. 6), metal coordination is
carbapenem deacylation by a range of class A β- completed by water molecules, one of which (the so-
lactamases [76] demonstrated that QM/MM ap- called “bridging” water) connects the two metal ions,
proaches could successfully discriminate between and the other (sometimes termed the “apical” water) is
enzymes with carbapenemase activity and acylen- bound to Zn2. The geometry of Zn1 is normally
zyme systems resistant to deacylation. considered to be tetrahedral; Zn2 has been described
When considered together with studies on other as trigonal bipyramidal or distorted square pyramidal.
class A carbapenemase systems such as those With the possible exception of the IMP enzymes [82],
described above, an accumulating body of evidence MBLs are isolated as zinc enzymes, although they are
suggests that activity against carbapenems im- active when reconstituted as other metallated forms
poses specific spatial requirements on the class A (e.g., see Refs. [83–85]), and other superfamily
β-lactamase active site that are necessary to orient members utilize a range of other metal ions [77,79].
bound carbapenems for hydrolysis, but may be As β-lactamases, the MBLs are notable for their
more stringent than those associated with other β- exceptionally broad spectrum of activity, which for the
lactam substrates. Specifically, a balance may exist binuclear enzymes encompasses penicillins,
Fig. 6. Active-site architecture of class B metallo-β-lactamases and their carbapenem complexes. Close-up views of
native metallo-β-lactamase active sites (left) and hydrolyzed carbapenem products (labeled) bound in the active sites
(right). (a) B1 NDM-1 (PDBs 5zgx and 5ypk; left and right, respectively). (b) B1 VIM-1 (5n5g and 5n5i). (c) B3 SMB-1 (3vpe
and 5b1u).
cephalosporins and carbapenems, although they show Cys, as found in the active site of B1 enzymes, is
little or no activity against monobactams [86,87]. unusual in catalytic, as opposed to structural, zinc
Compared to other superfamily members and indeed centers [88], as is the absence of any metal-bridging
other classes of binuclear metallohydrolases, MBLs protein group (e.g., a carboxylate side chain), although
exhibit several distinctive features. The presence of these are present in other members of the wider MBL
superfamily [89]. Understanding of the relationship larly P. aeruginosa [93], are disseminating widely on
between the different MBL classes and the wider mobile genetic elements. Indeed, the extent of NDM-1
enzyme superfamily continues to evolve; sequence dissemination on a variety of genetic supports has led
comparisons indicate that the subclass B3 MBLs are to its description as an “epidemic gene” and recognition
most closely related to other members of the super- as a major contributor to carbapenem failure in resistant
family; and it has been proposed that these are Enterobacteriaceae. In contrast, subclass B2 and B3
sufficiently distinct from the B1 and B2 MBL subclasses enzymes are, with a few exceptions where association
for them to be considered as a separate β-lactamase of B3 enzymes with mobile genetic elements has been
group [90]. For many years from their identification in identified [94,95], restricted to the chromosomes of
1966 [91], MBLs were considered as primarily of Gram-negative bacteria. The L1 enzyme of Stenotro-
biochemical, rather than clinical, interest, but this phomonas maltophilia [96], a highly resistant organism
situation has changed with the realization that several able to infect only severely compromised patients [97],
of the subclass B1 enzymes, notably NDM (New Delhi is probably the most clinically relevant of these.
MBL) in Enterobacteriaceae [92] and VIM (Verona The catalytic mechanism of MBLs (Fig. 7) has now
IMipenemase) in non-fermenting organisms, particu- been the subject of extensive investigation, to a
Fig. 7. Possible mechanism of carbapenem hydrolysis by binuclear class B metallo-β-lactamases. (a) Substrate
binding displaces Zn-bridging hydroxide to a terminal position (b) enabling attack upon the scissile carbonyl. (c) Anionic
intermediate with negative charge delocalized around the pyrroline ring resolves either (d) by protonation at C2 by bulk
water generating (e) the Δ1 pyrroline or (f) at N4 by incoming water at the bridging position generating (g) the Δ2-pyrroline
product. (Note that as shown Zn-bound “apical” water is displaced by substrate; some proposals [115] show this as moving
to the bridging position on substrate binding.)
great extent facilitated by the ability to substitute zinc spectroscopy, together with QM/MM and MD simula-
for more spectroscopically active metal ions such as tions, demonstrated that MBLs of all three classes
cobalt or cadmium, while retaining hydrolytic activity employ a common, branched, mechanism for carba-
(for review, see Ref. [98]). Against this, the diversity penem hydrolysis (Fig. 7), with bifurcation arising from
(sequence and structural variation) that exists within the possibility of protonating the anionic species either
even the same MBL subclass complicates the process at the amide N (with proton donation from metal-
of identifying the most salient features of β-lactam bound water at the “bridging” position) or at C2 from
breakdown that might constitute a common mecha- bulk solvent [115]. The role of Zn1 is then primarily to
nism. A further issue is the long-standing debate over deliver the nucleophile, as evidenced by the work of
the physiological importance of the binuclear center in the Page group, who demonstrated that hydrolytic
B1 MBLs. This arises through the differing affinities of activity of a series of metal-substituted MBL deriva-
the Zn1 and Zn2 sites (e.g., see Refs. [99,100]) that tives correlates with the position of the relevant metal
permit isolation of forms of the enzymes in which only ion in the activity series [83].
the Zn1 site is occupied [101,102], and is complicated Accumulation of structural (crystallographic) ev-
by the susceptibility of the Zn2-binding Cys residue to idence in support of mechanistic proposals for
oxidation [103–106]. The consensus view now holds MBLs has been impeded by the absence of
that the binuclear form of the enzyme is the most covalent species along the reaction pathway and
catalytically active and predominates under physio- consequent difficulty in trapping enzyme-bound
logical conditions [107]. In consequence, we here states. However, several publications [116–125]
focus on mechanistic studies of the binuclear forms of now describe structures of enzyme-bound com-
the B1 and B3 enzymes. plexes arising from reaction of MBLs with β-lactams
in crystallo, although (in common with many other
Mechanism of carbapenem hydrolysis by class B crystallographic studies of protein:ligand com-
β-lactamases plexes) the occupancy achieved and the confi-
dence with which the mode of ligand binding is
The studies of Waley and Benkovic provide the defined vary between them [126]. Complexes of
foundation of much current mechanistic understanding MBLs with hydrolyzed carbapenems have now
of MBLs. The former proposed, based on low- been described for the B1 enzymes NDM-1
temperature transient kinetic studies of Zn(II) and Co [117,119,121] (Fig. 6a) and VIM-1 [125] (Fig. 6b),
(II)-substituted enzymes, that reaction involved changes the B2 enzyme CphA [120] and the B3 enzyme
in metal coordination during the catalytic cycle and a SMB-1 [122] (Fig. 6c). Importantly, the consensus
branched kinetic mechanism [108,109]. The latter from these structures, as well as those of other
conclusively supported involvement of a metal-bound complexes of MBLs with hydrolyzed β-lactams
water (or hydroxide) nucleophile as opposed to [116–118,121,123], supports close approach of
alternatives such as an anhydride mechanism [110]. the ring-opened nitrogen to Zn2, consistent with
Benkovic and co-workers also proposed the key the concept of the populated anionic intermediate.
intermediate to be a ring-opened (i.e., product-like) However, in other aspects, available structures of
anionic species in which a negatively charged carbapenem complexes show considerable vari-
nitrogen is stabilized by proximity to Zn2, which is ability. A recent study of NDM-1 [119] observed both
therefore described as a “superacid” [110,111]. This the Δ2 and Δ1 tautomers of enzyme-bound ring-
differentiates the MBL-catalyzed reaction from pro- opened imipenem and meropenem, consistent with
posed mechanisms for other metallohydrolases (as the possibility of a branched mechanism, but proposed
reviewed in, e.g., Refs. [112,113]), which emphasize a linear reaction pathway whereby protonation occurs
stabilization by the metal center of tetrahedral oxyanion exclusively from bulk solvent. This was evidenced by
species formed after attack of a metal-bound water NMR experiments which, in contrast to those in the
nucleophile on the scissile bond. Although the anionic study above, identified only a single diastereomer of
intermediate was initially identified in hydrolysis of the the Δ1 hydrolysis product, and by the absence of
chromogenic cephalosporin nitrocefin [111], in which bound water in the Zn-bridging position, which is
formation of a conjugated system within the electron- instead occupied by the newly formed carboxylate at
withdrawing dinitrostyryl C3 substituent is proposed C7. By way of contrast, carbapenem complexes of
to help stabilize the negative charge, evidence is VIM-1 [125] (subclass B1) and SMB-1 [122] (B3)
accumulating for its existence in reactions with other exclusively show the Δ1 tautomer, but also contain a
β-lactam classes, in particular carbapenems, where water molecule in the bridging position interacting with
the ability of the pyrroline ring system to tautomerize is the 6α-hydroxyethyl group. Notably, these structures
considered to support formation of analogous species were obtained by prolonged exposure to carbapenem
by enabling delocalization of the negative charge rather than the shorter soaks used the NDM-1 study.
arising from β-lactam amide cleavage [84,114]. More Taken together, these data suggest that species
recently, a combination of UV–vis [utilizing both Zn(II) formed on opening of the β-lactam ring may feature
and Co(II)-substituted enzymes], NMR and X‐ray bidentate Zn1 coordination by the product carboxylate,
but that an incoming water molecule at the bridging Class C β-lactamases

position may cause the complex to reorganize,
possibly involving rotation around the carbapenem β-Lactamases of class C are widely distributed on
C5\C6 bond. As in other fields, in the absence of any the chromosomes of many Gram-negative species.
corroborating information (e.g., spectroscopic informa- Indeed, as the first β-lactamase to be identified, the
tion) collected in crystallo, assigning crystal structures E. coli enzyme occupies a unique position in the
as specific species in a reaction mechanism defined history of β-lactamase research. Many of the most
using solution data remains a challenging undertaking. important opportunistic Gram-negative pathogens
Crystallographic studies such as those above carry chromosomal genes encoding class C en-
have contributed much to our understanding of β- zymes, typically annotated as ampC, that under
lactam binding and hydrolysis by MBLs, but normal conditions are not expressed. However,
considerable uncertainties remain. Most notably, derepression of these, either as a result of mutation
structures are lacking for any MBL-bound states in or through induction by specific β-lactams, can lead to
which the β-lactam ring remains intact (i.e., high-level expression with consequent elevation of
Michaelis complexes) despite efforts to trap these MICs for susceptible β-lactams [133,134]. The clinical
using cadmium-substituted forms of NDM-1 [121]. relevance of class C enzymes is further enhanced by
Information on early stages of the reaction is then the dissemination of certain family members, such as
derived from spectroscopic studies, in particular the the CMY, FOX and DHA enzymes, on mobile genetic
combination of rapid freeze-quenching with elec- elements in both Enterobacteriaceae and non-
tron paramagnetic resonance or X‐ray analysis of fermenting species such as P. aeruginosa [135].
fine structure approaches. Such studies, which
have provided information at time points down to 10 Mechanism of class C β-lactamases
ms for carbapenem turnover by the BcII (B1)
[84,115] and L1 (B3) [127] MBLs, evidence key In many respects, the mechanism of class C
aspects of the proposed mechanism for binuclear enzymes has remained enigmatic, in large part due
MBLs, namely, flexible coordination of zinc ions to the lack of clear candidates for residues able to fulfill
with an increase in coordination number on asso- the role of general base, with consequent uncertainty
ciation with substrate, and flexibility of the metal surrounding both the acylation and deacylation halves
center as a whole as evidenced by increases in the of the reaction. In the absence of the Glu166 found in
zinc– zinc separation during the catalytic cycle. The the class A active site, initial crystallographic work
latter point is consistent with crystallographic data [136] inspired the suggestion, based on structural
[119]. An additional consideration is the existence comparisons with trypsin, that Tyr150 (a residue
of specific sequence/structural requirements for conserved among class C enzymes) could function
turnover of carbapenems that are more stringent as the general base for both acylation and deacyla-
than those for other substrates—a very recent deep tion, but such a mechanism would require this to exist
sequencing study of NDM-1 [128] identified a as tyrosinate (i.e., in the deprotonated form) through-
number of positions at which mutation affected out the reaction cycle and is not supported either by
carbapenem hydrolysis much more severely than NMR titration experiments [137] or by subsequent
that of other substrates. high-resolution crystal structures [138] that identify
Computational studies of β-lactam binding and this residue as neutral (i.e., remaining protonated) at
hydrolysis by MBLs are similarly constrained by the all stages of reaction. Alternative hypotheses then
absence of experimental structures for the Michae- focus on the role of Lys67 from the SBL SXXK motif
lis complex, as well as by the challenges associ- and/or require involvement of bound substrate. A
ated with accounting for the flexibility of further important mechanistic distinction from the
coordination geometry that is a feature of Zn(II) class A enzymes, identified from early crystal struc-
metalloenzyme systems [129]. In consequence, tures [139], is that the deacylating water molecule
the findings of QM/MM studies of carbapenem approaches from the opposite face of the acylenzyme
hydrolysis by binuclear MBLs vary in several carbonyl, that is, from the β- rather than the α direction.
aspects; notably, interactions involving the sub- While understanding of class C β-lactamase mecha-
strate carboxylate and Zn2, which in some cases nism remains incomplete, recent advances in this
are proposed to occur only on formation of anionic area have exploited the several high-quality crystal
species [130,131]; the identity of the nucleophile, in structures now available to apply increasingly sophis-
one study [121] considered to be bulk water rather ticated computational approaches to evaluate possi-
than the bridging hydroxide; and the protonation ble mechanisms.
route(s) for N4 [131]. It is nevertheless encouraging Computational approaches have been used to
that in a number of cases [131,132] simulations investigate both the acylation and deacylation steps
both predicted free energy barriers for carbapenem of the class C β-lactamase reaction, although most
breakdown and the structures of intermediates that have focused on the latter. Several investigations
closely approach those derived from experiment. have focused on the monobactam aztreonam, for
Fig. 8. Proposed deacylation mechanisms for class C β-lactamases. Figure shows possible schemes for hydrolysis of
generic cephalosporin acylenzyme. (a) Conjugate-base hypothesis where Lys67 deprotonates Tyr150 to activate
deacylating water molecule. (b) Substrate-activated hypothesis whereby substrate N deprotonates deacylating water.
Dashed lines denote hydrogen bonds.
which as a poor substrate, the crystal structure of the Proposals for the deacylation mechanism (Fig. 8)
wild-type acylenzyme complex is available [136]. MD now divide largely between those that involve the
simulations of AmpC and its Michaelis complex with substrate amide nitrogen (“substrate-assisted”) and
aztreonam in a range of protonation states [140] those that require coordinated proton transfers be-
supported Lys67 as the general base for acylation. A tween Lys67, Tyr150 and the deacylating water
QM/MM study of AmpC acylation by both aztreonam without direct involvement of substrate (“conjugate
and the good substrate cephalothin [141] (based on base”). The latter, first proposed on the basis of QM/
crystal structures of the noncovalent complex MM simulations [144] comparing the Enterobacter P99
determined by the Shoichet group [142]) also β-lactamase and the Streptomyces R60 DD carboxy-
supported Lys67 acting as the general base, without peptidase (a model PBP), was apparently disfavored
necessitating any role for substrate groups. Tyr150 by the relatively small effects upon deacylation of the
was instead proposed to be involved in protonation Lys67Arg mutation [145], but a more recent investiga-
of the β-lactam ring N atom, acting via either a bound tion [146] reported a more substantial loss of activity for
water molecule or the carboxylate group of bound this mutant and concluded that the role of Lys67 was
substrate. This mechanism is in agreement with more than simply electrostatic. A QM/MM study of
experimental investigations that show Tyr150 mu- deacylation of the good substrate cephalothin [147]
tants to retain activity [143], as well as mechanistic identifies Lys67 as responsible for proton transfer to
proposals based on crystal structures [142]. Ser64 in the final stage of the reaction, with Tyr150
Fig. 9. Class C β-lactamase active sites. (a) AmpC:imipenem complex (PDB:1LL5), imipenem acylenzyme covalently
attached to Ser64 (note the presence of putative deacylating water adjacent to Tyr150) and (b) ADC-68 active site (note
the residues 320 and 321 in the putative C-loop associated with carbapenem turnover). Distances (in Å) displayed as
dashed lines. Important residues are represented as sticks (labeled), and waters are shown as spheres.
undergoing transient deprotonation in the acylenzyme although a water molecule is observed in the
complex that enables its activation for deacylation. putative deacylating position and (due in part to its
These authors emphasize that the dynamic nature of approach from the β-direction) this is not involved in
the acylenzyme complex, with frequent proton trans- interactions with the imipenem 6α-hydroxyethyl
fers between these two residues, does not preclude group, the orientation of the acylenzyme within the
Tyr150 remaining largely protonated in the acylen- active site is considered unfavorable for facile
zyme state. The alternative model for deacylation, in deacylation. Intriguingly, this structure (obtained by
which the catalytic water is deprotonated by the β- relatively short soaking rather than cocrystallization)
lactam ring nitrogen, is supported by crystal structures shows imipenem bound as the Δ2 pyrroline form,
of acylenzyme complexes [148] and (at higher with the implication that tautomerization of the
resolution) of models of the deacylation transition carbapenem acylenzyme may occur slowly in
state [138], as well as by the observation that substrate class C enzymes compared to some other β-
analogues lacking an amine served as irreversible lactamases.
inhibitors [149], but is disfavored by the fact that class Recently, however, reports have emerged iden-
C β-lactamases can hydrolyze depsipeptides [150], tifying specific class C enzymes as capable of
which also lack an amide nitrogen. It is entirely possible hydrolyzing carbapenems (e.g., imipenem) in ad-
that both mechanisms may be employed, depending dition to extended-spectrum cephalosporin sub-
on the precise combination of enzyme and substrate strates. Crystal structures are now available for two
[146]. such enzymes, the plasmid-borne CMY-10 origi-
nally identified in Enterobacter aerogenes [155]
Carbapenemase activity of class C β-lactamases and the chromosomal ADC-68 enzyme from A.
baumannii (Fig. 9b [156]). In both cases, carba-
Class C β-lactamases are not usually regarded as penem turnover is considered to be promoted by a
possessing carbapenemase activity, and involve- 3-amino-acid deletion in the so-called R2 loop that,
ment in carbapenem resistance is generally consid- by expanding the active site, may relieve potential
ered to derive from the ability of strains that combine steric clashes with the extended C2 side chains of
reduced permeability (i.e., porin loss or mutation) many carbapenems. The presence in ADC-68 of
with over-expression of class C enzymes to se- glycine, rather than the more bulky arginine found
quester periplasmic carbapenems before they can in the parent ADC-1 enzyme (which does not
reach their PBP targets [151]. Consistent with this, hydrolyze carbapenems), in a second, shorter
affinities of class C β-lactamases for carbapenems loop (the C-loop) on the opposite face of the
are generally high [152,153], and the crystal substrate-binding cleft may exert a similar effect
structure of the imipenem acylenzyme formed by [156]. As no acylenzyme structure is yet available
E. coli AmpC (Fig. 9a [154]) reveals the carbonyl for a class C β-lactamase with carbapenem-
oxygen atom to be rotated approximately 180° away hydrolyzing activity, it remains unclear whether
from the oxyanion hole in a binding mode reminis- and how the sequence variations described above
cent of that observed with TEM-1 (see above). Thus, also promote deacylation by reorienting the
Fig. 10. Active sites of class D β-lactamases and carbapenem acylenzymes. (a) Native OXA-23 (PDB 4KOX; note the
hydrophobic bridge between Phe110 and Met221 and carboxylated Lys82; deacylating water is shown as a red sphere).
(b) OXA-23:meropenem acylenzyme [PDB 4JF4; note the carbapenem acylenzyme (yellow) in Δ1-pyrroline form]. (c)
OXA-24/40:doripenem acylenzyme [PDB 3PAE; note the carbapenem acylenzyme (cyan) in Δ2-pyrroline form].
Carbapenem acylenzymes (b and c) shown as sticks covalently attached to Ser79. Distances (in Å) displayed as dashed
lines. Note that, for consistency, residue numbering for all panels is as used by Smith et al. [185].
acylenzyme to facilitate attack by the deacylating evidenced by their utility in generating acylenzyme
water molecule. complexes for crystallographic studies). These results
indicate that other properties of the OXA active site,
besides the presence of the carboxylated lysine, also
Class D β-lactamases contribute to activation of the nucleophilic serine.
The OXA enzymes of class D are the most diverse Carbapenem hydrolysis by OXA β-lactamases
and in many respects the least well understood of all the
β-lactamases. While the first enzymes identified had Much recent effort in the field has concerned the
activity restricted to penicillins, the OXA class now interactions of OXA enzymes with carbapenems,
encompasses enzymes active against cephalosporins which remains an area of active investigation. Within
and carbapenems and with widely differing sensitivities the greater OXA family, five separate clades of
to inhibitors (for reviews, see Refs. [157,158]). Although enzymes are generally accepted as being able to
many members are chromosomal, dissemination of hydrolyze carbapenems [158], although it has been
plasmid-borne cephalosporinases in P. aeruginosa proposed, primarily on the basis of their ability to
[159], and more recently the spread of carbapenem- elevate carbapenem MICs in less permeable back-
hydrolyzing enzymes in A. baumannii [160] and in grounds such as Acinetobacter, that all OXA enzymes
Enterobacteriaceae (particularly K. pneumoniae [161]), can be considered as carbapenemases [172]. Of the
has increased the clinical significance of this class. The five recognized groups of OXA carbapenemases, four
recent identification of OXA enzymes in a variety of (OXA-23, OXA-24/40, OXA-51 and OXA-58) are
Gram-positive species [162] is further indication of the primarily associated with resistance in A. baumannii,
exceptionally wide distribution and diversity of these with the OXA-51 group comprising a range of intrinsic,
enzymes. predominantly chromosomal enzymes that can confer
Although plasmid-borne oxacillin resistance arising resistance when over-expressed as a result of, for
from carriage of OXA enzymes was described in the example, insertions in promoter sequences [173] and
1960s [28,163], and OXA enzymes responsible for a the others more usually associated with plasmids
range of resistance phenotypes continued to be [174,175]. In contrast, enzymes of the OXA-48 group
identified in subsequent years [157,158], it was not are found on plasmids in the Enterobacteriaceae [176],
until the turn of the millennium that the first OXA crystal where they represent a growing challenge to clinical
structures became available [164,165]. The break- microbiologists due to their increasingly common
through in understanding of OXA mechanism arose involvement in carbapenem failure and the difficulties
from the work of Mobashery and co-workers [166], who associated with their detection [161,177,178].
identified carboxylation of the conserved active-site Given the diversity within the OXA family, it is
lysine (the equivalent of Lys73 in class A enzymes) as perhaps not surprising that variability is evident in the
the key determinant of activity and demonstrated that interactions made by carbapenems with OXA carba-
this arose from reversible reaction with atmospheric penemases. OXA-24 was the first such enzyme for
carbon dioxide. (It was subsequently shown that which a crystal structure became available [179]; this
acylation of the related BlaR signal sensor responsible revealed a closed tunnel within the active site, created
for regulating expression of S. aureus BlaZ in response by close contacts between Tyr112 and Met223 on
to β-lactam challenge is similarly dependent on lysine opposite sides of the substrate-binding cleft, that
carboxylation [167,168].) This unexpected finding (prior was proposed to orient the carbapenem for hydrolysis
to this discovery, carboxylated lysine residues were via interactions with the 6α-hydroxyethyl group. The
known only as components of protein metal binding deleterious effects of mutations at these positions
centers) suggested carboxylated lysine as the clear confirmed the importance of these residues to
candidate for the general base for both the acylation and carbapenem turnover. However, a subsequent struc-
deacylation halves of the reaction. This conclusion is ture of a doripenem acylenzyme complex of OXA-24
supported by subsequent structural work (comparisons (isolated by mutation of Lys82) suggested instead that
with class A enzymes place the OXA lysine carboxylate Tyr112 and Met223 serve to constrain the orientation
in a near equivalent position to that of Glu166 [169]), by of the C3 substituent of bound carbapenem, in so
the inability of lysine mutants to support deacylation doing favoring retention of the Δ2 configuration for the
[170], and by the results of QM/MM simulations of OXA- pyrrolidine ring (Fig. 10c [180]). The Δ2 tautomer is
23 [171]. The relative hydrophobicity of the OXA active also observed in a doripenem complex of OXA-51
site, which contains markedly more non-polar surface [181], although here the presence of Trp at position
area than the equivalent region of class A β-lactamases 222 (equivalent to Met223) is proposed to impair
[164], is then proposed to contribute to activity by carbapenem binding [182]. A similar so-called “hydro-
modifying the lysine pKa to favor carboxylation. It phobic bridge” between the equivalent residues
should, however, be noted that in many [170] but not Phe110 and Met225 is also present in a structure of
all [166] cases, lysine mutants retain a residual (though OXA-58 complexed with the carbapenem analogue
greatly impaired) ability to support acylation (as 6α-hydroxymethyl penicillin [183], although this is not
Fig. 11. Interactions of β-lactamases with β-lactamase inhibitors. (a) Vaborbactam bound to class A CTX-M-15 (PDB 4xuz).
(b) Avibactam bound to CTX-M-15 (PDB 4hbu). (c) A cyclobutanone bound to subclass B2 metallo-β-lactamase SPM-1 (PDB
5ndb). (d) A bicyclic boronate bound to subclass B1 metallo-β-lactamase VIM-2 (PDB 5fqc). Inhibitor molecules and the protein
residues they interact, with are shown as sticks. Ligand interactions are shown as colored dashes.
evident in the uncomplexed enzyme [184] and is thus The multiple structural studies described above thus
proposed to form on substrate binding. Uncomplexed indicate that, while OXA carbapenemases of Acineto-
OXA-23 (Fig. 10a) also contains an equivalent bacter remain a heterogeneous group, the presence of
hydrophobic bridge, between Phe110 and Met 221, an active-site hydrophobic bridge represents a common
but in this case, the structure of a carbapenem structural feature that is an important contributor to
complex (here obtained at low pH, rather than by activity toward carbapenems. In contrast, OXA-48 is
mutation of lysine) showed binding of the Δ1 tautomer only distantly related to the Acinetobacter enzymes and
(Fig. 10b [185]). Very recently, a more comprehensive instead is closer in sequence to enzymes such as the
mutational study indicated that the effects of substi- OXA-10 group that are not usually considered as
tutions at individual bridge residues on carbapenem carbapenemases. Consistent with this, the crystal
turnover are substrate-dependent. Structures of imi- structure [187] shows that OXA-48 lacks an active-site
penem and meropenem complexes of a double hydrophobic bridge, an observation that may also
mutant lacking the hydrophobic bridge, however, explain the retention of activity against other substrates
both revealed the binding of the Δ1 pyrroline, but in (oxacillin) that are poor substrates for other OXA
the R, rather than S, form. It was concluded that in the carbapenemase types [188]. However, molecular
wild-type enzyme the bridge residues select against dynamics simulations of docked carbapenems support
the Δ1-R tautomer via steric hindrance but do not a role for some equivalent residues (OXA-48 Thr213
favor the Δ2 over the Δ1-S form [186]. and Arg214) in making interactions with the methyl
group of the carbapenem 6α-hydroxyethyl substituent. associated infections by β-lactamase-producing

Furthermore, exchanging the 10-residue loops con- organisms [195]. However, their limited spectrum of
necting strands β5 and β6 in OXA carbapenemases clinically useful activity, which is confined to a subset
(OXA-23, OXA-24 or OXA-48) onto the backbone of the of class A enzymes that, importantly, does not include
OXA-10 enzyme (not considered a carbapenemase) KPC [196], coupled with the emergence and dissemi-
resulted in acquisition of activity against imipenem nation of insusceptible β-lactamase types, necessi-
[189]. This result suggests that, despite differences in tates better treatment options for β-lactamase-
sequence and in the nature of their interactions with producing organisms. The search for more widely
bound carbapenem, the β5–β6 loop (that bears, e.g., effective β-lactamase inhibitors is given added impetus
Met221 of OXA-24) is a determinant of activity against by the continued weakness of the antibiotic discovery
carbapenems across the range of OXA enzymes. pipeline for Gram-negative bacteria [197,198]. Fortu-
Taken together, these studies identify a number of nately, in recent years, this process has borne fruit, with
factors contributing to carbapenem hydrolysis by OXA representatives of two new inhibitor classes now in the
enzymes. First, structures of carbapenem complexes clinic and further compounds and combinations in
[180,181,185] indicate that the active-site architecture, development.
particularly the hydrophobic bridge, imposes steric Space considerations, together with the availability
constraints upon the orientation and preferred tauto- of several excellent recent reviews [199–202], restrict
meric form of the carbapenem acylenzyme. Second, comments on inhibitors to a relatively brief con-
evidence both from modeled complexes [187] and sideration of some of the most important recent
experimental structures (in particular the observation of advances. Chief among these is the introduction of
alternative orientations for the 6α-hydroxymethyl group the diazabicyclooctanones (DBOs), of which avibac-
of 6α-hydroxymethyl penicillin in OXA-58 complex tam (formerly NXL-104; Fig. 1, 8) was the progenitor
structures [183]) also suggests that, as for the class A and first to reach the clinic as a combination with the
carbapenemases (see above), specific interactions oxyiminocephalosporin ceftazidime [203]. Avibactam
involving the carbapenem 6α-hydroxyethyl substituent is a mechanism-based non-β-lactam β-lactamase
may be important in positioning and orienting the inhibitor, based around a bicyclic core structure, that
deacylating water molecule. Lastly, it has also been is able to acylate the active site of serine β-lactamases
proposed [184,185] that steric clashes with the in a reversible manner [204]. This contrasts with
carbapenem 6α-hydroxyethyl group may lead to previous (β-lactam-based) inhibitor classes, where
expulsion of water from the active site on acylation by acylation is followed by rearrangement and fragmen-
carbapenems, requiring recruitment of a deacylating tation events that ultimately lead to irreversible inhibi-
water molecule from bulk solvent. As a result, tion, as liberation of compound from the active site
carbapenem turnover may also be dependent on normally leads to recyclization and regeneration of
access of water to the active site, such that conforma- active inhibitor. Avibactam is a potent inhibitor of β-
tional changes in the acylenzyme may be necessary to lactamases of classes A, including KPC enzymes, and
provide this [185]. Recent observations [190], however, C, and the ceftazidime combination is now approved for
demonstrate that OXA-48-catalyzed carbapenem conditions including complicated intra-abdominal and
breakdown can generate lactone species (most likely urinary tract infections, and hospital-acquired and
arising through intramolecular cyclisation of the acy- ventilator-associated pneumonias [205]. The success
lenzyme) alongside the usual hydrolysis products. This of avibactam has also stimulated development of a
indicates that for some (1β-methyl) substrates it is number of alternative DBOs, of which the Merck
possible to resolve the carbapenem acylenzyme compound relebactam (Fig. 1, 9 [206]) is at the most
without a requirement for a deacylating water molecule. advanced stage of development as a combination with
imipenem. In these compounds, the points of difference
occur in the C2, C3 and C4 substituents, which in some
β-Lactamase inhibitors cases results in addition of weak antibacterial activity,
arising through PBP inhibition, to complement β-
Alongside improvements to β-lactams them- lactamase inhibitory action [207,208,256].
selves, combinations of susceptible β-lactams with The success of DBOs has driven extensive inves-
mechanism-based β-lactamase inhibitors represent tigation of their interactions with β-lactamases of
the major strategy for combatting β-lactamase- multiple structural classes, with the result that inhibition
mediated resistance [191]. Since the discovery and kinetics [207,209–212], together with a number of
development of clavulanic acid (Fig. 1, 7 [192]) as an crystal structures [213–217] (Fig. 11b), are now
irreversible inhibitor of the most widely distributed class available for interactions of avibactam in particular
A enzymes (e.g., the TEM, SHV and CTX-M classes with representative enzymes of classes A, C and D. A
[193,194]), penicillin-inhibitor combinations (amoxicillin– particular focus of these efforts is the drive to
clavulanate, ampicillin–sulbactam, piperacillin– understand determinants of the fate of the acylenzyme
tazobactam) have found wide application as treat- complex, which can resolve by recyclization [204] but
ments for both community- and, particularly, healthcare- in some systems, particularly KPC, can also deacylate
to generate inactive compound [209]. Evidence from a importance. MBL inhibitor development has largely
number of high-resolution structural studies now focused on compounds that bind and/or chelate the
indicates that the avibactam acylenzyme is able to zinc ions of the active site [200–202]. The natural
adopt distinct conformations, differing in the distance compound aspergillomarasmine A inhibits the MBLs
between and relative orientation of the N6 and C7 NDM-1 and VIM-2 by chelating and removing the
atoms involved in the recyclisation reaction, and active site zinc ions and has been shown to be
whose relative occupancies may provide an indicator effective against NDM-1-expressing K. pneumoniae
of acylenzyme stability [218]. A further complication is in mice [234]. Inhibitors that strongly bind to the
the observation of desulfation of the acylenzyme in active site zinc ions include thiol-based compounds
complexes with KPC, which may be associated with such as the bisthiazolidines, small bicyclic com-
the greater propensity for deacylation in this system pounds that are effective against B1, B2 and B3
[204,214,219]. The extent to which modifications of the MBLs [235]. Phosphonate-containing compounds (6-
DBO scaffold and/or mutation of the enzyme target phosphonomethylpyridine-2-carboxylates) have also
influence these processes is a matter of much current been shown in vitro to inhibit B1 and B3 MBLs by
interest. As DBOs have only recently been introduced interacting with the di-zinc metal center, with inhibition
to the clinic, understanding of the propensity for constants in the nanomolar range [236]. More
resistance remains limited, but variants with reduced recently, virtual screening combined with NMR
DBO susceptibility have been generated in the resulted in compounds that bound in the active site
laboratory for both the KPC [219–221] and AmpC of VIM-2, but importantly did not ligand the metal ions
[222–224] enzymes, and a few reports of KPC [237]. To date, the most promising MBL inhibitors are
mutations in clinical strains unresponsive to DBO the bicyclic boronates (Fig. 1, 11), compounds that
therapy have now emerged [225–227]. have been shown to inhibit both SBLs and MBLs
Boronate-based compounds represent a second (Fig. 11d) by mimicking the tetrahedral oxyanion
novel scaffold for β-lactamase inhibitor develop- formed during β-lactam hydrolysis [238]. Their activity
ment. The propensity of boron to adopt a tetrahedral extends to NDM-1, and other B1 enzymes, as well as
geometry enables it to effectively mimic transient the SBLs CTX-M-15 and OXA-48 [239]. Indeed, a
tetrahedral species formed during hydrolytic reac- bicyclic boronate (VNRX-5133) in combination with
tions [228], with the result that for some years the fourth-generation cephalosporin cefepime is
boronates have found quite extensive application now in phase 1 clinical trials. However, we recently
both as inhibitors and as tool compounds in studies demonstrated that bicyclic boronates were ineffective
of β-lactamase action aimed at mimicking the against the B3 enzyme L1 [240], indicating that
tetrahedral oxyanion transition states of either the their combinations might not be effective against
acylation or deacylation reactions (for selected S. maltophilia. Identification of compounds such as
examples, see Refs. [138,229,230]). More recently, these with true pan β-lactamase inhibitory activity, that
however, development of cyclic boronates as broad- is, able to inhibit both MBLs and SBLs, remains an
spectrum inhibitors of serine β-lactamases [231] has ultimate goal in the field; the success of cyclic
resulted in introduction to the clinic of the combina- boronates justifies further exploration of β-lactams
tion of meropenem with the cyclic boronate vabor- [241] or analogues such as cyclobutanones [242,243]
bactam (Fig. 1, 10) as a treatment for complicated (Fig. 11c) where evidence exists that such broad-
urinary tract infections [232]. Vaborbactam potently spectrum activity is also achievable.
inhibits β-lactamases of classes A (Fig. 11a and c), Encouraging results have also emerged from recent
but activity does not extend to OXA enzymes or to investigations of alternative approaches to MBL
the MBLs [233]. inhibition. The concept of generating covalent inhib-
itors able to attach to the highly conserved active-site
Inhibition of MBLs Lys224 of B1 MBLs was first explored by Kurosaki
et al. [244], who demonstrated that Lys224 reacted
Introduction to the clinic of DBO and vaborbactam with activated esters of 3-mercapto propionic acid.
combinations dramatically increases treatment op- More recently, the selenium compound ebselen was
tions for serious infections by Gram-negative bacteria. shown to attach to Cys221 of NDM-1 [245]; the two
However, neither is effective against the class B approaches have subsequently been combined to
(metallo) enzymes, and these now constitute the create a dual covalent inhibitor, reacting with both
major challenge for β-lactamase inhibitor develop- Cys221 and Lys224, active against multiple MBLs of
ment. Indeed, despite extensive academic effort, no subclasses B1 and B2 [246]. A second rewarding
MBL inhibitor is yet close to the clinic. The advent of approach [247] has been to use colloidal bismuth
NDM as a highly mobile MBL in Enterobacteriaceae sulfate (identified from screens of metal compounds)
has dramatically increased the clinical importance of to displace Zn(II) from the NDM-1 active site,
MBLs, with the recent observation that MBLs can generating an inactive Bi(III) complex. Importantly,
weakly degrade avibactam (and by implication other ebselen and colloidal bismuth sulfate are respectively
DBOs) serving to further emphasize their growing in trials or clinical use as therapies for other conditions,
constituting grounds for optimism that the process of demands caution in extrapolation of results from a
gaining clinical approval may be simpler than for truly single model system to more general conclusions.
novel agents. Despite such progress, inhibition of The above comments are particularly relevant when
MBLs remains a challenging undertaking, not least considering carbapenem hydrolysis. While the last
due to the requirement for selectivity over other Zn(II) decade has provided great advances in understanding
metalloenzymes within and beyond (e.g., angiotensin- how enzymes considered as having carbapenemase
converting enzyme) the MBL superfamily. Achieving activity bind and hydrolyze these substrates, questions
potency across the range of MBL enzymes and remain. In particular, the relationship between pyrroline
subclasses is also not trivial; aside from the more tautomerization and acylenzyme hydrolysis (i.e., the
obvious divergence of subclass B3 enzymes (as extent and basis of the stability to deacylation of the Δ1
evidenced by the lack of activity of bicyclic boronates tautomer, particularly in the OXA enzymes) is yet to be
against L1) important variation also exists within established. While the Δ1-pyrroline is observed in
subclass B1, exemplified by the absence of Lys224 many crystal structures of β-lactamase:carbapenem
in enzymes such as VIM. acylenzymes, spectroscopic studies of SHV-1 [60]
remain as the strongest direct evidence for its
increased stability to hydrolysis compared to the Δ2
Concluding Remarks acylenzyme. The basis for this slow deacylation
remains uncertain: conformational changes associated
Several factors have combined to transform the with formation of the Δ1 species may variously
β-lactamase field in recent years. The ubiquity of reposition the acylenzyme carbonyl outside the oxya-
ESBLs has driven use of carbapenems as empiric nion hole (SHV-1 [59,60]), or reorient the 6α-
therapy for Gram-negative bacteria, leading to hydroxethyl group to enable interactions that deactivate
dissemination of carbapenemases, particularly in (e.g., with BlaC Glu166 [61]) or destabilize (e.g., with
Enterobacteriaceae, that challenge our ability to treat the carboxylated lysine of OXA enzymes [254]) the
many such infections. In addition to the ongoing deacylating water molecule. A further possibility is that
proliferation of the KPC enzymes, the association of the lack of a proton at N4 of the Δ1-acylenzyme
NDM-1 and OXA-48 with carbapenem resistance in perturbs the active-site hydrogen bonding networks
K. pneumoniae has provided new impetus for funda- that activate the general base for deacylation [180].
mental biochemical studies of β-lactamases of classes Furthermore, while carbapenemases of classes A and
B and D, respectively. Furthermore, the explosion in D may both adopt strategies to constrain the
genome sequence information has identified candidate orientation of the carbapenem 6α-hydroxyethyl
β-lactamases in organisms and environments not group, the implications of these for binding/activation
previously considered as likely to harbour such of the deacylating water molecule may differ in
enzymes [248,249], as well as in organisms formerly different systems. The interactions of carbapenems
regarded as niche pathogens that have attracted with β-lactamases of class C remain much less well
renewed attention as potential agents of bioterrorism understood, and the basis for both carbapenem
or biological warfare [102,250–252]. inhibition of the majority of these enzymes,and
Fundamental biochemical questions surrounding carbapenem turnover by exceptions such as CMY-
β-lactamase activity nevertheless remain. For en- 10 and ADC-68 remains unclear.
zymes of classes A and C, debate still surrounds the Aside from continuing investigations of underly-
respective acylation and deacylation mechanisms in ing biochemistry, future directions of β-lactamase
particular, while for the metallo-β-lactamases under- research will be in large part directed by the
standing of the early events in the reaction cycle emergence and dissemination of resistance phe-
remains far from complete, in no small part due to the notypes in the clinic. An open question is then how
absence of any covalent reaction intermediate. The the overall β-lactamase profile of bacterial patho-
situation is slowly improving as availability of more gens will respond to the clinical application of the
and better quality crystal structures and increased new wave of β-lactamase inhibitors. It may be
access to higher-performance computing facilities, reasonable to predict that MBL-producing strains
respectively, provide more accurate models for may become even more prevalent, given that these
mechanistically important species and permit appli- enzymes continue to avoid inhibitor combinations
cation of high-level computational methodologies to at or near to clinic, but reports of β-lactamase
systems of this size and complexity and over mutations resulting in insusceptibility to avibactam
biologically more relevant timescales. When coupled in clinical isolates carrying KPC [226,255] or CTX-
with technological advances in fields such as in M [225] enzymes, as well as in vitro generation of
crystallo spectroscopy [253], neutron diffraction and resistant mutants, demonstrate the existence of
SFX, the prospects for significant advancement in multiple potential routes to failure of DBO combi-
understanding of β-lactamase mechanism are good. nations. Of course, many precedents, including the
Against this, the extensive diversity of both enzyme emergence of KPC and NDM enzymes, also exist
and substrate in β-lactamase:β-lactam reactions to suggest that previously unknown enzymes with
distinctive biochemical properties may mobilize in

response to usage changes in β-lactams and
inhibitor combinations. Identifying the reasons †C.L.T. and P.H. contributed equally to this work.
behind the successful large-scale dissemination
of enzymes such as those above, when others with Abbreviations used:
similar biochemical properties remain confined to CTX-M, cefotaximase; DBO, diazabicyclooctane; ESBL,
occasional isolates or niche pathogens, may help extended spectrum β-lactamase; KPC, Klebsiella pneu-
to estimate the clinical significance of moniae carbapenemase; MBL, metallo-β-lactamase;
new sequences and enzymes as they continue to NDM, New Delhi metallo-β-lactamase; OXA, oxacillinase;
be discovered. It is a certainty that the selection PBP, penicillin-binding protein; QM/MM, quantum me-
pressure imposed by new β-lactams and, particu- chanics/molecular mechanics; SBL, serine β-lactamase;
larly, inhibitor combinations will change the land- SFX, serial femtosecond crystallography; SHV, sulfydryl
scape of β-lactamases, but much more difficult to variant; TEM, Temoneira (β-lactamase); VIM, Verona
predict the ways in which this will occur. imipenemase.
The long history and extensive literature of β-
lactamase research might suggest a mature field
lacking in scope for discovery science. We would References
instead argue that the response of the bacterial
population to the selection pressure imposed by [1] A. Fleming, On the antibacterial action of cultures of a
the continuing clinical importance of β-lactams and penicillium with special reference to their use in the isolation
of new β-lactam based therapies, the wealth of new of B. influenzae, Br. J. Exp. Pathol. 10 (1929) 226–236.
enzymes identified through the explosion in micro- [2] E.Y. Klein, T.P. Van Boeckel, E.M. Martinez, S. Pant, S.
bial sequence information, and the application of Gandra, S.A. Levin, et al., Global increase and geographic
new technological advances, together create rich convergence in antibiotic consumption between 2000 and
opportunities both for fundamental biochemistry 2015, Proc. Natl. Acad. Sci. U. S. A. 115 (2018) E3463-
E70.
and innovative drug discovery that are underpinned [3] H.C. Neu, beta-Lactam antibiotics: structural relationships
by the pressing need to overcome antimicrobial affecting in vitro activity and pharmacologic properties, Rev.
resistance. Infect. Dis. 8 (Suppl. 3) (1986) S237–S259.
[4] G. Brotzu, Ricerche su di un nuovo antibiotico. Lavori
dell'Istituto d'Igiene di Cagliari, 1948 4–18.
[5] G.G. Newton, E.P. Abraham, Degradation, structure and
some derivatives of cephalosporin N, Biochem. J. 58 (1954)
Acknowledgments 103–111.
[6] A.G. Brown, D. Butterworth, M. Cole, G. Hanscomb, J.D.
This work was supported by the UK Biotechnology Hood, C. Reading, et al., Naturally-occurring beta-lactamase
and Biological Sciences Research Council-funded inhibitors with antibacterial activity, J. Antibiot. 29 (1976)
South West Biosciences Doctoral Training Partnership 668–669.
(training grant reference BB/J014400/1, studentship [7] A. Imada, K. Kitano, K. Kintaka, M. Muroi, M. Asai, Sulfazecin
and isosulfazecin, novel beta-lactam antibiotics of bacterial
to C.L.T.), by the UK Medical Research Council
origin, Nature. 289 (1981) 590–591.
(MR/N0137941/1 for the GW4 BIOMED DTP [8] R.B. Sykes, C.M. Cimarusti, D.P. Bonner, K. Bush, D.M. Floyd,
awarded to the Universities of Bath, Bristol, Cardiff N.H. Georgopapadakou, et al., Monocyclic beta-lactam anti-
and Exeter, studentship to V.H.A.H.), and by the biotics produced by bacteria, Nature. 291 (1981) 489–491.
National Institute of Allergy and Infectious Diseases of [9] D.J. Tipper, J.L. Strominger, Mechanism of action of
the US National Institutes of Health under award penicillins—a proposal based on their structural similiarity
number R01AI100560 (P.H. and J.S.). The content is to acyl-D-alanyl–D-alanine, Proc. Natl. Acad. Sci. U. S. A.
solely the responsibility of the authors and does not 54 (1965) 1133–1141.
necessarily represent the official views of the National [10] E. Sauvage, F. Kerff, M. Terrak, J.A. Ayala, P. Charlier, The
penicillin-binding proteins: structure and role in peptidoglycan
Institutes of Health.
biosynthesis, FEMS Microbiol. Rev. 32 (2008) 234–258.
[11] G.N. Rolinson, S. Stevens, Microbiological studies on a new
Received 17 January 2019; broad-spectrum penicilin, “Penbritin”, Br. Med. J. 2 (1961)
Received in revised form 27 March 2019; 191–196.
Accepted 1 April 2019 [12] G.N. Rolinson, S. Stevens, F.R. Batchelor, J.C. Wood, E.B.
Available online 05 April 2019 Chain, Bacteriological studies on a new penicillin-BRL.
1241, Lancet. 2 (1960) 564–567.
Keywords: [13] W.F. Marshall, J.E. Blair, The cephalosporins, Mayo Clin.
β-lactam; Proc. 74 (1999) 187–195.
antimicrobial resistance; [14] D.T. King, S. Sobhanifar, N.C.J. Strynadka, The mecha-
nisms of resistance to β-lactam antibiotics, in: M. Gotte, A.
carbapenemase;
Berghuis, G. Matlashewski, M. Wainberg, D. Sheppard
enzyme mechanism;
(Eds.), Handbook of Antimicrobial Resistance, Springer, [34] J.M. Martin-Garcia, C.E. Conrad, J. Coe, S. Roy-Chowdhury,
New York, NY, 2017. P. Fromme, Serial femtosecond crystallography: a revolution
[15] K. Bush, Past and present perspectives on beta- in structural biology, Arch. Biochem. Biophys. 602 (2016)
lactamases, Antimicrob. Agents Chemother. 62 (2018). 32–47.
[16] E.P. Abraham, E. Chain, An enzyme from bacteria able to [35] C. Kupitz, J.L. Olmos Jr., M. Holl, L. Tremblay, K. Pande, S.
destroy penicillin, Nature. 146 (1940) 837. Pandey, et al., Structural enzymology using X‐ray free
[17] W.M. Kirby, Extraction of a highly potent penicillin inacti- electron lasers, Struct. Dyn. 4 (2017), 044003.
vator from penicillin resistant staphylococci, Science. 99 [36] J.L. Olmos Jr., S. Pandey, J.M. Martin-Garcia, G. Calvey, A.
(1944) 452–453. Katz, J. Knoska, et al., Enzyme intermediates captured “on
[18] J.F. Fisher, S. Mobashery, Beta-lactam resistance mecha- the fly” by mix-and-inject serial crystallography, BMC Biol.
nisms: Gram-positive bacteria and Mycobacterium tuber- 16 (2018) 59.
culosis, Cold Spring Harb. Perspect. Med. 6 (2016). [37] M.O. Wiedorn, D. Oberthur, R. Bean, R. Schubert, N.
[19] M. Rozwandowicz, M.S.M. Brouwer, J. Fischer, J.A. Werner, B. Abbey, et al., Megahertz serial crystallography,
Wagenaar, B. Gonzalez-Zorn, B. Guerra, et al., Plasmids Nat. Commun. 9 (2018) 4025.
carrying antimicrobial resistance genes in Enterobacteria- [38] Y. Chen, R. Bonnet, B.K. Shoichet, The acylation mecha-
ceae, J. Antimicrob. Chemother. 73 (2018) 1121–1137. nism of CTX-M beta-lactamase at 0.88 a resolution, J. Am.
[20] S.R. Partridge, S.M. Kwong, N. Firth, S.O. Jensen, Mobile Chem. Soc. 129 (2007) 5378–5380.
genetic elements associated with antimicrobial resistance, [39] S.J. Tomanicek, K.K. Wang, K.L. Weiss, M.P. Blakeley, J.
Clin. Microbiol. Rev. 31 (2018). Cooper, Y. Chen, et al., The active site protonation states of
[21] T. Naas, S. Oueslati, R.A. Bonnin, M.L. Dabos, A. Zavala, L. perdeuterated Toho-1 beta-lactamase determined by neu-
Dortet, et al., Beta-lactamase database (BLDB)—structure and tron diffraction support a role for Glu166 as the general base
function, J. Enzyme Inhib. Med. Chem. 32 (2017) 917–919. in acylation, FEBS Lett. 585 (2011) 364–368.
[22] K. Bush, G.A. Jacoby, A.A. Medeiros, A functional classifica- [40] J.C. Hermann, L. Ridder, A.J. Mulholland, L.D. Höltje,
tion scheme for beta-lactamases and its correlation with Identification of Glu166 as the general base in the acylation
molecular-structure, Antimicrob. Agents Chemother. 39 reaction of class A beta-lactamases through QM/MM
(1995) 1211–1233. modeling, J. Am. Chem. Soc. 125 (2003) 9590–9591.
[23] R.P. Ambler, The structure of beta-lactamases, Philos. [41] J.C. Hermann, C. Hensen, L. Ridder, A.J. Mulholland, L.D.
Trans. R. Soc. Lond. Ser. B Biol. Sci. 289 (1980) 321–331. Höltje, Mechanisms of antibiotic resistance: QM/MM model-
[24] I. Massova, S. Mobashery, Kinship and diversification of ing of the acylation reaction of a class A beta-lactamase with
bacterial penicillin-binding proteins and beta-lactamases, benzylpenicillin, J. Am. Chem. Soc. 127 (2005) 4454–4465.
Antimicrob. Agents Chemother. 42 (1998) 1–17. [42] J.C. Hermann, J. Pradon, J.N. Harvey, A.J. Mulholland,
[25] D.L. Paterson, Recommendation for treatment of severe High level QM/MM modeling of the formation of the
infections caused by Enterobacteriaceae producing tetrahedral intermediate in the acylation of wild type and
extended-spectrum beta-lactamases (ESBLs), Clin. Microbiol. K73A mutant TEM-1 class A beta-lactamase, J. Phys.
Infect. 6 (2000) 460–463. Chem. A 113 (2009) 11984–11994.
[26] Y. Zhang, X.L. Chen, A.W. Huang, S.L. Liu, W.J. Liu, N. [43] H. Adachi, T. Ohta, H. Matsuzawa, Site-directed mutants, at
Zhang, et al., Mortality attributable to carbapenem-resistant position 166, of RTEM-1 beta-lactamase that form a stable
Pseudomonas aeruginosa bacteremia: a meta-analysis of acyl-enzyme intermediate with penicillin, J. Biol. Chem. 266
cohort studies, Emerg. Microbes Infect. e27 (2016) 5. (1991) 3186–3191.
[27] L.K. Logan, R.A. Weinstein, The epidemiology of carbapenem- [44] N.C.J. Strynadka, H. Adachi, S.E. Jensen, K. Johns, A.
resistant enterobacteriaceae: the impact and evolution of a Sielecki, C. Betzel, et al., Molecular-structure of the acyl-
global menace, J. Infect. Dis. 215 (2017) S28–S36. enzyme intermediate in beta-lactam hydrolysis at 1.7 Angstrom
[28] N. Datta, P. Kontomichalou, Penicillinase synthesis con- resolution, Nature. 359 (1992) 700–705.
trolled by infectious R factors in Enterobacteriaceae, [45] J.F. Fisher, S.O. Meroueh, S. Mobashery, Bacterial resis-
Nature. 208 (1965) 239–241. tance to beta-lactam antibiotics: compelling opportunism,
[29] J. Chaves, M.G. Ladona, C. Segura, A. Coira, R. Reig, C. compelling opportunity, Chem. Rev. 105 (2005) 395–424.
Ampurdanes, SHV-1 beta-lactamase is mainly a chromo- [46] J.C. Hermann, L. Ridder, L.D. Höltje, A.J. Mulholland,
somally encoded species-specific enzyme in Klebsiella Molecular mechanisms of antibiotic resistance: QM/MM
pneumoniae, Antimicrob. Agents Chemother. 45 (2001) modelling of deacylation in a class A beta-lactamase, Org.
2856–2861. Biomol. Chem. 4 (2006) 206–210.
[30] A. Bauernfeind, H. Grimm, S. Schweighart, A new plasmidic [47] L.W. Tremblay, H. Xu, J.S. Blanchard, Structures of the
cefotaximase in a clinical isolate of Escherichia coli, Michaelis complex (1.2 A) and the covalent acyl interme-
Infection. 18 (1990) 294–298. diate (2.0 A) of cefamandole bound in the active sites of the
[31] R.P. Rapp, C. Urban, Klebsiella pneumoniae carbapene- Mycobacterium tuberculosis beta-lactamase K73A and
mases in Enterobacteriaceae: history, evolution, and micro- E166A mutants, Biochemistry. 49 (2010) 9685–9687.
biology concerns, Pharmacotherapy. 32 (2012) 399–407. [48] X. Pan, Y. He, J. Lei, X. Huang, Y. Zhao, Crystallographic
[32] Y. Chong, S. Shimoda, N. Shimono, Current epidemiology, snapshots of class A beta-lactamase catalysis reveal
genetic evolution and clinical impact of extended-spectrum structural changes that facilitate beta-lactam hydrolysis,
beta-lactamase-producing Escherichia coli and Klebsiella J. Biol. Chem. 292 (2017) 4022–4033.
pneumoniae, Infect. Genet. Evol. 61 (2018) 185–188. [49] V.G. Vandavasi, P.S. Langan, K.L. Weiss, J.M. Parks, J.B.
[33] T. Palzkill, Structural and mechanistic basis for extended- Cooper, S.L. Ginell, et al., Active-site protonation states in an
spectrum drug-resistance mutations in altering the specificity acyl-enzyme intermediate of a class A beta-lactamase with a
of TEM, CTX-M, and KPC beta-lactamases, Front. Mol. monobactam substrate, Antimicrob. Agents Chemother. 61
Biosci. 5 (2018) 16. (2017).
[50] P. Langan, V.G. Vandavasi, S.J. Cooper, K.L. Weiss, S.L. [65] F.K. Majiduddin, T. Palzkill, Amino acid sequence require-
Ginell, J.M. Parks, et al., Substrate binding induces ments at residues 69 and 238 for the SME-1 beta-lactamase
conformational changes in a class A b-lactamase that to confer resistance to beta-lactam antibiotics, Antimicrob.
prime it for catalysis, ACS Catal. 8 (2018) 2428–2437. Agents Chemother. 47 (2003) 1062–1067.
[51] S.O. Meroueh, J.F. Fisher, H.B. Schlegel, S. Mobashery, Ab [66] C.A. Smith, Z. Nossoni, M. Toth, N.K. Stewart, H. Frase, S.
initio QM/MM study of class A beta-lactamase acylation: dual B. Vakulenko, Role of the conserved disulfide bridge
participation of Glu166 and Lys73 in a concerted in class A carbapenemases, J. Biol. Chem. 291 (2016)
base promotion of Ser70, J. Am. Chem. Soc. 127 (2005) 22196–22206.
15397–15407. [67] F. Fonseca, E.I. Chudyk, M.W. van der Kamp, A. Correia, A.
[52] D.A. Nichols, J.C. Hargis, R. Sanishvili, P. Jaishankar, K. J. Mulholland, J. Spencer, The basis for carbapenem
Defrees, E.W. Smith, et al., Ligand-Induced proton transfer hydrolysis by class A beta-lactamases: a combined
and low-barrier hydrogen bond revealed by X‐ray crystal- investigation using crystallography and simulations, J. Am.
lography, J. Am. Chem. Soc. 137 (2015) 8086–8095. Chem. Soc. 134 (2012) 18275–18285.
[53] H. Kropp, L. Gerckens, J.G. Sundelof, F.M. Kahan, [68] C.A. Smith, H. Frase, M. Toth, M. Kumarasiri, K. Wiafe, J.
Antibacterial activity of imipenem: the first thienamycin Munoz, et al., Structural basis for progression toward the
antibiotic, Rev. Infect. Dis. 7 (Suppl. 3) (1985) S389–S410. carbapenemase activity in the GES family of beta-
[54] H. Yigit, A.M. Queenan, G.J. Anderson, A. Domenech- lactamases, J. Am. Chem. Soc. 134 (2012)
Sanchez, J.W. Biddle, C.D. Steward, et al., Novel 19512–19515.
carbapenem-hydrolyzing beta-lactamase, KPC-1, from a [69] S.C. Mehta, K. Rice, T. Palzkill, Natural variants of the KPC-2
carbapenem-resistant strain of Klebsiella pneumoniae, carbapenemase have evolved increased catalytic efficiency
Antimicrob. Agents Chemother. 45 (2001) 1151–1161. for ceftazidime hydrolysis at the cost of enzyme stability,
[55] R.L. Charnas, J.R. Knowles, Inhibition of the RTEM beta- PLoS Pathog. 11 (2015), e1004949.
lactamase from Escherichia coli. Interaction of enzyme with [70] K.M. Papp-Wallace, M. Taracila, J.M. Hornick, A.M. Hujer, K.M.
derivatives of olivanic acid, Biochemistry. 20 (1981) Hujer, A.M. Distler, et al., Substrate selectivity and a novel role
2732–2737. in inhibitor discrimination by residue 237 in the KPC-2 beta-
[56] C.J. Easton, J.R. Knowles, Inhibition of the RTEM beta- lactamase, Antimicrob. Agents Chemother. 54 (2010)
lactamase from Escherichia coli. Interaction of the enzyme 2867–2877.
with derivatives of olivanic acid, Biochemistry. 21 (1982) [71] K.M. Papp-Wallace, M. Taracila, C.J. Wallace, K.M. Hujer, C.R.
2857–2862. Bethel, J.M. Hornick, et al., Elucidating the role of Trp105 in the
[57] G. Zafaralla, S. Mobashery, Facilitation of the delta2 delta1 KPC-2 beta-lactamase, Protein Sci. 19 (2010) 1714–1727.
pyrroline tautomerization of carbapenem antibiotics by the [72] K.M. Papp-Wallace, M.A. Taracila, K.M. Smith, Y. Xu, R.A.
highly conserved arginine-244 of class A beta-lactamases Bonomo, Understanding the molecular determinants of
during the course of turnover, J. Am. Chem. Soc. 114 (1992) substrate and inhibitor specificities in the carbapenemase
1505–1506. KPC-2: exploring the roles of Arg220 and Glu276, Antimicrob.
[58] L. Maveyraud, L. Mourey, L.P. Kotra, J.D. Pedelacq, V. Agents Chemother. 56 (2012) 4428–4438.
Guillet, S. Mobashery, et al., Structural basis for clinical [73] O.A. Pemberton, X. Zhang, Y. Chen, Molecular basis of
longevity of carbapenem antibiotics in the face of challenge by substrate recognition and product release by the Klebsiella
the common class A beta-lactamases from the antibiotic- pneumoniae carbapenemase (KPC-2), J. Med. Chem. 60
resistant bacteria, J. Am. Chem. Soc. 120 (1998) 9748–9752. (2017) 3525–3530.
[59] M. Nukaga, C.R. Bethel, J.M. Thomson, A.M. Hujer, A. [74] G.A. Cortina, J.M. Hays, P.M. Kasson, Conformational
Distler, V.E. Anderson, et al., Inhibition of class A beta- intermediate that controls KPC-2 catalysis and beta-lactam
lactamases by carbapenems: crystallographic observation drug resistance, ACS Catal. 8 (2018) 2741–2747.
of two conformations of meropenem in SHV-1, J. Am. [75] I. Galdadas, S. Lovera, G. Perez-Hernandez, M.D. Barnes,
Chem. Soc. 130 (2008) 12656–12662. J. Healy, H. Afsharikho, et al., Defining the architecture of
[60] M. Kalp, P.R. Carey, Carbapenems and SHV-1 beta- KPC-2 carbapenemase: identifying allosteric networks to
lactamase form different acyl-enzyme populations in crystals fight antibiotics resistance, Sci. Rep. 8 (2018), 12916.
and solution, Biochemistry. 47 (2008) 11830–11837. [76] E.I. Chudyk, M.A. Limb, C. Jones, J. Spencer, M.W. van der
[61] L.W. Tremblay, F. Fan, J.S. Blanchard, Biochemical and Kamp, A.J. Mulholland, QM/MM simulations as an assay for
structural characterization of Mycobacterium tuberculosis carbapenemase activity in class A beta-lactamases, Chem.
beta-lactamase with the carbapenems ertapenem and Commun. (Camb.) 50 (2014) 14736–14739.
doripenem, Biochemistry. 49 (2010) 3766–3773. [77] C. Bebrone, Metallo-beta-lactamases (classification, activity,
[62] P. Swaren, L. Maveyraud, X. Raquet, S. Cabantous, C. genetic organization, structure, zinc coordination) and their
Duez, J.D. Pedelacq, et al., X-ray analysis of the NMC-A superfamily, Biochem. Pharmacol. 74 (2007) 1686–1701.
beta-lactamase at 1.64-A resolution, a class A carbapene- [78] F. Baier, N. Tokuriki, Connectivity between catalytic land-
mase with broad substrate specificity, J. Biol. Chem. 273 scapes of the metallo-beta-lactamase superfamily, J. Mol.
(1998) 26714–26721. Biol. 426 (2014) 2442–2456.
[63] W. Sougakoff, G. L'Hermite, L. Pernot, T. Naas, V. Guillet, [79] I. Pettinati, J. Brem, S.Y. Lee, P.J. McHugh, C.J. Schofield,
P. Nordmann, et al., Structure of the imipenem-hydrolyzing The chemical biology of human metallo-beta-lactamase fold
class A beta-lactamase SME-1 from Serratia marcescens, proteins, Trends Biochem. Sci. 41 (2016) 338–355.
Acta Crystallogr. D Biol. Crystallogr. 58 (2002) 267–274. [80] M.W. Crowder, J. Spencer, A.J. Vila, Metallo-beta-lactamases:
[64] W. Ke, C.R. Bethel, J.M. Thomson, R.A. Bonomo, F. van den novel weaponry for antibiotic resistance in bacteria, Acc. Chem.
Akker, Crystal structure of KPC-2: insights into carbapenemase Res. 39 (2006) 721–728.
activity in class A beta-lactamases, Biochemistry. 46 (2007) [81] T. Palzkill, Metallo-beta-lactamase structure and function,
5732–5740. Ann. N. Y. Acad. Sci. 1277 (2013) 91–104.
[82] T.J. Carruthers, P.D. Carr, C.-T. Loh, C.J. Jackson, G. [97] J.S. Brooke, Stenotrophomonas maltophilia: an emerging
Otting, Iron(III) located in the dinuclear metallo-β-lactamase global opportunistic pathogen, Clin. Microbiol. Rev. 25 (2012)
IMP-1 by pseudocontact shifts, Angew Chem. Int. Ed. Engl. 2–41.
53 (2014) 14269–14272. [98] M.R. Meini, L.I. Llarrull, A.J. Vila, Overcoming differences:
[83] A. Badarau, M.I. Page, The variation of catalytic efficiency the catalytic mechanism of metallo-beta-lactamases, FEBS
of Bacillus cereus metallo-beta-lactamase with different Lett. 589 (2015) 3419–3432.
active site metal ions, Biochemistry. 45 (2006) 10654–10666. [99] E.G. Orellano, J.E. Girardini, J.A. Cricco, E.A. Ceccarelli, A.
[84] M.F. Tioni, L. Llarrull, A.A. Poeylaut-Palena, M.A. Marti, M. J. Vila, Spectroscopic characterization of a binuclear metal
Saggu, G.R. Periyannan, et al., Trapping and characteriza- site in Bacillus cereus beta-lactamase II, Biochemistry. 37
tion of a reaction intermediate in carbapenem hydrolysis by (1998) 10173–10180.
B. cereus metallo-beta-lactamase, J. Am. Chem. Soc. 130 [100] S. Wommer, S. Rival, U. Heinz, M. Galleni, J.M. Frere, N.
(2008) 15852–15863. Franceschini, et al., Substrate-activated zinc binding of
[85] H. Yang, M. Aitha, A.M. Hetrick, T.K. Richmond, D.L. metallo-beta-lactamases: physiological importance of mono-
Tierney, M.W. Crowder, Mechanistic and spectroscopic nuclear enzymes, J. Biol. Chem. 277 (2002) 24142–24147.
studies of metallo-beta-lactamase NDM-1, Biochemistry. 51 [101] L.I. Llarrull, M.F. Tioni, J. Kowalski, B. Bennett, A.J. Vila,
(2012) 3839–3847. Evidence for a dinuclear active site in the metallo-beta-
[86] A.A. Poeylaut-Palena, P.E. Tomatis, A.I. Karsisiotis, C. lactamase BcII with substoichiometric Co(II). A new model
Damblon, E.G. Mata, A.J. Vila, A minimalistic approach to for metal uptake, J. Biol. Chem. 282 (2007) 30586–30595.
identify substrate binding features in B1 Metallo-beta- [102] M.J. Hawk, R.M. Breece, C.E. Hajdin, K.M. Bender, Z. Hu,
lactamases, Bioorg. Med. Chem. Lett. 17 (2007) A.L. Costello, et al., Differential binding of Co(II) and Zn(II) to
5171–5174. metallo-beta-lactamase Bla2 from Bacillus anthracis, J. Am.
[87] C.T. Lohans, J. Brem, C.J. Schofield, New Delhi metallo- Chem. Soc. 131 (2009) 10753–10762.
beta-lactamase 1 catalyzes avibactam and aztreonam [103] A. Carfi, E. Duee, M. Galleni, J.M. Frere, O. Dideberg, 1.85 Å
hydrolysis, Antimicrob. Agents Chemother. 61 (2017). resolution structure of the zinc (II) beta-lactamase from Bacillus
[88] T. Dudev, Y.L. Lin, M. Dudev, C. Lim, First-second shell cereus, Acta Crystallogr. D Biol. Crystallogr. 54 (Pt 3) (1998)
interactions in metal binding sites in proteins: a PDB survey 313–323.
and DFT/CDM calculations, J. Am. Chem. Soc. 125 (2003) [104] T.A. Murphy, L.E. Catto, S.E. Halford, A.T. Hadfield, W.
3168–3180. Minor, T.R. Walsh, et al., Crystal structure of Pseudomonas
[89] N. Mitic, S.J. Smith, A. Neves, L.W. Guddat, L.R. Gahan, G. aeruginosa SPM-1 provides insights into variable zinc
Schenk, The catalytic mechanisms of binuclear metallohy- affinity of metallo-beta-lactamases, J. Mol. Biol. 357 (2006)
drolases, Chem. Rev. 106 (2006) 3338–3363. 890–903.
[90] B.G. Hall, S. Salipante, M. Barlow, The metallo-beta- [105] I. Garcia-Saez, J.-D. Docquier, G.M. Rossolini, O. Dideberg,
lactamases fall into two distinct phylogenetic groups, J. Mol. The three-dimensional structure of VIM-2, a Zn-beta-
Evol. 57 (2003) 249–254. lactamase from Pseudomonas aeruginosa in its reduced
[91] L.D. Sabath, E.P. Abraham, Zinc as a cofactor for cephalos- and oxidised form, J. Mol. Biol. 375 (2008) 604–611.
porinase from Bacillus cereus 569, Biochem. J. 98 (1966) [106] A.M. Davies, R.M. Rasia, A.J. Vila, B.J. Sutton, S.M.
11C-3C. Fabiane, Effect of pH on the active site of an Arg121Cys
[92] K.K. Kumarasamy, M.A. Toleman, T.R. Walsh, J. Bagaria, mutant of the metallo-beta-lactamase from Bacillus cereus:
F. Butt, R. Balakrishnan, et al., Emergence of a new implications for the enzyme mechanism, Biochemistry. 44
antibiotic resistance mechanism in India, Pakistan, and the (2005) 4841–4849.
UK: a molecular, biological, and epidemiological study, [107] J.M. Gonzalez, M.R. Meini, P.E. Tomatis, F.J.M. Martin, J.A.
Lancet Infect. Dis. 10 (2010) 597–602. Cricco, A.J. Vila, Metallo-beta-lactamases withstand low Zn
[93] M.V. Edelstein, E.N. Skleenova, O.V. Shevchenko, J.W. (II) conditions by tuning metal-ligand interactions, Nat.
D'Souza, D.V. Tapalski, I.S. Azizov, et al., Spread of Chem. Biol. 8 (2012) 698–700.
extensively resistant VIM-2-positive ST235 Pseudomonas [108] R. Bicknell, S.G. Waley, Cryoenzymology of Bacillus cereus
aeruginosa in Belarus, Kazakhstan, and Russia: a longitudinal beta-lactamase-II, Biochemistry. 24 (1985) 6876–6887.
epidemiological and clinical study, Lancet Infect. Dis. 13 [109] R. Bicknell, A. Schaffer, S.G. Waley, D.S. Auld, Changes in
(2013) 867–876. the coordination geometry of the active-site metal during
[94] J. Wachino, H. Yoshida, K. Yamane, S. Suzuki, M. catalysis of benzylpenicillin hydrolysis by Bacillus cereus
Matsui, T. Yamagishi, et al., SMB-1, a novel subclass beta-lactamase II, Biochemistry. 25 (1986) 7208–7215.
B3 metallo-beta-lactamase, associated with ISCR1 and a [110] Z.G. Wang, W. Fast, S.J. Benkovic, On the mechanism of the
class 1 integron, from a carbapenem-resistant Serratia metallo-beta-lactamase from Bacteroides fragilis, Biochemistry.
marcescens clinical isolate, Antimicrob. Agents Che- 38 (1999) 10013–10023.
mother. 55 (2011) 5143–5149. [111] Z.G. Wang, W. Fast, S.J. Benkovic, Direct observation of an
[95] D. Yong, M.A. Toleman, J. Bell, B. Ritchie, R. Pratt, H. enzyme-bound intermediate in the catalytic cycle of the
Ryley, et al., Genetic and biochemical characterization of an metallo-beta-lactamase from Bacteroides fragilis, J. Am.
acquired subgroup B3 metallo-beta-lactamase gene, Chem. Soc. 120 (1998) 10788–10789.
blaAIM-1, and its unique genetic context in Pseudomonas [112] M. Hernick, C.A. Fierke, Zinc hydrolases: the mechanisms
aeruginosa from Australia, Antimicrob. Agents Chemother. of zinc-dependent deacetylases, Arch. Biochem. Biophys.
56 (2012) 6154–6159. 433 (2005) 71–84.
[96] T.R. Walsh, L. Hall, S.J. Assinder, W.W. Nichols, S.J. [113] N. Mitic, M. Miraula, C. Selleck, K.S. Hadler, E. Uribe, M.M.
Cartwright, A.P. MacGowan, et al., Sequence analysis of the Pedroso, et al., Catalytic mechanisms of metallohydrolases
L1 metallo-beta-lactamase from Xanthomonas maltophilia, containing two metal ions, Adv. Protein Chem. Struct. Biol.
Biochim. Biophys. Acta 1218 (1994) 199–201. 97 (2014) 49–81.
[114] J. Brem, W.B. Struwe, A.M. Rydzik, H. Tarhonskaya, I. [131] C.K. Das, N.N. Nair, Hydrolysis of cephalexin and merope-
Pfeffer, E. Flashman, et al., Studying the active-site loop nem by New Delhi metallo-beta-lactamase: the substrate
movement of the Sao Paolo metallo-beta-lactamase-1, protonation mechanism is drug dependent, Phys. Chem.
Chem. Sci. 6 (2015) 956–963. Chem. Phys. 19 (2017) 13111–13121.
[115] M.N. Lisa, A.R. Palacios, M. Aitha, M.M. Gonzalez, D.M. [132] K. Zhu, J. Lu, Z. Liang, X. Kong, F. Ye, L. Jin, et al., A quantum
Moreno, M.W. Crowder, et al., A general reaction mechanism mechanics/molecular mechanics study on the hydrolysis
for carbapenem hydrolysis by mononuclear and binuclear mechanism of New Delhi metallo-beta-lactamase-1, J. Comput.
metallo-beta-lactamases, Nat. Commun. 8 (2017) 538. Aided Mol. Des. 27 (2013) 247–256.
[116] J. Spencer, J. Read, R.B. Sessions, S. Howell, G.M. [133] C. Juan, G. Torrens, M. Gonzalez-Nicolau, A. Oliver,
Blackburn, S.J. Gamblin, Antibiotic recognition by binuclear Diversity and regulation of intrinsic beta-lactamases from
metallo-beta-lactamases revealed by X‐ray crystallography, non-fermenting and other Gram-negative opportunistic
J. Am. Chem. Soc. 127 (2005) 14439–14444. pathogens, FEMS Microbiol. Rev. 41 (2017) 781–815.
[117] D.T. King, L.J. Worrall, R. Gruninger, N.C. Strynadka, New [134] J.F. Fisher, S. Mobashery, The sentinel role of peptidoglycan
Delhi metallo-beta-lactamase: structural insights into beta- recycling in the beta-lactam resistance of the Gram-negative
lactam recognition and inhibition, J. Am. Chem. Soc. 134 Enterobacteriaceae and Pseudomonas aeruginosa, Bioorg.
(2012) 11362–11365. Chem. 56 (2014) 41–48.
[118] H. Feng, J. Ding, D. Zhu, X. Liu, X. Xu, Y. Zhang, et al., [135] G.A. Jacoby, AmpC beta-lactamases, Clin. Microbiol. Rev.
Structural and mechanistic insights into NDM-1 catalyzed 22 (2009) 161–182 (Table of Contents).
hydrolysis of cephalosporins, J. Am. Chem. Soc. 136 (2014) [136] C. Oefner, A. D'Arcy, J.J. Daly, K. Gubernator, R.L. Charnas,
14694–14697. I. Heinze, et al., Refined crystal structure of beta-lactamase
[119] H. Feng, X. Liu, S. Wang, J. Fleming, D.C. Wang, W. Liu, The from Citrobacter freundii indicates a mechanism for beta-
mechanism of NDM-1-catalyzed carbapenem hydrolysis is lactam hydrolysis, Nature. 343 (1990) 284–288.
distinct from that of penicillin or cephalosporin hydrolysis, Nat. [137] Y. Kato-Toma, T. Iwashita, K. Masuda, Y. Oyama, M.
Commun. 8 (2017) 2242. Ishiguro, pKa measurements from nuclear magnetic reso-
[120] G. Garau, C. Bebrone, C. Anne, M. Galleni, J.M. Frere, O.A. nance of tyrosine-150 in class C beta-lactamase, Biochem.
Dideberg, Metallo-beta-lactamase enzyme in action: crystal J. 371 (2003) 175–181.
structures of the monozinc carbapenemase CphA and its [138] Y. Chen, G. Minasov, T.A. Roth, F. Prati, B.K. Shoichet, The
complex with biapenem, J. Mol. Biol. 345 (2005) 785–795. deacylation mechanism of AmpC beta-lactamase at ultra-
[121] Y. Kim, M.A. Cunningham, J. Mire, C. Tesar, J. Sacchettini, high resolution, J. Am. Chem. Soc. 128 (2006) 2970–2976.
A. Joachimiak, NDM-1, the ultimate promiscuous enzyme: [139] E. Lobkovsky, P.C. Moews, H. Liu, H. Zhao, J.M. Frere, J.R.
substrate recognition and catalytic mechanism, FASEB J. Knox, Evolution of an enzyme activity: crystallographic
27 (2013) 1917–1927. structure at 2-A resolution of cephalosporinase from the
[122] J. Wachino, Y. Yamaguchi, S. Mori, W. Jin, K. Kimura, H. ampC gene of Enterobacter cloacae P99 and comparison
Kurosaki, et al., Structural insights into recognition of hydrolyzed with a class A penicillinase, Proc. Natl. Acad. Sci. U. S. A.
carbapenems and inhibitors by subclass B3 metallo-beta- 90 (1993) 11257–11261.
lactamase SMB-1, Antimicrob. Agents Chemother. 60 (2016) [140] N. Diaz, D. Suarez, T.L. Sordo, Molecular dynamics
4274–4282. simulations of class C beta-lactamase from Citrobacter
[123] H. Zhang, Q. Hao, Crystal structure of NDM-1 reveals a freundii: insights into the base catalyst for acylation,
common {beta}-lactam hydrolysis mechanism, FASEB J. 25 Biochemistry. 45 (2006) 439–451.
(8) (2011) 2574–2582. [141] R. Tripathi, N.N. Nair, Mechanism of acyl-enzyme complex
[124] H. Zhang, G. Ma, Y. Zhu, L. Zeng, A. Ahmad, C. Wang, et al., formation from the Henry–Michaelis complex of class C
Active-site conformational fluctuations promote the enzymatic beta-lactamases with beta-lactam antibiotics, J. Am. Chem.
activity of NDM-1, Antimicrob. Agents Chemother. 62 (2018). Soc. 135 (2013) 14679–14690.
[125] R. Salimraj, P. Hinchliffe, M. Kosmopoulou, J.M. Tyrrell, J. [142] B.M. Beadle, I. Trehan, P.J. Focia, B.K. Shoichet,
Brem, S.S. van Berkel, et al., Crystal structures of VIM-1 Structural milestones in the reaction pathway of an amide
complexes explain active site heterogeneity in VIM-class hydrolase: substrate, acyl, and product complexes of
metallo-beta-lactamases, FEBS J. 286 (1) (2018) 169–183. cephalothin with AmpC beta-lactamase, Structure. 10
[126] J.E. Raczynska, I.G. Shabalin, W. Minor, A. Wlodawer, M. (2002) 413–424.
Jaskolski, A close look onto structural models and primary [143] A. Dubus, S. Normark, M. Kania, M.G. Page, The role of
ligands of metallo-beta-lactamases, Drug Resist. Updat. 40 tyrosine 150 in catalysis of beta-lactam hydrolysis by
(2018) 1–12. AmpC beta-lactamase from Escherichia coli investigated
[127] J.D. Garrity, B. Bennett, M.W. Crowder, Direct evidence that by site-directed mutagenesis, Biochemistry. 33 (1994)
the reaction intermediate of metallo-beta-lactamase L1 is 8577–8586.
metal bound, Biochemistry. 44 (2005) 1078–1087. [144] B.F. Gherman, S.D. Goldberg, V.W. Cornish, R.A. Friesner,
[128] Z. Sun, L. Hu, B. Sankaran, B.V.V. Prasad, T. Palzkill, Mixed quantum mechanical/molecular mechanical (QM/
Differential active site requirements for NDM-1 beta- MM) study of the deacylation reaction in a penicillin binding
lactamase hydrolysis of carbapenem versus penicillin and protein (PBP) versus in a class C beta-lactamase, J. Am.
cephalosporin antibiotics, Nat. Commun. 9 (2018) 4524. Chem. Soc. 126 (2004) 7652–7664.
[129] C. Melia, S. Ferrer, V. Moliner, J. Bertran, Theoretical studies [145] D. Monnaie, A. Dubus, J.M. Frere, The role of lysine-67 in a
of the hydrolysis of antibiotics catalyzed by a metallo-beta- class C beta-lactamase is mainly electrostatic, Biochem. J.
lactamase, Arch. Biochem. Biophys. 582 (2015) 116–126. 302 (Pt 1) (1994) 1–4.
[130] R. Tripathi, A. Nair, Mechanism of meropenem hydrolysis [146] Y. Chen, A. McReynolds, B.K. Shoichet, Re-examining the
by New Delhi metallo-b-lactamase, ACS Catal. 5 (2015) role of Lys67 in class C beta-lactamase catalysis, Protein
2577–2586. Sci. 18 (2009) 662–669.
[147] R. Tripathi, N.N. Nair, Deacylation mechanism and kinetics lactamases: correlation with plasmid compatibility, J. Bacteriol.
of acyl-enzyme complex of class C beta-lactamase and 117 (1974) 56–62.
cephalothin, J. Phys. Chem. B 120 (2016) 2681–2690. [164] L. Maveyraud, D. Golemi, L.P. Kotra, S. Tranier, S.
[148] A. Patera, L.C. Blaszczak, B.K. Shoichet, Crystal structures Vakulenko, S. Mobashery, et al., Insights into class D
of substrate and inhibitor complexes with AmpC beta- beta-lactamases are revealed by the crystal structure of the
lactamase: possible implications for substrate-assisted OXA10 enzyme from Pseudomonas aeruginosa, Struct.
catalysis, J. Am. Chem. Soc. 122 (2000) 10504–10512. Fold. Des. 8 (2000) 1289–1298.
[149] A. Bulychev, I. Massova, K. Miyashita, S. Mobashery, [165] M. Paetzel, F. Danel, L. de Castro, S.C. Mosimann, M.G.
Nuances of mechanisms and their implications for evolution Page, N.C. Strynadka, Crystal structure of the class D beta-
of the versatile β-lactamase activity: from biosynthetic lactamase OXA-10, Nat. Struct. Biol. 7 (2000) 918–925.
enzymes to drug resistance factors, J. Am. Chem. Soc. 119 [166] D. Golemi, L. Maveyraud, S. Vakulenko, J.P. Samama, S.
(1997) 7619–7625. Mobashery, Critical involvement of a carbamylated lysine in
[150] Y. Xu, G. Soto, K.R. Hirsch, R.F. Pratt, Kinetics and mechanism catalytic function of class D beta-lactamases, Proc. Natl.
of the hydrolysis of depsipeptides catalyzed by the beta- Acad. Sci. U. S. A. 98 (2001) 14280–14285.
lactamase of Enterobacter cloacae P99, Biochemistry. 35 [167] D. Golemi-Kotra, J.Y. Cha, S.O. Meroueh, S.B. Vakulenko,
(1996) 3595–3603. S. Mobashery, Resistance to beta-lactam antibiotics and its
[151] W.H. Goessens, A.K. van der Bij, R. van Boxtel, J.D. Pitout, mediation by the sensor domain of the transmembrane
P. van Ulsen, D.C. Melles, et al., Antibiotic trapping by BlaR signaling pathway in Staphylococcus aureus, J. Biol.
plasmid-encoded CMY-2 beta-lactamase combined with Chem. 278 (2003) 18419–18425.
reduced outer membrane permeability as a mechanism of [168] K. Thumanu, J. Cha, J.F. Fisher, R. Perrins, S. Mobashery, C.
carbapenem resistance in Escherichia coli, Antimicrob. Wharton, Discrete steps in sensing of beta-lactam antibiotics by
Agents Chemother. 57 (2013) 3941–3949. the BlaR1 protein of the methicillin-resistant Staphylococcus
[152] H. Mammeri, H. Guillon, F. Eb, P. Nordmann, Phenotypic and aureus bacterium, Proc. Natl. Acad. Sci. U. S. A. 103 (2006)
biochemical comparison of the carbapenem-hydrolyzing 10630–10635.
activities of five plasmid-borne AmpC beta-lactamases, [169] L. Maveyraud, D. Golemi-Kotra, A. Ishiwata, O. Meroueh, S.
Antimicrob. Agents Chemother. 54 (2010) 4556–4560. Mobashery, J.P. Samama, High-resolution X‐ray structure
[153] A.M. Queenan, W. Shang, R. Flamm, K. Bush, Hydro- of an acyl-enzyme species for the class D OXA-10 beta-
lysis and inhibition profiles of beta-lactamases from lactamase, J. Am. Chem. Soc. 124 (2002) 2461–2465.
molecular classes A to D with doripenem, imipenem, and [170] K.D. Schneider, C.R. Bethel, A.M. Distler, A.M. Hujer, R.A.
meropenem, Antimicrob. Agents Chemother. 54 (2010) Bonomo, D.A. Leonard, Mutation of the active site carboxy-
565–569. lysine (K70) of OXA-1 beta-lactamase results in a deacylation-
[154] B.M. Beadle, B.K. Shoichet, Structural basis for imipenem deficient enzyme, Biochemistry. 48 (2009) 6136–6145.
inhibition of class C beta-lactamases, Antimicrob. Agents [171] J. Sgrignani, G. Grazioso, M. De Amici, Insight into the
Chemother. 46 (2002) 3978–3980. mechanism of hydrolysis of meropenem by OXA-23 serine-
[155] J.Y. Kim, H.I. Jung, Y.J. An, J.H. Lee, S.J. Kim, S.H. Jeong, beta-lactamase gained by quantum mechanics/molecular
et al., Structural basis for the extended substrate spectrum mechanics calculations, Biochemistry. 55 (2016)
of CMY-10, a plasmid-encoded class C beta-lactamase, 5191–5200.
Mol. Microbiol. 60 (2006) 907–916. [172] N.T. Antunes, T.L. Lamoureaux, M. Toth, N.K. Stewart, H.
[156] J.H. Jeon, M.K. Hong, J.H. Lee, J.J. Lee, K.S. Park, A.M. Frase, S.B. Vakulenko, Class D beta-lactamases: are they
Karim, et al., Structure of ADC-68, a novel carbapenem- all carbapenemases? Antimicrob. Agents Chemother. 58
hydrolyzing class C extended-spectrum beta-lactamase (2014) 2119–2125.
isolated from Acinetobacter baumannii, Acta Crystallogr. D [173] J.F. Turton, M.E. Ward, N. Woodford, M.E. Kaufmann, R.
Biol. Crystallogr. 70 (2014) 2924–2936. Pike, D.M. Livermore, et al., The role of ISAba1 in
[157] D.A. Leonard, R.A. Bonomo, R.A. Powers, Class D beta- expression of OXA carbapenemase genes in Acinetobacter
lactamases: a reappraisal after five decades, Acc. Chem. baumannii, FEMS Microbiol. Lett. 258 (2006) 72–77.
Res. 46 (2013) 2407–2415. [174] A. Bertini, L. Poirel, P.D. Mugnier, L. Villa, P. Nordmann, A.
[158] B.A. Evans, S.G. Amyes, OXA beta-lactamases, Clin. Carattoli, Characterization and PCR-based replicon typing
Microbiol. Rev. 27 (2014) 241–263. of resistance plasmids in Acinetobacter baumannii, Anti-
[159] A.M. Philippon, G.C. Paul, G.A. Jacoby, Properties of PSE-2 microb. Agents Chemother. 54 (2010) 4168–4177.
beta-lactamase and genetic basis for its production in [175] K.J. Towner, B. Evans, L. Villa, K. Levi, A. Hamouda, S.G.
Pseudomonas aeruginosa, Antimicrob. Agents Chemother. Amyes, et al., Distribution of intrinsic plasmid replicase genes
24 (1983) 362–369. and their association with carbapenem-hydrolyzing class D
[160] L. Poirel, P. Nordmann, Carbapenem resistance in Acine- beta-lactamase genes in European clinical isolates of
tobacter baumannii: mechanisms and epidemiology, Clin. Acinetobacter baumannii, Antimicrob. Agents Chemother.
Microbiol. Infect. 12 (2006) 826–836. 55 (2011) 2154–2159.
[161] L. Poirel, A. Potron, P. Nordmann, OXA-48-like carbape- [176] L. Poirel, R.A. Bonnin, P. Nordmann, Genetic features of
nemases: the phantom menace, J. Antimicrob. Chemother. the widespread plasmid coding for the carbapenemase OXA-
67 (2012) 1597–1606. 48, Antimicrob. Agents Chemother. 56 (2012) 559–562.
[162] M. Toth, N.T. Antunes, N.K. Stewart, H. Frase, M. [177] A. Stewart, P. Harris, A. Henderson, D. Paterson, Treatment
Bhattacharya, C.A. Smith, et al., Class D beta-lactamases of infections by OXA-48-producing Enterobacteriaceae,
do exist in Gram-positive bacteria, Nat. Chem. Biol. 12 (2016) Antimicrob. Agents Chemother. 62 (2018).
9–14. [178] M. Tomic Paradzik, D. Drenjancevic, A. Presecki-Stanko,
[163] R.W. Hedges, N. Datta, P. Kontomichalou, J.T. Smith, J. Kopic, J. Talapko, G. Zarfel, et al., Hidden carbapenem
Molecular specificities of R factor-determined beta- resistance in OXA-48 and extended-spectrum beta-
lactamase-positive Escherichia coli, Microb. Drug Resist. [194] A. Philippon, R. Labia, G. Jacoby, Extended-spectrum beta-
(2019) https://doi.org/10.1089/mdr.2018.0309. lactamases, Antimicrob. Agents Chemother. 33 (1989)
[179] E. Santillana, A. Beceiro, G. Bou, A. Romero, Crystal 1131–1136.
structure of the carbapenemase OXA-24 reveals insights [195] A.R. White, C. Kaye, J. Poupard, R. Pypstra, G. Woodnutt,
into the mechanism of carbapenem hydrolysis, Proc. Natl. B. Wynne, Augmentin (amoxicillin/clavulanate) in the
Acad. Sci. U. S. A. 104 (2007) 5354–5359. treatment of community-acquired respiratory tract infection:
[180] K.D. Schneider, C.J. Ortega, N.A. Renck, R.A. Bonomo, R. a review of the continuing development of an innovative
A. Powers, D.A. Leonard, Structures of the class D antimicrobial agent, J. Antimicrob. Chemother. 53 (Suppl. 1)
carbapenemase OXA-24 from Acinetobacter baumannii in (2004) i3–20.
complex with doripenem, J. Mol. Biol. 406 (2011) 583–594. [196] K.M. Papp-Wallace, C.R. Bethel, A.M. Distler, C.
[181] C.M. June, T.J. Muckenthaler, E.C. Schroder, Z.L. Klamer, Kasuboski, M. Taracila, R.A. Bonomo, Inhibitor resistance
Z. Wawrzak, R.A. Powers, et al., The structure of a in the KPC-2 beta-lactamase, a preeminent property of this
doripenem-bound OXA-51 class D beta-lactamase variant class A beta-lactamase, Antimicrob. Agents Chemother. 54
with enhanced carbapenemase activity, Protein Sci. 25 (2010) 890–897.
(2016) 2152–2163. [197] M.E. Falagas, A.D. Mavroudis, K.Z. Vardakas, The antibiotic
[182] C.A. Smith, N.T. Antunes, N.K. Stewart, H. Frase, M. pipeline for multi-drug resistant Gram negative bacteria: what
Toth, K.A. Kantardjieff, et al., Structural basis for can we expect? Expert Rev. Anti-Infect. Ther. 14 (2016)
enhancement of carbapenemase activity in the OXA-51 747–763.
family of class D beta-lactamases, ACS Chem. Biol. 10 [198] U. Theuretzbacher, S. Gottwalt, P. Beyer, M. Butler, L.
(2015) 1791–1796. Czaplewski, C. Lienhardt, et al., Analysis of the clinical
[183] S. Pratap, M. Katiki, P. Gill, P. Kumar, D. Golemi-Kotra, antibacterial and antituberculosis pipeline, Lancet Infect.
Active-site plasticity is essential to carbapenem hydrolysis by Dis. 19 (2) (2018) e40–e50.
OXA-58 class D beta-lactamase of Acinetobacter baumannii, [199] P.G. Ambrose, O. Lomovskaya, D.C. Griffith, M.N. Dudley,
Antimicrob. Agents Chemother. 60 (2016) 75–86. B. VanScoy, Beta-lactamase inhibitors: what you really
[184] C.A. Smith, N.T. Antunes, M. Toth, S.B. Vakulenko, Crystal need to know, Curr. Opin. Pharmacol. 36 (2017) 86–93.
structure of carbapenemase OXA-58 from Acinetobacter [200] C.M. Rotondo, G.D. Wright, Inhibitors of metallo-beta-
baumannii, Antimicrob. Agents Chemother. 58 (2014) lactamases, Curr. Opin. Microbiol. 39 (2017) 96–105.
2135–2143. [201] J.D. Docquier, S. Mangani, An update on beta-lactamase
[185] C.A. Smith, N.T. Antunes, N.K. Stewart, M. Toth, M. inhibitor discovery and development, Drug Resist. Updat.
Kumarasiri, M. Chang, et al., Structural basis for carbapene- 36 (2018) 13–29.
mase activity of the OXA-23 beta-lactamase from Acineto- [202] L.C. Ju, Z. Cheng, W. Fast, R.A. Bonomo, M.W. Crowder,
bacter baumannii, Chem. Biol. 20 (2013) 1107–1115. The continuing challenge of metallo-beta-lactamase inhibi-
[186] N.K. Stewart, C.A. Smith, N.T. Antunes, M. Toth, S.B. tion: mechanism matters, Trends Pharmacol. Sci. 39 (2018)
Vakulenko, Role of the hydrophobic bridge in the carbape- 635–647.
nemase activity of class D beta-lactamases, Antimicrob. [203] G.G. Zhanel, C.D. Lawson, H. Adam, F. Schweizer, S.
Agents Chemother. 63 (2) (2018) e02191-18. Zelenitsky, P.R. Lagace-Wiens, et al., Ceftazidime-avibactam:
[187] J.D. Docquier, V. Calderone, F. De Luca, M. Benvenuti, F. a novel cephalosporin/beta-lactamase inhibitor combination,
Giuliani, L. Bellucci, et al., Crystal structure of the OXA-48 Drugs. 73 (2013) 159–177.
beta-lactamase reveals mechanistic diversity among [204] D.E. Ehmann, H. Jahic, P.L. Ross, R.F. Gu, J. Hu, G. Kern,
class D carbapenemases, Chem. Biol. 16 (2009) et al., Avibactam is a covalent, reversible, non-beta-lactam
540–547. beta-lactamase inhibitor, Proc. Natl. Acad. Sci. U. S. A. 109
[188] C.M. June, B.C. Vallier, R.A. Bonomo, D.A. Leonard, R.A. (2012) 11663–11668.
Powers, Structural origins of oxacillinase specificity in class [205] F.F. Tuon, J.L. Rocha, M.R. Formigoni-Pinto, Pharmacolog-
D beta-lactamases, Antimicrob. Agents Chemother. 58 ical aspects and spectrum of action of ceftazidime-avibactam:
(2014) 333–341. a systematic review, Infection. 46 (2018) 165–181.
[189] F. De Luca, M. Benvenuti, F. Carboni, C. Pozzi, G.M. [206] T.A. Blizzard, H. Chen, S. Kim, J. Wu, R. Bodner, C. Gude,
Rossolini, S. Mangani, et al., Evolution to carbapenem- et al., Discovery of MK-7655, a beta-lactamase inhibitor for
hydrolyzing activity in noncarbapenemase class D beta- combination with Primaxin(R), Bioorg. Med. Chem. Lett. 24
lactamase OXA-10 by rational protein design, Proc. Natl. (2014) 780–785.
Acad. Sci. U. S. A. 108 (2011) 18424–18429. [207] K.M. Papp-Wallace, N.Q. Nguyen, M.R. Jacobs, C.R.
[190] C.T. Lohans, E. van Groesen, K. Kumar, C.L. Tooke, J. Bethel, M.D. Barnes, V. Kumar, et al., Strategic approaches
Spencer, R.S. Paton, et al., A new mechanism for beta- to overcome resistance against Gram-negative pathogens
lactamases: class D enzymes degrade 1beta-methyl using beta-lactamase inhibitors and beta-lactam en-
carbapenems through lactone formation, Angew. Chem. hancers: activity of three novel diazabicyclooctanes WCK
Int. Ed. Engl. 57 (2018) 1282–1285. 5153, Zidebactam (WCK 5107), and WCK 4234, J. Med.
[191] S.M. Drawz, R.A. Bonomo, Three decades of beta-lactamase Chem. 61 (2018) 4067–4086.
inhibitors, Clin. Microbiol. Rev. 23 (2010) 160–201. [208] B. Moya, I.M. Barcelo, S. Bhagwat, M. Patel, G. Bou, K.M.
[192] C. Reading, M. Cole, Clavulanic acid: a beta-lactamase- Papp-Wallace, et al., WCK 5107 (Zidebactam) and WCK 5153
inhiting beta-lactam from Streptomyces clavuligerus, Anti- are novel inhibitors of PBP2 showing potent “beta-lactam
microb. Agents Chemother. 11 (1977) 852–857. enhancer” activity against Pseudomonas aeruginosa, including
[193] C. Bebrone, P. Lassaux, L. Vercheval, J.S. Sohier, A. multidrug-resistant metallo-beta-lactamase-producing high-risk
Jehaes, E. Sauvage, et al., Current challenges in antimi- clones, Antimicrob. Agents Chemother. 61 (2017).
crobial chemotherapy: focus on ss-lactamase inhibition, [209] D.E. Ehmann, H. Jahic, P.L. Ross, R.F. Gu, J. Hu, T.F.
Drugs. 70 (2010) 651–679. Durand-Reville, et al., Kinetics of avibactam inhibition
against Class A, C, and D beta-lactamases, J. Biol. Chem. characterization of ceftazidime/avibactam-resistant mutants

288 (2013) 27960–27971. in Pseudomonas aeruginosa strains containing derepressed
[210] A.M. King, D.T. King, S. French, E. Brouillette, A. Asli, J.A. AmpC, J. Antimicrob. Chemother. 70 (2015) 1650–1658.
Alexander, et al., Structural and kinetic characterization of [224] D.M. Livermore, S. Mushtaq, M. Doumith, D. Jamrozy, W.W.
diazabicyclooctanes as dual inhibitors of both serine-beta- Nichols, N. Woodford, Selection of mutants with resistance
lactamases and penicillin-binding proteins, ACS Chem. or diminished susceptibility to ceftazidime/avibactam
Biol. 11 (2016) 864–868. from ESBL- and AmpC-producing Enterobacteriaceae,
[211] K.M. Papp-Wallace, M.D. Barnes, J. Alsop, M.A. Taracila, J. Antimicrob. Chemother. 73 (2018) 3336–3345.
C.R. Bethel, S.A. Becka, et al., Relebactam is a potent [225] A. Both, H. Buttner, J. Huang, M. Perbandt, C. Belmar
inhibitor of the KPC-2 beta-lactamase and restores imipe- Campos, M. Christner, et al., Emergence of ceftazidime/
nem susceptibility in KPC-producing enterobacteriaceae, avibactam non-susceptibility in an MDR Klebsiella pneu-
Antimicrob. Agents Chemother. 62 (2018). moniae isolate, J. Antimicrob. Chemother. 72 (2017)
[212] H. Xu, S. Hazra, J.S. Blanchard, NXL104 irreversibly 2483–2488.
inhibits the beta-lactamase from Mycobacterium tuberculo- [226] M.J. Giddins, N. Macesic, M.K. Annavajhala, S. Stump, S.
sis, Biochemistry. 51 (2012) 4551–4557. Khan, T.H. McConville, et al., Successive emergence of
[213] S.D. Lahiri, S. Mangani, T. Durand-Reville, M. Benvenuti, F. ceftazidime–avibactam resistance through distinct genomic
De Luca, G. Sanyal, et al., Structural insight into potent adaptations in blaKPC-2-harboring Klebsiella pneumoniae
broad-spectrum inhibition with reversible recyclization sequence type 307 isolates, Antimicrob. Agents Chemother.
mechanism: avibactam in complex with CTX-M-15 and 62 (2018).
Pseudomonas aeruginosa AmpC beta-lactamases, Anti- [227] G. Haidar, C.J. Clancy, R.K. Shields, B. Hao, S. Cheng, M.H.
microb. Agents Chemother. 57 (2013) 2496–2505. Nguyen, Mutations in blaKPC-3 that confer ceftazidime–
[214] N.P. Krishnan, N.Q. Nguyen, K.M. Papp-Wallace, R.A. avibactam resistance encode novel KPC-3 variants that
Bonomo, F. van den Akker, Inhibition of Klebsiella beta- function as extended-spectrum beta-lactamases, Antimicrob.
lactamases (SHV-1 and KPC-2) by avibactam: a structural Agents Chemother. 61 (2017).
study, PLoS One 10 (2015), e0136813. [228] H. Fu, H. Fang, J. Sun, H. Wang, A. Liu, J. Sun, et al.,
[215] S.D. Lahiri, S. Mangani, H. Jahic, M. Benvenuti, T.F. Durand- Boronic acid-based enzyme inhibitors: a review of recent
Reville, F. De Luca, et al., Molecular basis of selective progress, Curr. Med. Chem. 21 (2014) 3271–3280.
inhibition and slow reversibility of avibactam against class D [229] L.J. Rojas, M.A. Taracila, K.M. Papp-Wallace, C.R. Bethel,
carbapenemases: a structure-guided study of OXA-24 and E. Caselli, C. Romagnoli, et al., Boronic acid transition state
OXA-48, ACS Chem. Biol. 10 (2015) 591–600. inhibitors active against KPC and other class a beta-
[216] C.T. Lohans, D.Y. Wang, C. Jorgensen, S.T. Cahill, I.J. lactamases: structure–activity relationships as a guide to
Clifton, M.A. McDonough, et al., (13)C-carbamylation as a inhibitor design, Antimicrob. Agents Chemother. 60 (2016)
mechanistic probe for the inhibition of class D beta- 1751–1759.
lactamases by avibactam and halide ions, Org. Biomol. [230] D. Tondi, A. Venturelli, R. Bonnet, C. Pozzi, B.K. Shoichet,
Chem. 15 (2017) 6024–6032. M.P. Costi, Targeting class A and C serine beta-lactamases
[217] D.T. King, A.M. King, S.M. Lal, G.D. Wright, N.C. Strynadka, with a broad-spectrum boronic acid derivative, J. Med.
Molecular mechanism of avibactam-mediated beta- Chem. 57 (2014) 5449–5458.
lactamase inhibition, ACS Infect. Dis. 1 (2015) 175–184. [231] S.J. Hecker, K.R. Reddy, M. Totrov, G.C. Hirst, O.
[218] C. Pozzi, F. Di Pisa, F. De Luca, M. Benvenuti, J.D. Lomovskaya, D.C. Griffith, et al., Discovery of a cyclic
Docquier, S. Mangani, Atomic-resolution structure of a class boronic acid beta-lactamase inhibitor (RPX7009) with utility
C beta-lactamase and its complex with avibactam, Chem- vs class A serine carbapenemases, J. Med. Chem. 58
MedChem. 13 (2018) 1437–1446. (2015) 3682–3692.
[219] M.D. Barnes, M.L. Winkler, M.A. Taracila, M.G. Page, E. [232] G. Wu, E. Cheon, Meropenem–vaborbactam for the
Desarbre, B.N. Kreiswirth, et al., Klebsiella pneumoniae treatment of complicated urinary tract infections including
carbapenemase-2 (KPC-2), substitutions at ambler position acute pyelonephritis, Expert. Opin. Pharmacother. 19
Asp179, and resistance to ceftazidime-avibactam: unique (2018) 1495–1502.
antibiotic-resistant phenotypes emerge from beta- [233] O. Lomovskaya, D. Sun, D. Rubio-Aparicio, K. Nelson, R.
lactamase protein engineering, MBio. 8 (2017). Tsivkovski, D.C. Griffith, et al., Vaborbactam: spectrum of
[220] D.M. Livermore, M. Warner, D. Jamrozy, S. Mushtaq, W.W. beta-lactamase inhibition and impact of resistance mecha-
Nichols, N. Mustafa, et al., In vitro selection of ceftazidime– nisms on activity in Enterobacteriaceae, Antimicrob. Agents
avibactam resistance in Enterobacteriaceae with KPC-3 Chemother. 61 (2017).
carbapenemase, Antimicrob. Agents Chemother. 59 (2015) [234] A.M. King, S.A. Reid-Yu, W. Wang, D.T. King, G. De
5324–5330. Pascale, N.C. Strynadka, et al., Aspergillomarasmine A
[221] K.M. Papp-Wallace, M.L. Winkler, M.A. Taracila, R.A. overcomes metallo-beta-lactamase antibiotic resistance,
Bonomo, Variants of beta-lactamase KPC-2 that are Nature. 510 (2014) 503–506.
resistant to inhibition by avibactam, Antimicrob. Agents [235] P. Hinchliffe, M.M. Gonzalez, M.F. Mojica, J.M. Gonzalez, V.
Chemother. 59 (2015) 3710–3717. Castillo, C. Saiz, et al., Cross-class metallo-beta-lactamase
[222] S.D. Lahiri, M.R. Johnstone, P.L. Ross, R.E. McLaughlin, N. inhibition by bisthiazolidines reveals multiple binding modes,
B. Olivier, R.A. Alm, Avibactam and class C beta- Proc. Natl. Acad. Sci. U. S. A. 113 (2016) E3745–E3754.
lactamases: mechanism of inhibition, conservation of the [236] P. Hinchliffe, C.A. Tanner, A.P. Krismanich, G. Labbe, V.J.
binding pocket, and implications for resistance, Antimicrob. Goodfellow, L. Marrone, et al., Structural and Kinetic studies
Agents Chemother. 58 (2014) 5704–5713. of the potent inhibition of metallo-beta-lactamases by 6-
[223] S.D. Lahiri, G.K. Walkup, J.D. Whiteaker, T. Palmer, K. phosphonomethylpyridine-2-carboxylates, Biochemistry. 57
McCormack, M.A. Tanudra, et al., Selection and molecular (2018) 1880–1892.
[237] G.B. Li, M.I. Abboud, J. Brem, H. Someya, C.T. Lohans, S. [247] R. Wang, T.P. Lai, P. Gao, H. Zhang, P.L. Ho, P.C. Woo, et al.,
Y. Yang, et al., NMR-filtered virtual screening leads to non- Bismuth antimicrobial drugs serve as broad-spectrum metallo-
metal chelating metallo-beta-lactamase inhibitors, Chem. beta-lactamase inhibitors, Nat. Commun. 9 (2018) 439.
Sci. 8 (2017) 928–937. [248] H.K. Allen, L.A. Moe, J. Rodbumrer, A. Gaarder, J.
[238] J. Brem, R. Cain, S. Cahill, M.A. McDonough, I.J. Clifton, J. Handelsman, Functional metagenomics reveals diverse
C. Jimenez-Castellanos, et al., Structural basis of metallo- beta-lactamases in a remote Alaskan soil, ISME J. 3 (2009)
beta-lactamase, serine-beta-lactamase and penicillin- 243–251.
binding protein inhibition by cyclic boronates, Nat. Commun. [249] C. Brandt, S.D. Braun, C. Stein, P. Slickers, R. Ehricht, M.
7 (2016), 12406. W. Pletz, et al., In silico serine beta-lactamases analysis
[239] S.T. Cahill, R. Cain, D.Y. Wang, C.T. Lohans, D.W. Wareham, reveals a huge potential resistome in environmental and
H.P. Oswin, et al., Cyclic boronates inhibit all classes of beta- pathogenic species, Sci. Rep. 7 (2017), 43232.
lactamases, Antimicrob. Agents Chemother. 61 (2017). [250] I.C. Materon, A.M. Queenan, T.M. Koehler, K. Bush, T.
[240] K. Calvopina, P. Hinchliffe, J. Brem, K.J. Heesom, S. Palzkill, Biochemical characterization of beta-lactamases
Johnson, R. Cain, et al., Structural/mechanistic insights Bla1 and Bla2 from Bacillus anthracis, Antimicrob. Agents
into the efficacy of nonclassical beta-lactamase inhibitors Chemother. 47 (2003) 2040–2042.
against extensively drug resistant Stenotrophomonas [251] N.T. Antunes, H. Frase, M. Toth, S.B. Vakulenko, The class
maltophilia clinical isolates, Mol. Microbiol. 106 (2017) A beta-lactamase FTU-1 is native to Francisella tularensis,
492–504. Antimicrob. Agents Chemother. 56 (2012) 666–671.
[241] J.D. Buynak, H. Chen, L. Vogeti, V.R. Gadhachanda, C.A. [252] L.B. Randall, K. Dobos, K.M. Papp-Wallace, R.A. Bonomo,
Buchanan, T. Palzkill, et al., Penicillin-derived inhibitors that H.P. Schweizer, Membrane-bound PenA beta-lactamase of
simultaneously target both metallo- and serine-b-lactamases, Burkholderia pseudomallei, Antimicrob. Agents Chemother.
Bioorg. Med. Chem. Lett. 14 (2004) 1299–1304. 60 (2015) 1509–1514.
[242] J.W. Johnson, M. Gretes, V.J. Goodfellow, L. Marrone, M.L. [253] D. von Stetten, T. Giraud, P. Carpentier, F. Sever, M.
Heynen, N.C. Strynadka, et al., Cyclobutanone analogues Terrien, F. Dobias, et al., In crystallo optical spectroscopy
of beta-lactams revisited: insights into conformational (icOS) as a complementary tool on the macromolecular
requirements for inhibition of serine- and metallo-beta- crystallography beamlines of the ESRF, Acta Crystallogr. D
lactamases, J. Am. Chem. Soc. 132 (2010) 2558–2560. Biol. Crystallogr. 71 (2015) 15–26.
[243] M.I. Abboud, M. Kosmopoulou, A.P. Krismanich, J.W. Johnson, [254] K.D. Schneider, M.E. Karpen, R.A. Bonomo, D.A. Leonard, R.
P. Hinchliffe, J. Brem, et al., Cyclobutanone mimics of A. Powers, The 1.4 A crystal structure of the class D beta-
intermediates in metallo-beta-lactamase catalysis, Chemistry. lactamase OXA-1 complexed with doripenem, Biochemistry.
24 (2018) 5734–5737. 48 (2009) 11840–11847.
[244] H. Kurosaki, Y. Yamaguchi, T. Higashi, K. Soga, S. [255] R.K. Shields, L. Chen, S. Cheng, K.D. Chavda, E.G. Press,
Matsueda, H. Yumoto, et al., Irreversible inhibition of A. Snyder, et al., Emergence of ceftazidime–avibactam
metallo-beta-lactamase (IMP-1) by 3-(3-mercaptopropio- resistance due to plasmid-borne blaKPC-3 mutations during
nylsulfanyl)propionic acid pentafluorophenyl ester, Angew. treatment of carbapenem-resistant Klebsiella pneumoniae
Chem. Int. Ed. Engl. 44 (2005) 3861–3864. infections, Antimicrob. Agents Chemother. 61 (2017).
[245] J. Chiou, S. Wan, K.F. Chan, P.K. So, D. He, E.W. Chan, [256] T.F. Durand-Réville, S. Guler, J. Comita-Prevoir, B. Chen,
et al., Ebselen as a potent covalent inhibitor of New Delhi N. Bifulco, H. Huynh, et al., ETX2514 is a broad-spectrum
metallo-beta-lactamase (NDM-1), Chem. Commun. β-lactamase inhibitor for the treatment of drug-resistant
(Camb.) 51 (2015) 9543–9546. Gram-negative bacteria including Acinetobacter baumannii,
[246] C. Chen, Y. Xiang, K.W. Yang, Y. Zhang, W.M. Wang, J.P. Nat. Microbiol. 2 (2017) 17104, https://doi.org/10.1038/
Su, et al., A protein structure-guided covalent scaffold nmicrobiol.2017.104.
selectively targets the B1 and B2 subclass metallo-beta-
lactamases, Chem. Commun. (Camb.) 54 (2018)
4802–4805.

β-Lactamases and β-Lactamase Inhibitors in the 21st Century

Uploaded by

Copyright:

Available Formats

β-Lactamases and β-Lactamase Inhibitors in the 21st Century

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

β-Lactamases and β-Lactamase Inhibitors in the 21st Century

Uploaded by

Copyright:

Available Formats

Review

β-Lactamases and β-Lactamase Inhibitors in

Catherine L. Tooke † , Philip Hinchliffe † , Eilis C. Bragginton, Charlotte K. Colenso,

Correspondence to James Spencer: [email protected].

but that an incoming water molecule at the bridging Class C β-lactamases

group of the carbapenem 6α-hydroxyethyl substituent. associated infections by β-lactamase-producing

distinctive biochemical properties may mobilize in

against Class A, C, and D beta-lactamases, J. Biol. Chem. characterization of ceftazidime/avibactam-resistant mutants

You might also like