Altering the communication networks of multispecies microbial systems using a diverse toolbox of AI-2 analogues

Article abstract of DOI:10.1021/cb200524y

There have been intensive efforts to find small molecule antagonists for bacterial quorum sensing (QS) mediated by the universal QS autoinducer, AI-2. Previous work has shown that linear and branched acyl analogues of AI-2 can selectively modulate AI-2 signaling in bacteria. Additionally, LsrK-dependent phosphorylated analogues have been implicated as the active inhibitory form against AI-2 signaling. We used these observations to synthesize an expanded and diverse array of AI-2 analogues, which included aromatic as well as cyclic C-1-alkyl analogues. Species-specific analogues that disrupted AI-2 signaling in Escherichia coli and Salmonella typhimurium were identified. Similarly, analogues that disrupted QS behaviors in Pseudomonas aeruginosa were found. Moreover, we observed a strong correlation between LsrK-dependent phosphorylation of these acyl analogues and their ability to suppress QS. Significantly, we demonstrate that these analogues can selectively antagonize QS in single bacterial strains in a physiologically relevant polymicrobial culture.

Full text of DOI:10.1021/cb200524y

Articles
pubs.acs.org/acschemicalbiology
Altering the Communication Networks of Multispecies Microbial
Systems Using a Diverse Toolbox of AI-2 Analogues
Sonja Gamby,†,# Varnika Roy,‡,§,∥,# Min Guo,†,# Jacqueline A. I. Smith,† Jingxin Wang,† Jessica E. Stewart,∥
Xiao Wang,§ William E. Bentley,*,§,∥ and Herman O. Sintim*,†
‡
§
†Department of Chemistry and Biochemistry, Graduate Program in Molecular and Cell Biology, Institute for Bioscience and
∥
Biotechnology Research, and Fischell Department of Bioengineering, University of Maryland, College Park, Maryland 20742,
United States
S
* Supporting Information
ABSTRACT: There have been intensive efforts to find small molecule antagonists for
bacterial quorum sensing (QS) mediated by the “universal” QS autoinducer, AI-2. Previous
work has shown that linear and branched acyl analogues of AI-2 can selectively modulate
AI-2 signaling in bacteria. Additionally, LsrK-dependent phosphorylated analogues have
been implicated as the active inhibitory form against AI-2 signaling. We used these
observations to synthesize an expanded and diverse array of AI-2 analogues, which
included aromatic as well as cyclic C-1-alkyl analogues. Species-specific analogues that
disrupted AI-2 signaling in Escherichia coli and Salmonella typhimurium were identified.
Similarly, analogues that disrupted QS behaviors in Pseudomonas aeruginosa were found.
Moreover, we observed a strong correlation between LsrK-dependent phosphorylation of
these acyl analogues and their ability to suppress QS. Significantly, we demonstrate that
these analogues can selectively antagonize QS in single bacterial strains in a physiologically
relevant polymicrobial culture.
periplasmic protein LuxP.24 Besides LuxP, the detection of AI-2
acteria communicate through a phenomenon known as
by V. harveyi also requires another protein, LuxQ. LuxP has
been identified as the primary AI-2 receptor in V. harveyi and is
Bquorum sensing (QS), whereby neighboring cells secrete
and detect chemical signals known as autoinducers.1 Once a
a member of a large family of periplasmic binding proteins
whose members bind diverse ligands. When AI-2 binds
LuxPQ, a phospho-relay is triggered and bioluminescence is
induced.24 Another characterized receptor of AI-2 is LsrB,
found in S. typhimurium.25 The crystal structure of a non-
boronated chemically distinct form of AI-2 bound to LsrB has
been reported. In S. typhimurium and E. coli, the internalization
of AI-2 occurs as a non-borated cyclic form.25 Once inside
the cell, AI-2 is phosphorylated by the kinase LsrK. This
phosphorylated AI-2 then binds and derepresses LsrR, the
transcriptional regulator of QS genes in enteric bacteria
(Figure 6, panel a).
Because of the ubiquitous nature of AI-2, it is believed that
antagonists of AI-2 signaling would provide broader spectrum anti-
QS activities than molecules that inhibit the species-specific AI-1
signaling. The caveat, however, is that many bacteria in humans
play beneficial roles and these bacteria also signal with AI-2.
Therefore, finding anti-AI-2 agents that display some level of
selectivity could also be desired. If the receptor proteins for AI-2 in
different bacteria harbor the same active site architecture, then
finding selective small molecules that perturb AI-2 signaling in
some bacteria and not others could be challenging.
predetermined (threshold) concentration of autoinducer is
detected (typically in accordance with a threshold cell number
density, or quorum), bacteria can coordinate their gene expres-
sion and perform both pathogenic and symbiotic processes
such as biofilm formation, virulence factor production, and
bioluminescence induction.2,3 Recently there have been
intensive efforts by several groups to find small molecules
that can interrupt the communication among bacteria.4−10 It is
believed that disrupting bacterial communication and hence
virulence factor production would not put substantial evolu-
tionary pressure on bacteria to develop resistance.11,12 The
majority of studies, which investigated the inhibition of QS via
autoinducer analogues (AHL or oligopeptide), were performed
using monocultures.13−17 Of the three major autoinducer
classes (acyl homoserine lactones, AHLs, or AI-1’s for Gram-
negative bacteria, oligopeptides for Gram-positive bacteria, and
autoinducer-2 (AI-2), which is found in among Gram-negative
and Gram-positive bacteria), signaling via AI-218 is of particular
interest for the development of antiquorum sensing agents
because 4,5-dihydroxypentane-2,3-dione (DPD) is produced or
recognized by over 70 species.19 The key role of AI-2 in Vibrio
harveyi and the homologous QS systems of Escherichia coli and
Salmonella typhimurium have received significant attention.20−23
Two receptors of AI-2 have been identified: LuxP in
V. harveyi24 and LsrB in S. typhimurium.25 In V. harveyi, a
cyclic borate form of AI-2 was found to co-crystallize with the
Received: May 18, 2011
Accepted: March 20, 2012
Published: March 20, 2012
© 2012 American Chemical Society
1023
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030

ACS Chemical Biology
Articles
Several laboratories have investigated AI-2 analogues as anti-QS
agents in different organisms.10,26−28 For example, Janda et al. have
revealed that linear acyl analogues of AI-2 could inhibit AI-2
signaling in S. typhimurium.27 We found branched acyl analogues
with antagonist activity and further demonstrated that different
bacteria have different susceptibilities to acyl AI-2 analogues. We
also demonstrated the potential for species selectivity, noted
above.10 Importantly, LsrK phosphorylates these acyl analogues,
and it is the phosphorylated analogues that are likely the inhibitors
of phospho-AI-2 signaling. Interestingly, the longer chain acyl
groups confer structural similarity to the AI-1 family of
autoinducers (acylated homoserine lactones) and can trans-
verse cell membranes via diffusion.29 Our initial results have
prompted us to expand our study to include a more diverse set
of acyl AI-2 analogues, with the expectation that increasing the
diversity of the shapes and sizes of the C1-substituents could
shed light on the specificities and promiscuities of AI-2 based
signaling proteins and could potentially also lead to the
identification of more selective AI-2 modulators in diverse
bacterial species. Of interest would be the ability to selectively
modulate AI-2 signaling among specific bacteria in a poly-
microbial system. Herein, we describe the synthesis and bio-
logical evaluation of this expanded set of AI-2 analogues, both
biochemically and as interrupters of cross talk in a poly-
microbial system consisting of E. coli, S. typhimurium, and
P. aeruginosa (Table 1).
Table 1. List of Strains and Cells Used in This Study
strain,
plasmid, or
primer
relevant genotype and/or property
reference
Escherichia coli strains
W3110
wild type
laboratory
stock
LW7
W3110 ΔlacU160-tna2 ΔluxS:: Kan
W3110 ΔlacU169-tna2
ZK126 ΔlsrR::Kan
30
31
30
ZK126
LW8
Salmonella typhimurium strains
rpsl putRA:: Kan-lsr-lacZYA luxS:: T-POP
rpsl putRA:: Kan-lsr-lacZYA
P. aeruginosa strains
MET715
MET708
20
20
PAO1
wild type
laboratory
stock
plasmids
pLW11
pCT6
galK’-lacZYA transcriptional fusion vector,
containing lsrACDBFG promoter region,
Ampr
21
RESULTS AND DISCUSSION
■
pFZY1 derivative, containing lsrR and lsrR
32
32
promoter region fused with T7RPol, Apr
Synthesis of Cyclic and Aromatic AI-2 Analogues. C-1
pET-dsRED pET200 derivative, containing RFP, Kmr
alkyl AI-2 analogues (Figure 1, panel a) were synthesized
Figure 1. Analogue synthesis scheme and analogue structures. (a) Previously synthesized linear and branched analogues reported in refs 10 and 33,
respectively. (b) Newly synthesized cyclic and aromatic analogues (cyclopropyl- and cyclohexyl-DPD).
1024
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030

ACS Chemical Biology
Articles
following our previously reported diazocarbonyl aldol method-
ology.10 This methodology was adapted to make a variety of
cyclic and aromatic AI-2 analogues, starting from the requisite
acid chlorides, which are commercially available. Briefly, a
DBU-catalyzed addition of (OTBS)-acetaldehyde to diazo-
carbonyls (readily generated from the addition of diazomethane
to acid chlorides) afforded the protected diazo intermediates.
Without isolation of these products, TBAF-deprotection was
performed to afford diazodiol intermediates. Oxidation of these
diazodiols into DPD proceeded quantitatively with excess DMDO,
and the excess reagent was evaporated to give the desired cyclic
and aromatic DPD analogues. NMR characterization of newly
synthesized analogues is in Supporting Information. Figure 1
(panel b) depicts eight new AI-2 analogues; we tested their
effects on three different bacteria: E. coli, S. typhimurium, and
P. aeruginosa.
Effect of Cyclic and Aromatic DPD Analogues on β-
Galactosidase Production in E. coli and S. typhimurium.
Cyclic and aromatic analogues were tested for their modulation
of QS in E. coli and S. typhimurium. Both E. coli LW7 (luxS−)
and S. typhimurium MET715 (luxS−) synthesize β-galactosidase
in response to AI-2 by derepressing lsr transcription. The
addition of analogues to these cells yielded no increased LacZ
activity relative that of to mock addition controls, and hence
they had no lsr agonist activity (not shown). Results in Figure 2
depict responses to analogues (20 μM) added with 20 μM
AI-2 (synthetic DPD). The “Methyl (DPD)” data are a con-
trol, demonstrating the elicitation of lsr gene expression (QS
response) due to exogenously added chemically synthesized AI-
2; the “Cells Only” control depicts the background level of β-
galactosidase without addition of either analogue or AI-2 (leaky
transcription). In E. coli cultures, simultaneous addition of
20 μM DPD (eliciting response) and 20 μM cyclopentyl-DPD
(quenching response) resulted in over 60% reduction in lsr
promoter activity. Interestingly, nearly all of the cyclic analogues
tested (with the exception of cyclopropyl-DPD) attenuated the
QS response in E. coli. The aromatic analogues exhibited minimal
(furanosyl-DPD) or no (phenyl, fluorophenyl, and nitrophenyl)
effect. In S. typhimurium, which has significant homology in
operon structure, the effects were strikingly different, particularly
for the cyclic analogues. Cyclopentyl-DPD and cyclohexyl-DPD
elicited increased expression, suggesting synergistic agonist
activity due to the presence of both DPD and analogue. Effects
of the other cyclic compounds when added to Salmonella with
DPD were no different than DPD alone; this was distinctly
different than E. coli. On the other hand, with the exception of
nitrophenyl-DPD, results for the aromatic C-1 analogues were
identical to those of E. coli, exhibiting no apparent effect on the
QS response. Nitrophenyl-DPD addition resulted in increased
QS response in Salmonella.
Figure 2. Competitive inhibition of QS signaling by analogues
in the presence of stoichiometric amounts of DPD in E. coli and
S. typhimurium; AI-2 dependent β-galactosidase production in
E. coli LW7 pLW11 and S. typhimurium MET715 (both luxS−). (a)
Cyclic analogues (20 μM). (b) Aromatic analogues (20 μM)
(normalized to the native E. coli (LW7) response (1304 Miller
units) and the S. typhimurium (MET715) response (3480 Miller
units), respectively, following 20 μM DPD addition. Controls
denoted “Cells Only” represent the response of the luxS− cells
when no compounds were added; controls denoted “Methyl
(DPD)” indicate the response of luxS− cells to exogenously added
AI-2 (DPD). (* indicates p < 0.05 and ** indicates p < 0.01 for an
unpaired t test of the particular analogue as compared to the
“Methyl (DPD)” response).
our new data suggests that C1 acyl DPD analogues (both linear
and branched chain) may fit into an active site pocket of the AI-2
processing/binding proteins, whereas some of the cyclic
analogues do not have this capability. We investigated whether
the cyclic and aromatic analogues were also processed via LsrR
and LsrK.
Biological Evaluations of DPD Analogues on E. coli AI-2
Processing Enzymes (LsrK) and AI-2 Receptor (LsrR).
We have shown previously that analogues most effective in
inhibiting lsr expression in E. coli were also phosphorylated
by LsrK.10 The ability of E. coli LsrK to phosphorylate the
cyclic analogues (both alkyl and aromatic) was tested by
incubation with LsrK in vitro for 1 h using a method adapted
from Xavier et al.34 Cyclobutyl-DPD and cyclopentyl-DPD,
which were significantly phosphorylated (Figure 3, panel a), also
inhibit lsr expression in E. coli (Figure 2, panel a). Cyclohexyl-
DPD and CH2-cyclohexyl-DPD were weakly phosphorylated
Earlier work from our laboratories, using a variety of linear or
branched chain C1 DPD analogues, had indicated that of the
analogues affecting E. coli and S. typhimurium QS activity (e.g.,
isobutyl-DPD, Figure 2a) all worked through the native AI-2
signal processing system involving the kinase, LsrK, and the
transcriptional regulator, LsrR. We concluded that because
the analogues contained C1 chains that were dissimilar to the
native DPD, both E. coli LsrK and LsrR were somehow
promiscuous.10 Our new data with the C1 cyclic alkyl analogues
of AI-2 reveal that these E. coli AI-2 processing/binding pro-
teins are not entirely promiscuous since only two cyclic ana-
logues exhibited an effect. Similarly, the aromatic analogues
were largely ineffective. When coupled with our earlier work,10
1025
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030

ACS Chemical Biology
Articles
control, with the exception of the aromatic phenyl-DPD
(Figure 4, panel a).
Figure 4. Effects of analogues on (a) E. coli via LsrR and (b) P. aeruginosa
via production of pyocyanin. (a) AI-2 dependent β-galactosidase
production in E. coli LW8 pLW11 (luxS+, lsrR−) in response to 20 μM
analogue. (b) Effect of analogues (100 μM) on pyocyanin production in
P. aeruginosa PAO1. (* indicates p < 0.05 for an unpaired t test of the
particular analogue compared to the cells only response).
Identifying Modulators of P. aeruginosa QS. We
screened DPD analogues (100 μM) for activity in P. aeruginosa
PAO1 by monitoring pyocyanin production (see Figure 4b
and Supplementary Figure 2). Pyocyanin is a redox active
phenazine toxin produced by P. aeruginosa and has been shown
to have antibiotic properties35 as well as being a signal for
the up-regulation of QS-controlled genes in the stationary
phase.36 Phenyl- and heptyl-DPD showed some inhibition of
pyocyanin production (Figure 4, panel b and Supplementary
Figure 2).
Effect of Analogues in a Trispecies Synthetic Eco-
system. Our results suggesting species-specific modulation of
QS activity led us to test effects in mixed cell cultures. Isobutyl-
DPD (40 μM) and phenyl-DPD (40 μM) were added to a
trispecies synthetic ecosystem created by culturing E. coli,
S. typhimurium, and P. aeruginosa in the same tubes. In order to
differentiate and monitor the QS response from each organism,
different and orthogonal reporter strains were used: in E. coli,
AI-2 mediated dsRED expression; in S. typhimurium, AI-2
mediated β-galactosidase production; and in P. aeruginosa,
effects of AI-2 were examined for pyocyanin production. It has
previously been shown that isobutyl-DPD inhibited E. coli and
S. typhimurium simultaneously, but due to its lack of inhibitory
effects in P. aeruginosa, we suspected it would not serve well as
a broad-spectrum quorum silencer in our trispecies synthetic
ecosystem. Interestingly, it did inhibit all three species, but
P. aeruginosa to a much lesser extent (Figure 5). Phenyl-DPD,
Figure 3. In vitro phosphorylation of analogues by LsrK. Thin layer
chromatography (TLC) analysis of LsrK mediated analogue
phosphorylation. (a) Cyclic analogues: methyl (DPD), cyclobutyl-
DPD, cyclopentyl-DPD, cyclohexyl-DPD, CH2-cyclohexyl-DPD,
cycloheptyl-DPD. (b) Aromatic phenyl-, fluorophenyl-, and nitro-
phenyl-DPD treated with LsrK.
(Figure 3, panel a) and attenuated lsr expression in E. coli to a
lesser extent. Neither Cycloheptyl-DPD nor the aromatic
analogues were phosphorylated (Figure 3, panel a and b), nor
did they affect LsrR-controlled transcription in E. coli (Figure 2,
panel a). We evaluated IC50 values of all analogues; those that
were significantly phosphorylated had significantly lower IC50
than those not phosphorylated, (i.e., IC50 for cyclopentyl-DPD
was 2.4 μM compared to phenyl-DPD, which was >40 μM).
Isobutyl-DPD was the most efficient inhibitor, having an IC50 in
the nanomolar range. These observations support our earlier
proposal that the active DPD analogue species that are acting to
repress lsr expression are the phosphorylated forms.10
To test whether the phosphorylated analogues work through
repressor LsrR, we carried out studies in isogenic lsrR null
mutants. Previous work showed that DPD analogues did not
alter lsr expression in the absence of LsrR, which is the tran-
scriptional repressor of the circuit. In this work results are
similar; while there was more variability, there was no sta-
tistically significant difference between the analogues and the
1026
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030

ACS Chemical Biology
Articles
Figure 5. Effect of analogue and analogue cocktail in a trispecies synthetic ecosystem. (a) AI-2 dependent β-galactosidase production in
S. typhimurium MET708 (normalized to native S. typhimurium response in trispecies culture = 475 Miller units). (b) QS related pyocyanin production
in P. aeruginosa PAO1 (normalized to native P. aeruginosa response in trispecies culture = 2.5 μg/mL). (c) AI-2 dependent dsRED induction in E. coli
W3110 pCT6 dsRED, in response to isobutyl-DPD (40 μM) and phenyl-DPD (40 μM) individually and a cocktail of both analogues. Note, in this
entire system, no exogenous AI-2 was added. Instead, the enteric bacteria present in the ecosystem synthesized the AI-2 signal present.
on the other hand, was largely ineffective as a QS inhibitor for
E. coli (Figure 5, panel c) and S. typhimurium (Figure 5, panel a)
but retained some activity against pyocyanin production in
P. aeruginosa (Figure 5, panel b).
We found C1 alkyl chains needed at least 3-carbon length for
lsr inhibition in E. coli;15 we now suggest the shape as well as
the conformational flexibility of the C1 side chain is also impor-
tant. For cyclic compounds there was an apparent 5-carbon
maximum where analogues larger than cyclopentyl-DPD (i.e.,
cyclohexyl, CH2-cyclohexyl, cycloheptyl) were not phosphory-
lated by LsrK and were less effective on LsrR. This is in
agreement with our previous studies showing that phosphor-
ylation by E. coli’s LsrK was an essential checkpoint to first
establish effectiveness as a modulator of the lsr QS circuit. IC50
data (Table 2) supported this conclusion in that analogues that
While it would be tempting to state that there was a
synergistic effect of one inhibitor on the other (because the
responses of the cocktail were lower than the cases when each
analogue was added individually to the trispecies), the statistical
differences are insignificant. It can be stated, however, that
there was no diminution of the effect of one analogue due to
the presence of the other. That is, the effect of each analogue as
a quorum quencher was undisturbed by the presence of the
other analogue. A conclusion can be drawn, therefore, that a
cocktail of DPD analogues consisting of isobutyl- and phenyl-
DPD analogues could simultaneously disrupt QS signaling in
the three different bacteria simultaneously via a nontoxic
mechanism (see Supporting Information Figures 3, 4, 6 and 7).
Discussion. In light of the ubiquitous nature of AI-2, it has been
envisioned that an AI-2 inhibitor would offer a universal quench-
ing QS activity due to its prevalence among many bacteria in their
native environments. Although this “magic bullet” may offer relief to
the antibiotic resistance problem, there are many bacteria that form
our microflora and are vital to our immunity and health. Ultimately
anti-QS therapies must be able to offer a variety of options including
both species-specific and multiorganism QS targeting.
Table 2. Inhibitory Concentrations (IC50) of Select
a
Analogues
IC50 of
analogue (nM) error (nM) to inhibit standard error
to inhibit QS in (log IC50) QS in (log IC50)
E. coli E. coli S. typhimurium S. typhimurium
standard IC50 of analogue
analogue
isobutyl-DPD
butyl-DPD
54
181
193
146
244
ND
20000
>40000
>40000
>40000
>40000
116
ND
ND
ND
ND
2967
2443
cyclopentyl-DPD
heptyl-DPD
15940
>40000
phenyl-DPD
a
IC50 values reflect the concentration of analogues required to reduce
lsr expression (β-galactosidase production) to 50% in the presence of
20 μM exogenously added synthetic DPD. β-Galactosidase production
in the absence of analogue (only 20 μM DPD) was set as the 100%
response.
We have developed a set of C1 analogues that enhances our
understanding of the shapes and sizes of C1 substituted groups
that are tolerated by AI-2 processing enzymes, such as LsrK.33
1027
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030

ACS Chemical Biology
Articles
Figure 6. Quorum sensing pathways in E. coli, S. typhimurium and P. aeruginosa and AI-2 analogue points of action. (a) AI-2 based QS signaling
circuit in E. coli and S. typhimurium; effective analogues are phosphorylated by LsrK and work via altering repressor LsrR activity. (b) The two AI-1
based LuxI/LuxR QS signaling circuits in P. aeruginosa; LasI/R and RhlI/R that produce different AHL autoinducers, N-(3-oxododecanoyl)-HSL
(OdDHL) and N-butyryl-HSL (BHL), respectively, to control QS associated virulence factors. There remain no identified effectors modulated by
AI-2 or analogues, just phenomenological observation.
are not readily phosphorylated require higher concentrations in
order to obtain 50% lsr repression. In P. aeruginosa, where the
phosphorylation pathway is not involved, phenyl-DPD inhibited
QS related pyocyanin production via unknown mechanisms. The
inability of phenyl-DPD to completely inhibit pyocyanin
production is probably due to the fact that other pathways, such
as the AI-1 pathway, also control pyocyanin production.
Exogenous AI-2 can up-regulate genes important for
P. aeruginosa pathogenesis.37 P. aeruginosa utilizes the AI-1
controlled LuxI/LuxR QS signaling circuit (Figure 6, panel b),
but because they are often found among other Gram-negative
microbes capable of synthesizing AI-2, P. aeruginosa likely
detects their AI-2 to coordinate the expression of pathogenic
genes during infection.37 Although the mechanisms by which
AI-2 affects P. aeruginosa are unknown, this study has revealed
that AI-2 analogues, such as phenyl-DPD, could become lead
compounds for further development as P. aeruginosa antitoxin
inhibitors. Interestingly, we found some evidence that the
effects of phenyl-DPD were greater against P. aeruginosa when
cultivated in the mixed culture. We have not explored, however,
concentration dependence nor population dependence to see
whether this is a robust finding or an experimental artifact. We
do note, however, that V. harveyi QS signaling was dramatically
attenuated when in a mixed culture with E. coli.38
virulence factor production.39−41 This study also demonstrates
an initial foray into “quorum quenching” among mixed cultures,
and because the apparent effects of the key antagonists appear
to retain their independence, it may be feasible to interrogate
effector molecules, their concentrations, and differing species
and their own relative proportions among consortia as a means
to unravel these complex communication networks. Additional
efforts on developing meaningful in vitro testing platforms that
will enable a delineation of effects of antimicrobial concen-
tration and response time will be helpful.
METHODS
■
Bacterial Strains and Growth Conditions. Table 1 lists the
bacterial strains and plasmids used in this study. S. typhimurium and
E. coli strains were cultured in Luria−Bertani medium (LB, Sigma) at
37 °C with shaking at (250 rpm) unless otherwise noted. Antibiotics
were used for the following strains: kanamycin (60 or 100 μg/mL) for
S. typhimurium MET715 and S. typhimurium MET708; ampicillin (60
or 100 μg/mL) for E. coli LW7 pLW11; and ampicillin (50 μg/mL)
and kanamycin (50 μg/mL) for E. coli W3110 pCT6 dsRED .
Measurement of the QS Response (lsr Expression). The QS
response indicated by lsr gene expression was analyzed in pure culture
studies by culturing E. coli LW7 pLW11, E. coli ZK126 pLW11 and
S. typhimurium MET708, S. typhimurium MET715 overnight in LB
medium individually supplemented with appropriate antibiotics, as
specified above. These cells were diluted into fresh LB medium (with
antibiotics) and grown to an OD600 of 0.8−1.0 at 30 °C, 250 rpm.
Cells were collected by centrifugation at 10,000 × g for 10 min
and resuspended in 10 mM phosphate buffer. AI-2 (20 μM) and
the respective analogue (20 μM) were added to the E. coli or
In conclusion, this diverse library of AI-2 analogues has shed
light on the QS systems of E. coli, S. typhimurium, and
P. aeruginosa. In the future, these and other analogues might be
tailored with molecular precision to alter QS activity among
many clinically relevant species. That is, several studies have
revealed that among polymicrobial infections, AI-2 increases
1028
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030

ACS Chemical Biology
Articles
S. typhimurium suspension for 2 h at 37 °C. AI-2 dependent β-
galactosidase production was quantified by the Miller assay.42
In Vitro Phosphorylation of Analogues. LsrK was purified from
E. coli BL21 pET200-LsrK as described before.9 Phosphorylated
analogues were synthesized by incubating 1 μM LsrK with 40 μM ATP
(Roche), 0.2 Ci of [P32] ATP (Perkin-Elmer), 300 μM AI-2 or
analogue, and 200 μM MgCl2, in 25 mM phosphate buffer, pH 7.4 for
2 h. An aliquot (2.5 μL) was then spotted onto a cellulose TLC plate
(Selecto Scientific). The plate was developed using 0.8 M LiCl as the
solvent, air-dried, and developed via autoradiography.
(4) Lowery, C. A., Abe, T., Park, J., Eubanks, L. M., Sawada, D.,
Kaufmann, G. F., and Janda, K. D. (2009) Revisiting AI-2 quorum
sensing inhibitors: direct comparison of alkyl-DPD analogues and a
natural product fimbrolide. J. Am. Chem. Soc. 131, 15584−15585.
(5) Jakobsen, T. H., Bragason, S. K., Phipps, R. K., Christensen, L. D.,
van Gennip, M., Alhede, M., Skindersoe, M., Larsen, T. O., Høiby, N.,
Bjarnsholt, T., and Givskov, M. (2012) Food as a source for quorum
sensing inhibitors: iberin from horseradish revealed as a quorum
sensing inhibitor of Pseudomonas aeruginosa. Appl. Environ. Microbiol.
78, 2410−2421.
Measurement of the QS Response (Pyocyanin Production).
The cells were grown with the analogue at 3 mL total volume in
50 mL flasks in LB medium, with continuous shaking. Between 22 and
24 h, when the cells turned green after pyocyanin secretion, the
pigment was extracted. The pyocyanin quantification assay was
conducted as described.43 Two milliliters of chloroform was added
to the 3 mL culture and pipetted up and down. One milliliter of the
chloroform was transferred to a separate tube, and the pyocyanin was
re-extracted into 200 μL of 0.2 M HCl. The absorbance of the solution
was measured at OD 520 nm. To calculate the concentration of
pyocyanin extracted as μg/mL, the OD at 520 nm was multiplied by
17.07.43
Analyzing QS Response in the Synthetic Ecosystem.
S. typhimurium MET708, P. aeruginosa PAO1, and E. coli W3110
pCT6 dsRED were each cultured separately overnight in LB medium
supplemented with the appropriate antibiotic. P. aeruginosa PAO1,
S. typhimurium MET708, and E. coli W3110 pCT6 dsRED were diluted
(25 μL:2.5 μL:100 μL) from the overnight cultures, respectively, into a
single 2 mL final volume of fresh LB medium without antibiotics. The
co-culture was supplemented with the respective analogue at 40 μM
concentration or analogue cocktail initially and again after 2.5, 5, 9, and
18 h of growth. The S. typhimurium lacZ (β-galactosidase) activity was
measured after 4 h. The E. coli response was determined after 24 h, by
fixing the cells with 1:1 cold 4% paraformaldehyde and using flow
cytometric analysis. Samples were analyzed by flow cytometry (FACS
Canto II, BD 394 Biosciences), with 30,000 gated events analyzed per
sample. Pyocyanin was also extracted after 24 h of growth.
(6) Glansdorp, F. G., Thomas, G. L., Lee, J. K., Dutton, J. M.,
Salmond, G. P., Welch, M., and Spring, D. R. (2004) Synthesis and
stability of small molecule probes for Pseudomonas aeruginosa
quorum sensing modulation. Org. Biomol. Chem. 2, 3329−3336.
(7) Hodgkinson, J. T., Welch, M., and Spring, D. R. (2007) Learning
the language of bacteria. ACS Chem. Biol. 2, 715−717.
(8) Geske, G. D., O’Neill, J. C., and Blackwell, H. E. (2007) N-
Phenylacetanoyl-L-homoserine lactones can strongly antagonize or
superagonize quorum sensing in Vibrio fischeri. ACS Chem. Biol. 2,
315−319.
(9) Roy, V., Fernandes, R., Tsao, C. Y., and Bentley, W. E. (2010)
Cross species quorum quenching using a native AI-2 processing
enzyme. ACS Chem. Biol. 5, 223−232.
(10) Roy, V., Smith, J. A., Wang, J., Stewart, J. E., Bentley, W. E., and
Sintim, H. O. (2010) Synthetic analogs tailor native AI-2 signaling
across bacterial species. J. Am. Chem. Soc. 132, 11141−11150.
(11) Sintim, H. O., Smith, J. A., Wang, J., Nakayama, S., and Yan, L.
(2010) Paradigm shift in discovering next-generation anti-infective
agents: targeting quorum sensing, c-di-GMP signaling and biofilm
formation in bacteria with small molecules. Future Med. Chem. 2,
1005−1035.
(12) Roy, V., Adams, B. L., and Bentley, W. E. (2011) Developing
next generation antimicrobials by intercepting AI-2 mediated quorum
sensing. Enzyme Microb. Technol. 49, 113−123.
(13) George, E. A., Novick, R. P., and Muir, T. W. (2008) Cyclic
peptide inhibitors of staphylococcal virulence prepared by Fmoc-based
thiolactone peptide synthesis. J. Am. Chem. Soc. 130, 4914−4924.
(14) Lyon, G. J., Wright, J. S., Muir, T. W., and Novick, R. P. (2002)
Key determinants of receptor activation in the agr autoinducing
peptides of Staphylococcus aureus. Biochemistry 41, 10095−10104.
(15) Geske, G. D., Wezeman, R. J., Siegel, A. P., and Blackwell, H. E.
(2005) Small molecule inhibitors of bacterial quorum sensing and
biofilm formation. J. Am. Chem. Soc. 127, 12762−12763.
(16) Smith, K. M., Bu, Y. G., and Suga, H. (2003) Induction and
inhibition of Pseudomonas aeruginosa quorum sensing by synthetic
autoinducer analogs. Chem. Biol. 10, 81−89.
(17) Malladi, V. L., Sobczak, A. J., Maricic, N., Murugapiran, S. K.,
Schneper, L., Makemson, J., Mathee, K., and Wnuk, S. F. (2011)
Substituted lactam and cyclic azahemiacetals modulate Pseudomonas
aeruginosa quorum sensing. Bioorg. Med. Chem. 19, 5500−5506.
(18) Semmelhack, M. F., Campagna, S. R., Federle, M. J., and Bassler,
B. L. (2005) An expeditious synthesis of DPD and boron binding
studies. Org. Lett. 7, 569−572.
(19) Lowery, C. A., Dickerson, T. J., and Janda, K. D. (2008)
Interspecies and interkingdom communication mediated by bacterial
quorum sensing. Chem. Soc. Rev. 37, 1337−1346.
(20) Taga, M. E., Miller, S. T., and Bassler, B. L. (2003) Lsr-mediated
transport and processing of AI-2 in Salmonella typhimurium. Mol.
Microbiol. 50, 1411−1427.
(21) Wang, L., Li, J., March, J. C., Valdes, J. J., and Bentley, W. E.
(2005) luxS-dependent gene regulation in Escherichia coli K-12
revealed by genomic expression profiling. J. Bacteriol. 187, 8350−8360.
(22) Xavier, K. B., and Bassler, B. L. (2005) Regulation of uptake and
processing of the quorum-sensing autoinducer AI-2 in Escherichia coli.
J. Bacteriol. 187, 238−248.
(23) Surette, M. G., Miller, M. B., and Bassler, B. L. (1999) Quorum
sensing in Escherichia coli, Salmonella typhimurium, and Vibrio harveyi:
a new family of genes responsible for autoinducer production. Proc.
Natl. Acad. Sci. U.S.A. 96, 1639−1644.
ASSOCIATED CONTENT
* Supporting Information
This material is available free of charge via the Internet at
■
S
http://pubs.acs.org
AUTHOR INFORMATION
Corresponding Author
*E-mail: hsintim@umd.edu; bentley@umd.edu.
■
Author Contributions
#These authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank R. Fernandes for his helpful discussions.
Funding for this work was provided by the National Science
Foundation grants CHE0946988, CHE0746446, Camille
Dreyfus foundation and EFRI-735987, DTRA BO085PO008
and the R. W. Deutsch Foundation.
REFERENCES
■
(1) Waters, C. M., and Bassler, B. L. (2005) Quorum sensing: cell-to-
cell communication in bacteria. Annu. Rev. Cell Dev. Biol. 21, 319−346.
(2) Bassler, B. L., and Losick, R. (2006) Bacterially speaking. Cell
125, 237−246.
(3) Keller, L., and Surette, M. G. (2006) Communication in bacteria:
an ecological and evolutionary perspective. Nat. Rev. Microbiol. 4,
249−258.
1029
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030

ACS Chemical Biology
Articles
(24) Chen, X., Schauder, S., Potier, N., Van Dorsselaer, A., Pelczer, I.,
Bassler, B. L., and Hughson, F. M. (2002) Structural identification of a
bacterial quorum-sensing signal containing boron. Nature 415,
545−549.
(25) Miller, S. T., Xavier, K. B., Campagna, S. R., Taga, M. E.,
Semmelhack, M. F., Bassler, B. L., and Hughson, F. M. (2004)
Salmonella typhimurium recognizes a chemically distinct form of the
bacterial quorum-sensing signal AI-2. Mol. Cell 15, 677−687.
(26) Ni, N., Choudhary, G., Li, M., and Wang, B. (2008) Pyrogallol
and its analogs can antagonize bacterial quorum sensing in Vibrio
harveyi. Biorg. Med. Chem. Lett. 18, 1567−1572.
(42) Miller, J. (1972) Experiments in Molecular Genetics, Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, NY.
(43) Essar, D. W., Eberly, L., Hadero, A., and Crawford, I. P. (1990)
Identification and characterization of genes for a second anthranilate
synthase in Pseudomonas aeruginosa: interchangeability of the two
anthranilate synthases and evolutionary implications. J. Bacteriol. 172,
884−900.
(27) Lowery, C. A., Park, J., Kaufmann, G. F., and Janda, K. D.
(2008) An unexpected switch in the modulation of AI-2-based quorum
sensing discovered through synthetic 4,5-dihydroxy-2,3-pentanedione
analogues. J. Am. Chem. Soc. 130, 9200−9201.
(28) Tsuchikama, K., Lowery, C. A., and Janda, K. D. (2011) Probing
autoinducer-2 based quorum sensing: the biological consequences of
molecules unable to traverse equilibrium States. J. Org. Chem. 76,
6981−6989.
(29) Kamaraju, K., Smith, J., Wang, J., Roy, V., Sintim, H. O., Bentley,
W. E., and Sukharev, S. (2011) Effects on membrane lateral pressure
suggest permeation mechanisms for bacterial quorum signaling
molecules. Biochemistry 50, 6983−6993.
(30) Wang, L., Hashimoto, Y., Tsao, C. Y., Valdes, J. J., and Bentley,
W. E. (2005) Cyclic AMP (cAMP) and cAMP receptor protein
influence both synthesis and uptake of extracellular autoinducer 2 in
Escherichia coli. J. Bacteriol. 187, 2066−2076.
(31) Connell, N., Han, Z., Moreno, F., and Kolter, R. (1987) An E.
coli promoter induced by the cessation of growth. Mol. Microbiol. 1,
195−201.
(32) Tsao, C. Y., Hooshangi, S., Wu, H. C., Valdes, J. J., and Bentley,
W. E. (2010) Autonomous induction of recombinant proteins by
minimally rewiring native quorum sensing regulon of E. coli. Metab.
Eng. 12, 291−297.
(33) Smith, J. A., Wang, J., Nguyen-Mau, S. M., Lee, V., and Sintim,
H. O. (2009) Biological screening of a diverse set of AI-2 analogues in
Vibrio harveyi suggests that receptors which are involved in synergistic
agonism of AI-2 and analogues are promiscuous. Chem. Commun.
(Cambridge, U. K.), 7033−7035.
(34) Xavier, K. B., Miller, S. T., Lu, W., Kim, J. H., Rabinowitz, J.,
Pelczer, I., Semmelhack, M. F., and Bassler, B. L. (2007)
Phosphorylation and processing of the quorum-sensing molecule
autoinducer-2 in enteric bacteria. ACS Chem. Biol. 2, 128−136.
(35) Hassett, D. J., Charniga, L., Bean, K., Ohman, D. E., and Cohen,
M. S. (1992) Response of Pseudomonas aeruginosa to pyocyanin:
mechanisms of resistance, antioxidant defenses, and demonstration of
a manganese-cofactored superoxide dismutase. Infect. Immun. 60,
328−336.
(36) Dietrich, L. E., Price-Whelan, A., Petersen, A., Whiteley, M., and
Newman, D. K. (2006) The phenazine pyocyanin is a terminal
signalling factor in the quorum sensing network of Pseudomonas
aeruginosa. Mol. Microbiol. 61, 1308−1321.
(37) Duan, K. M., Dammel, C., Stein, J., Rabin, H., and Surette,
M. G. (2003) Modulation of Pseudomonas aeruginosa gene expression by
host microflora through interspecies communication. Mol. Microbiol.
50, 1477−1491.
(38) Xavier, K. B., and Bassler, B. L. (2005) Interference with AI-2-
mediated bacterial cell−cell communication. Nature 437, 750−753.
(39) Armbruster, C. E., Hong, W., Pang, B., Weimer, K. E., Juneau, R.
A., Turner, J., and Swords, W. E. (2010) Indirect pathogenicity of
Haemophilus influenzae and Moraxella catarrhalis in polymicrobial
otitis media occurs via interspecies quorum signaling. mBio 1 (3),
e00102−110.
(40) Merritt, J., Qi, F., Goodman, S. D., Anderson, M. H., and Shi, W.
(2003) Mutation of luxS affects biofilm formation in Streptococcus
mutans. Infect. Immun. 71, 1972−1979.
(41) Wen, Z. T., and Burne, R. A. (2004) LuxS-mediated signaling in
Streptococcus mutans is involved in regulation of acid and oxidative
stress tolerance and biofilm formation. J. Bacteriol. 186, 2682−2691.
1030
dx.doi.org/10.1021/cb200524y | ACS Chem. Biol. 2012, 7, 1023−1030