Rosmarinic acid is a homoserine lactone mimic produced by plants that activates a bacterial quorum-sensing regulator

Abstract

Quorum sensing is a bacterial communication mechanism that controls genes, enabling bacteria to live as communities, such as biofilms. Homoserine lactone (HSL) molecules function as quorum-sensing signals for Gram-negative bacteria. Plants also produce previously unidentified compounds that affect quorum sensing. We identified rosmarinic acid as a plant-derived compound that functioned as an HSL mimic. In vitro assays showed that rosmarinic acid bound to the quorum-sensing regulator RhlR of Pseudomonas aeruginosa PAO1 and competed with the bacterial ligand N-butanoyl-homoserine lactone (C4-HSL). Furthermore, rosmarinic acid stimulated a greater increase in RhlR-mediated transcription in vitro than that of C4-HSL. In P. aeruginosa, rosmarinic acid induced quorum sensing–dependent gene expression and increased biofilm formation and the production of the virulence factors pyocyanin and elastase. Because P. aeruginosa PAO1 infection induces rosmarinic acid secretion from plant roots, our results indicate that rosmarinic acid secretion is a plant defense mechanism to stimulate a premature quorum-sensing response. P. aeruginosa is a ubiquitous pathogen that infects plants and animals; therefore, identification of rosmarinic acid as an inducer of premature quorum-sensing responses may be useful in agriculture and inform human therapeutic strategies.

Full text

HOST-PATHOGEN INTERACTIONS
Rosmarinic acid is a homoserine lactone mimic
produced by plants that activates a bacterial
quorum-sensing regulator
Andrés Corral-Lugo, Abdelali Daddaoua, Alvaro Ortega, Manuel Espinosa-Urgel, Tino Krell*
Quorum sensing is a bacterial communication mechanism that controls genes, enabling bacteria to live as
communities, such as biofilms. Homoserine lactone (HSL) molecules function as quorum-sensing signals
for Gram-negative bacteria. Plants also produce previously unidentified compounds that affect quorum
sensing. We identified rosmarinic acid as a plant-derived compound that functioned as an HSL mimic.
In vitro assays showed that rosmarinic acid bound to the quorum-sensing regulator RhlR of Pseudomonas
aeruginosa PAO1 and competed with the bacter ial ligand N-butanoyl-homoserine lactone (C4-HSL).
Furthermore, rosmarinic acid stimulated a greater increase in RhlR-mediated transcription in vitro than
that of C4-HSL. In P. aeruginosa, rosmarinic acid induced quorum sensing–dep endent gene expression
and increased biofilm formation and the production of the virulence factors pyocyanin and elastase.
Because P. aeruginosa PAO1 infection induces rosmarinic acid secretion from plant roots, our results
indicate that rosmarinic acid secretion is a plant defense mechanism to stimulate a premature quorum-
sensing response. P. aeruginosa is a ubiquitous pathogen that infects plants and animals; therefore,
identification of rosmarinic acid as an inducer of premature quorum-sensing responses may be useful
in agriculture and inform human therapeutic strategies.
INTRODUCTION
Plants live in association with fungi and bacteria, and it is believed that
plant ev olution was influenced by the presence of these associated micro-
organisms (1). During this evolutio n, diverse signaling systems emerged that
permitted mutual plant-microorganism sensing. Quorum sensing (QS) is a
mechanism of communication betw een bacteria and is based on the synthesis,
detection, and response to QS signals (QSS). As cell density increases, QSS
accumulate in the environment and are sensed by bacterial proteins cal led
QS regulators, w hich in turn control the expre ssion of genes; the produc ts of
these genes direct activities that are beneficial w hen performed by groups of
bacteria acting in synchron y (2). Homoserine lactones (HSLs) produced by
Gram-negativ e bacteria are the best studied and possibly the most common
group of bacterial QSS (3). Frequentl y, pairs of genes encoding the HSL
synthase and the HSL-sensing transcriptional regulator are found close to
each other in bacterial genomes. In addition, many bacteria ha ve additional
paralogs of HSL-sensing regulator genes that are not associated with an HSL
synthase gene and that were consequently termed solo or orphan regulators (4).
Bacteria-to-plant and plant-to-bacteria signaling are also based on QS
systems. A proteomics study show ed that HSLs modulate the expression of
a large number of genes in the legume Medicag o truncatu la (5). Similarly, a
transcriptomic study rev ealed that C6-HSL, a bacterial QS molecule pro-
duced in the rhizosphere, changed gene expression in Arabidopsis thaliana
(6). HSL signaling processes are, in part, responsible for the induced systemic
resistance of plants tow ard bacterial pathog ens (7); these processes also
modulate plant growth (8).
In addition, different plants produce compounds that interfere with the
bacterial QS mechanism. Extracts (9, 10) and macerates of different plants,
plant parts, and seeds (11–15), as well as exudates from seeds (16)orseedlings
(17), interfere with bacterial QS mechanisms. Additionally, leaf washings
from 17 different plants stimulated or inhibited HSL-dependent activities
in bacteria (18). Furthermore, QS-dependent gene expression is altered when
pathogenic bacteria gro w in their host plants (19). Biofilm formation and
HSL production increase in the presence of different plant-deriv ed phenolic
compound s (20); how e ver , these compounds do not act as HSL mimics. In
contrast, most of the experiments using extracts from plants stimulated
rather than inhibited QS-dependent gene expression (1), with the data sug-
gesting that these plant compounds act as HSL mimics and bind to the
autoinducer-binding domain of QS regulators. A molecular docking study
identified rosmarinic acid (RA), naringin, morin, mangiferin, and chloro-
genic acid (21) as plant-deriv ed compounds that were predicted to bind to
QS regulators. Although in that study each of these compounds inhibited
QS-mediated phenotypes, suggesting that they function as QS antagonists,
potential toxic effects were not evaluated. Experimental confirmation of the
binding of an y of these compou nds to QS regulators has not b een done. The
algal compound lumichrome, which is a riboflavin deriv ati v e, stimulates
the activity of a QS regulator of Pseudomonas aeruginosa (22).
Here, w e used P. aeruginosa PAO1 as a model organism to screen for
plant-derived HSL mimics. This bacterium is a ubiquitous pathogen that
infects a wide range of species, including humans and different plants such
as barley, poplar tree (23), and lettuce (24, 25). As a model for studying
the effect of QS on pathogenic traits, P. aeruginosa has a multisignal QS system
that is based on the synthesis and detection of signals that belong to two dif-
ferent classes, namely, Pseudomonas HSLs and quinolone signals (26, 27).
The HSL response is mediated by two pairs of synthases and regulators—
the synthase LasI and the regulator LasR (LasI/LasR) and the synthase RhlI
and the regulator RhlR (RhlI/RhlR)—as w ell as by the orphan regulator
QscR. RhlI produces the signaling molecule N-butano yl-homoserine lactone
(C4-HSL), and LasI produces N-3-oxododecano yl-homoserine lactone (3-
Oxo-C12-HSL) (26). The P. aeruginosa QS system is hierarchically organized
with LasR at the top of the signaling cascade: LasR activ ation stimulates tran-
scription of multiple genes, including rhlR, rhlI,andlasI. The QS cascade
then modulates multiple QS phenotypes, including changes in the amounts
of elastase, pyocyanin, rhamnolipid, and hydro gen cyanide (28).
Department of Environmental Protection, Estación Experimental del Zaidín,
Consejo Superior de In vestigac iones Científicas, C/ Prof. Alb areda, 1,
18008 Granada, Spain.
*Corresponding author. E-mail: tino.krell@eez.csic.es
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 1
on January 14, 2016http://stke.sciencemag.org/Downloaded from

Here, w e used ligand-free LasR
and RhlR purified from Escheric hia
coli without added HSL (29)formi-
crocalorimetric binding studies of
plant-derived compounds that were
selected on the basis of in silico docking
experiments. We identified RA as a
plant-derived compound that binds with
nanomolar affinity to RhlR. In tran-
scription assays with RhlR, RA signif-
icantly more effectively stimulated
transcription at low er concentrations
than C4-HSL. In bioassa ys, RA, but not
the closely related compound chloro-
genic acid , stimulated biofilm formation
and the production of the virulence
factors p yocyanin and elastase. RA is
produced exclusivel y in plants and not
in bacteria (30, 31). Thus, these data
sho w ed that RA acts as a QS regulator
agonist, thereby providing the molecu-
lar identification of a plant QSS mimic.
RESULTS
RA binds to purified RhlR
with high affinity
A major limitation in the study of the
HSL-sensing regulators is their insta-
bility in the absence of HSL (32–34).
Because recombinant regulator pur-
ified from E. coli cultures binds HSL
added to the culture medium, using
most methods, the bound HSL co-
purifies with the protein, resulting
in partially saturated regulators and
thus hampering ligand-binding studies.
We dev eloped a method that enables
the purification of recombinant RhlR
and LasR without the addition of
HSL, thereby providing a system for performing ligand-binding analysis
(29). Recombinant RhlR and LasR isolated with this method in the ab-
sence of HSL bound C4-HSL and 3-Oxo-C12-HSL, respectively, with dis-
sociation constant (K
D
)valuesof1.66±0.4mM and 1.14 ± 0.2 mM, as
determined by isothermal titration calorimetry (ITC) (Fig. 1, A and B).
To identify potential ligands, w e conducted in silico docking experiments
of ligands present in a database of natural compounds to the structure of
LasR and a model of RhlR. We used the structu re of the LasR autoind uce r-
binding domain in complex with 3-Oxo-C12-HSL (PDB ID: 3IX3) to gen-
erate a homology model of the analogo us domain of RhlR, which could be
closely superimposed onto the template (Fig. 1C). We used 3-Oxo-C12 -
H SL a n d C 4-HS L as con t ro l s in the docking experiments (Table 1) and the n
w e screened the Natural Compounds subset of the ZINC compound database
(5391 compounds) and selected those of plant origin and with docking scores
below −8 for further analysis. Most of the tested compounds had lower
docking scores at RhlR or LasR when compared to those of their cognate
HSL ligands (Table 1).
Microcalorimetric binding studies with the selected compounds were
performed to assess binding to ligand-free LasR and RhlR. We found that
RA bound only to RhlR (Fig. 1D) and not to LasR (Fig. 1E). No other
selected compound bound to RhlR or LasR (T ab le 1).
We calculate that RA bound to RhlR with a K
D
of 0.49 ± 0.08 mM
and had small favorable enthalpy changes (DH = −0.4 ± 0.05 kcal/mol).
Similar to the HSL ligands, the binding stoichiometry was close to 1:1,
which can be observed as the point of inflection of the sigmoidal binding
curves in Fig. 1D with respect to the low er x axis. To assess the specificity
of this interaction, we titrated RhlR with chlorogenic acid, a compound
structurally similar to RA (Fig. 2) and that had a low docking score (Table 1).
Chlorogenic acid did not cause significant heat changes, indicating that
this compound did not bind RhlR (Fig. 1D).
The RhlR model containing the best fit of docked RA and C4-HSL showed
that both ligands ov erlap (Fig. 2). Although the docking simulations with
LasR predicted that RA could ov erlap with bound 3-Oxo-C12-HSL (Fig. 2),
the ITC studies sho we d an absence of binding of RA for LasR (Fig. 1E).
RA stimulates RhlR-mediated transcription
To determine whether RA beha ved as an agonist or antagonist, we conducted
in vitro transcription assays with a 490–base pair (bp) DNA fragment
Fig. 1. Microcalorimetric binding studies of LasR and RhlR. (A)Titration
of 8 mM RhlR with 100 mM C4-HSL. (B) Titration of 8 mMLasRwith
100 mM 3-Oxo-C12-HSL. (C) Structural superimposition of the homology
model of the autoinducer-binding domain of RhlR (in orange) with
the structure of the analogous domain from LasR [in pink; Protein
Data Bank (PDB) ID: 3IX3]. The alignment was done using Subcomb
(69). (D) Titration of 19 mM RhlR with 0.66 mM RA (I) or chlorogenic
acid (II). Lower panel plots the titration data for RA. (E) Titration of buffer (I) and 8 mM LasR (II) with 0.66 mM RA. For the
titration data (A, B, D, and E), the upper panels show the raw titration data and the lower panels are concentration-
normalized and dilution heat-corrected integrated peak areas of the titration data fitted with the “One binding site model”
of the MicroCal version of Origin.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 2
on January 14, 2016http://stke.sciencemag.org/Downloaded from

containing the hcnABC promoter , which is activate d by RhlR (35). We per-
formed the experiments with a constant RhlR concentration of 25 mMin
the presence of either equimolar or half-equimolar concentrations of C4-
HSL and RA. Because the K
D
values for both ligands are much less than
the protein concentration, we expect that protein saturation at equimolar
ligand concentrations is comparable and almost complete. Both C4-HSL
and RA stimulated RhlR-dependent tran-
scription (Fig. 3A), but quantification revealed
that RA exhibited a significantly greater ac-
tivity compared to C4-HSL when the re-
sponse to equal concentrations of ligand was
compared to the transcription detected in
the control (absence of ligand) condition
(Fig. 3B). In the presence of half-equimolar
C4-HSL or RA to RhlR concentrations, tran-
scriptional activity was less than at equimolar
concentrations of ligand to regulator, thus
indicating dose depen dence of the re-
sponse. Consistent with the lack of binding
to RhlR, chlorogen ic acid did not increase
transcription. These data show ed that com-
pared to the bacterial ligand C4-HSL, RA
was more effective at activating RhlR-
dependent transcription from this promoter .
RA stimulates QS gene
expression in vivo
To assess the capacity of RA to modulate
QS-dependent gene expression in vivo,
w e conducted experiments in E. coli and
P. aeruginosa. RhlR controls the expres-
sion of the gene encoding its cognate
HSL synthase RhlI (36). Therefore, we
transformed E. coli BL21 with either plas-
mid pPET28b-RhlR (RhlR expression
plasmid) or pPET28b (empty plasmid as
control) and pMULTIAHLPROM con-
taining an rhlI::lacZ transcriptional fusion.
b-Galactosidase measurements show ed statistically significant increases in
gene expression in the presence of C4-HSL or RA, whereas chlorogenic
acid did not stimulate expression (Fig. 4A).
To study gene expression in P. aeruginosa, we introduced the rhlI::
lacZ reporter plasmid into the lasI
−
/lasR
−
double mutant and measured
b-galactosidase activity in samples taken at different time intervals after
Fig. 2. In silico docking of RA to the autoinducer-binding domains
of RhlR or LasR. The RhlR autoinducer-binding domain is a homology
model (see Materials and Methods) and contains docked C4-HSL,
and the LasR structure (PDB ID: 3IX3) contains bound 3-Oxo-C12-
HSL. The best binding positions of docked ligands with minimal glide
score (XP G score) and glide energy (G energy) are displayed. The docking scores are shown in Table 1. The
structures of different ligands are shown in the lower part of the figure.
Table 1. Results from in silico docking and experimental
binding studies of plant-derived compounds to RhlR and LasR.
Shown are XP scores for the in silico docking of different ligands to a
homology model of the RhlR autoinducer domain and to the structure
of the analogous domain of LasR. Bin ding parameters are derived
from ITC exper imen ts of purified La sR and RhlR wit h the ligands
listed. Data shown are means and SD from three i ndependent
experiments.
Ligand
Docking XP score Binding parameters
RhlR LasR RhlR LasR
K
D
(mM)
DH
(kcal/mol)
K
D
(mM)
DH
(kcal/mol)
Isoorientin −12.33 −14.95 No binding No binding
Spiraeoside −12.69 −14.29 No binding No binding
Luteolin-galactoside −13.01 −13.37 No binding No binding
Propanolol −11.70 −12.36 No binding No binding
RA −8.13 −10.87 0.49 ± 0.1 −0.4 ± 0.05 No binding
Mangiferin −8.90 −10.47 No binding No binding
Morin −8.87 −9.57 No binding No binding
Chlorogenic acid −9.46 −7.42 No binding No binding
Naringin <4.0 −5.96 No binding No binding
3-Oxo-C12-HSL −7.39 −8.71 Not determined 1.14 ± 0.2 −13.6 ± 0.2
C4-HSL −4.77 −5.56 1.66 ± 0.4 −16.1 ± 0.2 Not determined
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 3
on January 14, 2016http://stke.sciencemag.org/Downloaded from

the addition of either DMSO, C4-HSL, or RA. The data indicated that
the bacteria exhibited differential kinetics in response to the two RhlR
ligands. Induction of the repor ter in the cultures exposed to RA peaked
within 1 hour and was significantly greater than that of the control and
the C4-HSL–exposed cultures at both 1 and 2 hours (Fig. 4B). At sub-
sequent time points, reporter activity decreased in the RA-containing
cultures. We predicted that the reduction in b-galactosidase activity
of RA-containing cultures after 2 hours indicated that RA was metabo-
lized. The activity of C4-HSL cells was comparable to that of the control
after 1 hour but was significantl y greater for time points 6 to 8 hours,
which ma y be due to slo w uptake.
We determined the dose-response relationships for the increase in gene
expression induced by C4-HSL and RA (Fig. 4C). Because of the differences
in kinetics, we took measurements for the concentrations of ligand tested
at 4 hours, the time point at which C4-HSL and RA induced similar amounts
of b-galactosidase acti vity (Fig. 4B). At concentrations of 1 to 100 mM, tran-
scriptional activities were comparable, which may be due to metabolization
ofRA(Fig.4C).However,at0.5,1,and2mM,theb-galactosidase activity
in response to RA was higher than that induced by C4-HSL (Fig. 4C). At
5 mM RA, w e observed a further increase in b-galactosidase activity, but
w e could not perform similar measurements with C4-HSL due to the solubility
limit of this compound. These data are consistent with the in vitro transcription
experiments and support the conclusion that the capacity of RA to stimulate
transcription is superior to that of C4-HSL.
To confirm that P. aeruginosa could metabolize RA, we analyzed
bacteria gro wn in minimal medium containing 1 to 10 mM RA as the only
carbon source. P. aeruginosa grewwith1to5mMRAastheonlycarbon
source (Fig. 5A), consistent with metabolism of this compound and sug-
gesting that metabolism of RA may be responsible for the reduction of its
gene induction activity o ver time. The bacteria did not grow in 10 mM RA.
Although our growth data are consistent with those of Annapoorani et al.
(21), they conflict with those of Walker et al.(31) who reported that RA is
toxic at low micromolar concentrations. Therefore, we examined cell
Fi g . 3 . The capaci ty of C4-HS L and RA to stimulate RhlR-mediated
transcription in vitro. (A) Representative acrylamide gel of an in vitro
transcription assay using a DNA fragment containing the hcnABC promoter
that is induced by RhlR (35). Conditions match those listed below the graph
in (B). Ligands tested included C4-HSL, RA, and chlorogenic acid. (B)
Densitometric analysis of in vitro transcription assays. Shown are means
and SD from three individual experiments. *P < 0.05, Student’s t test.
Fig. 4. RA- or C4-HSL–mediated activation of a QS reporter in bacteria.
(A) Transcriptional activation in E. coli. b-Galactosidase measurements
at 2 hours after induction of E. coli BL21 containing pET28b-RhlR (expression
plasmid for RhlR) or the empty expression plasmid and pMULTIAHLPROM
containing a rhlI::lacZ transcriptional fusion. The ligand concentrations were
100 mM. CA, chlorogenic acid. (B) Transcriptional activation over time by the
indicated ligands in P. aeruginosa lasI
−
/la sR
−
containing pMULTIAHLPROM.
Bars represent the b-galactosidase measurements at different time intervals
after the addition of dimethyl sulfoxide (DMSO; control), C4-HSL, or RA. The line
graphs represent growth curves of the corresponding cultures. (C) Concentration-
dependent transcriptional activation by the indicated ligands in P. aeru-
ginosa lasI
−
/lasR
−
containing pMULTIAHLPROM. Dose-response curves for
each ligand from samples taken 4 hours after the addition of the ligand.
Shown are means and SD from three independent experiments conducted
in duplicate. **P < 0.01, ***P < 0.001, Student’s t test.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 4
on January 14, 2016http://stke.sciencemag.org/Downloaded from

viability as a function of RA concentration. We found that viability was not
affec ted by RA concentrations up to 7.8 mM, whereas viability dropped at
15.6 mM (Fig. 5B). Furthermore, we confirmed that the presence of C4-
HSL and RA at the concentrations used for gene expression studies did
not change growth kinetics (Fig. 5C). These discrepancies in the RA tol-
erance of P. aeruginosa PAO1 may be due to differential evolution of the
strain in different laboratories.
The transcriptional reporter studies so far used the rhlI promoter . We
performed analogous experiments in P. aeruginosa PA O1 expressing lacZ
controlled by lasB (37), rhlA (38), or hcnABC (35), which are all induced
by RhlR. RA produced a significant increase in b-galactosidase activity
for each of the three reporters (Fig. 6, A to C), whereas control experi-
ments with bacteria transformed with the empty plasmid did not show an
increase in b-galactosidase activity upon RA addition (Fig. 6D).
RA increases biofilm formation, pyocyanin production,
and elastase synthesis
Increased biofilm formation and the production of the virulence factor
pyocyanin are characteristic features of HSL-mediated QS responses
(39, 40). We therefore tested these traits in P. aeruginosa grown in the
presence and absence of different concentrations of RA or chlorogenic
acid. We quantified pyoc yanin, which is green, by measuring the absorbance
at 520 nM and found that RA, but not chlorogenic acid , stimulated a dose-
dependent increase in the intensity of the green color when corrected for cell
density (Fig. 7A).
Vi sual inspection of culture tubes (inset in Fig. 7A) sho wed that RA also
stimulated biofilm formation. To quantify the effect of RA on biofilm for-
mation, we grew bacterial cultures in borosilicate glass tubes in the presence
or absence of either 2 mM RA or chlorogenic acid and quantified biofilm
formation at different time points. We found that RA stimulated biofilm
formation between 2 to 8 hours of gro wth, whereas chlorogenic acid had no
significant effect (Fig. 7B, upper). After 24 hours, the amount of biofilm
formed in these three conditions was comparable. The RA-mediated stim-
ulation of biofilm formation contradicted a previous report in which the
compound was used at higher concentrations than what we used here and
inhibited biofilm formation (21). To assess this discrepancy, w e determined
the dose-response relationship of the RA-mediated effect on biofilm (Fig. 7B,
lower). RA stimulated biofilm formation at concentrations up to 2 mM,
but abov e this concentration, biofilm formation was inhibited by RA.
Elastase synthesis is also stimulated by RhlR-mediated QS activity (37).
Our reporter analysis indicated that RA enhanced the activity of the
promoter controlling the expression of lasB, encoding elastase (Fig. 6A).
To verify whether this results in changes in elastase synthesis, we measured
elastas e activity in P. aeruginosa grown in the absence and presence of RA.
RA stimulated the amount of elastase activity in density-normalized cultures
(Fig. 7C).
DISCUSSION
Multiple lines of evidence indicate that plants produce HSL mimics,
which interfere with the bacterial QS system (9–17). However, little
information is available regarding the molecular identity of the active
compounds that are responsible for this interference (1).
We report the identification and characterization of a plant com-
pound that directly interacts with a bacterial QS regulator, stimulating
its transcriptional activity. RA bound purified RhlR with a higher affinity
than C4-HSL, and this translated into a greater stimulatory activity of
RA on RhlR-m ediated tran scription and gene expression compared
with that induced by C4-HSL. Further more, RA stimulated biofilm
formation and the synthesis of pyocyanin and elastase, which are
Fig. 5. The effect of RA on P. aeruginos a PAO1 survival and growth. (A)Growth
curve of P. aeruginosa PAO1 in M9 minimal medium supplemented with the
indicated concentration of RA. Shown are means and SD from three
experiments conducted in triplicate. (B)GrowthinLBmediumsupplemented
with the indicated concentrations of RA (and the corresponding DMSO-
containing controls). Cell survival after 24 hours was determined by plating
out on solid medium and cell counting. Shown are means and SD from three
independent experiments each conducted in quintuplicate. (C)ImpactofRA
on bacterial growth on rich medium. Shown is a growth curve in LB medium
supplemented with different concentrations of C4-HSL or RA. Shown are
means and SD from three experiments conducted in triplicate.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 5
on January 14, 2016http://stke.sciencemag.org/Downloaded from

phenotypic characteristics that are regulated
by QS mechanisms in P. aeruginosa (40, 41).
Because interference with bacterial QS
mechanisms by plant-derived compounds
has been observed for various different
plant patho gens, it was proposed that this rep-
resents a plant defense strategy (1). HSL
mi mi c s acti n g as agonists of HSL-mediated
sensing may decrease pathogenicity because
these mimics would stimulate premature ex-
pression of genes encoding proteins involved
in QS-controlled functions (1). Molecular
identification of HSL mimics will enable their
application in medicine and agriculture, for
example, in the generation of pathogen-resistant
plants. However , is this mimicking of a bac-
terial QSS by RA of physiological relevance?
Walker et al.(31) monitored the consequences
of sw eet basil root infection by P. aeruginosa
strains PA O1 and PA14 and determined that
infection with either strain induced RA se-
cretion. RA concentration in root exudates
gradually increased, reaching a maximum
of ~40 mM 6 days after infection (31). Thus,
plant infection triggers RA release, suggesting
that RA release forms part of a plant defense
strategy and supporting the model that plant-
derived HSL mimics decreased pathogenicity
by stimulating premature QS-responsive gene
expression (1). We found that RA was both a
gro wth substrate and a signaling molecule.
Bacterial consumption of RA ma y pro vide
Fig. 6. Effect of RA on the expression from other QS-regulated promoters. (A to D) b-Galactosidase
measurements of P. aeruginosa PAO1 containing the plasmids pb01 (lasB-lacZ)(A),pb02 (rhlA-lacZ)
(B), pME2823 (hcnA-lacZ) (C), or pQF50 (empty vector) (D). Measurements were made 2 hours after
the addition of RA or the corre spondi ng amou nt of DM SO. Shown a re means and SD from three
independent experiments conducted in duplicate. **P < 0.01, ***P < 0.001, Student’s t test.
Fig. 7. RA induces QS-regulated phenotypes. (A) Pyocyanin production
measured in P. aeruginosa PAO1 grown in the presence of different con-
centrations of RA or chlorogenic acid. Shown are means and SD from three
independent experiments of the absorbance at 520 nm (pyocyanin) of P.
aeruginosa supernatants relative to the absorbance at 660 nm (cell density).
The inset shows culture tubes after 8 hours of growth. (The amount of DMSO
equivalent to that added in the 2 mM RA condition was added as a control
to the first tube “0.”)(B) Biofilm formation quantified in the presence and
absence of RA or chlorogenic acid at different time points. Shown are
means and SD from three independent experiments. *P < 0.05, **P <
0.01, ***P <0.001,Student’s t test. The inset shows representative crystal
violet–stained tubes containing an 8-hour culture of P. aeruginosa in the
presence and absence of RA or chlorogenic acid. In the lower part, the
tubes show the RA concentration dependence of biofilm formation (after
8 hours). (C) Elastase synthesis measured as enzymatic activity of P. aeruginosa
PAO1 cultures grown in the presence or absence of RA. Data are normalized
for culture density. Shown are means and SD from three individual ex-
periments conducted in triplicate. **P < 0.01, Student’s t test.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 6
on January 14, 2016http://stke.sciencemag.org/Downloaded from

a mechanism to eliminate the signaling effects of this compound, which is
consistent with the kinetics of gene induction by RA that w e observed.
Reports of the toxic effects of RA to P. aeruginosa PAO1 differ. Our data
and those reported by Annapoorani et al.(21) indicate an effectiv e inhibitory
concentration of 2.1 mM with regard to bacterial growth, whereas W alker et al.
(31) reported minimal inhibitory concentration of 8 mM. These discrepan-
cies may be due to differential strain evolution in different laboratories.
Annapoorani et al.(21) show e d a reduction of P. aeruginosa PA O1 biofilm
formation and elastase activity in the presence of RA and a number of rel at e d
compounds, including chlorogenic acid; how e v er , our data show ed no ac-
tivity for chlorogenic acid in binding to or stimulating RhlR or in promoting
biofilm formation, and we found that RA functioned as an HSL mimic.
P. aeruginosa is one of the pathogens that infect a wide range of species,
including plants and animals (25). Common pathogenic mechanisms enable
bacteria to infect phylog eneticall y different hosts, and there are also parallels
in the key features underlying host defense responses in plants, in vertebrates,
and mammalian hosts (25). The production of HSL mimics is considered a
plant defense strategy. The detailed knowledge of plant defense mechanisms
ma y enable the dev elopment of strategies to protect human from this path-
ogen. The effect of HSL mimics, including R, on the virulence properties
to ward mammals is of interest.
RA is synthesized by many plants (42) and can accumulate to high con-
centrations (43). RA has multiple biological activities, including antibacterial
(31), antivira l (44), antiallergy (45), anticarcinogen ic (46), antigenotoxic (47),
anti-inflammatory (48), and antioxidant effects (49) activity. In addition,
RA was found to be effectiv e against amyloid-b peptide–induced neuro-
toxicity that is associated with Alzheimer’sdisease(50), reduces atopic derma-
titis (51), and protects keratinocytes from ultraviolet radiation damage (52).
Consequentl y, plants rich in RA are used as medicinal herbs and by the food
industry (43).
Solo or orphan QS regulators recognize plant-derived HSL mimics (1).
Solo QS regulators are abundant in plant-associated bacteria, which supports
the view that they are inv olved in interkingdom signaling between plants and
bacteria (1). Our data sho wed that RhlR has a double function and mediates
responses to bacterial HSL and plant-derived RA, thus indicating that rec-
ognition of HSL mimics is not limited to solo QS regulators.
Sev eral synthetic HSL and non-HSL ligands hav e been identified that
modulate the activity of P. aeruginosa QS regulators (53–58). These com-
pounds behaved as either agonists or antagonists. Structurally, the agonists
(53, 58) are similar to RA because part of these compounds were linear with
aromatic moieties at each extension of the molecule.
Here, we identified RA as an HSL mimic. Indeed, RA mimicked the
action of C4-HSL in vitro and in vivo despite being dissimilar structurally.
RA is not the only plant-deriv ed HSL mimic. Gao et al.(16) estimated that
there are 15 to 20 separable compounds with the capacity to affect HSL-based
QS processes. Although RA can serve as a lead compound, this work dem-
onstrated that mimics can have a structure very different from that of the
co gnate ligand.
MATERIALS AND METHODS
Materials
The strains and plasmids used in this study are pro vided in Table 2. HSL and
plant-deri ved compounds (Table 1) w ere purchased from Sigma-Aldrich.
Purification of LasR and RhlR
The DNA fragments encoding Las R and RhlR were amplified using
the primers 5′-GTTTAAGAAGAACGT
GCTAGCATGGCCTTG-3′
and 5′-CTGAGAG
GGATCCTCAGAGAGTAAT AAGAC-3′ (LasR)
and 5′-TATCGA
GCTAGCCTTACTGCAATGAGGAATGAC-3′ and
5′-C
GAGCTCTGCGCTTCAGATGAGACC-3′ (RhlR), respectively.
These primers contained restriction sites (underlined) for Nhe IandBam
HI (LasR) and Nhe IandSac I (RhlR). Pol ymerase chain reaction (PCR)
products were digested with these enzymes and cloned into the expression
plasmid pET28b(+). E. coli BL21 (DE3) was transformed with the resulting
plasmids, and cultures were gro wn in LB medium supplemented with ka-
nam ycin (50 mg/ml) at 37°C until an OD
660
(optical density at 660 nm) of
0.4. The temperature w as then low ered to 18°C, and gro wth continued until
an OD
660
of 0.6 to 0.8, at which point protein production was induced by
the addition of 0.1 mM isopropyl-b-
D-thio g alactop yranoside. Growth was
continued at 18°C ov ernight, and cells w ere harvested by centrifugation at
10,000g for 30 min. Cell pellets were resuspended in buffer A [20 mM
tris-HCl, 0.1 mM EDTA, 500 mM NaCl, 10 mM imidazole, 5 mM b-
mercaptoethanol, 5% (v/v) glycerol, 1 mM dithiothreitol (DTT) (pH 7.8)]
and broken by French press at 1000 psi. After centrifugation at 20,000g for
1 hour , the supernatant was loaded onto a 5-ml HisTrap column (Amersham
Biosciences) previously equilibrated with buffer A. The column was then
washed with buffer A containing 45 mM imidazole before protein elution
with a linear gradient (20 min) of 45 to 500 mM imidazole in buffer A at a
flo w of 1 ml/min. Protein-containing fractions w ere pooled and dialyzed into
bufferB[50mMtris-HCl,500mMNaCl,1mMDTT(pH7.8)]and
Table 2. Strains and plasmids used in this study. Antibiotic resistance: Ap, ampicillin; Gm, gentamicin; Km, kanamycin; Tc, tetracycline;
Cb, carbenicillin.
Strains or plasmids Relevant characteristics Reference
P. aeruginosa PAO1 Wild type, prototroph; Ap
r
(70)
P. aeruginosa PAO1
lasI
−
/lasR
−
Double mutant in lasI and lasR genes Gm
r
Personal gift, M. Cámara
(University of Nottingham)
E. coli BL21 (DE3) F
−
, ompI, hsdS
B
(r
−
B
m
−
B
)(71)
pET28b(+) Km
R
, protein expression vector Novagen
pET28b-LasR Km
R
, pET28b(+) derivative containing lasR gene This work
pET28b-RhlR Km
R
, pET28b(+) derivative containing rhlR gene This work
pMULTIAHLPROM Tc
R
, broad-host-range plasmid containing 8-luxI type
promoters fused to a promoter lacZ gene
(63)
pQF50 Cb
R
, broad-host-range transcriptional fusion vector (68)
pb01 Cb
R
, pQF50 derivative containing lasB-lacZ transcriptional fusion (68)
pb02 Cb
R
, pQF50 derivative containing rhlA-lacZ transcriptional fusion (68)
pME2823 Cb
R
, pKT240 derivative containing hcnA-lacZ transcriptional fusion (35)
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 7
on January 14, 2016http://stke.sciencemag.org/Downloaded from

applied to a HiPrep26/60 Sephacryl S-200 high resolution gel filtration col-
umn previousl y equilibrated with the same buffer. Protein was eluted by a
constant flow (1 ml/min) of buffer B at 4°C.
Isothermal titration calorimetry
Experiments were conducted on a VP-microcalorimeter (MicroCal) at 25°
and 30°C. Proteins were dialyzed in analysis buffer [50 mM K
2
HPO
4
/
KH
2
PO
4
, 150 mM NaCl, 1 mM DTT (pH 7.8)]. HSL ligands were pre-
pared at a concentration of 1 mM in DMSO and subsequently diluted 1:10
with analysis buffer. The corresponding amount of DMSO [10% (v/v)]
was added to the dialyzed protein sample. Typically, 8 to 15 mM protein
was titrated with 0.1 mM HSL solution. Control experiments involved the
titration of dialysis buffer containing 10% (v/v) DMSO with HSL ligand.
For the non-HSL ligands, a solution of 660 mM was prepared directly in
dialysis buffer and used for the titration of the dialyzed protein. The mean
enthalpies measured from the injection of ligands into the buffer were
subtracted from raw titration data before data analysis with the MicroCal
version of Origin.
Molecular docking, homology modeling,
and structural alignment
The atomic structure of LasR was obtained from the PDB (www.pdb.org;
PDB ID: 3IX3). The structure was refined and optimized with the Protein
Preparation Wizard of the Schrödinger Suite (Schrödinger Suite 2012 Pro-
tein Preparation Wizard; Epik version 2.3, Schrödinger, LLC, New York,
NY, 2012; Impact version 5.8, Schrödinger, LLC, New York, NY, 2012;
Prime version 3.1, Schrödinger, LLC, New York, NY, 2012). Ligands were
obtained from the Natural Compounds (Metabolites) subset of the ZINC
database (59), opt imi ze d by LigPrep (LigPrep, version 2.5, Schrödinger ,
LLC, New York, NY, 2012), and then submitted to virtual screening
docking experiments to LasR using the Glide dock SP mode (Glide, version
5.8, Schrödinger , LLC, New York, NY, 2012) (60). The best hits we re sub-
sequently docked using the Glide dock XP mod e. A homo lo g y model of
the autoinducer domain of RhlR was generated using Swiss-Model (61)and
the structure of QscR (PDB ID: 3SZT) as template. RA was docked onto
this structure in the Glide dock XP mode. The structural alignment of the
autoinducer domains of LasR and RhlR was generated by PyMOL (PyMOL
Molecular Graphics System, version 1.5.0.4, Schrödinger, LLC).
In vitro transcription assay
A 490-bp DN A fragment of the hcnABC promoter of P. aeruginosa (35)
was amplified by PCR using the primers 5′-GCACTGAGTCGGACAT-
GACGGAA-3′ and 5′-CGTGTTGACGTTCAAGAAGGTGCATTGC-
3′ and used as a template for these assays. Transcription reactions (20 ml)
w ere performed in binding buffer [20 mM tris-HCl, 50 mM KCl, 1 mM
EDTA, 1 mM DTT, 10% (v/v) glycerol (pH 7.8)] containing 50 nM E. coli
RN A polymerase holoenzyme saturated with s
70
sigma factor (Epicentre
Technolo gies), 5 nM linear hcnABC DNA, 25 mM RhlR, and different ef-
fector molecules (12.5 or 25 mM C4-HSL, RA, and chlorogenic acid). The
mixtures were incubated at 30°C for 20 min before the addition of 0.1 mM
adenosine triphosphate, cytidine triphosphate, and guanosine triphosphate;
0.05 mM uridine triphosphate (UTP); and 3.6 mCi of [a-
32
P]UTP (10 mCi/ml)
(1 Ci = 37 GBq). After incubation for 50 min, the reactions were stopped
by transferring them to a 95°C thermoblock and then subsequently chilled at
4°C, at which point 4 ml of formamide sequencing dye was added. Samples
w ere separated on 6.5% (w/v) polyacrylamide gel electrophoresis gels for
2 hours. Gels were dried and then exposed on a phosphorimager, and the
resulting images were processed with the Quantity One software 4.6.2
(Bio-Rad Laboratories). The densitometric analysis was carried out using
the program ImageJ (62).
Gene expression studies
Gene expression experiments were conducted with E. coli BL21 harboring
pET28b-RhlR (expression plasmid for RhlR) or pET28b (empty plasmid
as a control) and pMULTIAHLPR OM containing an rhlI::lacZ transcrip-
tional fusion (63), as well as with P. aeruginosa PAO 1 lasI
−
/lasR harboring
pMULTIAHLPR OM. E. coli BL21 were grown in LB containing tetracy-
cline (10 mg/ml) and kanamycin (50 mg/ml), and P. aeruginosa PAO 1 lasI
−
/lasR
in LB containing tetracycline (40 mg/ml) and gentamicin (20 mg/ml) at 37°C
o v ernight. Stock solutions of RA and chlorogenic acid were prepared in
DMSO (100%) and diluted in water to the desired concentration, whereas
C4-HSL solutions were prepared in 10% (v/v) DMSO. Fresh LB medium
was then inoculated with the resulting cultures (1:100 dilution), grown for
1 hour, and then diluted twofold twice at 30-min intervals to ensure proper
dilution of accumulated b-galactosidase after overnight growth. The result-
ing culture was then grown for another hour before induction with different
ligands. To rule out nonspecific effects of DMSO (present in the stock solu-
tion), control experiments were performed in which the amount of DMSO
corresponding to that present in C4-HSL–,RA-,orchlorogenicacid–containing
cultures was added. Growth was continued at 37°C, and samples w ere tak-
en t different time points for the determination of b-galactosidase in per-
meabilized whole cells as described in (64). For dose-response experiments,
the b-galactosidase activity was measured 4 hours after induction. Data
sho wn are means and SD from at least three independent experiments.
To explore the effect of RA on other RhlR-regulated promoters, plasmids
containing lacZ fusions were transferred to the wild-type strain by electro-
poration. These pla smids were pß01 (lasB::lacZ), pß02 (rhlA::lacZ),
pME3823 (hcnA::lacZ), and the insert-free pQF50, which has served to
construct the former two plasmids (Table 2). Experimental conditions
w ere as those described abo v e except that cultures w ere gro wn for 6 hours
after dilution before the induction with different ligands.
Minimal inhibitory concentration assay
These assays were performed in 96-w ell plates using a modified version of
theprotocolreportedin(65). Wells of a 96-w ell plate w ere filled with 200 ml
of LB containing different amounts of RA (added using a 250 mM stock
solution in DMSO). Control experiments contained the corresponding
amounts of DMSO . Wells were inoculated with 10 ml of an ov ernight culture
of P. aeruginosa PAO1 in LB medium. Plate was incubated at 37°C for
24 hours, at which point the viable cell amount was determined by plating
out cells on LB agar medium and counting.
Growth experiments
To assess the potential of the bacterium to use RA as sole growth substrate,
sterile honeycomb plates (Bioscreen C) containing 200 ml of M9, supple-
mented with 1 to 10 mM RA, were inoculated with an ov ernight culture of
P. aeruginosa PAO1 gro wn in M9 minimal medium (66) containing 5 mM
citrate at 37°C. Cultures w ere grown in a Bioscreen C (Thermo Fisher Sci-
entific) instrument under constant shaking at 37°C during which time the
OD
660
was measured in 1-hour intervals. To assess the effect of different
RA concentrations on P. aeruginosa gro wth, honeycomb plates were filled
with LB medium containing 1 to 100 mM C4-HSL or RA, and cultures were
carried out as described abov e.
Biofilm formation
Overnight cultures of P. aeruginosa PAO1 were grown at 37°C and used
to inxoculate borosilicate glass tubes containing 2 ml of LB medium (supple-
mented with either 2 mM RA or chlorogenic acid) to an initial OD
660
of 0.05.
Both compounds we re added as 143 mM solutions in DMSO , and the
corresponding control experiments were conducted to assess the effect
of the equiv alent amount of DMSO on biofilm formation. Cultures were
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 8
on January 14, 2016http://stke.sciencemag.org/Downloaded from

incubated in a Stuart SB3 tube rotator for 2, 4, 6, 8, and 24 hours at 30°C,
with an angle of 45° at 40 rpm. Biofilms formed were visualized by
crystal violet (0.4%) staining and quantified by solubilizing the dye with
30% acetic acid and measuring the absorbance at 540 nm (67). Data
sho wn are means and SD from three experiments conducted in duplicate.
Quantification of pyocyanin production
Cultures of P. aeruginosa PAO1 were grown in LB at 37°C o v ernight and
used to inoculate glass tubes containing 2 ml of LB medium to an initial
OD
660
of 0.05. Stock solutions of 143 mM RA or chlorogenic acid were
prepared in DMSO , and aliquots w ere then added to the tubes to final con-
ce ntrations of 0.5 to 2 mM. The amount of DMSO corresponding to the
experiment at 2 mM was added to the control tube. Growth was continued ,
and pyocyanin production was determined after 8 hours. The OD
660
of
cultures was determined before centrifugation of cultures at 13,000 rpm for
5min.TheOD
520
(indicativ e of py ocyanin production) was measured , and
va lues w ere normalized with the cell density (OD
660
). LB media containing
either RA or chlorogenic acid w ere used as blanks.
Elastolysis assay
Elastase activity in P. aeruginosa PAO1 cultures was determined using a
modified version of the elastin–Congo red (ECR) assa y (68). Cells were
grown with shaking in LB medium at 37°C ov ernight and used to inoc-
ulate glass tubes containing 3 ml of LB medium to an initial OD
660
of
0.05. Growth was continued for another 6 hours, and cultures were in-
ducedwith500mM RA. The equiv alent amount of DMSO was added to
control tubes. Gro wth was continued until 24 hours, and 1 ml of cell sus-
pension was centrifuged at 13,000 rpm for 15 min. The resulting super-
natant was added to tubes containing 10 mg of ECR (Sigma) and 1 ml
of buffer [0.1 M tris-HCl, 1 mM CaCl
2
(pH 7.0)]. Tubes were incubated
at 37°C with shaking (150 rpm) for 24 hours. The reaction was stopped
by the addition of 1 ml of sodium phosphate buffer (0.7 M; pH 6.0).
Residual, solid ECR was remo ved b y centrifugation, and the OD
492
of the
supernatant was measured. Sho wn are means and SD from three replicates
conducted in triplicate.
REFERENCES AND NOTES
1. V. Venturi, C. Fuqua, C hemical signaling between plants and plant-pathogenic
bacteria. Annu. Rev. Phytopathol. 51,17–37 (2013).
2. S. T. Rutherford, B. L. Bassler, Bacterial quorum sensing: Its role in virulence and
possibilities for its control. Cold Spring Harb. Perspect. Med. 2, a012427 (2012).
3. C. Fuqua, M. R. Parsek, E. P. Greenberg, Regulation of gene expression by cell-to-cell
communication: Acyl-ho moserine lac tone quorum sensing. Annu. Rev. Genet. 35,
439–468 (2001).
4. J. F. González, V. Venturi, A novel widespread interkingdom signaling circuit. Trends
Plant Sci. 18, 167–174 (2013).
5. U. Mathesius, S. Mulders, M. Gao, M. Teplitski, G. Caetano-Anollés, B. G. Rolfe,
W. D. B auer, Extensive and specific responses of a eukaryote to bacterial quorum-
sensing signals. Proc. Natl. Acad. Sci. U.S.A. 100, 1444–1449 (2003).
6. U. von Rad, I. Klein, P. I. Dobrev, J. Kottova, E. Zazimalova, A. Fekete, A. Hartmann,
P. Schmitt-Kopplin, J. Durner, Response of Arabidopsis thaliana to N-hexanoyl-DL-
homoserine-lactone, a bacterial quorum sensing molecule produced in the rhizo-
sphere. Planta 229,73–85 (2008).
7. A. Schikora, S. T. Schenk, E. Stein, A. Molitor, A. Zuccaro, K.-H. Kogel, N-a cyl-homoserine
lactone confers resistance toward biotrophic and hemibiotrophic pathogens via altered
activation of AtMPK6. Plant Physiol. 157, 1407–1418 (2011).
8. S. T. Schenk, E. Stein, K.-H. Kogel, A. Schikora, Arabidopsis growth and defense are
modulated by bacterial quorum sensing molecules. Plant Signal. Behav. 7,178–181 (2012).
9. G. Degrassi, G. Devescovi, R. Solis, L. Steindler, V. Venturi, Oryza sativa rice plants
contain molecules that activate different quorum-sensing N-acyl homoserine lactone
biosensors and are sensitive to the specific AiiA lactonase. FEMS Microbiol. Lett. 269,
213–220 (2007).
10. A. A. Tolmacheva, E. A. Rogozhin, D. G. Deryabin, Antibacterial and quorum sensing
regulatory activities of some traditional Eastern-European medicinal plants. Acta
Pharm. 64, 173–186 (2014).
11. S. Ferlug a, V. Venturi, OryR is a LuxR-family p rotein involved in in terkingd om
signaling between pathogenic Xanthomonas oryzae pv. oryzae and rice. J. Bacteriol.
191, 890–897 (2009).
12. S. Ferluga, J. Bigirimana, M. Höfte, V. Venturi, A LuxR homologue of Xanthomonas
oryzae pv. oryzae is required for optimal rice virulence. Mol. Plant Pathol. 8, 529–538
(2007).
13. J. F. González, M. P. Myers, V. Venturi, The inter-kingdom solo OryR regulator of Xan-
thomonas oryzae is important for motility. Mol. Plant Pathol. 14, 211–221 (2013).
14. S. Subramoni, J. F. Gonzalez, A. Johnson, M. Péchy-Tarr, L. Rochat, I. Paulsen, J. E. Loper,
C. Keel, V. Venturi, Bacterial sub family of LuxR re gulators that respond to plant
compounds. Appl. Environ. Microbiol. 77, 4579–4588 (2011).
15. F. Perez-Montano, I. Jimenez-Guerrero, R. Contreras Sanchez-Matamoros, F. J. Lopez-Baena,
F. J. Ollero, M. A. Rodriguez-Carvajal, R. A. Bellogin, M. R. Espuny, Rice and bean
AHL-mimic quorum-sensing signals specifically interfere with the capacity to form bio-
films by plant-associated bacteria. Res. Microbiol. 164, 749–760 (2013).
16. M. Gao, M. Teplitski, J. B. Robinson, W. D. Bauer, Production of substances by Med-
icago truncatula that affect bacterial quorum sensing. Mol. Plant Microbe Interact. 16,
827–834 (2003).
17. M. Teplitski, J. B. Robinson, W. D. Bauer, Plants secrete substances that mimic bacterial
N-acyl homoserine lactone signal activities and affect population density-dependent be-
haviors in associated bacteria. Mol. Plant Microbe Interact. 13, 637–648 (2000).
18. K. Karamanoli, S. E. Lindow, Disruption of N-acyl homoserine lactone-mediated cell
signaling and iron acquisition in epiphytic bacteria by leaf surface compounds. Appl.
Environ. Microbiol. 72, 7678–7686 (2006).
19. L. Zhang, Y. Jia, L. Wang, R. Fang, A proline iminopeptidase gene upregulated in
planta by a LuxR homologue is essential for pathogenicity of Xanthomonas campes-
tris pv. campestris. Mol. Microbiol. 65, 121–136 (2007).
20. V. Plyuta, J. Zaitseva, E. Lobakova, N. Zagoskina, A. Kuznetsov, I. Khmel, Effect of
plant phenolic compounds on biofilm formation by Pseudomonas aeruginosa. APMIS
121, 1073–1081 (2013).
21. A. Annapoorani, V. Umamages waran, R. Parameswari, S. K. Pandian, A. V. Ravi,
Computational discovery of putative quorum sensing inhibitors against LasR and
RhlR receptor pro teins of Pseudomonas aeruginosa. J. Comput. Aided Mol. Des.
26, 1067–1077 (2012).
22. S. Rajamani, W. D. Bauer, J. B. Robinson, J. M. Farrow III, E. C. Pesci, M. Teplitski,
M. Gao, R. T. Sayre, D. A. Phillips, The vitamin riboflavin and its derivative lumichrome
activate the LasR bacterial quorum-sensing receptor. Mol. Plant Microbe Interact. 21,
1184–1192 (2008).
23. C. Attila, A. Ueda, S. L. G. Cirillo, J. D. Cirillo, W. Chen, T. K. Wood, Pseudomonas
aeruginosa PAO1 virulence factors and poplar tree response in the rhizosphere. Microb.
Biotechnol. 1,17–29 (2008).
24. L. G. Rahme, F. M. Ausubel, H. Cao, E. Drenkard, B . C. G oumnerov, G. W. Lau,
S. Mahajan-Miklos, J. Plotnikova, M.-W. Tan, J. Tsongalis, C. L. Walendziewicz,
R. G. Tompkins, Plants and animals share functionally common bacterial virulence
factors. Proc. Natl. Acad. Sci. U.S.A. 97, 8815–8821 (2000).
25. H. Cao, R. L. Baldini, L. G. Rahme, Common mechanisms for pathogens of plants
and animals. Annu. Rev. Phytopathol. 39, 259–284 (2001).
26. M. Schuster, D. J. Sexton, S. P. Diggle, E. P. Greenberg, Acyl-homoserine lactone
quorum sensing: From evolu tion to application. Annu. Rev. Microbiol. 67,43–63
(2013).
27. E. C. Pesci, J. B. J. Milbank, J. P. Pearson, S. McKnight, A. S. Kende, E. P. Greenberg,
B. H. Iglewski, Quinolone signaling in the cell-to-cell communication system of Pseu-
domonas aeruginosa. Proc. Natl. Acad. Sci. U.S.A. 96, 11229–11234 (1999).
28. J. Lee, L. Zhang, The hierarchy quorum sensing network in Pseudomonas aeruginosa.
Protein Cell 6,26–41 (2015).
29. A. Corral-Lugo, A. Daddaoua, A. Ortega, B. Morel, A. I. Díez-Peña, M. Espinosa-Urgel,
T. Krell, Purification of functional recombinant LasR and RhlR of Pseudomonas aeruginosa
from Escherichia coli cultures. Protein Sci. (2015).
30. H. P. Bais, T. S. Walker, H. P. Schweizer, J. M. Vivanco, Root specific elicitation and
antimicrobial activity of rosmarinic acid in hairy root cultures of Ocimu m basilicum.
Plant Physiol. Biochem. 40, 983–995 (2002).
31. T.S.Walker,H.P.Bais,E.Déziel,H.P.Schweizer,L.G.Rahme,R.Fall,J.M.Vivanco,
Pseudomonas aeruginosa-plant root interactions. Pathogenicity, biofilm formation, and
root exudation. Plant Physiol. 134, 320–331 (2004).
32. K.-I. Oinuma, E. P. Greenberg, Acyl-homoserine lactone binding to and stability of the
orphan Pseudomonas aeruginosa quorum-sensing signal receptor QscR. J. Bacteriol.
193, 421–428 (2011).
33. M. J. Bottomley, E. Muraglia, R. Bazzo, A. Carfi, Molecular insights into quorum
sensing in the human pathogen Pseudomonas aeruginosa from the structure of the
virulence regulator LasR bound to its autoinducer. J. Biol. Chem. 282, 13592–13600
(2007).
34. M. Schuster, M. L. Urbanowski, E. P. Greenberg, Promoter specificity in Pseudomonas
aeruginosa quorum sensing revealed by DNA binding of purified LasR. Proc. Natl. Acad.
Sci. U.S.A. 101, 15833–15839 (2004).
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 9
on January 14, 2016http://stke.sciencemag.org/Downloaded from

35. G. Pessi, D. Haas, Transcriptional control of the hydrogen cyanide biosynthetic genes
hcnABC by the anaerobic regulator ANR and the quorum-sen sing regulators LasR
and RhlR in Pseudomonas aeruginosa. J. Bacteriol. 182, 6940–6949 (2000).
36. E. C. Pesci, J. P. Pearson, P. C. Seed, B. H. Iglewski, Regulation of las and rhl quorum
sensing in Pseudomonas aeruginosa. J. Bacteriol. 179, 3127–3132 (1997).
37. J. P. Pearson, E. C. Pesci, B. H. Iglewski, Roles of Pseudomonas aeruginosa las and
rhl quorum-sensing systems in control of elastase and rhamnolipid biosynthesis
genes. J. Bacteriol. 179, 5756–5767 (1997).
38. G. Medina, K. Juárez, B. Valderrama, G. Soberón-Chávez, Mechanism of Pseudo-
monas aeruginosa RhlR transcriptional regulation of the rhlAB promoter. J. Bacteriol.
185, 5976–5983 (2003).
39. Y. Sakuragi, R. Kolter, Quorum-sensing regulation of the biofilm matrix genes (pel)of
Pseudomonas aeruginosa. J. Bacteriol. 189, 5383–5386 (2007).
40. D. G. Davies, M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, E. P. Greenberg,
The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280,
295–298 (1998).
41. C. T. O’Loughlin, L. C. Miller, A. Siryaporn, K. Drescher, M. F. Semmelhack, B. L. Bassler,
A quorum-sensing inhibitor blocks Pseudomonas aeruginosa virulence and biofilm forma-
tion. Proc. Natl. Acad. Sci. U.S.A. 110, 17981–17986 (2013).
42. M. Petersen, Y. Abdullah, J. Benner, D. Eberle, K. Gehlen, S . Hücherig, V. Janiak,
K. H. Kim, M. Sander, C. Weitzel, S. Wolters, Evolution of rosmarinic acid bio-
synthesis. Phytochemistry 70,1663–1679 (2009).
43. M. Shekarchi, H. Hajimehdipoor, S. Saeidnia, A. R. Gohari, M. P. Hamedani, Com-
parative study of rosmarinic acid content in some plants of Labiatae family. Pharmacogn.
Mag. 8,37–41 (2012).
44. V. Swarup, J. Ghosh, S. Ghosh, A. Saxena, A. Basu, Antiviral and anti-inflammatory
effects of rosmarinic acid in an experimental murine model of Japanese encephalitis.
Antimicrob. Agents Chemother. 51, 3367–3370 (2007).
45. C. Sanbongi, H. Takano, N. Osakabe, N. Sasa, M. Natsume, R. Yanagisawa, K.-i. Inoue,
K. Sadakane, T. Ichinose, T. Yoshikawa, Rosmarinic acid in perilla extract inhibits allergic
inflammation induced by mite allergen, in a mouse model. Clin. Exp. Allergy 34, 971–977
(2004).
46. N. Osakabe, A. Yasuda, M. Natsume, T. Yoshikawa, Rosmarinic acid inhibits epidermal
inflammatory responses: Anticarcinogenic effect of Perilla frutescens extract in the murine
two-stage skin model. Carcinogenesis 25, 549–557 (2004).
47. M. Mladenović,S.Matić,S.Stanić,S.Solujić, V. Mihailović,N.Stanković,J.Katanić,
Combining molecular docking and 3-D pharmacophore generation to enclose the in vivo
antigenotoxic activity of naturally occurring aromatic compounds: Myricetin, quercetin,
rutin, and rosmarinic acid. Biochem. Pharmacol. 86, 1376–1396 (2013).
48. M.-A. Kang, S.-Y. Yun, J. Won, Rosmarinic acid inhibits Ca
2+
-dependent pathways of
T-cell antigen receptor-mediated signaling by inhibiting the PLC-g 1 and Itk activity.
Blood 101, 3534–3542 (2003).
49. G.-T. Liu, T.-M. Zhang, B.-e. Wang, Y.-W. Wang, Protective action of seven natural
phenolic compounds against peroxidative damage to biomembranes. Biochem . Pharmacol .
43, 147–152 (1992).
50. T. Iuvone, D. De Filippis, G. Esposito, A. D’Amico, A. A. Izzo, The spice sage and its
active ingredient rosmarinic acid protect PC12 cells from amyloid-b peptide-induced
neurotoxicity. J. Pharmacol. Exp. Ther. 317, 1143–1149 (2006).
51. J. Lee, E. Jung, J. Koh, Y. S. Kim, D. Park, Effect of rosmarinic acid on atopic der-
matitis. J. Dermatol. 35, 768–771 (2008).
52. J. Psotova, A. Svobodova, H. Kolarova, D. Walterova, Photoprotective properties of
Prunella vulgaris and rosmarinic acid on human keratinocytes. J. Photochem. Photobiol.
B 84, 167–174 (2006).
53. B. R. Borlee, G. D. Geske, H. E. Blackwell, J. Handelsman, Identification of synthetic
inducers and inhibitors of the quorum-sensing regulator LasR in Pseudomonas aeruginosa
by high-throughput screening. Appl. Environ. Microbiol. 76, 8255–8258 (2010).
54. M. E. Mattmann, G. D. Geske, G. A. Worzalla, J. R. Chandler, K. J. Sappington,
E. P. Greenberg, H. E. Blackwell, Synthetic ligands that activate and inhibit a quorum-
sensing regulator in Pseudomonas aeruginosa. Bioorg. Med. Chem. Lett. 18, 3072–3075
(2008).
55. G. D. Geske, R. J. Wezeman, A. P. Siegel, H. E. Blackwell, Small molecule inhibitors
of bacterial quorum sensing and biofilm formation. J. Am. Chem. Soc. 127, 12762–12763
(2005).
56. G. D. Geske, J. C. O’Neill, D. M. Miller, M. E. Mattmann, H. E. Blackwell, Modulation of
bacterial quorum sensing with synthetic ligands: Systematic evaluation of N-acylated
homoserine lactones in multiple species and new insights into their mechanisms of
action. J. Am. Chem. Soc. 129, 13613–13625 (2007).
57. N. Rabin, A. Delago, B. Inbal, P. Krief, M. M. Meijler, Tailor-made LasR agonists modula te
quorum sensing in Pseudomonas aeruginosa. Org. Biomol. Chem. 11, 7155–7163
(2013).
58. S. Skovstrup, S. T. Le Quement, T. Hansen, T. H. Jakobsen, M. Harmsen, T. Tolker-Nielsen,
T. E. Nielsen, M. Givskov, O. Taboureau, Identification of LasR ligands through a virtual
screening approach. ChemMedChem 8,157–163 (2013).
59. J. J. Irwin, T. Sterling, M. M. Mysinger, E. S. Bolstad, R. G. Coleman, ZINC: A free tool
to discover chemistry for biology. J. Chem. Inf. Model. 52, 1757–1768 (2012).
60. R.A.Friesner,J.L.Banks,R.B.Murphy,T.A.Halgren,J.J.Klicic,D.T.Mainz,
M. P. Repasky, E. H. Knoll, M. Shelley, J. K. Perry, D. E. Shaw, P. Francis, P. S. Shenkin,
Glide: A new approach for rapid, accurate docking and scoring. 1. Method and assess-
ment of docking accuracy. J. Med. Chem. 47, 1739–1749 (2004).
61. K. Arno ld, L. Bordoli, J. Kopp, T. S chwede, The SWISS-MO DEL workspace: A
web-based environment for protein structure homology modelling. Bioinformatics
22, 195–201 (2006).
62. M. D. Abràmoff, P. J. Magalhães, S. J. Ram, Image processing with ImageJ. Biophotonics
Int. 11,36–42 (2004).
63. L. Steindler, G. Devescovi, S. Subramoni, V. Venturi, A versatile plasmid biosensor
useful to identify quorum sensing LuxR-family orphans in bacterial strains. J. Microbiol.
Methods 73, 273–275 (2008).
64. M. Espinosa-Urgel, J.-L. Ramos, Cell density-dependent gene contributes to efficient
seed colonization by Pseudomonas putida KT2440. Appl. Environ. Microbiol. 70,
5190–5198 (2004).
65. T.-F. Mah, E stablishing the minimal bacteri cidal concentration of an antimicrobial
agent for planktonic cells (MBC-P) and biofilm cells (MBC-B). J. Vis. Ex p. e50854
(2014).
66. M.-A. Abril, C. Michan, K. N. Timmis, J. L. Ramos, Regulator and enzyme specificities
of the TOL plasmid-encoded upper pathway for degradation of aromatic hydrocar-
bons and expansio n of the substrate range of the pathway. J. Bacteriol. 171,
6782–6790 (1989).
67. M. Martínez-Gil, D. Romero, R. Kolter, M. Espinosa-Urgel, Calcium causes multimer-
ization of the large adhesin LapF and modulates biofilm formation by Pseudomonas
putida. J. Bacteriol. 194, 6782–6789 (2012).
68. T. Ishida, T. Ikeda, N. Takiguchi, A. Kuroda, H. Ohtake, J. Kato, Inhibition of quorum
sensing in Pseudomonas aerugin osa by N-acyl cyclopentylamides. Appl. Environ.
Microbiol. 73,3183–3188 (2007).
69. M. B. Kozin, D. I. Svergun, Automated matching of high- and low-resolution structural
models. J. Appl. Cryst. 34,33–41 (2001).
70. C. K. Stover, X. Q. Pham, A. L. Erwin, S. D. Mizoguchi, P. Warrener, M. J. Hickey,
F. S. L. Brinkman, W. O. Hufnagle, D. J. Kowalik, M. Lagrou, R. L. Garber, L. Goltry,
E. Tolentino, S. Westbrock-Wadman, Y. Yuan, L. L. Brody, S. N. Coulter, K. R. Folger,
A. Kas, K. Larbig, R. Lim, K. Smith, D. Spencer, G. K.-S. Wong, Z. Wu, I. T. Paulsen,
J. Reizer, M. H. Saier, R. E. W. Hancock, S. Lory, M. V. Olson, Complete genome
sequence of Pseudomo nas aeruginosa PAO1, an opportunistic pathogen. Nature
406, 959–964 (2000).
71. F. W. Studier, B. A. Moffatt, Use of bacteriophage T7 RNA polymerase to direct se-
lective high-level expression of cloned genes. J. Mol. Biol. 189, 113–130 (1986).
Acknowledgments: We th ank J. Kato for provid ing plasmid s pQF50, pb01, and pb02 ;
S. Heeb for providing plasmid pME2823; and M. Cámara for plasmid pMULTIAHLPROM
and P. aeruginosa PAO1 lasI
−
/lasR
−
. Funding: This work was supported by the FEDER
funds and the Fondo Social Europeo through grants from the Junta de Andalucía (grants
P09-RNM-4509 and CVI-7335 to T.K. and CVI-7391 to M.E.-U.) and the Spanish Ministry for
Economy and Competitiveness (grants BIO2010-16937 and BIO2013-42297 to T.K. and
grant BFU2010-17946 to M.E.-U.). Author contributions: A.C.-L., A.D., and A.O. conducted
research and analyzed the data, and M.E.-U. and T.K. designed the experiments, interpreted
the data, and wrote the manuscript. Competing interests: The authors declare that they
have no competing interests.
Submitted 2 February 2015
Accepted 11 December 2015
Final Publication 5 January 2016
10.1126/scisignal.aaa8271
Citation: A. Corral-Lugo, A. Daddaoua, A. Ortega, M. Espinosa-Urgel, T. Krell,
Rosmarinic acid is a homoserine lactone mimic produced by plants that activates a
bacterial quorum-sensing regulator. Sci. Signal. 9, ra1 (2016).
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 5 January 2016 Vol 9 Issue 409 ra1 10
on January 14, 2016http://stke.sciencemag.org/Downloaded from

(409), ra1. [doi: 10.1126/scisignal.aaa8271]9Science Signaling
Espinosa-Urgel and Tino Krell (January 5, 2016)
Andrés Corral-Lugo, Abdelali Daddaoua, Alvaro Ortega, Manuel
plants that activates a bacterial quorum-sensing regulator
Rosmarinic acid is a homoserine lactone mimic produced by
This information is current as of January 14, 2016.
The following resources related to this article are available online at http://stke.sciencemag.org.
Article Tools
http://stke.sciencemag.org/content/9/409/ra1
article tools:
Visit the online version of this article to access the personalization and
References
http://stke.sciencemag.org/content/9/409/ra1#BIBL
This article cites 69 articles, 30 of which you can access for free at:
Permissions
http://www.sciencemag.org/about/permissions.dtl
Obtain information about reproducing this article:
reserved.
DC 20005. Copyright 2016 by the American Association for the Advancement of Science; all rights
American Association for the Advancement of Science, 1200 New York Avenue, NW, Washington,
(ISSN 1937-9145) is published weekly, except the last December, by theScience Signaling
on January 14, 2016http://stke.sciencemag.org/Downloaded from