Design, Synthesis, and Biochemical Characterization of Non-Native Antagonists of the Pseudomonas aeruginosa Quorum Sensing Receptor LasR with Nanomolar IC50 Values

Article abstract of DOI:10.1021/acsinfecdis.9b00518

Quorum sensing (QS), a bacterial cell-to-cell communication system mediated by small molecules and peptides, has received significant interest as a potential target to block infection. The common pathogen Pseudomonas aeruginosa uses QS to regulate many of its virulence phenotypes at high cell densities, and the LasR QS receptor plays a critical role in this process. Small molecule tools that inhibit LasR activity would serve to illuminate its role in P. aeruginosa virulence, but we currently lack highly potent and selective LasR antagonists, despite considerable research in this area. V-06-018, an abiotic small molecule discovered in a high-throughput screen, represents one of the most potent known LasR antagonists but has seen little study since its initial report. Herein, we report a systematic study of the structure-activity relationships (SARs) that govern LasR antagonism by V-06-018. We synthesized a focused library of V-06-018 derivatives and evaluated the library for bioactivity using a variety of cell-based LasR reporter systems. The SAR trends revealed by these experiments allowed us to design probes with 10-fold greater potency than that of V-06-018 and 100-fold greater potency than other commonly used N-acyl-l-homoserine lactone (AHL)-based LasR antagonists, along with high selectivities for LasR. Biochemical experiments to probe the mechanism of antagonism by V-06-018 and its analogues support these compounds interacting with the native ligand-binding site in LasR and, at least in part, stabilizing an inactive form of the protein. The compounds described herein are the most potent and efficacious antagonists of LasR known and represent robust probes both for characterizing the mechanisms of LuxR-type QS and for chemical biology research in general in the growing QS field.

Full text of DOI:10.1021/acsinfecdis.9b00518

Subscriber access provided by UNIV OF CAMBRIDGE
Article
Design, synthesis, and biochemical characterization of
non-native antagonists of the Pseudomonas aeruginosa
quorum sensing receptor LasR with nanomolar IC50 values
Daniel E. Manson, Matthew C. O'Reilly, Kayleigh E. Nyffeler, and Helen E. Blackwell
ACS Infect. Dis., Just Accepted Manuscript • DOI: 10.1021/acsinfecdis.9b00518 • Publication Date (Web): 09 Feb 2020
Downloaded from pubs.acs.org on February 11, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
Design, synthesis, and biochemical characterization of non-native an-
tagonists of the Pseudomonas aeruginosa quorum sensing receptor
LasR with nanomolar IC50 values
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Daniel E. Manson, Matthew C. O’Reilly,† Kayleigh E. Nyffeler, and Helen E. Blackwell*
Department of Chemistry, University of Wisconsin–Madison, 110 University Ave., Madison, WI
53706 USA; blackwell@chem.wisc.edu
Quorum sensing (QS), a bacterial cell-to-cell communication system mediated by small molecules
and peptides, has received significant interest as a potential target to block infection. The common
pathogen Pseudomonas aeruginosa uses QS to regulate many of its virulence phenotypes at high
cell densities, and the LasR QS receptor plays a critical role in this process. Small molecule tools
that inhibit LasR activity would serve to illuminate its role in P. aeruginosa virulence, but we cur-
rently lack highly potent and selective LasR antagonists, despite considerable research in this area.
V-06-018, an abiotic small molecule discovered in a high-throughput screen, represents one of the
most potent known LasR antagonists, but has seen little study since its initial report. Herein, we re-
port a systematic study of the structure-activity relationships (SARs) that govern LasR antagonism
by V-06-018. We synthesized a focused library of V-06-018 derivatives and evaluated the library
for bioactivity using a variety of cell-based LasR reporter systems. The SAR trends revealed by
these experiments allowed us to design probes with 10-fold greater potency than V-06-018 and 100-
fold greater potency than other commonly used N-acyl L-homoserine lactone (AHL)-based LasR
antagonists, along with high selectivities for LasR. Biochemical experiments to probe the mecha-
nism of antagonism by V-06-018 and its analogs support these compounds interacting with the na-
tive ligand-binding site in LasR and, at least in part, stabilizing an inactive form of the protein. The
compounds described herein are the most potent and efficacious antagonists of LasR known, and
represent robust probes both for characterizing the mechanisms of LuxR-type QS and for chemical
biology research in general in the growing QS field.
1
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 2 of 39
1
2
3
4
5
6
Keywords: N-acyl L-homoserine lactone; bacterial communication; intercellular signaling; LuxR-
7
8
type receptor; small molecule probes; virulence
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
Microbial resistance to antibiotics is emerging faster than new treatments are being developed, set-
ting the stage for a public health crisis.1-2 As traditional antibiotics become less effective, interest
has arisen in attenuating virulence via interference with nonessential pathways.3 Inhibition of
quorum sensing (QS), a mode of bacterial communication dependent on the exchange of chemical
signals, has been shown to reduce virulence phenotypes in multiple human pathogens without af-
fecting cell viability.4-6 Accordingly, it has attracted significant interest as a potential anti-virulence
strategy for combatting bacterial infections.7-8 Our laboratory9-11 and others12-15 are interested in the
development of small molecule and peptide probes to dissect the mechanisms of QS and their roles
in infection.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The prototypical QS circuit in Gram-negative bacteria is the LuxI/LuxR synthase/receptor
pair, first discovered in the marine symbiont Vibrio fischeri.6 At low cell density, a LuxI-type en-
zyme synthesizes the QS signal, an N-acyl L-homoserine lactone (AHL), at a low basal rate. These
low-molecular weight molecules can freely diffuse out of the cell, although in certain cases they are
also actively exported.16 The concentration of AHL signal is largely proportional to cell density
(and this correlation is highly dependent on the environment), but as a bacterial community grows,
the level of AHL signal in the local environment likewise increases (Figure 1A). At high cell densi-
ties, the intracellular AHL concentration is sufficient for productive binding of the AHL to its cog-
nate LuxR-type receptor, a transcription factor. The activated receptor:ligand complex then typi-
cally dimerizes and binds to DNA, which subsequently alters gene expression levels to promote
group-beneficial behaviors. In pathogenic bacteria, these behaviors can include the production of
toxic virulence factors and biofilm. Typically, once a “quorum” is achieved, expression of the LuxI-
type synthase is also increased, amplifying AHL production in a positive “autoinduction” feedback
loop.17
3
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 4 of 39
1
2
3
4
5
6
7
8
9
Pseudomonas aeruginosa is an opportunistic pathogen that regulates many aspects of viru-
lence using QS. This bacterium has a high rate of resistance to traditional antibiotics and causes in-
fections that are especially dangerous for individuals with cystic fibrosis (CF), burn victims, and
AIDS patients. The QS system in P. aeruginosa is relatively complex (Figure 1B),18 consisting of
two LuxI/LuxR pairs (LasI/LasR and RhlI/RhlR) along with an orphan LuxR-type receptor (QscR),
which lacks a related synthase and native AHL signal. LasI synthesizes N-(3-oxo-dodecanoyl)-L-
homoserine lactone (OdDHL), which targets LasR but also strongly activates QscR. RhlI synthe-
sizes N-butyryl-L-homoserine lactone (BHL), which targets RhlR. Additionally, P. aeruginosa has a
LysR-type receptor, PqsR, which is unrelated to LuxR-type receptors and uses 2-heptyl-3-hydroxy-
4-(1H)-quinolone (i.e., the Pseudomonas quinolone signal (PQS)) as its ligand. These four QS sys-
tems are intimately linked and control different aspects of P. aeruginosa virulence that are highly
dependent on the environment (Figure 1B).9 LasR plays a central role in the QS hierarchy. For in-
stance, LasR directly regulates the production of virulence factors such as elastase, alkaline prote-
ase, and exotoxin, and regulates rhamnolipid, HCN, and pyocyanin production via control of the rhl
and pqs systems.18 Biofilm, a major virulence phenotype in P. aeruginosa, is also regulated by LasR
via the rhl and pqs systems.19 In turn, LasR and RhlR are repressed by QscR, which again is
strongly activated via LasR’s native signal, OdDHL.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
ACS Paragon Plus Environment

Page 5 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1: (A) General schematic of LuxI/LuxR-type quorum sensing (QS) in Gram-negative bacteria. (B)
Simplified view of QS in P. aeruginosa. LasI/R and RhlI/R are LuxI/R homologues. QscR is an “orphan”
LuxR-type receptor and responds to OdDHL. PqsR is a LysR-type receptor that responds to the Pseudomo-
nas quinolone signal (PQS). AHL synthases are omitted for clarity. (C) Structures of native agonist OdDHL
(EC50 = 139 nM), non-AHL antagonist V-06-018 (IC50 = 5.2 µM), non-AHL agonist TP1-P (EC50 = 71 nM),
and representative, synthetic AHL antagonist 4-bromo PHL (IC50 = 116 µM); potency values all obtained in
the same P. aeruginosa LasR reporter (from ref. 30).20
The connection between QS and virulence in P. aeruginosa, and in other Gram-negative
bacterial pathogens, has motivated the development of small molecules and macromolecules capa-
ble of inhibiting LuxI-type synthases,21 destroying or sequestering AHL signals,22 or blocking the
5
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 6 of 39
1
2
3
4
5
6
7
8
9
binding of AHL signal to LuxR-type receptor.23 The latter competitive inhibition strategy has seen
the most study to date, with significant contributions by the Spring,7 Bassler,24 Greenberg,25 and
Meijler15 labs, as well as our lab.26 Due to its prominent position in the P. aeruginosa QS system
(vide supra), much of the effort devoted to identifying small molecule modulators of QS in P. aeru-
ginosa has focused on LasR. The majority of the known synthetic ligands that modulate LasR were
identified by making systematic changes to the lactone “head group” and acyl “tail group” of
OdDHL (e.g., 4-bromo PHL; Figure 1C).27-28 However, these past efforts have failed to yield com-
pounds that antagonize LasR with both high efficacies and potencies.29 To our knowledge, none of
these AHL analogs have lower than double-digit micromolar (µM) IC50 values in reporter gene as-
says of LasR activity in P. aeruginosa.30 These IC50 values contrast with the nanomolar (nM) EC50
value of LasR’s native ligand, OdDHL, and those of other non-native agonists (e.g., the triphenyl
derivative TP-1; Figure 1C).25-26 The poor antagonism potencies for AHL analogs may be due, at
least in part, to reliance on the AHL scaffold, which has several major liabilities for probe mole-
cules. AHLs are susceptible to lactone hydrolysis, enzymatic degradation, and active efflux by P.
aeruginosa.16, 31-32 These drawbacks make the development of non-AHL antagonists of LasR, and
other LuxR-type receptors, highly desirable.30 That said, conversion of non-AHL scaffolds known
to strongly agonize LasR (e.g., TP-1) into antagonists (i.e., “mode switching”) has also not provided
sub-µM LasR antagonists so far,26 underscoring the challenges of this process.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
High-throughput screens of small molecule libraries provide another pathway to identify
non-AHL LasR antagonists.33 One such screen by Greenberg and coworkers in 2006 revealed the
compound V-06-018, a β-keto amide with a phenyl head group and a nine carbon tail (Figure 1C).33
V-06-018 is a relatively potent LasR antagonist in both E. coli and P. aeruginosa LasR reporter
strains (single digit micromolar IC50) and has been shown to inhibit genes and phenotypes related to
virulence in P. aeruginosa.9, 33 The phenyl head group and aliphatic acyl tail of V-06-018 resemble
6
ACS Paragon Plus Environment

Page 7 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
that of the homoserine lactone head group and acyl tail of LasR’s native ligand, OdDHL (Figure
1C). However, as V-06-018 lacks a lactone moiety, it is not susceptible to hydrolysis or enzymatic
cleavage by AHL lactonases.31-32 A prior study of ours also revealed that V-06-018 is not actively
effluxed from P. aeruginosa by the promiscuous MexAB-OprM efflux pump, which is known to
efflux both native and non-native AHLs with long acyl tails.16 Despite these desirable qualities, V-
06-018 has seen practically no scrutiny from a structure–function perspective and no substantive use
as a chemical probe since its initial report over a decade ago.30 We reasoned that the V-06-018 scaf-
fold could provide entry into LasR antagonists with improved potencies along with robust physical
properties, and in the current study we report our findings with regard to the first structure-function
analysis of this scaffold. Our combined cell-based assays, synthesis, and iterative compound design
revealed a set of new LasR antagonists based on V-06-018 with potencies, efficacies, and receptor
selectivities in P. aeruginosa that, to our knowledge, surpass all known compounds reported to date.
Follow on biochemical experiments on these compounds and V-06-018 support a mechanism of an-
tagonism by which they interact with the OdDHL-binding site in LasR and, at least in part, stabilize
an inactive form of the protein.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
V-06-018 is selective for LasR over RhlR and QscR in P. aeruginosa
We began our study by exploring the selectivity of V-06-018 for LasR over the other two LuxR-
type receptors (RhlR and QscR) in P. aeruginosa, as other than its antagonistic activity in LasR,30
this profile was unknown. In view of the overlapping activities of these three receptors in P. aeru-
ginosa (see Figure 1B), small molecule tools that are selective for LasR (or indeed any of these re-
ceptors) are of significant interest for use as mechanistic probes in this pathogen. We submitted V-
06-018 to reporter gene assays in E. coli to examine its antagonistic activity (in competition with the
7
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 8 of 39
1
2
3
4
5
6
7
8
9
receptors’ native or preferred ligand) and agonistic activity (alone) in LasR, RhlR, and QscR, using
our previously reported methods (see Experimental Section). In these reporter assays in a heterolo-
gous background (i.e., E. coli), each of the receptors was examined in isolation from the others, al-
lowing for clearer selectivity profiles to be defined relative to using analogous P. aeruginosa re-
porter systems. Receptor activity was monitored via b-galactosidase production. These experiments
revealed V-06-018 was only an antagonist of LasR, displayed no activity (as either an antagonist or
agonist) in RhlR, and was only a very weak antagonist QscR at the highest concentrations tested
(see Figure S1). This high receptor selectivity profile rendered the V-06-018 scaffold even more
compelling for new LasR antagonist development in P. aeruginosa.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
An efficient synthesis of V-06-018 and analogs
We next sought to devise a synthetic route to V-06-018 that was scalable and adaptable to analog
synthesis. The only previously reported synthesis of V-06-018 gave the molecule in 5% yield, albeit
in one step.24 That synthesis involved refluxing ethyl benzoyl acetate and nonylamine in ethanol.
We reasoned the low yield for this reaction could be due to imine formation; therefore, we decided
to protect the ketone in ethyl benzoyl acetate as a ketal (e.g., 2 à 3; Scheme 1), and then saponified
the ester to access the carboxylic acid (4). Standard carbodiimide-mediated amide bond coupling
(via EDC) of the acid with nonylamine proceeded smoothly to yield amide 5. Deprotection of the
ketone furnished V-06-018 in 44% yield over four steps, in quantities typically greater than 100 mg.
This synthetic route was advantageous as it could be easily modified to generate V-06-018 analogs
with alternate tail groups (R’ in Scheme 1) through the coupling of different amines. In turn, alter-
nate head groups could be incorporated by coupling different carboxylic acid building blocks (4),
many of which are readily accessible from acylation reactions of substituted acetophenones using
8
ACS Paragon Plus Environment

Page 9 of 39
ACS Infectious Diseases
1
2
3
diethyl carbonate as an electrophile (e.g., 1 à 2; Scheme 1).34 We introduced both modifications in
4
5
6
our subsequent synthesis of a focused library of V-06-018 analogs.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1: Synthesis of V-06-018 and related analogs. Reagents over arrows: a = NaH, (C2H5)2CO3, THF, Δ;
b = C2H6O2, p-TsOH, benzene, Δ, Dean-Stark trap; c = 1:1 LiOH (1M, aq.), THF; d = EDC·HCl, DMAP,
H2NR’, CH2Cl2; e = p-TsOH, acetone. See Experimental Section and SI for additional details.
Structure-informed design of a V-06-018 analog library
We approached our design of V-06-018 analogs by first considering the binding mode of OdDHL to
LasR (Figure 2). The reported X-ray structure of OdDHL bound to the LasR ligand-binding domain
(LBD) indicates that the lactone, amide, and keto functionality in OdDHL can make several hydro-
gen bonds with residues in the LasR ligand-binding site (e.g., Tyr 56, Trp 60, Asp 73, and Ser
129).35 In view of their structural similarity (see Figure 1C), it is not unreasonable to assume that V-
06-018 could target the same binding site on LasR as OdDHL. We therefore were interested in syn-
thesizing analogs that could either gain or lose the ability to make the same hydrogen bonding con-
tacts as OdDHL, to examine their effects on V-06-018 activity. As the phenyl head group of V-06-
018 cannot engage in a hydrogen bond with LasR, we synthesized a series of analogs via Scheme 1
with alternate head groups (8, 12, 13, and 17–21; Figure 3) that either place a heteroatom in a posi-
tion to potentially accept, or in the case of phenols 17 and 18, accept and/or donate a hydrogen
bond.
9
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 10 of 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2: Three-dimensional (A) and two-dimensional (B) images of the OdDHL-binding site in the [LasR
LBD:OdDHL]2 co-crystal structure (PDB ID: 2UV0).36 Dashed lines indicate putative hydrogen bonds be-
tween the labeled residues or water (shown as a red ball in part A) and OdDHL. OdDHL in part A is shown
with carbon in grey, oxygen in red, and nitrogen in blue.
10
ACS Paragon Plus Environment

Page 11 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3: Library of V-06-018 analogs. Systematic changes were made to the head, tail, and linker regions
of V-06-018 (see text). Compound 26 in this series, comprised of a phenyl head and nine carbon tail, is V-
06-018.
To examine LasR’s tolerance for increased steric bulk on V-06-018’s headgroup, we synthe-
sized napthyl derivative 10 (Figure 3). We also synthesized a variety of analogs with halogenated
aryl headgroups (7, 9, 11, and 14-16) to explore electronic effects on activity. Within this set, com-
pounds 9, 12, and 13 were also inspired by work reported by Spring and coworkers, who found that
11
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 12 of 39
1
2
3
4
5
6
7
8
9
related molecules with these head groups were efficacious inhibitors of the production of QS-
regulated virulence factors in P. aeruginosa.14 To alter the electronics and hydrogen-bonding ability
of the V-06-018 headgroup without significantly increasing its size, we constructed a set of analogs
with heterocyclic, aromatic headgroups (19-21).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Turning to the tail group of V-06-018, we again looked to OdDHL for guidance. The im-
portance of hydrophobic contacts between ligands and the OdDHL acyl tail binding pocket in LasR
has been noted (i.e., at residues Ala 127 and Leu 130),37-38 and AHL-based LasR agonists decrease
in potency as their tails decrease from 12 carbons in length.39 To examine the importance of tail
length for V-06-018’s antagonistic activity, we introduced five to twelve carbon tails via the amine
coupling in Scheme 1, yielding compounds 22-29 (Figure 3; compound 26 is V-06-018). To mimic
the molecular architecture of known AHL30 and TP-type26 antagonists of LasR, we included several
derivatives with cyclic tail groups (30-32, 34 and 35 ). In addition, we examined an analog with a
sec-butyl tail (33, racemic) to evaluate LasR’s tolerance for bulk at the position vicinal to the V-06-
018 amide nitrogen. Lastly, to evaluate the importance of the heteroatoms in the “linker” region be-
tween the headgroup and tail, we synthesized diketone 36 and amide 37. Compound 38, a constitu-
tional isomer of V-06-018, was reported previously by our lab;40 we included it here for comparison
and to further expand our SAR analyses.
12
ACS Paragon Plus Environment

Page 13 of 39
ACS Infectious Diseases
1
2
3
Evaluation of the V-06-018 library for LasR antagonism
4
5
6
7
8
We examined the activity of the V-06-018 library for LasR antagonism using a P. aeruginosa mu-
tant strain (PAO-JP2, ΔlasIrhlI) that lacks the ability to synthesize OdDHL (or BHL) and contains a
green fluorescent protein (GFP) reporter plasmid to examine LasR activity.16, 41 We used a P. aeru-
ginosa LasR reporter as opposed to the E. coli LasR reporter introduced above, as we were most in-
terested in the activity of the compounds (and their eventual use as probes) in the native organism.
Further, as we previously showed that V-06-018 is not subject to active efflux by the MexAB-OprM
pump in P. aeruginosa,16 we wanted to examine if these close analogs were also active in the pres-
ence of this pump. In this P. aeruginosa reporter system, compounds capable of LasR antagonism
should reduce GFP production, and this loss can be quantitated by fluorescence (see Experimental
Section). To start, we screened the library for LasR antagonism at a concentration of 10 µM in com-
petition against 150 nM OdDHL. Analogs with substituents on the head group were found to be
generally less efficacious as LasR antagonists relative to V-06-018 (compounds 7–18, Figure 4A),
suggestive that bulkier V-06-018 analogs may not be as well accommodated in the AHL binding
site, regardless of their hydrogen bonding ability. Decreasing the size of the headgroup and includ-
ing a polar atom was more fruitful. Two of the analogs based on five-membered heterocycles, furan
19 and thiophene 20, had equivalent efficacy to V-06-018 (~90% LasR antagonism). Not all hetero-
cycles were effective as headgroups, however; thiazole 21 lost efficacy relative to V-06-018.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 14 of 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: Primary LasR antagonism screening data in P. aeruginosa reporter PAO-JP2 for the (A) head
group and (B) tail group and linker modified V-06-018 analogs. Compounds were screened at 10 µM in the
presence of 150 nM OdDHL. Bacteria treated with 150 nM OdDHL only was defined as 100% LasR activ-
ity/0% LasR antagonism; conversely, bacteria treated with DMSO only (i.e., vehicle) was defined as 0%
LasR activation/100% LasR antagonism. Error bars indicate SD of n ³ 3 trials.
Turning to the tail group modified V-06-018 analogs, we found that only compounds with
unbranched, acyclic alkyl tails were efficacious LasR antagonists (e.g., 27–29, Figure 4B). No com-
pounds with cyclic moieties or branching (i.e., 30-35) in their tails were capable of antagonizing
LasR by more than 50%. The length of the tail was also important; analogs 27-29, with 10- to 12-
carbon tails, were equally as efficacious as V-06-018. The shorter tail analogs 22-25, however, an-
tagonized LasR by less than 50%. These data suggest that binding interactions between LasR and
these truncated V-06-018 analogs may have been reduced due to the lack of hydrophobic contacts
(again, shown to be important for LasR:OdDHL binding).37-38 Modifications to the linker region
also resulted in less active analogs. All three linker-modified compounds (36-38) lost efficacy rela-
tive to V-06-018, implicating the presence and position of the amide in V-06-018 as critical to LasR
14
ACS Paragon Plus Environment

Page 15 of 39
ACS Infectious Diseases
1
2
3
4
5
antagonism. Overall, these primary screening data indicated that only subtle alterations to the head
and tail groups of V-06-018, and not the linker group, were tolerated for strong LasR antagonism.
6
7
8
9
Dose-response antagonism analysis of primary screening hits
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To obtain a quantitative measure of compound potency, we performed dose-response analyses on
the compounds that antagonized LasR ≥ 90% at 10 μM (19, 20, and 27-29) using the same P. aeru-
ginosa LasR reporter strain and calculated their IC50 values (Table 1). We were excited to observe
that each of these analogs was more potent than V-06-018. Increasing the length of the V-06-018
tail from 10 to 12 carbons (i.e., as in 27-29) led to a ~3-4-fold increase in potency. The heterocyclic
analogs were also stronger LasR antagonists than V-06-018; furan 19 was approximately two-fold
more potent than V-06-018, and thiophene 20 was closer to five-fold.
15
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 16 of 39
1
2
3
Table 1. Potency and maximum LasR inhibition (efficacy) data for selected compounds in P. aeruginosa
4
5
6
IC50
95% CI
(μM)b
Maximum Inhibition
(%)c
compound
(μM)a
7
8
9
V-06-018 (26) 2.3
(1.7 – 3.1)
(0.8 – 1.8)
(0.3 – 0.6)
(0.5 – 0.9)
(0.4 – 0.7)
(0.5 – 1.0)
(0.2 – 0.3)
(0.2 – 0.3)
(2.0 – 7.1)
(0.1 – 0.2)
(0.1 – 0.2)
(0.5 – 0.8)
89
96
84
93
92
91
83
85
89
91
93
84
19
20
27
28
29
39
40
41
42
43
44
1.2
0.5
0.7
0.5
0.7
0.2
0.2
3.8
0.2d
0.2d
0.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aFor details of PAO-JP2 reporter strain, see Experimental Section. aAntagonism experiments performed by competing
the compounds against OdDHL (1) at its approximate EC50 (150 nM for PAO-JP2) and inhibitory activity was measured
relative to receptor activation at this EC50. IC50 values determined by testing compounds over a range of concentrations
b
(0.64 nM – 50 µM). All assays performed in triplicate. CI = confidence interval. 95% CIs calculated from the SEM of
c
d
n ≥ 3 trials. Denotes the best-fit value for the bottom of the computed dose-response curve. Compound exhibited non-
monotonic dose-response behavior. Reported IC50 corresponds to the antagonism portion of the curve. Full antagonism
dose response curves are shown in Figure S2.
Second-generation V-06-108 analogs and LasR agonism profiles
Encouraged by the antagonistic activity profiles of our initial set of compounds, we designed and
synthesized a set of “hybrid” second-generation V-06-018 analogs that combined features of the
most active compounds. These compounds were comprised of a furan or thiophene head group
united with 10, 11, or 12 carbon tails (compounds 39-44; see Figure 2), and were synthesized and
evaluated for LasR antagonism in P. aeruginosa as described above. The second-generation com-
pounds displayed a variety of activities in the LasR antagonism assay (listed in Table 1). Notably,
furan derivatives 39 and 40, containing 10 or 11 carbon tails, respectively, were more potent than
their parent compounds and were each 10-fold more potent than V-06-018. The 12-carbon furan an-
alog 41, however, lost activity relative to its parent compounds.
16
ACS Paragon Plus Environment

Page 17 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
We note that the thiophene analogs of 39 and 40, compounds 42 and 43, displayed non-mon-
otonic partial agonism behavior in the LasR dose-response assays; 30 30 2929 namely, at concentrations
below 2 µM these compounds antagonized LasR, while at concentrations above 2 µM they agonized
LasR. We have reported this activity profile for a series of ligands in reporter assays of LuxR-type
proteins to date.27-28, 30 The antagonist portions of their dose-response curves indicated that 42 and
43 were each highly potent at lower concentrations, with IC50 values 10-fold lower than that of V-
06-018. Interestingly, thiophene analog 44, differing by only one methylene unit than 43, lacked ob-
servable non-monotonic activity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The discovery that two of the hybrid compounds were non-monotonic partial LasR agonists
prompted us to measure dose-response agonism curves for all our most potent compounds (Figure
S3). V-06-018 and compounds 27-29, comprised of phenyl headgroups, did not activate LasR. We
also screened our first-generation library for LasR agonism at a single concentration (100 µM) and
found that none of the analogs with phenyl headgroups activated LasR; however, thiophene 20
weakly agonized LasR (to 20%; Figure S4). We found that furans 39 and 40 could very weakly ago-
nize LasR (7% and 4%, respectively) at the highest concentration screened (50 µM). Relative to 39
and 40, thiophenes 42 and 43 were stronger LasR agonists at 50 µM (30% and 22%, respectively),
which matched their activity profile at this concentration in the dose-response antagonism analysis
(as described above).
Activation in this cell-based reporter assay requires LasR to initiate transcription of gfp. This
process requires LasR to adopt a conformation capable of homodimerization and productive DNA
binding. Our results suggest that, at sufficiently high compound concentration, these furan and thio-
phene ligands can make contacts with LasR (either directly or indirectly via some other target) that
promotes this process. However, contacts with just the head groups of 39, 40, 42, and 43 are pre-
sumably insufficient, as compounds 41 and 44, comprised of the same furan and thiophene head
17
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 18 of 39
1
2
3
4
5
6
7
8
9
groups, respectively, yet linked to a twelve-carbon tail, failed to activate LasR even at high concen-
trations. These results suggest that contacts with the tail—specifically, a tail of 9–11 carbons—
along with the head group are necessary for LasR agonism by this ligand class at high concentra-
tions. Whether these ligands target the OdDHL binding site or another site on LasR, or another fac-
tor altogether, to promote LasR activation at these concentrations remains to be determined.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
E. coli reporter assays indicate V-06-018 and analogs act directly via LasR
We next examined if our improved V-06-018 analogs elicit their antagonistic activity via acting di-
rectly on LasR using an E. coli LasR reporter system (see Experimental Section).42-44 As high-
lighted above, LasR is directly and indirectly regulated by other QS systems in P. aeruginosa, and
thus activity profiles in the P. aeruginosa LasR reporter are a measure of this inter-regulated net-
work. To address this question, we obtained dose-response curves for all of the compounds in Table
1 in an E. coli LasR reporter strain, and found that their relative efficacies and potencies largely
tracked between the E. coli and P. aeruginosa reporters (Figure S5, Table S2). This alignment be-
tween the P. aeruginosa and E. coli reporter data suggests that these compounds elicit their effects
via direct interactions with LasR. We note that all of our antagonists were less efficacious and po-
tent against LasR in the E. coli reporter relative to P. aeruginosa. For example, the lead compound
40 was only four-fold more potent than V-06-018 in E. coli vs. being 10-fold more potent in P. ae-
ruginosa. This reduction in potency also obscured the non-monotonic effects observed above for
compounds 42 and 43. We postulate that this reduction in potency in E. coli is an artifact of differ-
ences in LasR expression levels between the two reporter systems (non-native level in E. coli vs.
native level in P. aeruginosa).45 With more LasR present, higher concentrations of ligands are pre-
sumably required to inhibit LasR activity. Critically, the stronger efficacies and potencies of these
18
ACS Paragon Plus Environment

Page 19 of 39
ACS Infectious Diseases
1
2
3
4
5
V-06-018 derived antagonists in the native host background will increase their utility as probe mol-
ecules.
6
7
8
9
We were also curious to see if the new antagonists, like V-06-018, were selective for LasR
over RhlR and QscR in P. aeruginosa. Screening representative compounds (39 and 40) in the E.
coli RhlR and QscR reporter systems showed that 40 is highly LasR selective, with no observable
activity in either RhlR or QscR (Figure S6). Compound 39 was found to be inactive in RhlR and,
similar to V-06-018, only a weak QscR antagonist (~35% inhibition) at the very highest concentra-
tion tested. These results further underscore the receptor selectivity profile of the V-06-018 scaffold
and the value of these compounds as chemical tools to study QS in P. aeruginosa.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
P. aeruginosa reporter data support a competitive mechanism of LasR antagonism for V-06-
018 and related compounds
We were interested to determine if V-06-018 and our new lead antagonists were acting as competi-
tive LasR antagonists, and examined this question by testing them against OdDHL at varying con-
centrations in the P. aeruginosa LasR reporter assay. The observed potency of a competitive LasR
antagonist should vary with OdDHL concentration, as both molecules are competing for space in
the same ligand-binding site. We obtained antagonism dose response curves for V-06-018 and one
of our lead compounds (40, which did not display non-monotonic behavior) in competition with
OdDHL at 150 nM, 1 µM, and 10 µM (Figure 5). We observed an OdDHL-concentration-depend-
ent decrease in the potency of both compounds. The relative potency trends for V-06-018 and 40
were also maintained, with compound 40 significantly more potent than V-06-018 at 150 nM and 1
µM. Unlike V-06-018, compound 40 was still capable of antagonizing LasR (to 55%) even in the
presence of 10 µM OdDHL. These results are supportive of the ability of V-06-018 and its close an-
alogs to act as competitive antagonists of LasR.
19
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 20 of 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5: Dose-response LasR antagonism curves for V-06-018 and analog 40 in P. aeruginosa PAO-JP2.
Dose-response curves of V-06-018 (black triangles) and 40 (red squares) in competition with (A) 150 nM,
(B) 1 µM, and (C) 10 µM OdDHL. V-06-018 has IC50 values of 2.3 and 3.9 µM vs. 0.15 and 1 µM OdDHL,
respectively; 40 has IC50 values of 0.2 and 0.7 µM vs. 0.15 and 1 µM OdDHL, respectively. IC50 values
could not be calculated for these compounds in competition with 10 µM OdDHL (curves in part C).
Antagonists and non-classical partial agonist 42 solubilize LasR
We sought to further characterize the interactions between V-06-018 and related analogs with LasR,
to understand how they engender receptor antagonism. Very little is known about the molecular
mechanisms that lead to antagonism of LuxR-type receptors by small molecules, largely due to the
20
ACS Paragon Plus Environment

Page 21 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
instability of these proteins in vitro even in the presence of their native AHL ligand.46 LasR requires
OdDHL throughout the production and purification process to be isolated, and has proven intracta-
ble to structural studies in full length form.36, 47-48 In principle, antagonists of LuxR-type proteins
can operate by binding either in place of an AHL signal, or to a hypothetical, allosteric binding site.
Once bound, antagonists can then cause antagonism by further destabilizing the protein (as has been
shown for QscR and LasR)38, 47, 49 or by forming soluble complexes that are either incapable of di-
merization or binding to DNA (as has been shown for CviR and LasR),50,51-52 or presumably combi-
nations of these mechanisms (and potentially others). We were curious to investigate whether solu-
ble LasR could be isolated when it was produced in the presence of V-06-018 or our new antago-
nists, or if it was destabilized in their presence relative to OdDHL. To test these questions, we pro-
duced LasR in E. coli grown in the presence of no compound (DMSO control) or 50 µM OdDHL,
V-06-018, 40, or 42 (see Experimental Methods). After 16 h of protein production, we lysed the E.
coli cells and separated the whole cell (WC) and soluble (S) lysate on an SDS-PAGE gel (Figure 6;
quantitative analysis of the bands in the gel is provided in Table S3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Characterization of LasR via SDS-PAGE gel in the presence of different ligands. Whole Cell
(WC) and soluble (S) portions of E. coli cell lysates with LasR overexpressed in the presence of DMSO or
50 µM of OdDHL, V-06-018, 40, or 42. LasR has a mol. wt. of 27.9 kD.
As expected, we did not obtain any LasR in the soluble fraction of cells grown without ex-
ogenous compound (DMSO), while we obtained soluble LasR in the culture grown with exogenous
OdDHL (S band ~30% as intense as WC band; Figure 6). These data recapitulate the finding that
21
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 22 of 39
1
2
3
4
5
6
7
8
9
LasR requires a ligand to be soluble in vitro.53 We detected soluble bands for LasR produced in the
presence of V-06-018 and furan 40. The bands were four-fold smaller than that of OdDHL (~7% as
intense as WC band, vs. ~30% for OdDHL, Table S3), suggesting that these ligands do not solubil-
ize LasR to the same extent as OdDHL. This result correlates with the previous report of Schneider
and co-workers demonstrating that certain synthetic AHL-type antagonists (along with the close V-
06-018 analog 38) form soluble complexes with LasR, albeit in less amounts than OdDHL.51
Schneider went on to show that these complexes were unable to bind to LasR’s target DNA using
electrophoretic mobility shift assays (EMSAs), which allows for the interpretation that these ligands
can stabilize an inactive LasR complex (e.g., incapable of dimerization or DNA binding). We also
observed thiophene 42 solubilize LasR. The soluble band for 42 was more intense than those ob-
served for V-06-018 and 40, and comparable to OdDHL (~30%). We note that 42 has a non-mono-
tonic activity profile in the P. aeruginosa reporter assay and is capable of weak LasR agonism at
higher concentrations; the larger quantity of LasR isolated in this experiment relative to V-06-018
and 40 (at 50 µM concentration) is therefore interesting and could arise due to this agonistic activity
profile. Collectively, these SDS-PAGE data support the hypothesis that V-06-018 and related ana-
logs act at LasR antagonists, at least in part, via inducing a soluble but inactive conformation of
LasR. The reduced amount of protein in these soluble fractions relative to OdDHL suggests that V-
06-018 and 40 may also cause antagonism by promoting LasR unfolding (i.e., destabilizing the re-
ceptor); thus, more than one mechanism of antagonism is likely operative. Further biochemical
(e.g., EMSAs) and structural experiments are required to test these mechanistic hypotheses and are
ongoing in our laboratory.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22
ACS Paragon Plus Environment

Page 23 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
LasR mutants reveal residues critical for activation and inhibition by synthetic ligands
The results of the competitive LasR antagonism dose response assays, E. coli reporter assays, and
protein production experiments outlined above suggest that V-06-018 and the lead analogs target
LasR and interact with the OdDHL binding site to cause antagonism. In view of our original com-
pound design, we were curious as to whether the residues in LasR that are known to govern
LasR:OdDHL interactions (Figure 2) were also important to LasR antagonism by the V-06-018 lig-
and class, and applied a method utilized previously in our laboratory involving LasR mutants with
modifications to the OdDHL binding site.35-36, 54 In this past work, a set of LasR single-point mu-
tants were generated in which residues implicated in hydrogen bonding interactions with OdDHL
were converted to residues incapable of hydrogen bonding but approximately the same steric size
(e.g., Tyr à Phe). The mutant LasR proteins were then tested for activity using a LasR reporter
plasmid in an E. coli host background (analogous to the E. coli LasR reporter assay system above).
Compounds showing reduced activity in these mutants relative to wild-type LasR then can be postu-
lated to make a contact with LasR that depends on the mutated residue. We tested V-06-018 and fu-
ran 39 at 100 μM in three LasR mutants with modifications to residues that make hydrogen-bonds
to OdDHL (Tyr 56, Trp 60, and Ser 129; see Figure 3).54 Notably, all of these single-point LasR
mutants (Y56F, W60F, and S129A) are still functional in the reporter assay, but are less active than
wild-type LasR (as measured via reduced OdDHL potencies; Figure S7), reflective of the im-
portance of these LasR:OdDHL interactions for activation. (As noted above, antagonists display re-
duced efficacy in general in this heterologous background relative to the native (P. aeruginosa) re-
porter system.)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
V-06-018 was found to antagonize all three LasR mutants to a significantly lesser extent
than wild-type LasR (Figure 7A). The same trend was true for furan 39. Tyr 56 and Ser 129 are be-
lieved to form hydrogen bonds with the amide carbonyl of OdDHL (Figure 2), and potentially could
23
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 24 of 39
1
2
3
4
5
6
7
8
9
bind to one of the two linker carbonyl oxygens in V-06-018 and its analogs.36 Trp 60 hydrogen
bonds with the lactone carbonyl oxygen of OdDHL, and it may be capable of hydrogen bonding
with the furan oxygen of 39. An analogous hydrogen-bond to the head-group of V-06-018 is not
possible, but the lower activity of V-06-018 in the W60F LasR mutant suggests that Trp 60 interacts
in some other manner with V-06-018 to enforce antagonism. Further studies are necessary to pin-
point the specific molecular interactions that govern LasR antagonism by these two ligands. Never-
theless, these experiments with LasR mutants support V-06-018 and new antagonist 39 interacting
with the OdDHL binding site in LasR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
24
ACS Paragon Plus Environment

Page 25 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7: (A) LasR mutant antagonism data for V-06-018 and lead compound 39. Compounds tested at 100
µM against OdDHL at its approximate EC50 value in the specific E. coli LasR reporter strain (as indicated on
the X-axis). (B) LasR mutant agonism data for V-06-018, 39, and 43. Compounds tested at 100 µM. For an-
tagonism experiments, 100% is defined as the EC50 concentration of OdDHL in that specific LasR reporter
strain (see Figure S6); for agonism experiments, 100% is defined as the activity of 100 µM OdDHL in that
specific LasR reporter strain. Significance was assessed via a one-way ANOVA: **** = p < 0.0001; *** = p
< 0.001; ** = p < 0.01; * = < 0.05. ns = no significant difference.
We also were curious to learn whether alternations to these LasR residues could impact the
ability of our compounds to agonize LasR. Therefore, we examined the agonistic activities V-06-
018, furan 39, and thiophene 43 in the three LasR mutant reporter strains at 100 µM; thiophene 43
was included in these agonism assays due to its non-monotonic agonism profile (see above). We
25
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 26 of 39
1
2
3
4
5
6
7
8
9
were surprised to find that all three compounds agonized the LasR Y56F mutant to a significantly
greater extent than wild-type LasR. For example, V-06-018, which does not agonize wild-type
LasR, activated LasR Y56F to ~60% (relative to OdDHL) at 100 µM. In view of this unexpected
result, we screened the remainder of our lead compounds in this LasR mutant reporter and found
that they all were capable of activating the LasR Y56F mutant to some extent (from 9–56% at 100
µM; Figure S8). V-06-018 and 39 also agonized the LasR S129A mutant significantly more than
wild-type LasR. These results suggest that removing the hydrogen bonds donated by Tyr 56 or Ser
129, or reducing sterics at these positions, may allow these V-06-018 type ligands more freedom to
adjust their position in the LasR OdDHL binding pocket and adopt new contacts that engender LasR
agonism as opposed to antagonism. None of our compounds were found to agonize the LasR W60F
mutant; in fact, 43 lost agonistic activity in that mutant relative to wild-type LasR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In our laboratory’s prior mutational studies of LasR, we observed compound 38 (Figure 3), a
LasR antagonist and constitutional isomer of V-06-018, could agonize both the LasR Y56F and
W60F mutants. We termed this transition from antagonist to agonist “Janus” behavior (after the
two-faced Roman god).35 Here, we observed V-06-018 and compound 39 exhibit analogous “Janus”
behavior in Y56F and S129A, but not in W60F (like 38). These results suggest that chemical modi-
fication of either the ligand (via chemical synthesis; i.e., V-06-018 à 38 or 39) or LasR (via muta-
genesis of at least these three residues) is sufficient to alter contacts between the ligand and receptor
to allow for either agonism or antagonism, or the degree thereof, and that these changes to molecu-
lar contacts are likely very subtle. The implications of these findings—specifically, that single point
mutations can convert potent LasR antagonists into agonists—on the propensity for resistance to
arise in P. aeruginosa to LasR antagonists did not escape our attention. We do note that the agonis-
tic activity of these compounds is quite low (relative to OdDHL in wild type LasR). Additional ex-
periments are required to explore the possibility of LasR mutants to arise naturally upon sustained
26
ACS Paragon Plus Environment

Page 27 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
treatment with V-06-018 or related analogs. However, our lab and others has shown previously that
resistance to QS inhibitors, even if it was to develop, should be slow to spread through and not
overtake a population of bacteria,55-56 supporting the continued search for such compounds. Moreo-
ver, the ability of V-06-018, 38, 39, and 43 to agonize the LasR mutants suggests that structural
studies of these LasR mutant:ligand complexes could be particularly noteworthy, as they could illu-
minate the mechanisms by which these ligands both agonize LasR mutants and antagonize wild-
type LasR. The heightened stability of LasR:agonist complexes relative to LasR:antagonist com-
plexes could significantly enable such structural studies.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SUMMARY AND CONCLUSIONS
The work reported herein was motivated by the need for chemical probes of a key QS receptor,
LasR, in the opportunistic pathogen P. aeruginosa. Despite considerable research to date, antago-
nists with sub-micromolar potencies, high efficacies, and selectivities for LasR over the other QS
circuits in P. aeruginosa have been elusive. We performed the first structure-function analysis of
the small molecule V-06-018, a promising yet unstudied LasR antagonist emerging from a high-
throughput screen reported over 10 years ago.33 We developed a versatile and efficient synthetic
route to V-06-018, produced a focused library of analogs using this route to explore the headgroup,
linker, and tail portions of V-06-018, and evaluated the library for LasR modulatory ability using
cell-based reporter systems. These screening data revealed stringent SARs for LasR antagonism by
this ligand scaffold, including the requirement for a linear, alkyl tail group between nine to 12 car-
bons in length, an amide in the linker, an intolerance for substitution on the aryl head group, and a
tolerance of certain 5-membered heterocyclic head groups. These SARs allowed us to design and
synthesize second-generation LasR antagonists with nanomolar IC50 values in P. aeruginosa (e.g.,
27
ACS Paragon Plus Environment

ACS Infectious Diseases
Page 28 of 39
1
2
3
4
5
6
7
8
9
39 and 40). These compounds represent, to our knowledge, the most potent and efficacious syn-
thetic antagonists of LasR to be reported, with IC50 values in P. aeruginosa 10-fold lower than V-
06-018 and at least 100-fold lower than other AHL-based ligands.48 We note that we discovered
these analogs after synthesizing fewer than 40 compounds; further development of the V-06-018
scaffold would likely yield even more potent compounds.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Our results indicate that the V-06-018 scaffold is quite selective for LasR over the other two
LuxR-type receptors in P. aeruginosa, with 39, 40, and V-06-018 showing neither antagonistic nor
agonistic activity in RhlR, 40 being inactive in QscR, and 39 and V-06-018 showing only modest
antagonistic activity in QscR at the very highest concentrations tested. This activity profile is signif-
icant because the ability to selectively attenuate LasR activity in the midst of the highly inter-regu-
lated QS system of this pathogen will facilitate mechanistic studies, and highlights the value of
these V-06-018 analogs as chemical tools to study QS in P. aeruginosa.
We also report herein our investigations into the mechanism by which V-06-018 and related
compounds modulate LasR activity. In the course of these studies, certain analogs were found to
display interesting dual activity profiles—capable of strong LasR antagonism at nanomolar levels,
yet LasR agonism at micromolar levels (i.e., non-monotonic partial agonists)—and we were in-
trigued by their mechanisms of action as well. Examination of the lead compounds against OdDHL
at various concentrations and in an E. coli LasR reporter support a mechanism by which they bind
competitively with OdDHL and interact directly with LasR. V-06-018 and furan antagonist 39 were
found to be significantly less efficacious in LasR mutants that lack key residues in the ligand-bind-
ing site shown to make hydrogen-bonding contacts with OdDHL. This result is congruent with these
compounds binding in the same site on LasR as or near to OdDHL. Protein production studies of
LasR in the presence of V-06-018, furan-based antagonist 40, and thiophene-based antagonist 42
demonstrated that these compounds support folding of the protein into a soluble form, suggestive
28
ACS Paragon Plus Environment

Page 29 of 39
ACS Infectious Diseases
1
2
3
4
5
6
7
8
9
that they may stabilize an inactive form of the protein, analogous to the mechanism of CviR antago-
nism by the chlorolactone AHL analog (CL).50 V-06-018 and 40 also appear to reduce the amount
of soluble LasR relative to 42 or the native agonist OdDHL, indicating that receptor destabilization
could also contribute to the mechanism of inactivation by certain of these compounds. Finally,
study of V-06-018 and furan-based antagonist 39 revealed that they were each capable of shifting
from LasR antagonists to agonists in a LasR mutant lacking a single hydrogen-bonding motif in the
ligand-binding site (e.g., Tyr 56à Phe 56; removal of the Tyr hydroxyl). This finding indicates that
subtle interactions of these ligands with LasR can have dramatic effects on receptor activity and
suggests a novel route for exploring the mechanisms of this ligand class via structural studies of
LasR mutant:ligand complexes. Overall, this study has provided a set of highly potent LasR antago-
nists that should find broad use as chemical probes of QS in P. aeruginosa, a robust chemical route
to generate these compounds, and new insights into the mechanisms of LasR antagonism. These
compounds and insights expand the understanding of LuxR-type QS in this important opportunistic
pathogen.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MATERIALS AND METHODS
Chemistry
All chemicals were obtained from Sigma-Aldrich, Agros Organics, or TCI America. All reagents
and solvents were used without further purification except for hexane, ethyl acetate, and dichloro-
methane, which were distilled prior to use. Analytical thin-layer chromatography (TLC) was per-
formed on 250 µm glass backed silica plates with F-254 fluorescent indicator from Silicycle. Visu-
alization was performed using UV light and iodine. All new compounds were fully characterized for
purity and identity; see SI for characterization data. Compound stock solutions were prepared in
DMSO at appropriate concentrations and stored at -4 °C prior to use.
29
ACS Paragon Plus Environment