Identification of small molecules blocking the pseudomonas aeruginosa type iii secretion system protein pcrv

Article abstract of DOI:10.3390/biom11010055

Pseudomonas aeruginosa is an opportunistic bacterial pathogen that employs its type III secretion system (T3SS) during the acute phase of infection to translocate cytotoxins into the host cell cytoplasm to evade the immune system. The PcrV protein is located at the tip of the T3SS, facilitates the integration of pore-forming proteins into the eukaryotic cell membrane, and is required for translocation of cytotoxins into the host cell. In this study, we used surface plasmon resonance screening to identify small molecule binders of PcrV. A follow-up structure-activity relationship analysis resulted in PcrV binders that protect macrophages in a P. aeruginosa cell-based infection assay. Treatment of P. aeruginosa infections is challenging due to acquired, intrinsic, and adaptive resistance in addition to a broad arsenal of virulence systems such as the T3SS. Virulence blocking molecules targeting PcrV constitute valuable starting points for development of next generation antibacterials to treat infections caused by P. aeruginosa.

Full text of DOI:10.3390/biom11010055

biomolecules
Article
Identification of Small Molecules Blocking the Pseudomonas
aeruginosa Type III Secretion System Protein PcrV
Charlotta Sundin *, Michael Saleeb, Sara Spjut, Liena Qin and Mikael Elofsson
Department of Chemistry, Umeå University, SE-901 87 Umeå, Sweden; michael.saleeb@umu.se (M.S.);
sara.spjut@umu.se (S.S.); lnqin@hotmail.com (L.Q.); mikael.elofsson@umu.se (M.E.)
* Correspondence: charlotta.sundin@umu.se; Tel.: +46-90-786-54-93
Abstract: Pseudomonas aeruginosa is an opportunistic bacterial pathogen that employs its type III
secretion system (T3SS) during the acute phase of infection to translocate cytotoxins into the host
cell cytoplasm to evade the immune system. The PcrV protein is located at the tip of the T3SS,
facilitates the integration of pore-forming proteins into the eukaryotic cell membrane, and is required
for translocation of cytotoxins into the host cell. In this study, we used surface plasmon resonance
screening to identify small molecule binders of PcrV. A follow-up structure-activity relationship
analysis resulted in PcrV binders that protect macrophages in a P. aeruginosa cell-based infection assay.
Treatment of P. aeruginosa infections is challenging due to acquired, intrinsic, and adaptive resistance
in addition to a broad arsenal of virulence systems such as the T3SS. Virulence blocking molecules
targeting PcrV constitute valuable starting points for development of next generation antibacterials
to treat infections caused by P. aeruginosa.
Keywords: Pseudomonas aeruginosa; type III secretion system; PcrV; virulence inhibitors; screening;
surface plasmon resonance; infection
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
ꢀꢁꢂꢃꢄꢅꢆ
Citation: Sundin, C.; Saleeb, M.;
1. Introduction
Spjut, S.; Qin, L.; Elofsson, M.
Infectious diseases continue to constitute a threat to public health and a serious cause
Identification of Small Molecules
of morbidity and mortality worldwide. The alarming emergence rate of antimicrobial
Blocking the Pseudomonas aeruginosa
type III Secretion System Protein PcrV.
resistance is estimated to result in 10 million annual deaths with a cost of $100 trillion by
2050 [1]. The gram-negative bacterium Pseudomonas aeruginosa is an opportunistic pathogen
and one of the most frequent causes of urinary tract, bloodstream, and burn wound
Biomolecules 2021, 11, 55. https://
doi.org/biom11010055
infections as well as hospital-acquired pneumonia and cystic fibrosis-associated infections
worldwide [2]. P. aeruginosa is a “superbug” with a unique capacity to develop resistance [3].
This is due to a combination of acquired, intrinsic, and adaptive resistance. Acquired
resistance results from horizontal gene transfer and mutations leading to reduced uptake,
efflux pump overexpression, target mutations, and expression of antibiotic modifying
Received: 26 November 2020
Accepted: 29 December 2020
Published: 4 January 2021
Publisher’s Note: MDPI stays neu-
tral with regard to jurisdictional clai-
ms in published maps and institutio-
nal affiliations.
enzymes such as extended spectrum
β-lactamases. The intrinsic resistance stems from
a generally low outer membrane permeability,
β-lactamase production, and constitutive
expression of efflux pumps. Adaptive resistance is the result of triggering factors such as
antibiotics, biocides, polyamines, pH, anaerobiosis, cations, and carbon sources as well
as social behavior in biofilm formation. These factors modulate the expression of genes
that lead to increased resistance. Taken together, these features have resulted in multi-
drug-resistant P. aeruginosa strains for which no effective antibiotic treatment is available,
and these strains are becoming more frequent [4]. The ability of P. aeruginosa to rapidly
acquire or develop resistance against multiple classes of antimicrobials narrows down
and complicates the selection of antimicrobial therapy, which is crucial in optimizing
Copyright: © 2021 by the authors. Li-
censee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and con-
ditions of the Creative Commons At-
tribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
clinical outcome [
are associated with increased rates of morbidity and mortality, particularly in critically ill
and immunocompromised individuals such as cystic fibrosis and cancer patients [ ]. In
2,5]. Furthermore, infections caused by antibiotic-resistant P. aeruginosa
2
Biomolecules 2021, 11, 55. https://doi.org/10.3390/biom11010055
https://www.mdpi.com/journal/biomolecules

Biomolecules 2021, 11, 55
2 of 17
such patients, approaches targeting virulence factors have the potential to prevent infection
or to augment the effect of conventional antibiotics.
P. aeruginosa uses a broad arsenal of virulence factors, such as type IV pili, flagella,
quorum sensing, biofilm formation, and different secretion systems (type I–IV and VI)
to adapt to and survive within diverse harsh environmental settings and under minimal
nutritional requirements [6–10]. Several of these systems, predominantly quorum sensing,
biofilm formation, and type III and VI secretion systems, have been exploited in the search
for novel therapeutic agents [11] A recent example is the report on tanshinones, a class of
natural products blocking T3SS biogenesis in P. aeruginosa in vitro and in vivo [12].
Like several other gram-negative pathogens, P. aeruginosa utilizes its type III secretion
system (T3SS) during the acute phase of infection to inject effector proteins (toxins) into
the eukaryotic cytoplasm to subvert the host and to evade the immune defense [13]. To
date, four known toxins, exoenzymes, secreted by P. aeruginosa T3S have been described:
ExoU, a potent phospholipase; ExoS and ExoT, closely related bifunctional enzymes with
ADP-ribosylation and GTPase activity; and ExoY, an adenylate cyclase [14,15]. The core
of the T3SS is a nanosyringe structure that consists of a basal body spanning the bacterial
membrane and an extracellular needle-like structure protruding from the bacterial surface
and designed to deliver the effector proteins upon contact with a host cell (Figure 1) [16,17].
Although, the exact mechanism of the translocation remains poorly understood, it is
believed that translocation of the effector protein is triggered by contact with the host cell.
The PcrV protein forms an oligomeric ring structure, most likely a pentamer, at the tip
of the needle that is crucial for a functional T3SS. [18
,19] In addition, two hydrophobic
translocator proteins, PopB and PopD [20 22], assemble with the PcrV protein at the tip
–
of the T3SS needle and are inserted into the host cell membrane to form a functional pore
(Figure 1) [23–26].
Figure 1. Schematic representation of the type III secretion system (T3SS): the T3SS has a syringe-
like structure with two rings spanning the bacterial membranes and a needle protruding from the
surface with the PcrV protein at its tip. Upon contact with the eukaryotic cell, the exoenzymes are
translocated into the eukaryotic cell via the translocation-pore (PopB and PopD), which is dependent
on the PcrV protein for its function. A PcrV inhibitor will block translocation, and no toxins will be
transported into the eukaryotic cell.
One approach to develop novel antibacterial agents is to block pathogenicity by in-
hibiting virulence instead of targeting bacterial viability [27]. Several authors suggest that
a low probability for development of resistance is one of the key advantages of using
virulence systems as targets for novel anti-infectives [28,29]. Resistance to compounds
targeting virulence factors will most likely not evolve and spread in endogenous bacte-
rial flora, since these bacteria do not express the specific virulence targets. The T3SS of
P. aeruginosa and other gram-negative pathogens have been explored as targets for the
development of virulence-blocking antibacterial agents [30,31]. A number of synthetic and

Biomolecules 2021, 11, 55
3 of 17
natural compounds have been demonstrated to block the T3SS of P. aeruginosa, resulting in
attenuation of infection in vitro and in vivo [32–37]. However, most of these compounds
have been identified by phenotypic screening and the mode of action at the molecular level
remains unknown.
PcrV is an attractive therapeutic target for the development of virulence blocking
compounds against P. aeruginosa. Several studies have shown that PcrV is essential for a
functional T3SS and intoxication of the host cell. Deletion of the pcrV gene (
in a mutant that is incapable of delivering the T3S toxins into the eukaryotic cells, thus
abolishing in vitro and in vivo cytotoxicity [15 25 33 38]. In addition, it was found that
∆pcrV) results
,
, ,
active and passive immunization against PcrV greatly increase the survival rate in animal
models of P. aeruginosa-induced lung infection [39]. Since PcrV is present on the exterior of
the bacterium, the challenge of entry into the bacterial cell and the action of efflux pumps
are circumvented. Importantly, the T3SS in P. aeruginosa is essential to block macrophage
phagocytosis and the anti-PcrV antibody was demonstrated to successfully protect the
macrophage from infection [35
antibodies were shown to protect against P. aeruginosa infection in a variety of animal
models [39 43] and anti-PcrV antibodies have been advanced to clinical trials [18 44 45].
,39]. Later, many monoclonal and polyclonal anti-PcrV
–
, ,
Despite the essential role of PcrV in translocation mechanism, to date, there is no structural
information, e.g., crystal structures available of the complete PcrV protein although a
homology model has been built based on the crystal structure of LcrV from Yersinia [46], and
to the best of our knowledge, small molecules blocking PcrV have not been described. Two
recent studies describe surface plasmon resonance (SPR) screening against the analogues in
Shigella (IpaD) and in Salmonella (SipD), resulting in low affinity binders; however, virulence
blocking activity of these compounds have not been established [47,48].
Surface plasmon resonance (SPR) is a powerful and versatile biophysical technique
for the detection of small molecule-biomolecule interactions, complementary with other
techniques such as nuclear magnetic resonance (NMR) spectroscopy and X-ray crystallogra-
phy [49]. SPR is sensitive and allows kinetic and thermodynamic evaluation, which makes
it an excellent choice for the detection of low molecular weight analytes (e.g., 100–300 Da)
and low-affinity interactions (i.e., high µM to mM). Moreover, one advantage that most
SPR technologies offer nowadays is the availability of multiple biosensor channels, which
allows the study of multiple proteins in parallel.
In this study, we describe SPR screening of a library of 7600 diverse small molecules
to identify nontoxic PcrV binders with efficacy in a cell-based P. aeruginosa infection model
at µM concentrations that can potentially be used as starting points for the development of
a virulence blocking molecule.
2. Materials and Methods
2.1. Protein Expression and Purification
The PcrV gene (amino acid 20-294) was cloned into NcoI/Acc65I sites in the pETHis1a
plasmid and transformed into the competent E. coli strain BL(DE3) (Novagen), generating a
PcrV-protein lacking the first 19 amino acids and having a His-tag in the N-terminal. Three
milliliters of overnight cultures of BL21(DE3) (pETHis1a, pcrV) grown in Luria Broth (LB)
containing 50 µg/mL Kanamycin were added to 600 ml TYM-5052 medium, grown for 4 h;
shook at 37 ◦C; and after the temperature was shifted to 26 ◦C, incubated overnight. The
culture was then centrifuged, and pellet was washed in phosphate buffered saline (PBS)
and then frozen for 24 h, after which lysis buffer was added and the sample was sonicated.
Purification was carried out using gravity flow on Econo-Pac columns (Bio-Rad) packed
with 2 mL of Ni2+ charged nitrilotriacetic acid (NTA) agarose (Qiagen) and equilibrated
with lysis buffer. The column was washed with 10
×
the bead volume using wash buffer II,
and the PcrV protein was eluted using 5 the bead volume of elution buffer. For all buffer
×
ingredients, see the Supplementary Materials. The eluate was dialyzed in PBS, loaded on a
12% sodium dodecyl sulfate (SDS) gel, and stained using Coomassie. A very strong band

Biomolecules 2021, 11, 55
4 of 17
was seen corresponding to PcrV, and the concentration was measured to 6.87 mg/mL. The
purified PcrV protein was then used in SPR screening.
2.2. Surface Plasmon Resonance
The SPR screening was performed using the ProteOnTM XPR36 Protein Interaction
Array System (Bio-Rad Laboratories Inc., Hercules, CA, USA) with ProteOn GLH sensor
chip (Bio-Rad Laboratories Inc., Hercules, CA, USA), and the results were analyzed with the
ProteOn Manager softwareTM (Bio-Rad Laboratories Inc., Hercules, CA, USA), Excel 2016
◦
(Microsoft Office), and GraphPad Prism 7.04. All SPR experiments were performed at 25 C.
The multichannel module of the ProteOn system was in the vertical position during protein
immobilization. The proteins were attached to the surface of a GLH sensor chip via amine
coupling chemistry. A buffer scouting was performed to elucidate the optimal binding
conditions of PcrV to the GLH sensor chip. First, channels 1–4 on the chip surface were
activated using a 1:1 mixture of 40 mM 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
hydrochloride (EDAC) (Bio-Rad Laboratories Inc., Hercules, CA, USA) and 10 mM N-
hydroxysulfosuccinimide (sulfo-NHS, Bio-Rad Laboratories Inc., Hercules, CA, USA) for
300 s at 30
10 mM NaOAc (pH 4, 4.5, and 5) to 0.05 mg/mL. PcrV was immobilized in channel 1–3
for 540 s at 30 L/min. Channel 4 was flushed with 10 mM NaOAc (pH 5) for 540 s at
30 L/min. As deactivating agent, a 1:1 mixture of 1 M ethanolamine and milli-Q filtered
water was used. The deactivating solution was injected over all four channels for 300 s at
30 L/min. Phosphate buffered saline with Tween (PBST; 10 mM phosphate buffer pH
µL/min. Next, the proteins PcrV (33,398.6 Da, 6.37 mg/mL) was diluted in
µ
µ
µ
7.4, 140 mM NaCl, 2.7 mM KCl, and 0.05% Tween 20) was used as the running buffer. The
final amount of immobilized protein was measured in resonance units (RU) relative to the
signal of the baseline prior activation. Sodium acetate of pH 5 and 4.5 gave comparable
results (21,000 and 20,000 RU), while sodium acetate of pH 4 gave lower immobilization
(16,000 RU) of PcrV to the surface. Immobilization of PcrV diluted in sodium acetate of pH
4.5 in one channel and of pH 5 in one channel were used in the beginning of the screening
procedure (for plate 1–26). Since no difference was found between the two pH values, the
PcrV protein was diluted in sodium acetate at pH 5 only (in two channels) from plate 27 and
the rest of the screen. The immobilization procedure for the screening was performed as
described above with the addition that the control protein carbonic anhydrase isoenzyme
II from bovine erythrocytes (CAII, Sigma-Aldrich, article number 2522) was immobilized
in one channel that was activated and deactivated as described above. CAII (1 mg/mL
in PBS) was deactivated when diluted in 10 mM sodium acetate (pH 4) to 0.05 mg/mL.
Immobilization level of the deactivated CAII was 18,000 RU. Deactivated CAII was used as
a control protein according to the recommendations by the ProteOn manufacturer (Bio-Rad
Laboratories Inc., Hercules, CA, USA). As running buffer, we employed PBS (pH 7.4, 0.05%
Tween 20).
After ligand immobilization, the multichannel module of the ProteOn system was
shifted to the horizontal position. The screening compound library consisted of 7600 com-
pounds (95 plates and 80 compounds/plate) provided by the Laboratories for Chemical
Biology Umeå (LCBU), Umeå University. The compounds were selected to cover the chemi-
cal space occupied of around 40,000 diverse compounds (Diverset, ChemBridge, San Diego,
CA, USA). The compounds had a molecular weight between 250.17 and 449.97 g/mol and
cLogP between
−
5.62 and 11.12 and, thus, contained fragments as well as lead-like and
drug-like molecules. The analytes, i.e., the small molecule screening compounds, were
prepared from DMSO stock solution (5 mM) in PBST buffer (0.05% Tween 20) to achieve a
final concentration of 100 µM with a final DMSO concentration of 5% in 96-well plates (Pro-
teon Standard Microplates, 98 wells, Bio-Rad Laboratories Inc., Hercules, CA, USA). The
plates were sealed (ProteOn Microplate sealing film, Bio-Rad Laboratories Inc., Hercules,
CA, USA) when the plate was prepared. The running buffer was PBST (0.05% Tween 20)
containing 5% DMSO. The analytes were injected at a flow rate of 100
µL/min, contact time
was 60 s, and dissociation time was 60–120 s. Solvent correction was performed continu-

Biomolecules 2021, 11, 55
5 of 17
ously during screening to account for bulk response of DMSO. Each sensor chip was used
for up to a week, allowing immobilization and then screening of 8–16 plates. The data was
baseline-aligned, processed by channel-referencing with channel 4, and solvent-corrected.
The data, i.e., the response, was then exported to Excel (Microsoft Office), where it was
adjusted for molecular weight of the analyte (and multiplied with 100). Molecules with
two times higher binding to PcrV compared to the control protein, CAII, were considered
as selective binders, exemplified in Figure 2. Occasionally, air bubbles in the diluted sample
wells interfered.
Table 1. KD and biological evaluation of the synthesized compounds.
Viability Assay (% of Uninfected Control)
SPR KD (µM) a
ID
H1
Structure
200 µM
100 µM
50 µM
25 µM
106 ± 36
95 ± 41
102, 127
28%
27%
20%
12%
16% b
nd
1
2
3
21%
12%
14%
>600
nd
4
5
>600
>600
nd
nd
6
>600
nd
7
8
9
>600
>600
>600
nd
nd
nd
10
>600
nd

Biomolecules 2021, 11, 55
6 of 17
Table 1. Cont.
Viability Assay (% of Uninfected Control)
SPR KD (µM) a
ID
Structure
200 µM
100 µM
50 µM
25 µM
11
>600
nd
12
13
>600
nd
431 ± 110
20%
21%
26%
18%
23%
19%
13%
14
148 ± 62
32%
15
16
153
17% c
nd
24% c
17% c
12% c
>600
−2% b
17
536
7%
3%
10%
18
19
20
42 ± 12
121 ± 45
71, 131
Toxic e
Toxic e
9%
15%
15%
26% c
27% c
15%
12% c
18% c
21
22
67, 88
62, 77
30% c
28% c
36% c
29% c

Biomolecules 2021, 11, 55
7 of 17
Table 1. Cont.
Viability Assay (% of Uninfected Control)
SPR KD (µM) a
ID
Structure
200 µM
100 µM
50 µM
25 µM
9% b
23
24
94, 103
20%
18%
17%
>600
nd
25
26
27
102, 164
93 ± 31
118 ± 62
34%
13%
22%
17%
25%
11%
11%
15%
9%
10%
13%
−6% b
28
29
127, 351
Toxic e
110 ± 83
14% c
23% c
21%
12% c
25%
8% c
16%
−16% b
30
113 ± 31
31
120, 138
17%
38%
18%
33%
17%
21%
11%
22%
32 d
61 ± 10
33 d
63 ± 4
36%
38%
24%
24%
a
nd: not determined. KD has been determined by equilibrium analysis in the ProteOn Manager softwareTM
(Bio-Rad Laboratories Inc.). KD values consists of a mean KD
±
standard deviation (except from substances
with duplicates runs or less) calculated from three or more independent experiments. Standard deviation was
√
calculated using (( (x
−
M)2)/(n
−
1)), where x refers to the individual data points, M is the mean, and n is the
∑
number of data points. For compound where n < 3, the individual KD values are given. The compound is toxice
b
c
towards eukaryotic cells, which might affect the results negatively. The compound is fluorescent, which might
d
e
affect the results positively. Racemic compound. <80% viability compared to the DMSO control.

Biomolecules 2021, 11, 55
8 of 17
Figure 2. The x-axis shows the binding levels to the immobilized deactivated control protein arbonic
anhydrase isoenzyme II from bovine erythrocytes (CAII), and the y-axis shows the binding levels
(the average of two channels) to the immobilized PcrV protein for compounds from one exemplified
screening plate. The binding levels (in resonance units (RU)) were blank subtracted, molecular weight
adjusted, and multiplied by 100. The values on the y-axis show an average value of the binding levels
to PcrV proteins immobilized in two channels while the binding level on the x-axis is from CAII
immobilized in one channel. Two compounds that bind to PcrV with a higher selectivity than to CAII
are shown as filled squares (“selective PcrV binders”, one is H1 (Table 1)). One selective binder to
CAII are marked with a filled circle (“selective CAII binder”). The remaining compounds are shown
as triangles.
The analytes, i.e., the hits from the screening campaign that did not inhibit growth of
P. aeruginosa, and synthesized analogs were further evaluated in dose-response experiments
using SPR technique. First, the PcrV protein was immobilized in two channels as described
above while keeping a blank channel which was activated and deactivated as described
above. The multichannel module of the ProteOn system was in horizontal position during
the small molecule interaction studies. They were serially diluted to (400,) 200, 100, 50, 25,
12.5, 6.25, and (3.125); or 100, 50, 25, and 12.5 µM; or 100, 66.7, 44.4, 29.6, 19.7, and 13.2 µM
in PBST (0.05% Tween 20) with a total concentration of 5% DMSO in 96-well plates. The run-
ning buffer was PBST (0.05% Tween 20) containing 5% DMSO. The analytes were injected
at a flowrate of 100 µL/min, contact time was 60 s, and dissociation time was 60–120 s.
Solvent correction was performed continuously during the experiments to account for bulk
response of DMSO. The data was then baseline-aligned, processed by channel referencing
with the blank channel, and solvent-corrected. The equilibrium dissociation constant (KD)
values were determined from equilibrium analysis in the ProteOn Manager softwareTM
Solubility problems frequently hampered KD determination since the small molecules
could not always be dissolved in concentrations high enough to reach a plateau. The KD
.
values in Table 1 consist of an averaged KD
three independent experiments (except from substances where no mean KD or standard
±
standard deviation calculated from at least
deviation of KD is given) from one of the channels. Standard deviation was calculated
√
using (( (x
−
M)2)/(n
−
1)), where x refers to the individual data points, M is the mean,
∑
and n is the number of data points.

Biomolecules 2021, 11, 55
9 of 17
1
2.3. H NMR Binding Experiment
A solution containing 100
µM of the small molecule was prepared from 20 mM
stock solution of the tested compound in DMSO and diluted in PBS (pH 7.4) containing
trimethylsilyl propanoic acid (TMSP) as an internal standard. This solution was then split
into halves and used for preparation of the test and the reference samples as follows: the test
sample was prepared by adding 24 µL of 0.211 mM PcrV protein stock solution, resulting
in a final protein concentration of 10 µM, whereas the reference sample was prepared by
adding the corresponding volume of PBS buffer only before being transferred to 5-mm
NMR tubes. The relaxation-edited experiments shown in Figure 3D were recorded on
Bruker 600 MHz, and data were acquired at 298 K. Data from 256 scans were accumulated,
and a Carr-Purcell-Meiboom-Gill (cpmg) spin-lock of 200 ms was used. Excitation sculpture
was applied for water suppression.
Figure 3.
6.25, and 0
98.3
M. Rmax was 71.3 RU, and Chi2 was 10.0 RU. (
protects macrophages from T3SS-mediated P. aeruginosa toxicity (wt): uninfected cells and infection with a pcrV mutant
were used as controls, and standard deviation was calculated by Gaussian approximation. (
) 1H NMR relaxation-edited
binding experiment showing the aromatic region for compound H1 at 100 M with (bottom trace) and without 10
PcrV (upper trace): equilibrium analysis was performed in the ProteOn Manager softwareTM (Bio-Rad Laboratories Inc.,
Hercules, CA, USA). ( ) Bacterial growth assay where P. aeruginosa was grown in the presence of compound H1 at 200,
(
A
µ
) Surface plasmon resonance data for compound H1 binding to PcrV at 400, 200 (duplicate), 100, 50, 25, 12.5,
M. ( ) Equilibrium analysis of compound H1 from binding levels at 45-55 s in : KD was determined to be
) Infection assay with dose-response analysis of compound H1 that
B
A
µ
C
D
µ
µM
E
100, 50, and 25 µM and OD600 was measured regularly for 8 h: each value represents the mean value of a triplicate; to the
control wells, 1% DMSO was added (DMSO). The panels show one representative experiment.

Biomolecules 2021, 11, 55
10 of 17
2.4. Cell Viability Assay
The cell viability assay was performed essentially as described before [35]. J774
cells were seeded in 96-well micro-titer plates (Nunc), 100 L/well, to a final number of
50,000 cells/well in Dulbecco’s modified Eagle’s medium (DMEM) with glutamax (Merck)
supplemented with 10% fetal calf serum (FCS) and 3 g/mL gentamicin and incubated
µ
µ
for 16–18 h. Overnight cultures of the P. aeruginosa strains PAK and PAKpcrV [10] were
grown in Luria Broth (LB) on a rotary shaker at 220 rpm at 37 ◦C. Prior to infection, the
bacteria were diluted 1:10 in DMEM + glutamax and incubated for 1 h at 37 ◦C (shaking).
The J774 cells were washed once with PBS, then 30
FCS supplemented with the potential binders at concentrations of 50, 100, 200, and 400
were added to the wells in triplicates. The infection was initiated by adding 30 L of the
µL of DMEM with glutamax and 10%
µM
µ
diluted bacterial solutions (where the OD600 had been diluted to 0.0008) to each well to
initiate infection. This infection results in a final bacterial concentration of OD600 = 0.0004
and a multiplicities of infection (MOI) of approximately 8, with final concentrations of the
compounds at 25, 50, 100, and 200
µM in 60
µL of DMEM + glutamax and 5% FCS. In the
primary screening, the compounds were run at one concentration (100
µM) in duplicates.
As an infection control, the wild-type strain PAK with 1% DMSO was added, and as a
positive control, PAKpcrV with 1% DMSO was added into four wells each. After 3 h and
30 min, 10 µL of UptiBlue Viable Cell Counting Kit (Interchime, France) was added to
all wells. The infection was followed under the light microscope and the fluorescence
(excitation 535 nm and emission 595 nm) was measured at 4, 5, and 6 h in a microplate
reader (Synergy
™ H4, Biotek). The results were calculated as percentage of living cells
compared to uninfected control, where the uninfected control was set to 100% living cells
and the infected control was set to 0% living cells. As a control, compounds were added to
uninfected cells in a parallel experiment and viability was calculated as above to screen for
cell toxicity. A compound was considered toxic if the viability was <80% of the uninfected
control after 6 h of incubation and if a visual change of cell morphology was observed
by microscopy. Z’ was calculated for each plate to ensure good quality throughout the
screening. Only results from plates with a Z’ > 0.4 were counted. Standard deviation was
calculated with Gaussian approximation when appropriate and shown as error bars in
Figures 3C and 4A.
Figure 4. Compounds 32 and 33 inhibit cell infection without affecting bacterial growth. (A) Infection assay with dose-
response analysis of compound 32 and 33 that protects macrophages from T3SS-mediated P. aeruginosa toxicity (wt):
uninfected cells and infection with a pcrV mutant were used as controls, and standard deviation was calculated by Gaussian
approximation. Each value represents the mean value of a triplicate. (
B) Bacterial growth assay where P. aeruginosa was
grown in the presence of compounds 32 and 33 at 200, 100, 50, and 25 M and OD600 was measured regularly for 8 h: each
µ
value represents the mean value of a triplicate; to the control wells, 1% DMSO was added (DMSO).
2.5. Bacterial Toxicity Assay
Overnight culture of the P. aeruginosa strain PAK was diluted in LB (Luria Broth) to an
OD600 of 0.2. The bacteria were added into microtiter plates, and 25, 50, 100, and 200 M of
µ

Biomolecules 2021, 11, 55
11 of 17
the potential binders were added in triplicates to the bacteria with a final concentration of
1% DMSO. OD600 was measured every hour for seven hours in a microplate reader (Victor,
Perkin Elmer). Wells with 1% DMSO without substance were used as positive controls.
3. Results and Discussion
3.1. Screening and Hit Validation
Amino acid 20-294 of PcrV were cloned and expressed in Escherichia coli, and subse-
quent purification furnished pure PcrV in high yield (Figure S1). The protein was soluble
in phosphate-buffered saline (PBS), able to form secondary structures according to circular
dichroism (CD) spectroscopy measurements (Figure S2) and to bind to its cognate chap-
erone, PcrG, using NMR spectroscopy (data not shown), and was therefore considered
suitable for SPR screening.
Assay development and screening were performed on a ProteOn XPR36 SPR instru-
ment (Bio-RadTM Laboratories Inc., Hercules, CA, USA) that offers screening in 96-well
plates, allowing analysis of up to 36 different interactions in parallel. The target protein
PcrV was immobilized as duplicates on two different channels, and a deactivated control
protein, carbonic anhydrase (CAII, Sigma-Aldrich), was immobilized as a negative control
on a third channel. This makes it possible to distinguish between selective and nonselective
PcrV binders. To allow blank subtraction, a fourth channel was treated in the same way as
the other three channels (activation and deactivation) but without protein immobilization.
The proteins were immobilized covalently with amine coupling chemistry to carboxylic
acid functionalized chips activated with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide
hydrochloride (EDAC) and N-hydroxysulfosuccinimide (sulfo-NHS). A library composed
of 7600 small organic compounds selected from a ChemBridge diversity set dissolved in
dimethylsulfoxide (DMSO) were screened at 100 µM in PBS with 0.05% Tween 20 (PBST)
at a final DMSO concentration of 5%. Compounds with two times higher binding to PcrV
than to the deactivated control protein CAII were considered as potential binders, while the
best binders had more than five times higher binding to PcrV than to CAII, as exemplified
in Figure 2. The screening resulted in 395 potential PcrV binders that were tested for
biological activity in an infection assay at 100 µM [35].
The T3SS is important for P. aeruginosa infection of eukaryotic cells, and a T3SS
activator mutant or a mutant lacking the T3SS tip protein PcrV are not able to infect
eukaryotic cells [10
P. aeruginosa strain PAK, and the potential PcrV binders were added to the wells at a
final concentration of 100 M to screen for inhibition of infection. The T3SS-mediated
,15]. The macrophage cell line, J774, was infected with the wild-type
µ
cytotoxic effect on the eukaryotic cells was measured by using a viability staining method
(Uptiblue, Interchim), and the results were measured after 4, 5, and 6 h of infection. The
pcrV-mutant (PAKpcrV) [10], which is not able to infect macrophages due to deletion of
the PcrV protein, was used as a control. Out of the 395 tested compounds, 53 inhibited the
infection significantly (>3 standard deviations) and were chosen for further validation.
The binding affinity constants (KD) for binding of the compounds to PcrV were
determined by SPR dose-response experiments as exemplified for H1 (averaged KD
106 ± 36 µM; Figure 3A,B, Table 1, and Table S1). The 53 compounds were also eval-
uated in the macrophage cell infection assay at four different concentrations: 200, 100, 50,
and 25 µM, as exemplified for H1 (Figure 3C). The toxicity of the binders was also tested
towards P. aeruginosa since PcrV binders should function as virulence blockers, i.e., block
virulence without affecting bacterial growth (Figure 3E).
To confirm the binding data obtained from the SPR, a secondary qualitative label-free
1
1
binding experiment based on H-NMR was performed [50]. In this H-NMR binding
experiment, changes in relaxation time of resonances from small molecules (<600 Da) upon
binding to the large protein (PcrV) were observed, as exemplified by the signal reduction
for H1 (Figure 3D).
Taken together, the results from the validation of the binders resulted in 20 compounds
that bind the PcrV protein according to NMR spectroscopy and show dose-dependent

Biomolecules 2021, 11, 55
12 of 17
binding in SPR. In addition, these compounds showed biological activity by inhibiting
cell infection with low toxicity towards bacteria and eukaryotic cells. The identity and the
purity of the 20 potential PcrV binders that showed efficacy in the cell infection assay were
then validated using liquid chromatography mass spectrometry (LCMS). The ChEMBL
database [47,51] was used for cheminformatics and bioactivity mining using substructure
and similarity searches. Resynthesizing the most promising compounds along with a few
analogues and subsequent evaluation in SPR and in the cell infection assay confirmed H1
(Table 1) as the most promising starting point for further investigation.
3.2. Design, Synthesis and Structure–Activity Relationships
To probe the binding interactions between H1 and PcrV and to establish structure-
activity relationships, a number of analogues were designed and synthesized as outlined
in Schemes 1–3.
Scheme 1. Williamson ether synthesis and modification of the carboxylic acid functionality: the reagents and conditions
were ( ) Br(CH2)nCO2Et, K2CO3, KI, (CH3)2CO, reflux, 18–24 h; ( ) LiOH H2O, tetrahydrofuran (THF)/CH3OH/H2O
(3:1:1), 65 ◦C microwave irradiation (MWI), 15 min, 54–94% over two steps; (
) Br(CH2)nCO2Et, K2CO3, CH3CN, reflux,
24 h; ( ) NaOH (aq., 1M), THF/CH3OH, 40–80% over two steps; ( ) HATU, R’NHR”, dimethylformamide (DMF), room tem-
perature (RT), 18–20 h, 47–82%; (
a
b
·
c
d
e
f
) R’NHR”, 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethylaminium tetrafluoroborate (TBTU),
◦
DMF, RT, 18–20 h, 40–65%; (
0 C, 1 h, 79%.
g
) BH3.(CH3)2S, THF, 0 C-RT, 18 h, 74%; and (
h) 2,6-Lutidine, Me3SiCH2N2, CH2Cl2,/CH3OH,
◦
Scheme 2. Reagents and conditions: (
a
) Propargyl alcohol, Pd(PPh3)2Cl2 (5 mol%), CuI (3 mol%), triethylamine (TEA)/THF
) Pd/C 10%, H2, EtOAc (ethyl acetate)/CH3OH, RT, 2 h, 59%;
) 4-bromo-2,3-dimethylphenol, K2CO3, KI, (CH3)2CO, reflux, 18 h; and (
◦
(3:1), 40 C, microwave irradiation (MWI), 30 min, quant.; (
b
(
c
) CBr4, PPh3, CH2Cl2, 0 ◦C, 30 min, 94%; (
d
e)
◦
LiOH·H2O, THF/CH3OH/H2O (3:1:1), 65 C, MWI, 15 min, 85% (over two steps).
Scheme 3. Reagents and conditions: (
a) Allyl bromide, K2CO3, KI, (CH3)2CO, reflux, 18 h, 88%;
(b
) Ethyl diazoacetate, Rh2(OAc)2 (1 mol%), CH2Cl2, RT, 18 h, 68% overall yield (2:1 racemic E:Z
diastereoisomers which were separated by chromatography); and (
c
) LiOH H2O, THF/CH3OH/H2O
·
(3:1:1), 65 ◦C, MWI, 15 min, 78% (Z) and 27% (E) after preparative high-performance liquid chro-
matography (HPLC).
The central ether linker in all compounds
ether synthesis from the corresponding phenol derivatives in 40–94% yields as outlined
in Scheme 1. Compounds 18 27 with terminal amide or sulfonamide were synthesized
from the corresponding acid and amine using 1-[bis(dimethylamino)methylene]-1H-1,2,3-
1–33 were synthesized via Williamson
–

Biomolecules 2021, 11, 55
13 of 17
triazolo[4,5-b]pyridinium 3-oxid hexafluoro- phosphate (HATU) or 1H-benzotriazole-1-
yl)-1,1,3,3-tetramethylaminium tetrafluoroborate (TBTU) as coupling agents (Scheme 1).
Alcohol 28 and ester 29 were prepared from the corresponding acid via reduction with
borane or esterification with trimethylsilyldiazomethane, respectively. Compound 30 was
synthesized by employing Sonogashira coupling and alkyne reduction followed by Appel
reaction and O-alkylation (Scheme 2).
Compounds 32 and 33 with cyclopropyl linker were synthesized from the correspond-
ing allyloxy substrate by applying intermolecular carbenoid cyclopropanation reaction
with dirhodium tetraacetate and ethyl diazoacetate [52] that afforded the desired substance
in a diastereomeric mixture in 2:1 (E/Z) ratio, which was subjected to chromatography that
allowed the separation of the E and Z diastereoisomers in their racemic form and an overall
yield of 68% (Scheme 3).
All analogues were evaluated for dose-dependent binding to PcrV using SPR (Table 1);
compounds with dose-dependent binding were tested for virulence blocking activity in
the macrophage infection assay and for toxicity towards bacteria and eukaryotic cells.
Substitution: Based on the possibility that the aromatic ring of H1 is involved in a
hydrophobic or halogen interaction [53], we designed and synthesized analogues with
various substituents on different positions of the aromatic ring. We observed that dihalo-
genated aromatics with a hydrophobic character such as 2,4-dichloro (
) showed binding affinities < 127 M, while the 4-chloro ( ) and 4-iodo (
fluoro ( ) showed no binding at the examined concentrations. Electron-donating groups
with a resonance effect located ortho ( ), meta ( ), or para ( ) to the linker impaired binding.
Analogues carrying one electron withdrawing group in the meta ( ) position as well as
polar functionalities with the capacity to form hydrogen bonds such as carboxylic acid
10) showed no binding to PcrV. The same result was obtained with a heteroaromatic
1) or 3,5-dicholoro
(2
µ
3
4) or 4-bromo-2-
5
6
7
8
9
(
compound (11). Taken together, these results suggest a preference for an aromatic ring with
hydrophobic and electron withdrawing substituents. Accordingly, a few trisubstituted
analogues were synthesized by shuffling the dimethyl groups around the ring (12
14) while keeping a bromo or cyano substituent in the para position. The 2,6-dimethyl
pattern (12 and 13) reduced binding, while hydrophobic substituents on positions 2,3 (H1
or 3,5 (14) resulted in binding affinities < 150 M. To explore the possibility of extending
the aromatic group further, we tested a larger substituent, such as p-cyclohexyl (15), benzyl
16), or phenoxy (17) on the aromatic ring. Only the p-cyclohexyl substituent was tolerated
, 13, and
)
µ
(
with KD of 153 µM (15).
Carboxylic acid functionality: To study the importance of the carboxylic acid moiety
and the possibility to extend the structure via this functionality, we synthesized analogues
with amide (18–24), sulfonamide (25–27), alcohol (28), ester (29), and aromatic carboxylic
(
30) functionalities. Analogues with carboxamide (18) or N-methyl carboxamide (19
)
showed improved or similar binding affinity to the hit carboxylic acid (H1). N-substituted
amides carrying carboxylic acid at different spatial arrangements such as racemic nipecotic
acid (20), phenylalanine (21), pipecolic acid (22), isonipecotic acid (23), 2-aminobenzoic
acid (24), and ortho-linked benzoic acid (30) showed binding with KD values in the range of
62 to 131 µM with the exception of 24 that failed to bind PcrV. The underlying reason might
be its ability to form a stable intramolecular hydrogen bond with the amide linker, as noted
in the 1H-NMR spectroscopy for this compound, thus affecting its overall conformation.
Interestingly, the H1 analogue with an ester functionality (29) that lacks the ionic head and
the hydrogen bonding was capable of binding to PcrV with KD of 110
µM. A similar result
was obtained with the corresponding alcohol (28) with KD of 127 and 351
µM, respectively.
Acyl sulfonamide isosteres (25–27) showed affinities with KD in a range of 93–164 µM.
Linker: A small set of compounds in which we varied the length and conformation
of the linker carrying the carboxylic acid moiety were synthesized. A one-atom longer
linker (31) showed similar affinity to PcrV (KD 120 and 138
µM, respectively) as the Hit
compound H1, whereas a conformationally locked linker with a racemic Z or E-cyclopropyl

Biomolecules 2021, 11, 55
14 of 17
group (32 and 33) showed improved affinity with KD of around 60
µM. Full exploration of
the potential of linker optimization would be facilitated by access to the structure of PcrV.
3.3. Anti-Virulence Properties
All analogues that showed dose-dependent responses in the SPR binding experi-
ments were then evaluated for their anti-virulence efficacy in the infection assay, and
their inhibitory effect were calculated (Table 1). Many analogues displayed no or modest
anti-virulence activity without clear dose-response (e.g., 20 and 31), and for some, this
could be linked to eukaryotic cell toxicity (e.g., 18 and 19) (<80% viability compared to
DMSO control) or assay interference due to autofluorescence (e.g., 15 and 22). However,
a set of compounds proved to be as efficacious as or better than H1 such as the close
analogue 14, the carboxylic acid isostere 25, and compounds 32 and 33 with a cyclopropyl
modified linker. Importantly, the most promising analogues 32 and 33 display SPR KD
around 61–63
(Figure 4). Furthermore, this data suggest that stereochemistry can be explored to further
optimize efficacy. Compound is a typical chlorophenoxy herbicide with adverse effects in
µM and dose-dependent inhibition of the infection up to 38% at 200 µM
1
humans, and it is possible that the toxic effects observed for some of the compounds can be
traced to this similarity [54].
Nevertheless, a clear correlation between binding affinity and efficacy in the infection
model could not be established. The biophysical SPR method is simplified in comparison to
the infection model where PcrV is present in its native form as a T3SS multiprotein complex
in a context involving both living bacteria and eukaryotic cells. The infection assay is also
affected by the cell toxicity of the substance and serum binding, while the binding assay is
not; this might also affect the correlation between the results for some of the compounds.
4. Conclusions
In this study, we develop and describe a screening funnel to identify nontoxic PcrV
binders that block T3SS-mediated virulence in a P. aeruginosa cell infection assay as well as
to perform a structure-activity relationship analysis with the aim to increase the potency
and to lower the toxicity of the hit compound. The analysis indicates the importance of an
appropriately substituted aromatic ring of the hit compound H1 and that the linker and
carboxylic acid functionality can be explored to reach more potent nontoxic compounds
with anti-virulence properties. Importantly, the compounds are small in size, display good
aqueous solubility, are amendable to medicinal chemistry, and could be considered as good
starting points for further development. Continued optimization would greatly benefit
from access to the PcrV structure. These compounds thus hold promise to be further
developed into compounds with efficacy in vivo. Ultimately, small molecule virulence
blocking agents targeting PcrV can be used in preventive care, as standalone therapy or
in combination with conventional antibiotics, and can thereby improve our capabilities to
treat P. aeruginosa infections.
Supplementary Materials: The following are available online at https://www.mdpi.com/2218-2
73X/11/1/55/s1, Figure S1: Purification of PcrV, Figure S2: Stability of PcrV, Table S1: SPR dose-
response curves for compounds with determined KD, Table S2: cLogP for all compounds, buffers, and
medium used during protein purification chemical synthesis, general chemistry, synthetic methods,
and analytical data for all compounds.
Author Contributions: Conceptualization, C.S., M.S., S.S., and M.E.; methodology, C.S., M.S., and
S.S.; validation, C.S., M.S., S.S., and M.E.; formal analysis, C.S., M.S., and S.S.; investigation, C.S., M.S.,
L.Q., and S.S.; resources, M.E.; data curation, C.S., M.S., and S.S.; writing—original draft preparation,
C.S. and M.S.; writing—review and editing, C.S., M.S., S.S., and M.E.; visualization, C.S., M.S., and
S.S.; supervision, M.E.; project administration, C.S.; funding acquisition, M.E. All authors have read
and agreed to the published version of the manuscript.
Funding: This project was funded by the Swedish Foundation for Strategic Research, grant number
SB12-0022.

Biomolecules 2021, 11, 55
15 of 17
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data is contained within the article and the Supplementary Materials.
Acknowledgments: We thank Mattias Hedenström and the Knut and Alice Wallenberg foundation
program NMR for Life for technical support with the NMR experiments, Yngve Östberg for support
with cloning of PcrV, Åke Forsberg for scientific advice, the Laboratories for Chemical Biology Umeå
(LCBU) at Umeå University for access to screening compounds.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
O’Neill, J. Antimicrobial Resistance: Tackling a Crisis for the Health and Wealth of Nations 2014. Available online:
https://amr-review.org/sites/default/files/AMR%20Review%20Paper%20-%20Tackling%20a%20crisis%20for%20the%20
health%20and%20wealth%20of%20nations_1.pdf (accessed on 11 November 2020).
2.
3.
4.
5.
Lister, P.D.; Wolter, D.J.; Hanson, N.D. Antibacterial-resistant Pseudomonas aeruginosa: Clinical impact and complex regulation of
chromosomally encoded resistance mechanisms. Clin. Microbiol. Rev. 2009, 22, 582–610. [CrossRef]
Breidenstein, E.B.; de la Fuente-Nunez, C.; Hancock, R.E. Pseudomonas aeruginosa: All roads lead to resistance. Trends Microbiol.
2011, 19, 419–426. [CrossRef] [PubMed]
Potron, A.; Poirel, L.; Nordmann, P. Emerging broad-spectrum resistance in Pseudomonas aeruginosa and Acinetobacter baumannii:
Mechanisms and epidemiology. Int. J. Antimicrob. Agents 2015, 45, 568–585. [CrossRef] [PubMed]
Sader, H.S.; Huband, M.D.; Castanheira, M.; Flamm, R.K. Pseudomonas aeruginosa antimicrobial susceptibility results from four
years (2012 to 2015) of the international network for optimal resistance monitoring program in the United States. Antimicrob.
Agents Chemother. 2017, 61, e02252-16. [CrossRef] [PubMed]
6.
7.
8.
9.
Bleves, S.; Viarre, V.; Salacha, R.; Michel, G.P.; Filloux, A.; Voulhoux, R. Protein secretion systems in Pseudomonas aeruginosa: A
wealth of pathogenic weapons. Int. J. Med. Microbiol. 2010, 300, 534–543. [CrossRef]
Haiko, J.; Westerlund-Wikström, B. The role of the bacterial flagellum in adhesion and virulence. Biology 2013, 2, 1242–1267.
[CrossRef]
Leighton, T.L.; Buensuceso, R.N.; Howell, P.L.; Burrows, L.L. Biogenesis of Pseudomonas aeruginosa type IV pili and regulation of
their function. Environ. Microbiol. 2015, 17, 4148–4163. [CrossRef]
Rasmussen, T.B.; Givskov, M. Quorum-sensing inhibitors as anti-pathogenic drugs. Int. J. Med. Microbiol. 2006, 296, 149–161.
[CrossRef]
10. Sundin, C.; Wolfgang, M.C.; Lory, S.; Forsberg, A.; Frithz-Lindsten, E. Type IV pili are not specifically required for contact
dependent translocation of exoenzymes by Pseudomonas aeruginosa. Microb. Pathog. 2002, 33, 265–277. [CrossRef]
11. Shao, X.; Xie, Y.; Zhang, Y.; Liu, J.; Ding, Y.; Wu, M.; Wang, X.; Deng, X. Novel therapeutic strategies for treating Pseudomonas
aeruginosa infection. Expert Opin. Drug Discov. 2020, 15, 1403–1423. [CrossRef]
12. Feng, C.; Huang, Y.; He, W.; Cheng, X.; Liu, H.; Huang, Y.; Ma, B.; Zhang, W.; Liao, C.; Wu, W.; et al. Tanshinones: First-in-class
inhibitors of the biogenesis of the type 3 secretion system needle of Pseudomonas aeruginosa for antibiotic therapy. ACS Cent. Sci.
2019, 5, 1278–1288. [CrossRef] [PubMed]
13. Roy-Burman, A.; Savel, R.H.; Racine, S.; Swanson, B.L.; Revadigar, N.S.; Fujimoto, J.; Sawa, T.; Frank, D.W.; Wiener-Kronish, J.P.
Type III protein secretion is associated with death in lower respiratory and systemic Pseudomonas aeruginosa infections. J. Infect.
Dis. 2001, 183, 1767–1774. [CrossRef] [PubMed]
14. Hauser, A. The type III secretion system of Pseudomonas aeruginosa: Infection by injection. Nat. Rev. Microbiol. 2009, 7, 654–665.
[CrossRef] [PubMed]
15. Sundin, C.; Thelaus, J.; Broms, J.E.; Forsberg, A. Polarisation of type III translocation by Pseudomonas aeruginosa requires PcrG,
PcrV and PopN. Microb. Pathog. 2004, 37, 313–322. [CrossRef] [PubMed]
16. Cornelis, G.R. The type III secretion injectisome, a complex nanomachine for intracellular ‘toxin’delivery. Biol. Chem. 2010
391, 745–751. [CrossRef] [PubMed]
,
17. Portaliou, A.G.; Tsolis, K.C.; Loos, M.S.; Zorzini, V.; Economou, A. Type III Secretion: Building and operating a remarkable
nanomachine. Trends Biochem. Sci. 2015, 41, 175–189. [CrossRef]
18. Sawa, T.; Ito, E.; Nguyen, V.H.; Haight, M. Anti-PcrV antibody strategies against virulent Pseudomonas aeruginosa. Hum. Vaccines
Immunother. 2014, 10, 2843–2852. [CrossRef]
19. Caroline, G.; Eric, F.; Bohn, Y.-S.T.; Sylvie, E.; Attree, I. Oligomerization of PcrV and LcrV, protective antigens of Pseudomonas
aeruginosa and Yersinia pestis. J. Biol. Chem. 2008, 283, 23940–23949. [CrossRef]
20. Frithz-Lindsten, E.; Holmström, A.; Jacobsson, L.; Soltani, M.; Olsson, J.; Rosqvist, R.; Forsberg, Å. Functional conservation of
the effector protein translocators PopB/YopB and PopD/YopD of Pseudomonas aeruginosa and Yersinia pseudotuberculosis. Mol.
Microbiol. 1998, 29, 1155–1165. [CrossRef]
21. Dacheux, D.; Goure, J.; Chabert, J.; Usson, Y.; Attree, I. Pore-forming activity of type III system-secreted proteins leads to oncosis
of Pseudomonas aeruginosa-infected macrophages. Mol. Microbiol. 2001, 40, 76–85. [CrossRef]

Biomolecules 2021, 11, 55
16 of 17
22. Allmond, L.R.; Karaca, T.J.; Nguyen, V.N.; Nguyen, T.; Wiener-Kronish, J.P.; Sawa, T. Protein binding between PcrG-PcrV and
PcrH-PopB/PopD encoded by the pcrGVH-popBD operon of the Pseudomonas aeruginosa type III secretion system. Infect. Immun.
2003, 71, 2230–2233. [CrossRef] [PubMed]
23. Mueller, C.; Broz, P.; Cornelis, G. The type III secretion system tip complex and translocon. Mol. Microbiol. 2008, 68, 1085–1095.
[CrossRef] [PubMed]
24. Schoehn, G.; Di Guilmi, A.M.; Lemaire, D.; Attree, I.; Weissenhorn, W.; Dessen, A. Oligomerization of type III secretion proteins
PopB and PopD precedes pore formation in Pseudomonas. EMBO J. 2003, 22, 4957–4967. [CrossRef] [PubMed]
25. Goure, J.; Broz, P.; Attree, O.; Cornelis, G.R.; Attree, I. Protective anti-V antibodies inhibit Pseudomonas and Yersinia translocon
assembly within host membranes. J. Infect. Dis. 2005, 192, 218–225. [CrossRef] [PubMed]
26. Goure, J.; Pastor, A.; Faudry, E.; Chabert, J.; Dessen, A.; Attree, I. The V antigen of Pseudomonas aeruginosa is required for assembly
of the functional PopB/PopD translocation pore in host cell membranes. Infect. Immun. 2004, 72, 4741–4750. [CrossRef]
27. Rasko, D.A.; Sperandio, V. Anti-virulence strategies to combat bacteria-mediated disease. Nat. Rev. Drug Discov. 2010, 9, 117–128.
[CrossRef]
28. Clatworthy, A.E.; Pierson, E.; Hung, D.T. Targeting virulence: A new paradigm for antimicrobial therapy. Nat. Chem. Biol. 2007
3, 541–548. [CrossRef]
,
29. Marra, A. Targeting virulence for antibacterial chemotherapy: Identifying and characterising virulence factors for lead discovery.
Drugs R D 2006, 7, 1–16. [CrossRef]
30. Lyons, B.J.E.; Strynadka, N.C.J. On the road to structure-based development of anti-virulence therapeutics targeting the type III
secretion system injectisome. Medchemcomm 2019, 10, 1273–1289. [CrossRef]
31. Anantharajah, A.; Mingeot-Leclercq, M.P.; Van Bambeke, F. Targeting the type three secretion system in Pseudomonas aeruginosa.
Trends Pharmacol. Sci. 2016, 37, 734–749. [CrossRef]
32. Anantharajah, A.; Faure, E.; Buyck, J.M.; Sundin, C.; Lindmark, T.; Mecsas, J.; Yahr, T.L.; Tulkens, P.M.; Mingeot-Leclercq, M.P.;
Guery, B.; et al. Inhibition of the injectisome and flagellar type III secretion systems by INP1855 impairs Pseudomonas aeruginosa
pathogenicity and inflammasome activation. J. Infect. Dis. 2016, 214, 1105–1116. [CrossRef] [PubMed]
33. Uusitalo, P.; Hagglund, U.; Rhoos, E.; Scherman Norberg, H.; Elofsson, M.; Sundin, C. The salicylidene acylhydrazide INP0341
attenuates Pseudomonas aeruginosa virulence in vitro and in vivo. J. Antibiot. (Tokyo) 2017, 70, 937–943. [CrossRef] [PubMed]
34. Zetterstrom, C.E.; Hasselgren, J.; Salin, O.; Davis, R.A.; Quinn, R.J.; Sundin, C.; Elofsson, M. The resveratrol tetramer (-)-
hopeaphenol inhibits type III secretion in the gram-negative pathogens Yersinia pseudotuberculosis and Pseudomonas aeruginosa.
PLoS ONE 2013, 8, e81969. [CrossRef] [PubMed]
35. Sundin, C.; Zetterstrom, C.E.; Vo, D.D.; Brkljaca, R.; Urban, S.; Elofsson, M. Exploring resveratrol dimers as virulence blocking
agents—Attenuation of type III secretion in Yersinia pseudotuberculosis and Pseudomonas aeruginosa. Sci. Rep. 2020, 10, 2103.
[CrossRef] [PubMed]
36. Ngo, T.D.; Ple, S.; Thomas, A.; Barette, C.; Fortune, A.; Bouzidi, Y.; Fauvarque, M.O.; Pereira de Freitas, R.; Francisco Hilario,
F.; Attree, I.; et al. Chimeric protein-protein interface inhibitors allow efficient inhibition of type III secretion machinery and
Pseudomonas aeruginosa virulence. ACS Infect. Dis. 2019, 5, 1843–1854. [CrossRef]
37. Aiello, D.; Williams, J.D.; Majgier-Baranowska, H.; Patel, I.; Peet, N.P.; Huang, J.; Lory, S.; Bowlin, T.L.; Moir, D.T. Discovery and
characterization of inhibitors of Pseudomonas aeruginosa type III secretion. Antimicrob. Agents Chemother. 2010, 54, 1988–1999.
[CrossRef] [PubMed]
38. Audia, J.P.; Lindsey, A.S.; Housley, N.A.; Ochoa, C.R.; Zhou, C.; Toba, M.; Oka, M.; Annamdevula, N.S.; Fitzgerald, M.S.; Frank,
D.W. In the absence of effector proteins, the Pseudomonas aeruginosa type three secretion system needle tip complex contributes to
lung injury and systemic inflammatory responses. PLoS ONE 2013, 8, e81792. [CrossRef]
39. Sawa, T.; Yahr, T.L.; Ohara, M.; Kurahashi, K.; Gropper, M.A.; Wiener-Kronish, J.P.; Frank, D.W. Active and passive immunization
with the Pseudomonas V antigen protects against type III intoxication and lung injury. Nat. Med. 1999, 5, 392. [CrossRef]
40. Imamura, Y.; Yanagihara, K.; Fukuda, Y.; Kaneko, Y.; Seki, M.; Izumikawa, K.; Miyazaki, Y.; Hirakata, Y.; Sawa, T.; Wiener-Kronish,
J. Effect of anti-PcrV antibody in a murine chronic airway Pseudomonas aeruginosa infection model. Eur. Respir. J. 2007, 29, 965–968.
[CrossRef]
41. Faure, K.; Fujimoto, J.; Shimabukuro, D.W.; Ajayi, T.; Shime, N.; Moriyama, K.; Spack, E.G.; Wiener-Kronish, J.P.; Sawa, T. Effects
of monoclonal anti-PcrV antibody on Pseudomonas aeruginosa-induced acute lung injury in a rat model. J. Immune Based Ther.
Vaccines 2003, 1, 2. [CrossRef]
42. Song, Y.; Baer, M.; Srinivasan, R.; Lima, J.; Yarranton, G.; Bebbington, C.; Lynch, S. PcrV antibody–antibiotic combination improves
survival in Pseudomonas aeruginosa-infected mice. Eur. J. Clin. Microbiol. Infect. Dis. 2012, 31, 1837–1845. [CrossRef] [PubMed]
43. Shime, N.; Sawa, T.; Fujimoto, J.; Faure, K.; Allmond, L.R.; Karaca, T.; Swanson, B.L.; Spack, E.G.; Wiener-Kronish, J.P. Therapeutic
0
administration of anti-PcrV F (ab ) 2 in sepsis associated with Pseudomonas aeruginosa. J. Immun. 2001, 167, 5880–5886. [CrossRef]
[PubMed]
44. Thanabalasuriar, A.; Surewaard, B.G.; Willson, M.E.; Neupane, A.S.; Stover, C.K.; Warrener, P.; Wilson, G.; Keller, A.E.; Sellman,
B.R.; DiGiandomenico, A. Bispecific antibody targets multiple Pseudomonas aeruginosa evasion mechanisms in the lung vasculature.
J. Clin. Investig. 2017, 127, 2249–2261. [CrossRef] [PubMed]

Biomolecules 2021, 11, 55
17 of 17
45. Milla, C.E.; Chmiel, J.F.; Accurso, F.J.; VanDevanter, D.R.; Konstan, M.W.; Yarranton, G.; Geller, D.E.; Group, K.S. Anti-PcrV
antibody in cystic fibrosis: A novel approach targeting Pseudomonas aeruginosa airway infection. Pediatr. Pulmonol. 2014
49, 650–658. [CrossRef] [PubMed]
,
46. Sato, H.; Hunt, M.L.; Weiner, J.J.; Hansen, A.T.; Frank, D.W. Modified needle-tip PcrV proteins reveal distinct phenotypes relevant
to the control of type III secretion and intoxication by Pseudomonas aeruginosa. PLoS ONE 2011, 6, e18356. [CrossRef] [PubMed]
47. McShan, A.C.; Anbanandam, A.; Patnaik, S.; De Guzman, R.N. Characterization of the binding of hydroxyindole, indoleacetic
acid, and morpholinoaniline to the Salmonella type III secretion system proteins SipD and SipB. ChemMedChem 2016, 11, 963–971.
[CrossRef] [PubMed]
48. Dey, S.; Anbanandam, A.; Mumford, B.E.; De Guzman, R.N. Characterization of small molecule scaffolds that bind to the Shigella
type III secretion system protein IpaD. ChemMedChem 2017, 12, 1534–1541. [CrossRef]
49. Neumann, T.; Junker, H.; Schmidt, K.; Sekul, R. SPR-based fragment screening: Advantages and applications. Curr. Top. Med.
Chem. 2007, 7, 1630–1642. [CrossRef]
50. Hajduk, P.J.; Olejniczak, E.T.; Fesik, S.W. One-dimensional relaxation- and diffusion-edited NMR methods for screening com-
pounds that bind to macromolecules. J. Am. Chem. Soc. 1997, 119, 12257–12261. [CrossRef]
51. Gaulton, A.; Bellis, L.J.; Bento, A.P.; Chambers, J.; Davies, M.; Hersey, A.; Light, Y.; McGlinchey, S.; Michalovich, D.; Al-Lazikani,
B. ChEMBL: A large-scale bioactivity database for drug discovery. Nucleic Acids Res. 2011, 40, D1100–D1107. [CrossRef]
52. Davies, H.M.L.; Antoulinakis, E.G. Intermolecular metal-catalyzed carbenoid cyclopropanations. In Organic Reactions; John Wiley
& Sons, Inc.: Hoboken, NJ, USA, 2004; pp. 1–326.
53. Sirimulla, S.; Bailey, J.B.; Vegesna, R.; Narayan, M. Halogen Interactions in Protein-Ligand Complexes: Implications of halogen
bonding for rational drug design. J. Chem. Inf. Model 2013, 53, 2781–2791. [CrossRef] [PubMed]
54. Bradberry, S.M.; Watt, B.E.; Proudfoot, A.T.; Vale, J.A. Mechanisms of toxicity, clinical features, and management of acute
chlorophenoxy herbicide poisoning: A review. J. Toxicol. Clin. Toxicol. 2000, 38, 111–122. [CrossRef] [PubMed]