Structural and Functional Study of the Klebsiella pneumoniae VapBC Toxin-Antitoxin System, including the Development of an Inhibitor That Activates VapC

Article abstract of DOI:10.1021/acs.jmedchem.0c01118

Klebsiella pneumoniae is one of the most critical opportunistic pathogens. TA systems are promising drug targets because they are related to the survival of bacterial pathogens. However, structural information on TA systems in K. pneumoniae remains lacking; therefore, it is necessary to explore this information for the development of antibacterial agents. Here, we present the first crystal structure of the VapBC complex from K. pneumoniae at a resolution of 2.00 ?. We determined the toxin inhibitory mechanism of the VapB antitoxin through an Mg2+ switch, in which Mg2+ is displaced by R79 of VapB. This inhibitory mechanism of the active site is a novel finding and the first to be identified in a bacterial TA system. Furthermore, inhibitors, including peptides and small molecules, that activate the VapC toxin were discovered and investigated. These inhibitors can act as antimicrobial agents by disrupting the VapBC complex and activating VapC. Our comprehensive investigation of the K. pneumoniae VapBC system will help elucidate an unsolved conundrum in VapBC systems and develop potential antimicrobial agents.

Full text of DOI:10.1021/acs.jmedchem.0c01118

pubs.acs.org/jmc
Article
Structural and Functional Study of the Klebsiella pneumoniae VapBC
Toxin−Antitoxin System, Including the Development of an Inhibitor
That Activates VapC
Sung-Min Kang,⊥ Chenglong Jin,⊥ Do-Hee Kim,⊥ Yuno Lee, and Bong-Jin Lee*
Cite This: https://dx.doi.org/10.1021/acs.jmedchem.0c01118
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Klebsiella pneumoniae is one of the most critical opportunistic pathogens. TA systems are promising drug targets
because they are related to the survival of bacterial pathogens. However, structural information on TA systems in K. pneumoniae
remains lacking; therefore, it is necessary to explore this information for the development of antibacterial agents. Here, we present
the first crystal structure of the VapBC complex from K. pneumoniae at a resolution of 2.00 Å. We determined the toxin inhibitory
mechanism of the VapB antitoxin through an Mg2+ switch, in which Mg2+ is displaced by R79 of VapB. This inhibitory mechanism of
the active site is a novel finding and the first to be identified in a bacterial TA system. Furthermore, inhibitors, including peptides and
small molecules, that activate the VapC toxin were discovered and investigated. These inhibitors can act as antimicrobial agents by
disrupting the VapBC complex and activating VapC. Our comprehensive investigation of the K. pneumoniae VapBC system will help
elucidate an unsolved conundrum in VapBC systems and develop potential antimicrobial agents.
The four major types of TA systems are classified by the
mechanism by which the antitoxin inhibits the toxin and by the
nature of the antitoxin. In type I and III TA systems, antitoxins
are RNA molecules that regulate the cellular levels of toxins. An
RNA antitoxin inhibits the translation of an mRNA toxin (type
I) and inhibits the activity of a protein toxin by direct binding
(type III). However, antitoxins in type II and IV TA systems are
proteins. These protein antitoxins bind directly to a protein
toxin and inhibit the activity of the toxin (type II) or neutralize
the toxic activity of the toxin without direct binding, such as by
triggering opposite effects to the target (type IV). There are only
single cases of type V and VI TA systems, each of which has a
different regulatory principle.9
1. INTRODUCTION
Klebsiella pneumoniae is one of the most critical opportunistic
pathogens. It is widespread in the environment, including in
water, on solids or on leaf surfaces. K. pneumoniae is responsible
for infections of the upper respiratory tract, urinary tract, and
bloodstream and causes pneumonia.1 Currently, K. pneumoniae
is considered a significant threat to the health of people
worldwide. It possesses an extraordinarily plastic genome and a
great tendency to develop antibiotic resistance. In particular,
drug-resistant K. pneumoniae, with resistance even to carbape-
nems, has emerged as a global problem.2
The overall biological functions of toxin−antitoxin (TA)
systems cover a wide variety. The initial role identified for a TA
system was plasmid maintenance.3 Shortly afterward, similar
functions of TA systems, such as postsegregational killing, were
also observed.4 Later, it was shown that the TA systems mediate
programmed cell death from stresses such as antibiotic
treatment, phage infection, DNA damage, oxidative stress, and
high temperature.5,6 Recently, it has been shown that TA
systems are likely to function in the formation of bacterial
persister cells.7,8
Received: June 30, 2020
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
© XXXX American Chemical Society
J. Med. Chem. XXXX, XXX, XXX−XXX
A

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Figure 1. Overall structural information on K. pneumoniae VapBC (PDB ID: 7BY3). (A−C) Antitoxin VapB subunits are shown in yellow and the toxin
VapC subunits are shown in light purple. (A) Heterotetrameric assembly of the VapBC complex containing an intermolecular interface formed by the
dimerization of VapC. (B) Heterotetrameric assembly of the VapBC complex containing an intermolecular interface formed by the dimerization of the
N-terminal DNA-binding domain of VapB. The DNA-binding domain of VapB is labeled. (C) Schematic diagram of the VapBC heterodimer showing
the secondary structural architecture. (D) Secondary structure analysis of the N-terminal DNA-binding domain of VapB. Prediction data using the
PSIPRED server are presented with structural information on the N-terminus of VapB in the crystal structure.
The bacterial type II TA system contains two proteins, a
biophysically stable toxin and a relatively labile antitoxin. In the
normal state, the expression levels of the toxin and antitoxin are
well-harmonized, so bacteria can normally survive because the
antitoxin counteracts any harmful effect of the toxin.10 However,
the labile antitoxin degrades faster under stress conditions,
including antibiotic treatment, nutrient starvation, and unfav-
orable temperature. Then, the toxin is freely released from the
antitoxin, and the bacteria suffer damage because of the toxicity
of the released toxin.11
Type II TA systems are more broadly distributed in
pathogenic bacteria than in nonpathogenic bacteria.12 This
suggests that TA systems are closely related to bacterial
pathogenicity. Additionally, TA systems are not found in any
eukaryotes and are therefore a promising drug target for
bacterial pathogens.11 Dozens of structures from the TA systems
of pathogenic bacteria have been solved in structural and
functional studies; however, the structural information on TA
systems in K. pneumoniae is still unknown.13 To gain molecular
insights into and validate the K. pneumoniae TA system as a
future drug target, it is essential to explore the structure of
Klebsiella pneumonia TA systems.
VapBC is one of the largest families of type II TA systems. To
date, at least 11 complex structures have been identified in the
VapBC TA system: Neisseria gonorrheae FitAB,14 Shigella flexneri
VapBC;15 Rickettsia felis VapBC2;16 Caulobacter crescentus
VapC1;17 Haemophilus influenza VapC1;18 and Mycobacterium
tuberculosis VapBC3,19 VapBC5,20 VapBC11,21 VapBC15,22
VapC26,23 and VapBC30.24 These VapBC complexes highlight
the great structural complexity of VapB, which contains a
variable DNA-binding domain and flexible region, whereas
VapC exhibits a conserved rigid domain.
The VapC toxin contains a pilus retraction protein (PilT) N-
terminal (PIN) domain, which is crucial for ribonuclease
(RNase) activity. These PIN domains comprise approximately
130 amino acid proteins and contain a strictly conserved
negatively charged acidic quartet of residues coordinated by
Mg2+ in the active site. On the other hand, the cognate antitoxin
VapB folds into a ribbon−helix−helix (RHH) DNA-binding
motif at the N-terminus and wraps the active site of VapC via its
C-terminal helices.23 As VapB sterically blocks the catalytic site
of VapC and acts as a tight binding inhibitor, the molecules that
inhibit the binding between VapB and VapC can generate and
release the free toxin. There are previous reports about these
inhibitory molecules acting as binding competitors, leading to
growth arrest or ultimately cell death in other pathogenic
bacteria.23,25 The exploitation of TA systems as an antibacterial
strategy via artificial activation of the toxin has been proposed as
a potential approach for antibiotics therapy.26 In some recent
studies, sequence-specific antisense peptide nucleic acid
oligomers and peptide-based inhibitors led to bacterial cell
death, highlighting this strategy.21,27
Here, we present the first crystal structure of the VapBC
complex from K. pneumoniae at a resolution of 2.00 Å. First,
through structural analysis, we found that R79 of VapB displaces
the coordinated Mg2+ from the neighboring D9, E43, and D90 of
VapC to D9, D90, and D111 of VapC and abolishes the RNase
activity of VapC. This finding was further verified by site-
directed mutagenesis and in vitro RNase activity experiments.
We designed peptides mimicking the binding interface of the
VapBC complex and confirmed the complex disruption ability of
peptides by the successful generation of RNase activity of the
VapC toxin. Additionally, using the structural information, a
total of 400 small molecules were screened. The final small-
B
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Figure 2. Structural comparison of K. pneumoniae VapBC with its homologues. (A) Sequence alignment of K. pneumoniae VapC with its homologues.
The secondary structural elements of these proteins are shown above the alignment. Highly conserved residues are highlighted in red and yellow.
Among them, the acidic active-site residues and residual residues are marked as red and blue triangles, respectively. (B) Electrostatic surface of VapC
(left) (PDB ID: 7BY3). Positive and negative charges are colored blue and red, respectively. An acidic active site of VapC is shown enlarged in cartoon
and stick representations (right) showing the locations of seven conserved residues in the structure of each homologue. Seven conserved residues are
shown in stick representation. (C) Comparison of the VapBC complexes (PDB IDs: in the order of 7BY3, 6A7V, 5X3T, 5K8J, and 6NKL). Each
VapBC complex is shown in the same direction based on the VapC structure. For VapC, the same colors as in Figure 2B are used.
molecule candidate is thought to be a potent structure-based
inhibitor with complex disrupting efficacy. This comprehensive
investigation of the K. pneumoniae VapBC system will help us to
decipher unsolved conundrums in VapBC systems and develop
potential antimicrobial agents against K. pneumoniae.
wavelength anomalous diffraction (SAD) data for the SeMet-
labeled VapBC complex were refined [Protein Data Dank
(PDB) ID: 7BY2] and used to solve the phase problem. Then,
the native data were refined (PDB ID: 7BY3) by molecular
replacement employing the structure of the SeMet-labeled
VapBC complex as the search model.
As a result of the structural calculation of K. pneumoniae
VapBC from native data, the overall conformation of the VapBC
complex was obtained. The VapB antitoxin is divided into an N-
terminal part containing a DNA-binding domain and a C-
terminal part containing a toxin-binding region, which blocks
the active site of the VapC toxin. Interestingly, the phased
structure of VapBC from the SAD data showed an incomplete
complex structure in which the C-terminus of VapB was
digested; thus, the active site of VapC was not inhibited by
VapB. Detailed new findings related to this observation will be
2. RESULTS AND DISCUSSION
2.1. Overall Information on the VapBC Complex. First,
we validated the K. pneumonia VapBC TA system experimen-
tally. VapC expression resulted in considerable restraint of cell
growth and a decrease in the colony-forming unit (cfu).
However, suppressed cell viability was almost fully rescued by
the expression of VapB (Figure S1). The bacterium used for
validation of the Klebsiella pneumonia VapBC TA system is
Escherichia coli. To solve the structure of VapBC, the single-
C
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Figure 3. Active site and RNase activities of VapC. (A) Two forms of the VapBC complex structure: open form (left) (PDB ID: 7BY2) and closed form
(right) (PDB ID: 7BY3). (B) Two forms of the active site are presented with a cartoon diagram of VapC. Four conserved residues are shown in stick
representation. Active-site conformations are denoted as black dotted squares and enlarged. The 2mFo-DFc electron density map of the Mg2+ site
contouring at σ (calculated with PHENIX) is also displayed. In the open form of the VapC active site, the coordinated Mg2+ is located in the midst of
residues D9, E43, and D90. In the closed form of the VapC active site, R79 of VapB triggers the displacement of Mg2+. Consequently, R79 of VapB fills
the original location of Mg2+, and Mg2+ moves to the middle of D9, D90, and D111. (C) In vitro RNase activities of VapC, VapC mutants, and VapBC.
Proteins were treated with EDTA to remove metal ions, and 10 mM Mg2+ was then added to proteins. In the assay, 10 μM proteins were used, and a
general RNase inhibitor (40 units) was used to prevent unintended contamination. The kinetics for the RNase activities of these proteins were
recorded for 1 h. The data shown are representative of three independent experiments.
described in the third subsection of the Results and Discussion
section.
VapC26 (PDB ID: 5X3T),23 C. crescentus VapC1 (PDB ID:
5K8J),17 and Hemophilus influenza VapC1 (PDB ID: 6NKL)18
were compared with that of K. pneumoniae VapC. The alignment
revealed seven highly conserved residues (Figure 2A). The first,
third, fourth, and sixth conserved residues are negatively charged
and acidic residues, such as aspartate and glutamate, which make
up the active site of the toxin. Among the remaining three
residues, the locations of the second threonine and seventh
phenylalanine are closely related to the location of the active site,
and hydrogen bonding and hydrophobic interactions between
these residues and the active-site residues support and
contribute to the building and structural integrity of the active
site (Figure 2B).
Despite the low sequence similarity (13.7% on average) with
the four homologues, K. pneumoniae VapC shares structural
similarity with the other VapC proteins. VapC toxins adopt a
canonical α/β/α sandwich fold composed of five parallel β-
strands flanking seven α-helices. VapC toxins also similarly
feature the PIN domain, which is essential for ribonuclease
activity.32 The conserved acidic residues in PIN domain proteins
coordinate with divalent metals and trigger catalytic activity
based on acid−base reactions.33
In the crystal structures, we observed two types of
intermolecular interfaces, which originated from crystal packing.
The first is a toxin dimerization-driven heterotetramer (Figure
1A). In this case, the interaction between α-helices in the
adjacent toxin monomers is a main binding force. The other is an
antitoxin dimerization-driven heterotetramer (Figure 1B),
which is mediated by the DNA-binding domain of the antitoxin.
A schematic diagram of the VapBC heterodimer is helpful for
understanding the spatial arrangement of the secondary
structural elements that make up the heterodimeric architecture
(Figure 1C). In our native structure, the first β-strand of the
VapB N-terminus was not observed because of the poor electron
density in the map. According to the prediction by PSIPRED,28
VapB is expected to have an RHH DNA-binding domain29 in
the order β1−α1−α2 (Figure 1D) at the N-terminus.
2.2. Structural Insights into K. pneumoniae VapBC
Compared with Its Homologues. To identify the structural
characteristics of K. pneumoniae VapBC, a comparative analysis
was performed with its structural homologues, and the complex
structure was solved. To reflect the latest information,
comparisons were made with relatively recently reported
homologues since 2017. First, an amino acid sequence
alignment of the VapC toxins using Clustal Omega 1.2.130 and
ESPript 3.031 was carried out. In this alignment, the primary
sequences of M. tuberculosis VapC11 (PDB ID: 6A7V)21 and
The four homologues were also compared with K. pneumoniae
VapBC to determine the differences in the VapB structures and
VapB- and VapC-binding patterns (Figure 2C). In comparison,
the VapC structures had similar folds; however, the binding
patterns of the VapC toxin with its cognate VapB and the VapB
D
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Figure 4. Heterodimeric interface of the VapBC complex (PDB ID: 7BY3) and efficacy of complex disruption using the designed peptides. (A,B)
Residues involved in interaction networks are shown in stick representation. Schematic diagrams of the hydrophilic and hydrophobic interactions are
provided in the table. (A) Intermolecular interaction network of the VapB α3 helix. (B) Intermolecular interaction network of the VapC α1 helix.
(C,D) RNase activities resulting from peptides added to the VapBC complex. Activities generated by peptides are plotted in proportion to that of
VapC. The pretreatments for the assay were the same as those described in Figure 3C. In these complex disruption assays, 100 μM peptides were added
to 10 μM proteins, and their complex disruption efficacies were evaluated. Data are represented as the mean of three independent replicates SD. (C)
Efficacy of peptides designed based on the sequence of the α3 helix of VapB. (D) Efficacy of peptides designed based on the sequence of the α1 helix of
VapC.
DNA-binding domains were quite different. The C-terminal
toxin-binding region of VapB is intrinsically extremely flexible;
thus, structural folding and integrity are achieved only after
binding to the toxin.23 For this reason, the α-helical contents and
the loop contents of each VapB in each binding region between
these VapB antitoxins and VapC toxins seem to have large
differences. In addition, there are two different DNA-binding
domains in VapB, including the RHH29 and AbrB-type34
domains. Interestingly, DNA-binding domains in VapB differ
not only in their type of domain but also within the same domain
type, and the orientation of the domain varies considerably
depending on the curve at the hinge region. This emphasizes
that the structural variability of VapB proteins causes a diversity
of transcription regulation mechanisms and further explains the
specificity in the recognition of their own promoter regions.
2.3. RNase Activity of VapC and Unveiling the
Inhibition Mechanism of VapB. As mentioned in the first
subsection, the structure obtained from the SAD data is in an
open form with an active site exposed because of the natural
digestion of the C-terminus of VapB (Figure 3A). On the other
hand, the native structure obtained from the native crystal shows
a closed form with an active site in which the toxin-binding
region of VapB is completely preserved (Figure 3A). In
accordance with the innate nature of VapC, its RNase activity
is only shown when divalent metal ions such as Mg2+ are present.
K. pneumoniae VapC showed RNase activity only when Mg2+
was added to the protein solution. In the open form without
VapB interference in the active site, Mg2+ is located in the midst
of the conserved active-site residues (D9, E43, and D90) of
VapC. However, in the closed form, VapB covers the active site
of the toxin. R79 of VapB exactly blocks the original Mg2+ site
and displaces Mg2+ to the center of residues D9, D90, and D111
of VapC, causing the active site to switch from the open state to
the closed state (Figure 3B). This inhibitory mechanism of the
active site is a novel TA system mechanism. This inhibitory
mechanism of the active site is first observed through two states
of crystal structures.
To expand the importance of these results, we mutated each
residue comprising the active site of VapC and measured the
RNase activities of each VapC mutant (Figure 3C). From the
graphs, which express the kinetics for 1 h and show a
proportional plot for VapC, the residue with the largest
contribution to activity was E43, and the residue with the least
contribution was D111. In conclusion, as R79 of VapB causes
the displacement of Mg2+, VapC active-site inhibition is
achieved by completely abolishing the contribution of E43 to
the RNase activity. However, based on the mutation experi-
ments, when E43 of VapC is mutated to alanine, VapC still has
approximately 40% of its RNase activity. This suggests that the
RNase activity of VapC depends on the position of Mg2+, which
could be displaced by VapB, as well as the contribution of each
residue constituting the active site.
2.4. Exploration of Peptides That Activate VapC by
Disrupting the Interface between VapB and VapC. If the
formation of the TA complex is artificially inhibited, toxins freely
released from the TA complex can arrest the growth or
eventually cause the death of bacteria.11 Therefore, to design
peptides capable of disrupting the K. pneumoniae VapBC
E
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Figure 5. Surface pockets on VapC (PDB ID: 7BY3) and efficacy of final compounds expressed with molecular docking. (A) Surface of VapC with a
cartoon representation of VapB. The negatively charged active-site cavity and surface cleft interface pockets are linked through the α2−α3 loop of
VapB. The compound-binding site of VapC is located on the surface cleft interface pockets represented as the interface pocket valley. (B) RNase
activities of compounds when the VapBC complex was added. Activities generated by compounds are plotted in proportion to that of VapC. The
activities of compounds without the VapBC complex were also measured. Pretreatments for these assays were the same as those described in Figure 3C.
In these complex disruption assays, compound (1, 3.16, and 10 μM) was added to 10 μM protein, and the complex disruption efficacy was evaluated.
The activity was measured based on the initial velocity during the first hour. Data are represented as the mean of three independent replicates SD.
(C,D) Results of molecular docking of the compounds on the VapC structure (PDB ID: 7BY3) with the detailed interaction networks. (C) Interaction
data for compound 1. (D) Interaction data for compound 2. In the cartoon diagram, ligand-binding residues of VapC are shown in stick and labeled. 2D
ligand-VapC interaction diagrams were generated using LigPlot+.38 Compounds 1 and 2 are shown in green and pink, respectively. The residues
involved in HBs are shown, and HBs are represented by blue dashes. The residues making nonbonded contacts with the ligand are shown in the red
spoked arcs.
complex, the interaction networks between VapB and VapC
were analyzed based on the results of analyses from the proteins,
interfaces, structures, and assemblies (PISA)35 and protein
interaction calculator (PIC)36 servers. The results showed that
the α3 helix of VapB (Figure 4A) and α1 helix of VapC (Figure
4B) were most important in the dimerization interaction
between VapB and VapC.
Therefore, several peptides were designed based on the
sequence of the α3 helix of VapB and the α1 helix of VapC
(Table S3). When a peptide is added to the VapBC complex, the
inhibition efficacy of the peptide can be estimated by measuring
the RNase activity generated by the free VapC, which is released
by the disruption of the VapBC complex. In the complex
disruption assay, although the α3 helix of VapB consists of G71−
F
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
K80 (GMEYQRQLRK), the VapB71−78 peptide consisting of
G71−L78 (GMEYQRQL), which excludes R79 of VapB that
interacts with three active-site residues (D9, E43, and D90) of
VapC, showed high efficacy. Y74 and Q75 of VapB interact with
E43 and D9 of VapC, respectively. As a result, when Y74 and
Q75 were individually mutated to alanine, the VapB71−78, Y74A
peptide showed the highest efficacy among all of the designed
peptides. The VapB71−78, Y74A, Q75A peptide, in which both Y74
and Q75 are mutated to an alanine, showed a lower efficacy than
the individual alanine-mutated VapB peptides. This might be
due to defects in the structural integrity of the α helix because of
multiple alanine substitutions (Figure 4C).
(Figure S2B) because they are designed to bind to the binding
interface between VapB and other regions of VapC, which is not
the active site of VapC.
These two compounds had a higher efficacy at 1 μM, which is
one hundredth of the peptide concentration (100 μM)
previously used. Therefore, compounds 1 and 2 are very
efficient at disrupting the VapBC complex. Compounds 1 and 2
exhibited IC50 values of 0.42 and 0.40 μM, respectively (Figure
S3). For better understanding, the results of molecular docking
simulations of compounds 1 (Figure 5C) and 2 (Figure 5D) are
also presented. The carbonyl group of compound 1 interacted
with the positively charged amino group on the side chain of
K50 by forming a strong HB. Residues E43, S18, and A48 also
formed hydrogen and carbon−hydrogen bond interactions.
Residues Y51, I14, and V46 formed π-alkyl and alkyl
interactions. Similar to the HB of compound 1, the sulfonyl
group of compound 2 was also found to have a strong HB with
the positively charged amino group in the side chain of K50.
Additionally, residues T10, N11, G47, and A48 formed carbon−
hydrogen bonds and amide−π-stacked interactions. Residues
I14 and V46 were involved in the formation of π−alkyl and alkyl
interactions, respectively. F17 and Y51 formed π−π stacking
interactions with the fluorobenzene ring of the compound. In
conclusion, we suggest that these two final compounds are
potential candidates for controlling drug-resistant K. pneumo-
niae. This is the first report on small molecules controlling a
bacterial TA system.
2.6. Insights into the Possibility of Competitive
Inhibitor and Autoregulation Based on Binding Affinity.
To further verify the peptides and small molecules selected
through this study, the binding affinities of VapC with VapB,
peptides, and small molecules were measured by isothermal
titration calorimetry (ITC) (Figure S4). VapB was most
strongly coupled with VapC (Kd = 3 nM), and peptides and
small molecules also showed affinities with VapC in the
nanomolar range (Kd = 110−170 nM for peptides and 22−35
nM for small molecules) (Figure S4). Identifications of these
affinities confirm that the peptides were successfully rationally
designed as competitive inhibitors for the VapB-binding region
and that the proposed docking models of small molecules are
reliable.
When free VapC was treated with VapB-mimicking peptides,
the overall RNase activity measured was slightly increased, but
the pattern was similar to that resulting from VapBC treated with
peptides (Figures 4C and S2A). Because the designed peptides
are made by mimicking the binding interface of VapB and VapC,
arithmetically, the more residues that participate in the binding
of VapB and VapC are present in the peptides, the greater the
complex disruption capability should be. However, if a peptide
contains the residues that interact with the nucleic acid-binding
site of VapC, which is the active site of VapC, the disruption
capability is inhibited. Moreover, the lack of structural integrity
due to multiple alanine substitutions (both Y74A and Q75A)
had little effect on the activity of VapC. In summary, among the
residues involved in the binding of VapB and VapC included in
the peptide, the more residues in the peptide that interact with
the residues constituting the active site of VapC, the more the
activity of VapC is unintentionally inhibited. In particular, as
shown in the mutational study, this phenomenon is more
pronounced when the peptide contains a residue that interacts
with D43, which is most related with the activity of VapC.
In the VapBC complex structure, the α1 helix of VapC is
composed of T10−F17 (TNILIDLF). Although D9 and S18 are
not included in the α1 helix of VapC, the VapC9−18 peptides,
including D9 and S18, which interact with VapB, showed lower
efficacy than the peptide consisting of only the α1 helix (Figure
4D). As a result, the importance of the α-helical nature in the
artificial disruption of the TA complex was reconfirmed.37
2.5. Exploration of Small Molecules That Activate
VapC by Occluding the Interface Pocket. In parallel, based
on the information on the interface between K. pneumoniae
VapB and VapC, a total of 400 small-molecule candidates
capable of binding to the VapB-binding pocket of the VapC
surface were selected by virtual screening. The majority of these
compounds are distributed in an MW range of 300−500 Da with
an average of ∼5.1 HB acceptors, ∼2.2 HB donors, and 6.8
rotatable bonds. The predicted A log P values for most of the
compounds are in the range of 0.5−5, with an average value of
2.8. Each compound was designed to disrupt the VapBC
complex by binding not to the active-site cavity of VapC but to
the cleft through the α2−α3 loop of VapB, linking the interface
pockets (Figure 5A).
The affinities of the peptides VapB71−78, VapB71−78, Y74A, and
VapB71−78, Q75A to VapC were 110 nM, 170 nM, and 150 nM,
respectively (Figure S4). By mutation of Y74 and Q75, some
interactions with VapC disappear and the affinity with VapC
decreases. However, Y74 and Q75 from VapB interact with the
active site of VapC. Therefore, VapB71−78, Y74A and
VapB71−78, Q75A show higher activities than VapB71−78 because
of the decrease in inhibitions of VapC activity, despite their
decreased binding affinity with VapC.
In type II TA systems, the antitoxin and toxin proteins play
important roles in transcriptional autoregulation.39,40 To
investigate the molecular mechanism of transcriptional regu-
lation, the 28-bp DNA sequence at the promoter region was
titrated into VapB and VapBC. The affinity of VapBC to DNA
(Kd = 1.26 μM) was fivefold higher than that of VapB to DNA
(Kd = 5.38 μM) (Figure S4). This indicates that VapB serves as a
primary repressor in autoregulation by directly binding to its
own promoter, while VapC acts as a corepressor, enhancing the
interaction between VapB and the operator site by forming a
complex with VapB.
Using a similar assay method to that in the previous section,
the inhibition efficacies of the compounds were ranked
according to the amount of RNase activity by the free VapC
released by the disruption of the VapBC complex. A total of 400
compounds were narrowed to 25 according to inhibition efficacy
ranking through triplicate tests, and these 25 small molecules
were further narrowed through additional triplicate testing to
the two compounds with the highest efficacy, named compound
1 and compound 2 (Figure 5B). These two final compounds did
not affect the RNase activity of VapC when added to free VapC
G
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
3. CONCLUSIONS
pubs.acs.org/jmc
Article
HKL2000.41 The detailed statistical information is shown in Table S2.
During the structural calculation, PHENIX42 was used to automatically
build the model, and Coot43 was utilized to refine it. REFMAC and
PHENIX42,44 were employed to obtain the plausible Rwork and Rfree
values. MolProbity45 was used to validate the overall geometry. PyMOL
The structure of the VapBC complex in this study is the first
reported structure of a TA complex in K. pneumoniae. The VapC
toxin inhibition mechanism by R79 of VapB displacing Mg2+ was
revealed through two crystal structures, the open and closed
forms of the active site. This is a novel finding first identified in
the field of bacterial TA systems. Through the rational design of
peptides mimicking the VapBC-binding interface, which can
disrupt the VapBC complex, the artificial activation of the toxin
as an antibacterial strategy can be applied to K. pneumoniae.
Furthermore, by discovering small compounds that effectively
inhibit the formation of the VapBC complex even at low
concentrations, controlling drug-resistant K. pneumoniae
through modification with inhibitors may be possible.
̈
(the PyMOL Molecular Graphics System, Schrodinger, LLC., Cam-
bridge, MA, USA) was used to generate figures.46
4.3. In Vitro RNase Assay for VapC and VapC Mutants. The
RNase activities of VapC and VapC mutants were confirmed using an
RNase Alert Kit (Integrated DNA Technologies, Inc.). In this assay,
each well contained 50 pm of synthetic RNA containing a fluorophore
at one end and a quencher group at the other end. If the prepared RNA
is contacted by RNase, the quencher is detached. Then, the resulting
fluorescence (RFU) during the 1 h reaction period was detected using a
SPECTRAmax GEMINI XS spectrofluorometer (Molecular Devices)
at 490 and 520 nm as the excitation and emission wavelengths,
respectively. For this assay, the purified VapC and VapC mutants were
treated with 10 mM EDTA to remove metal ions and exchanged with a
purification buffer containing 10 mM Mg2+. In each experiment, the
concentration of both VapC and VapC mutants was 10 μM. Forty units
of RiboLock (Thermo Scientific) were added to each well to prevent
contamination. RiboLock does not inhibit prokaryotic RNases I and H.
VapC is similar in architecture to RNase H and is thus not affected by
RiboLock.33 The final concentration of RNA was 5 pM, and the buffer
components were 20 mM HEPES, pH 7.5, and 400 mM NaCl.
4.4. Complex Disruption Assay by the Peptides Mimicking
the VapBC Interface. To disrupt the VapBC complex and activate the
VapC toxin, several peptides mimicking the binding interface of VapBC
were designed (Table S3). The peptides were designed through
structural analysis and synthesized by ANYGEN (http://www.anygen.
com). In these assays, the peptide (100 μM) was added to the VapBC
complex (10 μM), and the RFU from the activated VapC toxin was
measured. Other experimental protocols used for the peptides were the
same as those used for the RNase activity test of VapC and VapC
mutants.
4.5. Structure-Based Pharmacophore Generation and Mo-
lecular Docking Studies. The structure-based drug design was
carried out with a structure-based pharmacophore (SBP) model and
molecular docking simulation to virtually screen the primary hit
compounds. SBP modeling, which extracts the chemical features from
the protein−ligand complex structure, was performed to generate an
SBP model of the crystal structure of the VapBC complex (PDB ID:
7BY3). Receptor-ligand pharmacophore generation was used in the
pharmacophore generation tools of Discovery Studio (DS) 2018
software (DASSAULT SYSTEMES, 2017) with the default parameters.
The SBP model, having four significant chemical features [two
hydrogen-bond (HB) donors, one HB acceptor, and one hydrophobic
interaction], was designed by considering only the VapB α2−α3 loop
region (W64, C67, and Y74 of VapB) as a replacement ligand to inhibit
the protein−protein interaction. The generated SBP model was used to
retrieve a total of ∼610,000 Korea Chemical Bank (KCB) compounds
and select ∼16,000 compounds by filtering.
4. EXPERIMENTAL SECTION
4.1. Cloning, Protein Purification, and Mutation. For the
structural determination of the VapBC complex, the genes encoding K.
pneumoniae (strain MGH78578) VapB (kpn_04185) and VapC
(kpn_04186) were amplified by PCR using the primers VapB-F/
VapB-R1 and VapC-F/VapC-R (Table S1). The restriction enzymes
Nde1 and Xho1 were used to doubly cleave all of the PCR products and
vectors. Then, the cleaved PCR products were ligated to pET-21a
(kpn_04185) and pET-28a (kpn_04186). The pET-28a vector, used
f o r V a p C ( k p n _ 0 4 1 8 6 ) , h a s a n N - t e r m i n a l t a g
(MGSSHHHHHHSSGLVPRGSH). The cloned plasmids were
cotransformed into competent E. coli BL21(DE3) cells (Novagen).
The bacterial cells were grown at 37 °C using Luria broth (LB) until the
optical density at 600 nm (OD600) reached 0.6. The cells were then
induced by 0.5 mM IPTG and further incubated for 4 h at 37 °C. The
induced cells were centrifuged at 11,355g and suspended in a buffer
containing 20 mM Tris-HCl, pH 7.9, and 500 mM NaCl with 5%
glycerol. The suspended cells were lysed by ultrasonication and
centrifuged at 28,306g. After centrifugation, the supernatants
containing soluble proteins were loaded onto an Ni2+ affinity open
column (Bio-Rad) equilibrated in advance with buffer A. The bound
proteins were washed with buffer A containing 100 mM imidazole and
eluted by an imidazole gradient (150−500 mM). The eluted proteins
were diluted with a buffer containing 20 mM HEPES, pH 7.5, and 100
mM NaCl and further purified using a HiTrap Q HP anion-exchange
chromatography column (GE Healthcare) with a NaCl gradient (200−
800 mM). These eluted proteins were exchanged with a buffer
containing 20 mM HEPES, pH 7.5, and 400 mM NaCl by size-
exclusion chromatography on a HiLoad Superdex 200 prep-grade
column (GE Healthcare) and concentrated to 20 mg/mL using an
Amicon Ultra Centrifugal Filter Unit (Millipore).
For the purification of the selenomethionine (SeMet)-labeled
VapBC complex, the same procedure described above was used, except
that the cells were grown in SeMet Medium Base plus Nutrient Mix
(Molecular Dimensions) containing extra SeMet.
Subsequently, the molecular docking simulation was performed with
the resulting compounds and 7155 diverse set compounds,
representing a total of ∼610,000 KCB compounds. The target-binding
site of VapC was set from the cocrystal structure of VapB, especially the
α2−α3 loop region with a 14 Å radius sphere. The 3D structure of
VapC was refined and protonated by the Prepare protein tool
implemented in DS 2018. To perform the docking simulation,
CDOCKER47 was used in DS 2018. CDOCKER is a CHARMm-
based docking tool that uses a powerful ligand flexibility algorithm and a
refinement process that performs random rotation and grid-based
simulated annealing at high temperatures. The resulting hit compounds
were ranked by the score, which considers both fit value and negative
CDOCKER interaction energy (including van der Waals and
electrostatic interactions).
To create VapC mutations, forward (F) and reverse (R) primers for
D9A, E43A, D90A, and D111A were used (Table S1). The mutations
were created using the EZchange Site-Directed Mutagenesis Kit
(Enzynomics). PCR products were ligated to pET-28a with an N-
terminal tag (MGSSHHHHHHSSGLVPRGSH). The VapC and
mutated VapC proteins were produced by identical steps to those of
native VapBC complexes. The purities of the purified proteins were
verified by sodium dodecyl sulphate−polyacrylamide gel electro-
phoresis at every step.
4.2. Crystallization, Data Collection, and Processing. To
crystallize the VapBC complex, the sitting drop vapor diffusion method
was employed using the Wizard Classic crystallization screen series
(Rigaku) at 20 °C. The final crystallization condition was 0.1 M sodium
citrate, pH 5.5, and 200 mM lithium sulfate. Crystals were
cryoprotected with 20% glycerol and flash-cooled in liquid nitrogen.
The diffraction data were collected using an ADSC Quantum Q270r
CCD detector at beamlines 5C and 11C of Pohang Light Source,
Republic of Korea. All raw data were scaled and processed using
4.6. Assay for Complex Disruption of Small Molecules That
Occlude the Interface Pocket. To disrupt the VapBC complex and
activate the VapC toxin, 400 small molecules were initially selected by
virtual screening using the pharmacophore model and molecular
H
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
docking. These 400 small molecules were provided by KCB dissolved in
dimethyl sulfoxide solution in 384-well plates. In this virtual screening,
small molecules that are expected to occlude the interface pocket of the
VapBC complex were explored and listed in the order of their virtual
fitness scores as criteria for selecting initial candidates. Small molecules
were added to the VapBC complex (10 μM), and their efficacies for
activating VapC were tested. Other experimental protocols were the
same as those described for the complex disruption assay by the
peptides. The synthetic procedures of the final selected two compounds
are described in the Supporting Information. The purity of all
compounds was >95%, as determined by liquid chromatography−
mass spectrometry.
4.7. Growth Assay and Colony-Forming Unit Measurement.
The genes encoding VapC were additionally cloned into pBAD33. The
cloned plasmid was transformed into competent E. coli BL21 (DE3)
cells containing previously cloned genes encoding VapB. This
compatible plasmid pair (pBAD33 and pET-21a) was grown overnight
at 37 °C in M9 minimal medium plus 25 μg/mL chloramphenicol and
50 μg/mL ampicillin. The overnight cultures were diluted 50-fold and
further grown for the next 5 h. At an OD600 of 0.2, 0.2% arabinose was
added. For the growth assay, 0.5 mM IPTG was added after 1 h. For the
cfu measurement, every 30 min, cell samples were collected and
streaked onto LB medium plates with 0.2% glucose and 0.5 mM IPTG.
The plates were incubated for 16 h before counting.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jmedchem.0c01118
Author Contributions
⊥S.-M.K., C.J., and D.-H.K. contributed equally to this paper.
Funding
This research was funded by the National Research Foundation
of Korea (NRF) grants funded by the Korean government
( g r a n t n u m b e r s : 2 0 1 8 R 1 A 2 A 1 A 1 9 0 1 8 5 2 6 ,
2018R1A5A2024425, 2019R1C1C1002128, and
2019R1I1A1A01057713) and the 2019 BK21 Plus Project for
Medicine, Dentistry and Pharmacy. This work was also funded
by the International Program (ICR-20-05) of Institute for
Protein Research (IPR), Osaka University.
Notes
The authors declare no competing financial interest.
The structures from this study have been deposited to the PDB
website under the PDB IDs 7BY2 (SeMet) and 7BY3 (native).
ACKNOWLEDGMENTS
■
We thank the beamline (BL) staff members at Pohang Light
Source (BL-5C, BL-7A, and BL-11C), Korea, and SPring-8
(BL44XU) under the Collaborative Research Program of the
Institute for Protein Research, Osaka University (proposal no.
2019B6972). The chemical library used in this study was kindly
provided by the Korea Chemical Bank (http://www.chembank.
org) of the Korea Research Institute of Chemical Technology.
This work was supported by MSIT and PAL, Korea.
4.8. Isothermal Titration Calorimetry. ITC was carried out using
a MicroCal200 (GE Healthcare). Concentrations of proteins and DNA
were determined using NanoDrop (Thermo) at 280 nm and 260 nm,
respectively. Stock solutions of peptides and small molecules were
prepared at the appropriate concentrations. Titrations were performed
at 25 °C. Data were calculated by subtracting the background and fitted
to a single-site-binding model with the origin-based software provided
using MicroCal200.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c01118.
ABBREVIATIONS
■
■
sı
TA, toxin−antitoxin; PilT, pilus retraction protein; PIN, pilus
retraction protein N-terminal; RNase, ribonuclease; LB, Luria
broth; IPTG, isopropyl-β-D-thiogalactopyranoside; SeMet,
selenomethionine; F, forward; R, reverse; RFU, resulting
fluorescence; SBP, structure-based pharmacophore; PDB,
Protein Data Bank; DS, Discovery Studio; HB, hydrogen
bond; KCB, Korea Chemical Bank; cfu, colony-forming unit;
ITC, isothermal titration calorimetry; HTH, ribbon−helix−
helix; PISA, proteins, interfaces, structures, and assemblies; PIC,
protein interaction calculator; BL, beamline
Growth assay and cfu measurement; effects of the
peptides and small molecules on free VapC; inhibition
curves of compounds; ITC assays; primers used for
cloning and producing mutations; data collection and
refinement statistics for SeMet-substituted and native
structures; peptides used to disrupt the binding interface
of VapBC; and synthetic procedures (PDF)
Molecular formula strings (CSV)
Hit compounds and peptide analysis reports (PDF)
REFERENCES
■
́
(1) Fernandez-García, L.; Blasco, L.; Lopez, M.; Bou, G.; García-
Contreras, R.; Wood, T.; Tomas, M. Toxin-Antitoxin Systems in
AUTHOR INFORMATION
Corresponding Author
Bong-Jin Lee − The Research Institute of Pharmaceutical Sciences,
College of Pharmacy, Seoul National University, Seoul 08826,
Korea; orcid.org/0000-0001-7896-2961; Phone: +82 2
880 7869; Email: lbj@nmr.snu.ac.kr; Fax: +82 2 872 3632
■
Clinical Pathogens. Toxins 2016, 8, 227.
(2) Gomez-Simmonds, A.; Uhlemann, A.-C. Clinical Implications of
Genomic Adaptation and Evolution of Carbapenem-Resistant
Klebsiella pneumoniae. J. Infect. Dis. 2017, 215, S18−S27.
(3) Ogura, T.; Hiraga, S. Mini-F Plasmid Genes that Couple Host Cell
Division to Plasmid Proliferation. Proc. Natl. Acad. Sci. U.S.A. 1983, 80,
4784−4788.
(4) Gerdes, K.; Rasmussen, P. B.; Molin, S. Unique Type of Plasmid
Maintenance Function: Postsegregational Killing of Plasmid-Free Cells.
Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 3116−3120.
(5) Hazan, R.; Sat, B.; Engelberg-Kulka, H. Escherichia coli mazEF-
Mediated Cell Death is Triggered by Various Stressful Conditions. J.
Bacteriol. 2004, 186, 3663−3669.
(6) Sat, B.; Hazan, R.; Fisher, T.; Khaner, H.; Glaser, G.; Engelberg-
Kulka, H. Programmed Cell Death in Escherichia coli: Some
Antibiotics can Trigger mazEF Lethality. J. Bacteriol. 2001, 183,
2041−2045.
(7) Wang, X.; Wood, T. K. Toxin-Antitoxin Systems Influence Biofilm
and Persister Cell Formation and the General Stress Response. Appl.
Environ. Microbiol. 2011, 77, 5577−5583.
Authors
Sung-Min Kang − The Research Institute of Pharmaceutical
Sciences, College of Pharmacy, Seoul National University, Seoul
08826, Korea
Chenglong Jin − The Research Institute of Pharmaceutical
Sciences, College of Pharmacy, Seoul National University, Seoul
08826, Korea
Do-Hee Kim − College of Pharmacy and Interdisciplinary
Graduate Program in Advanced Convergence Technology &
Science, Jeju National University, Jeju 63243, Korea
Yuno Lee − Korea Chemical Bank, Korea Research Institute of
Chemical Technology, Daejeon 34114, Korea
I
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
̈
́
(8) Dorr, T.; Vulic, M.; Lewis, K. Ciprofloxacin Causes Persister
Formation by Inducing the TisB Toxin in Escherichia coli. PLoS Biol.
2010, 8, No. e1000317.
tuberculosis VapBC26 Toxin-Antitoxin System based on a Structural
Study: Insights into Unique Binding and Antibiotic Peptides. Nucleic
Acids Res. 2017, 45, 8564−8580.
(9) Harms, A.; Brodersen, D. E.; Mitarai, N.; Gerdes, K. Toxins,
Targets, and Triggers: An Overview of Toxin-Antitoxin Biology. Mol.
Cell 2018, 70, 768−784.
(10) Park, S. J.; Son, W. S.; Lee, B.-J. Structural Overview of Toxin-
Antitoxin Systems in Infectious Bacteria: A Target for Developing
Antimicrobial Agents. Biochim. Biophys. Acta 2013, 1834, 1155−1167.
(11) Kang, S.-M.; Kim, D.-H.; Jin, C.; Lee, B.-J. A Systematic
Overview of Type II and III Toxin-Antitoxin Systems with a Focus on
Druggability. Toxins 2018, 10, 515.
(12) Xie, Y.; Wei, Y.; Shen, Y.; Li, X.; Zhou, H.; Tai, C.; Deng, Z.; Ou,
H.-Y. TADB 2.0: An Updated Database of Bacterial Type II Toxin-
Antitoxin Loci. Nucleic Acids Res. 2018, 46, D749−D753.
(13) Burley, S. K.; Berman, H. M.; Bhikadiya, C.; Bi, C.; Chen, L.; Di
Costanzo, L.; Christie, C.; Dalenberg, K.; Duarte, J. M.; Dutta, S.; Feng,
Z.; Ghosh, S.; Goodsell, D. S.; Green, R. K.; Guranovic, V.; Guzenko,
D.; Hudson, B. P.; Kalro, T.; Liang, Y.; Lowe, R.; Namkoong, H.;
Peisach, E.; Periskova, I.; Prlic, A.; Randle, C.; Rose, A.; Rose, P.; Sala,
R.; Sekharan, M.; Shao, C.; Tan, L.; Tao, Y.-P.; Valasatava, Y.; Voigt,
M.; Westbrook, J.; Woo, J.; Yang, H.; Young, J.; Zhuravleva, M.;
Zardecki, C. RCSB Protein Data Bank: Biological Macromolecular
Structures Enabling Research and Education in Fundamental Biology,
Biomedicine, Biotechnology and Energy. Nucleic Acids Res. 2019, 47,
D464−D474.
(24) Lee, I.-G.; Lee, S. J.; Chae, S.; Lee, K.-Y.; Kim, J.-H.; Lee, B.-J.
Structural an Functional Studies of the Mycobacterium tuberculosis
VapBC30 Toxin-Antitoxin System: Implications for the Design of
Novel Antimicrobial Peptides. Nucleic Acids Res. 2015, 43, 7624−7637.
(25) Kim, D.-H.; Kang, S.-M.; Park, S. J.; Jin, C.; Yoon, H.-J.; Lee, B.-J.
Functional Insights into the Streptococcus pneumoniae HicBA Toxin-
Antitoxin System Based on a Structural Study. Nucleic Acids Res. 2018,
46, 6371−6386.
(26) Williams, J. J.; Hergenrother, P. J. Artificial Activation of Toxin-
Antitoxin Systems as an Antibacterial Strategy. Trends Microbiol. 2012,
20, 291−298.
́
́
(27) Rownicki, M.; Pienko, T.; Czarnecki, J.; Kolanowska, M.;
Bartosik, D.; Trylska, J. Artificial Activation of Escherichia coli mazEF
and hipBA Toxin-Antitoxin Systems by Antisense Peptide Nucleic
Acids as an Antibacterial Strategy. Front. Microbiol. 2018, 9, 2870.
(28) Buchan, D. W. A.; Jones, D. T. The PSIPRED Protein Analysis
Workbench: 20 Years on. Nucleic Acids Res. 2019, 47, W402−W407.
(29) Schreiter, E. R.; Drennan, C. L. Ribbon-Helix-Helix Tran-
scription Factors: Variations on a Theme. Nat. Rev. Microbiol. 2007, 5,
710−720.
́
́
(30) McWilliam, H.; Li, W.; Uludag, M.; Squizzato, S.; Park, Y. M.;
Buso, N.; Cowley, A. P.; Lopez, R. Analysis Tool Web Services from the
EMBL-EBI. Nucleic Acids Res. 2013, 41, W597−W600.
(31) Robert, X.; Gouet, P. Deciphering Key Features in Protein
Structures with the New ENDscript Server. Nucleic Acids Res. 2014, 42,
W320−W324.
(14) Mattison, K.; Wilbur, J. S.; So, M.; Brennan, R. G. Structure of
FitAB from Neisseria gonorrhoeae Bound to DNA Reveals a Tetramer
of Toxin-Antitoxin Heterodimers Containing Pin Domains and
Ribbon-Helix-Helix Motifs. J. Biol. Chem. 2006, 281, 37942−37951.
(15) Dienemann, C.; Bøggild, A.; Winther, K. S.; Gerdes, K.;
Brodersen, D. E. Crystal Structure of the VapBC Toxin-Antitoxin
Complex from Shigella Flexneri Reveals a Hetero-Octameric DNA-
Binding Assembly. J. Mol. Biol. 2011, 414, 713−722.
(32) Arcus, V. L.; Rainey, P. B.; Turner, S. J. The PIN-Domain Toxin-
Antitoxin Array in Mycobacteria. Trends Microbiol. 2005, 13, 360−365.
(33) Arcus, V. L.; McKenzie, J. L.; Robson, J.; Cook, G. M. The PIN-
Domain Ribonucleases and the Prokaryotic VapBC Toxin-Antitoxin
Array. Protein Eng., Des. Sel. 2011, 24, 33−40.
́
(16) Mate, M. J.; Vincentelli, R.; Foos, N.; Raoult, D.; Cambillau, C.;
̈
(34) Coles, M.; Djuranovic, S.; Soding, J.; Frickey, T.; Koretke, K.;
Ortiz-Lombardía, M. Crystal Structure of the DNA-Bound VapBC2
Antitoxin/Toxin Pair from Rickettsia felis. Nucleic Acids Res. 2012, 40,
3245−3258.
(17) Bendtsen, K. L.; Xu, K.; Luckmann, M.; Winther, K. S.; Shah, S.
A.; Pedersen, C. N. S.; Brodersen, D. E. Toxin Inhibition in C.
Crescentus VapBC1 is Mediated by a Flexible Pseudo-Palindromic
Protein Motif and Modulated by DNA Binding. Nucleic Acids Res. 2017,
45, 2875−2886.
(18) Molinaro, A. L.; Kashipathy, M. M.; Lovell, S.; Battaile, K. P.;
Coussens, N. P.; Shen, M.; Daines, D. A. Crystal Structure of VapBC-1
from Nontypeable Haemophilus influenzae and the Effect of PIN
Domain Mutations on Survival during Infection. J. Bacteriol. 2019, 201,
No. e00026.
(19) Min, A. B.; Miallau, L.; Sawaya, M. R.; Habel, J.; Cascio, D.;
Eisenberg, D. The Crystal Structure of the Rv0301-Rv0300 VapBC-3
Toxin-Antitoxin Complex from M. tuberculosis Reveals a Mg(2)(+)
Ion in the Active Site and a Putative RNA-Binding Site. Protein Sci.
2012, 21, 1754−1767.
(20) Miallau, L.; Faller, M.; Chiang, J.; Arbing, M.; Guo, F.; Cascio,
D.; Eisenberg, D. Structure and Proposed Activity of a Member of the
VapBC Family of Toxin-Antitoxin Systems. VapBC-5 from Mycobac-
terium tuberculosis. J. Biol. Chem. 2009, 284, 276−283.
(21) Deep, A.; Tiwari, P.; Agarwal, S.; Kaundal, S.; Kidwai, S.; Singh,
R.; Thakur, K. G. Structural, Functional and Biological Insights into the
Role of Mycobacterium tuberculosis VapBC11 Toxin-Antitoxin
System: Targeting a tRNase to Tackle Mycobacterial Adaptation.
Nucleic Acids Res. 2018, 46, 11639−11655.
Truffault, V.; Martin, J.; Lupas, A. N. AbrB-Like Transcription Factors
Assume a Swapped Hairpin Fold that is Evolutionarily Related to
Double-Psi Beta Barrels. Structure 2005, 13, 919−928.
(35) Krissinel, E.; Henrick, K. Inference of Macromolecular
Assemblies from Crystalline State. J. Mol. Biol. 2007, 372, 774−797.
(36) Tina, K. G.; Bhadra, R.; Srinivasan, N. PIC: Protein Interactions
Calculator. Nucleic Acids Res. 2007, 35, W473−W476.
(37) Luong, H. X.; Kim, D.-H.; Lee, B.-J.; Kim, Y.-W. Effects of Lysine-
to-Arginine Substitution on Antimicrobial Activity of Cationic Stapled
Heptapeptides. Arch. Pharmacal Res. 2018, 41, 1092−1097.
(38) Laskowski, R. A.; Swindells, M. B. LigPlot+: Multiple Ligand-
Protein Interaction Diagrams for Drug Discovery. J. Chem. Inf. Model.
2011, 51, 2778−2786.
(39) Kedzierska, B.; Lian, L. Y.; Hayes, F. Toxin-Antitoxin Regulation:
Bimodal Interaction of YefM-YoeB with Paired DNA Palindromes
Exerts Transcriptional Autorepression. Nucleic Acids Res. 2007, 35,
325−339.
(40) Overgaard, M.; Borch, J.; Jørgensen, M. G.; Gerdes, K.
Messenger RNA Interferase RelE Controls relBE Transcription by
Conditional Cooperativity. Mol. Microbiol. 2008, 69, 841−857.
(41) Otwinowski, Z.; Minor, W. [20] Processing of X-Ray Diffraction
Data Collected in Oscillation Mode. Methods Enzymol. 1997, 276,
307−326.
́
(42) Adams, P. D.; Afonine, P. V.; Bunkoczi, G.; Chen, V. B.; Davis, I.
W.; Echols, N.; Headd, J. J.; Hung, L.-W.; Kapral, G. J.; Grosse-
Kunstleve, R. W.; McCoy, A. J.; Moriarty, N. W.; Oeffner, R.; Read, R.
J.; Richardson, D. C.; Richardson, J. S.; Terwilliger, T. C.; Zwart, P. H.
PHENIX: A Comprehensive Python-Based System for Macro-
molecular Structure Solution. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2010, 66, 213−221.
(22) Das, U.; Pogenberg, V.; Subhramanyam, U. K. T.; Wilmanns, M.;
Gourinath, S.; Srinivasan, A. Crystal Structure of the VapBC-15
Complex from Mycobacterium tuberculosis Reveals a Two-Metal Ion
Dependent PIN-Domain Ribonuclease and a Variable Mode of Toxin-
Antitoxin Assembly. J. Struct. Biol. 2014, 188, 249−258.
(43) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features and
Development of Coot. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010,
66, 486−501.
(23) Kang, S.-M.; Kim, D.-H.; Lee, K.-Y.; Park, S. J.; Yoon, H.-J.; Lee,
S. J.; Im, H.; Lee, B.-J. Functional Details of the Mycobacterium
J
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
́
(44) Murshudov, G. N.; Skubak, P.; Lebedev, A. A.; Pannu, N. S.;
Steiner, R. A.; Nicholls, R. A.; Winn, M. D.; Long, F.; Vagin, A. A.
REFMAC5 for the Refinement of Macromolecular Crystal Structures.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 355−367.
(45) Chen, V. B.; Arendall, W. B., 3rd; Headd, J. J.; Keedy, D. A.;
Immormino, R. M.; Kapral, G. J.; Murray, L. W.; Richardson, J. S.;
Richardson, D. C. MolProbity: All-Atom Structure Validation for
Macromolecular Crystallography. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2010, 66, 12−21.
(46) Delano, W. L. The PyMOL Molecular Graphics System; DeLano
Scientific: San Carlos, CA, 2002.
(47) Wu, G.; Robertson, D. H.; Brooks, C. L., 3rd; Vieth, M. Detailed
Analysis of Grid-based Molecular Docking: A Case Study of
CDOCKER-A CHARMm-Based MD Docking Algorithm. J. Comput.
Chem. 2003, 24, 1549−1562.
K
https://dx.doi.org/10.1021/acs.jmedchem.0c01118
J. Med. Chem. XXXX, XXX, XXX−XXX