Fluorescent labeling agents for quorum-sensing receptors (FLAQS) in live cells

Article abstract of DOI:10.1002/chem.201301387

Lighten up: A selective fluorescent-labeling agent for quorum sensing (FLAQS) can be used for the visualization of the communication pathway of bacteria in live cells (see figure). This represents a new, operationally simple, fast, and inexpensive tool for the imaging of quorum-sensing receptors by using fluorescently labeled signaling-molecule analogues. Copyright

Full text of DOI:10.1002/chem.201301387

COMMUNICATION
DOI: 10.1002/chem.201301387
Fluorescent Labeling Agents for Quorum-Sensing Receptors (FLAQS) in
Live Cells
Josꢀ Gomes,[a] Natalie Huber,[a] Alexander Grunau,[b] Leo Eberl,*[b] and
Karl Gademann*[a]
Live-cell fluorescent labeling of bacterial cell-to-cell com-
munication pathways has emerged as a challenging problem.
The use of labeled ligands with a fluorescent tag represents
the simplest method of choice. However, such approaches
are often accompanied by insufficient binding affinity to-
wards the receptor and competition problems with the natu-
ral agonists. Herein, the first example of a fluorescently la-
beled bacterial quorum-sensing signaling compound is pre-
sented. The synthetic labeling agent mimics the activity of
the natural agonist throughout a large concentration range,
and excellent labeling of bacterial quorum-sensing receptors
in live cells was achieved. This operationally simple, fast,
and inexpensive method was successfully applied to the se-
lective labeling of the Burkholderia cenocepacia quorum-
sensing receptor CepR. Furthermore, selective labeling was
achieved in mixed bacterial cultures, demonstrating the po-
tential of this approach as a very powerful tool to visualize
quorum sensing in bacteria in their natural habitat.
dinate expression of virulence factors in a cell-density-de-
pendent manner, which is referred to as quorum sensing
(QS).[8–10] Bacteria utilize QS to synchronize their behavior
to regulate functions that benefit the entire population, such
as biofilm development, synthesis of virulence factors, and
the production of antibiotics and extracellular hydrolytic en-
zymes.[11]
QS in Gram-negative bacteria is mediated by the ex-
change of diffusible small-molecule signals termed autoin-
ducers (AI).[12] The most commonly used signaling mole-
cules are N-acyl-l-homoserine lactones (AHLs).[13] Evidence
has emerged that AHLs are also recognized by eukaryotes
and induce specific responses often affecting the immune
system of the organism.[14,15] The QS system of B. cenocepa-
cia is comprised of the LuxR-family AHL receptor CepR
and the LuxI-type CepI synthase, which directs the synthesis
of N-octanoyl homoserine lactone (C8-AHL) and minor
amounts of N-hexanoyl homoserine lactone (C6-AHL) sig-
naling molecules.[16,17] The CepIR system regulates multiple
functions, including virulence, biofilm formation, swarming
motility, and the production of proteases, siderophores, and
antifungal compounds.[18]
B. cenocepacia, a member of a bacterial group collectively
referred to as the B. cepacia complex (Bcc), has emerged as
an important pathogen for patients suffering from cystic fib-
rosis (CF) or immunocompromised persons.[1,2]These patho-
gens are known to form mixed biofilms with Pseudo
ACHTUNGTRENNUNG
monas
Several Burkholderia identification methods have been
developed including fluorescence in situ hybridization
(FISH)[19] and rRNA gene-based PCR assays.[20] However,
in spite of the importance of QS for virulence and biofilm
development, knowledge of the temporal and spatial pro-
duction of AHL signaling molecules within biofilms or
within the infection host is scarce. This is largely due to the
lack of techniques to visualize AHL-mediated communica-
tion at the single-cell level. Although GFP-based biosensors
have been used for this purpose,[21] their use is restricted to
genetically engineered experimental model settings. An al-
ternative approach was recently presented by Meijler and
co-workers, who reported on an aniline-catalyzed two-step
aeruginosa in the lungs of CF patients[3]and are often associ-
ated with reduced survival and the risk of developing a fatal
pneumonia known as cepacia syndrome.[4–6] B. cenocepacia is
considered to be the most problematic species, particularly
because some strains have the ability to spread epidemically
between CF individuals.[7] Many pathogens have developed
mechanisms to overcome the host defenses and remain in-
visible until a critical population density is reached. Typical-
ly, these bacteria use a cell-to-cell signaling system to coor-
[a] J. Gomes,+ N. Huber,+ Prof. Dr. K. Gademann
Department of Chemistry, University of Basel
National Centre of Competence in Research “Chemical Biology”
St. Johanns-Ring 19, 4056 Basel (Switzerland)
E-mail: karl.gademann@unibas.ch
labeling strategy for the visualization of the Pseudo
ACHTUNGTRENNUNGmonas
AHCTUNGTREGaNNNU eruginosa LasR QS receptor in live cells by using a selec-
tive bio-orthogonal ligation.[22] This method is based on the
selective covalent binding of the isothiocyanate functional-
ized AHL to LasR and good labeling quality was achieved
by using the double-mutant strain P. aeruginosa JP2 (lasI/
rhlI deleted). However, this approach is restricted to the ab-
sence of native AHLs and cannot be applied for the investi-
gation of bacteria–host communication pathways. A more
general approach by using a labeled ligand with a fluores-
[b] A. Grunau, Prof. Dr. L. Eberl
Department of Microbiology, University of Zurich
Institute of Plant Biology
Winterthurerstrasse 190, 8057 Zurich (Switzerland)
E-mail: leberl@botinst.uzh.ch
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201301387.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

cent tag represents an operationally simpler labeling
method, but this often results in decreased affinity of the
modified ligand for its target protein. Furthermore, competi-
tion with native AHLs constitutes a challenging problem.
The design of our target molecules was guided by recent
observations that the termini of certain AHL compounds
can be modified while retaining their agonistic activity. In
fact, terminal modification[23] of AHL fragments with cate-
cholate groups allowed their immobilization on surfaces and
concomitant induction of QS in the GFP-based biosensor
P. putida F117 (pAS-C8),[21] in which cepR is recombinantly
overexpressed. We hypothesized if we could take advantage
of this molecular plasticity and introduce fluorescent labels
at the terminal position of AHL derivatives with the goal of
selectively labeling quorum sensing in bacterial strains.
Herein, we report the development of a fluorescent-labeling
agent for QS receptors (FLAQS) and demonstrate selective
tagging of CepR in B. cenocepacia. This highly selective and
operationally simple method for live-cell labeling of QS re-
ceptors by using a fluorescent AHL-analogue represents a
new in situ imaging method for AHL-mediated communica-
tion pathways.
Scheme 1. Reaction conditions: a) Ref. [24]; b) N,N’-disuccinimidyl car-
bonate, pyridine, MeCN, 6 h, RT; c) l-homoserine lactone (HSL), NEt3,
MeCN, 7 h, RT; d) CH2Cl2, trifluoroacetic acid (TFA), 5 h, RT; e) MeCN,
15 h, RT, 71% (over four steps).
A first challenge consisted in finding a suitable fluoro-
phore that could be used for visualization with the GFP-
based P. putida F117 (pAS-C8) biosensor to monitor the cor-
relation between activity and labeling. The advantage of this
reporter strain, in which the AHL synthase gene ppuI is
knocked out, is the absence of AHL production, preventing
competition with added synthetic AHLs. In addition, a
better labeling compared with the wild type (WT) can be
expected in this sensor, because CepR is overexpressed. We
synthesized various FLAQS containing different rho-
ACHTUNGTRENNUNGdamine B dyes. Rhodamine B was selected due to its high
luminescence quantum yield and red fluorescence, which
can be clearly separated from the green fluorescence of the
biosensor. Most fluorescent hybrids proved to be difficult to
purify, and decomposition was observed. The only com-
pound that could be purified by preparative reverse-phase
HPLC without degradation was hybrid 1, which could be
synthesized in five steps from commercially available rho-
Figure 1. Fluorescence spectrum of FLAQS 1; a=absorption; b=emis-
sion.
A
Having achieved the desired fluorescence properties with
FLAQS 1, we proceeded to compare the activities of the
synthetic and the natural agonists with respect to binding to
the receptor proteins. The GFP fluorescence of the P. putida
F117 (pAS-C8) biosensor induced by both compounds over
a broad range of concentrations (58 pm to 488 mm) were
compared. The results with the natural agonist C8-AHL are
in good agreement with literature (blue in Figure 2) with a
maximum activity at a concentration of 15 mm.[21] A propor-
tional decrease of activity and concentration was observed
at lower concentrations. In contrast, higher concentrations
resulted in inhibition of GFP synthesis. FLAQS 1 shows a
very similar profile compared to C8-AHL (red in Figure 2)
with an activity maximum at a concentration of 4 mm. As a
negative control, the P. putida F117 (pAS-C8) sensor was
also incubated with only the fluorophore FLU 1 under iden-
tical conditions, and no activity could be observed at any
desired fluorescently labeled AHL analogue was obtained
with an overall yield of 71% from the known fluorescent
dye Flu 1.[24] The rhodamine moiety of this hybrid cannot
build the nonfluorescent spirolactone form due to the pres-
ence of the non-nucleophilic secondary amide making it pH
independent and water soluble. The fluorescence properties
of FLAQS 1 were evaluated in aqueous medium by record-
ing the fluorescence spectrum (Figure 1). Both bands have
almost no overlap leading to improved fluorescence proper-
ties of hybrid 1 relative to rhodamine B. The absorption
maximum is located at l=560 nm, and only little absorption
occurs at 510 nm, which is the maximum GFP emission
wavelength. This important attribute should preclude fluo-
rescence quenching, which otherwise could negatively influ-
ence the activity assays. The maximum emission of FLAQS
1 is located in the red spectral area at l=600 nm.
&
2
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Quorum-Sensing Receptors (FLAQS) in Live Cells
COMMUNICATION
cent detection. To label B. cenocepacia H111 cultures were
incubated in the presence of 244 mm FLAQS 1 for 3 h, sub-
sequently washed, and examined by confocal microscopy.
We were pleased to find that the entire bacterial population
could be labeled with the red FLAQS 1 (Figure 3a). The
red fluorescence was found to be uniformly distributed in
the cell, suggesting that the CepR receptor is a soluble cyto-
plasmic protein. B. cenocepacia H111 cells were incubated
with different concentration of FLAQS 1 to determine the
required minimal concentration to achieve good labeling. A
good yield of red-fluorescence labeling was achieved with a
concentration of 31 mm (Figure 3b), whereas only minor red
fluorescence was seen at 8 mm, and almost no fluorescence
was detected at a concentration of 2 mm (see the Supporting
Information).
Various control experiments were performed to exclude
random binding of FLAQS 1 to the bacterial population.
The double knockout mutant B. cenocepacia H111DcepRI,
in which both cepR and cepI have been inactivated, showed
no fluorescence labeling after incubation under the same
conditions used for the WT (Figure 3c). The presence of the
receptor CepR appears to be necessary for labeling. We
were also pleased that in the absence of the QS receptor, no
unspecific binding was detected. In contrast, enhanced red-
fluorescence labeling could be observed with a strain over-
expressing cepR as compared to the WT (see the Supporting
Information). Labeling of E. coli XL1-Blue, which does not
encode an AHL-dependent QS system, also showed no fluo-
rescence under the same incubation conditions (Figure 3d).
As an additional control experiment, B. cenocepacia H111
was incubated with the fluorophore Flu 1 alone to confirm
that the bioactive moiety is indeed responsible for the selec-
tive binding towards CepR. As was expected, no labeling of
the cells could be detected (Figure 3e). This clearly indicates
Figure 2. Activity tests with the P. putida F117 (pAS-C8) biosensor com-
paring FLAQS 1, Flu 1, and the natural C8-AHL (data is reported as
meanÆSEM, N=3).
concentration (purple in Figure 2). These results demon-
strate that FLAQS 1 shares the activity of C8-AHL over the
desired concentration range and possesses the desired fluo-
rescent properties for our bioassays.
As a next challenge, we addressed the labeling of the
wild-type strain B. cenocepacia H111. This strain was isolat-
ed from a CF patient, and its cepIR QS system was shown
to control biofilm formation and swarming motility.[25] The
use of this strain is especially advantageous due to its lack
of the CciR/I QS system that is present in some B. cenocepa-
cia strains.[26] The labeling of the wild type is more problem-
atic than the reporter strain, because the receptor is not
overexpressed, and competition between the natural agonist
and the synthetic FLAQS can create challenges for fluores-
Figure 3. Imaging of different strains incubated with fluorescent probes. B. cenocepacia H111 incubated in the presence of A) 244 mm or B) 31 mm
FLAQS 1. B. cenocepacia H111DcepRI incubated in the presence of C) 244 mm FLAQS 1. E. coli XL1-Blue incubated in the presence of D) 244 mm
FLAQS 1. B. cenocepacia H111 incubated in the presence of E) 244 mm Flu 1. Phase-contrast images are shown at the top, red-fluorescence images at the
bottom. Scale bars are 7 mm.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

L. Eberl, K. Gademann et al.
that the rhodamine derivative by itself is not binding to the
bacterial cells.
tively labeled by using FLAQS with varying alkyl-chain
lengths. We are currently investigating P. aeruginosa selec-
tive FLAQS, which, in combination with FLAQS 1, could
lead to a very fast, simple, and inexpensive tool to analyze
sputa of CF patients. Furthermore, as FLAQS possess simi-
lar properties as the natural agonists, they could be used as
a new and potentially useful method for investigations of
pathogenic and symbiotic interactions between bacteria and
their eukaryotic hosts.
Our results demonstrate that easily synthesized FLAQS
can be successfully used for the operationally simple and se-
lective labeling of QS receptors in B. cenocepacia H111, a
pathogen isolated from a CF patient. This represents a new,
operationally simple, fast, and inexpensive tool for the imag-
ing of QS receptors by using fluorescently labeled AHL an-
alogues. Furthermore, application of FLAQS 1 suggested
the presence of CepR as a soluble cytoplasmic protein. The
selectivity of FLAQS 1 has been applied to specific QS la-
beling in a mixed bacterial culture. We think that FLAQS
could be utilized for the fast analysis of QS in various envi-
ronmental and clinical samples, such as, for example, sputa
from CF patients and samples of plant roots to deliver a
better understanding of bacteria–host communication path-
ways.
We next wanted to investigate if FLAQS 1 actually binds
to the same target site as the natural AHL molecule. To
evaluate this hypothesis, competition assays between the
synthetic and the natural ligands were performed. B. cenoce-
pacia H111 was incubated with the same concentration of
FLAQS (244 mm) and varying concentrations of pure C8-
AHL (1 and 100 mm). Microscopic inspection of the cultures
revealed a significant decrease in labeling when the concen-
tration of C8-AHL was increased (see the Supporting Infor-
mation). These results are in agreement with the hypothesis
that both compounds compete for the same binding pocket
of the CepR receptor in B. cenocepacia H111.
Finally, we wanted to test if the cepR overexpression
strain H111 (cepR+) and the GFP-tagged cepR knockout
mutant H111DcepR:GFP can be differentiated by our CepR
labeling approach. The mixed strains were incubated with
FLAQS 1 for 3 h and visualized by fluorescence microscopy.
The results clearly demonstrate that the GFP-tagged knock-
out mutant could not be labeled with FLAQS 1, whereas
the strain-overexpressing CepR was clearly labeled
(Figure 4). Based on these results, we expect that FLAQS
Acknowledgements
We thank the Swiss National Foundation (200021-144028, PE0022-
117136, 31003 A-122013, and 143773) for financial support. Part of this
work was supported by the NCCR “Chemical Biology”.
Keywords: Burkholderia cenocepacia
· cell recognition ·
fluorescent probes · lactones · quorum sensing
[1] P. Drevinek, E. Mahenthiralingam, Clin. Microbiol. Infect. 2010, 16,
821–830.
[2] E. Mahenthiralingam, A. Baldwin, C. G. Dowson, J. Appl. Micro-
biol. 2008, 104, 1539–1551.
[3] L. Saiman, G. Cacalano, A. Prince, Infect. Immun. 1990, 58, 2578–
2584.
[4] E. Mahenthiralingam, T. A. Urban, J. B. Goldberg, Nat. Rev. Micro-
biol. 2005, 3, 144–156.
[5] A. M. Jones, Thorax 2004, 59, 948–951.
[6] O. C. Tablan, W. J. Martone, C. F. Doershuk, R. C. Stern, M. J. Tho-
massen, J. D. Klinger, J. W. White, C. L. A, W. R. Jarvis, Chest 1987,
91, 527–532.
[7] T. Coenye, P. Vandamme, J. R. W. Govan, J. J. LiPuma, J. Clin. Mi-
crobiol. 2001, 39, 3427–3436.
[8] J. Engebrecht, K. Nealson, M. Silverman, Cell 1983, 32, 773–781.
[9] M. Boyer, F. Wisniewski-Dyꢁ, FEMS Microbiol. Ecol. 2009, 70, 1–
19.
[10] B. L. Bassler, R. Losick, Cell 2006, 125, 237–246.
[11] L. Eberl, K. Riedel, Proteomics 2011, 11, 3070–3085.
[12] G. D. Geske, J. C. OꢂNeill, H. E. Blackwell, Chem. Soc. Rev. 2008,
37, 1432.
Figure 4. Imaging of mixed bacterial populations with FLAQS 1: A)
Phase contrast, B) red fluorescence, C) GFP fluorescence, and D) overlay
of B) and C) of a bacterial mixture of the CepR-overexpressing strain
B. cenocepacia H111 (cepR+) and the GFP-tagged QS mutant B. cenoce-
pacia H111DcepR:GFP with FLAQS 1 (244 mm). Scale bars are 7 mm.
will be able to label bacteria based on their unique AHL
signaling system. Therefore, FLAQS might also be applied
for the study of bacterial interspecies communication path-
ways based on a given AHL chemotype. Although our work
focused on agonists of C8-AHL, we are confident that other
QS receptors using different AHL molecules may be selec-
[13] A. M. Stevens, Y. Queneau, L. Soulꢃre, S. V. Bodman, A. Doutheau,
Chem. Rev. 2011, 111, 4–27.
&
4
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Quorum-Sensing Receptors (FLAQS) in Live Cells
COMMUNICATION
[14] V. V. Kravchenko, G. F. Kaufmann, J. C. Mathison, D. A. Scott,
A. Z. Katz, D. C. Grauer, M. Lehmann, M. M. Meijler, K. D. Janda,
R. J. Ulevitch, Science 2008, 321, 259–263.
[21] K. Riedel, M. Hentzer, O. Geisenberger, B. Huber, A. Steidle, H.
Wu, N. Høiby, M. Givskov, S. Molin, L. Eberl, Microbiology 2001,
147, 3249–3262.
[15] M. Cooley, S. R. Chhabra, P. Williams, Chem. Biol. 2008, 15, 1141–
1147.
[22] J. Rayo, N. Amara, P. Krief, M. M. Meijler, J. Am. Chem. Soc. 2011,
133, 7469–7475.
[16] S. Lewenza, B. Conway, E. P. Greenberg, P. A. Sokol, J. Bacteriol.
1999, 181, 748–756.
[23] J. Gomes, A. Grunau, A. K. Lawrence, L. Eberl, K. Gademann,
Chem. Commun. 2013, 49, 155–157.
[17] S. Inhꢄlsen, C. Aguilar, N. Schmid, A. Suppiger, K. Riedel, L. Eberl,
Microbiologyopen 2012, 1, 225–242.
[18] P. A. Sokol, R. J. Malott, K. Riedel, L. Eberl, Future Microbiol.
2007, 2, 555–563.
[19] A. R. Brown, J. R. W. Govan, J. Clin. Microbiol. 2007, 45, 1920–
1926.
[24] P. Moody, M. E. B. Smith, C. P. Ryan, V. Chudasama, J. R. Baker, J.
Molloy, S. Caddick, ChemBioChem 2012, 13, 39–41.
[25] B. Huber, K. Riedel, M. Hentzer, A. Heydorn, A. Gotschlich, M.
Givskov, S. Molin, L. Eberl, Microbiology 2001, 147, 2517–2528.
[26] A. Baldwin, P. A. Sokol, J. Parkhill, E. Mahenthiralingam, Infect.
Immun. 2004, 72, 1537–1547.
[20] K. Stoecker, C. Dorninger, H. Daims, M. Wagner, Appl. Environ.
Microbiol. 2010, 76, 922–926.
Received: April 12, 2013
Published online: && &&, 0000
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

L. Eberl, K. Gademann et al.
Lighten up: A selective fluorescent-
labeling agent for quorum sensing
(FLAQS) can be used for the visuali-
zation of the communication pathway
of bacteria in live cells (see figure).
This represents a new, operationally
simple, fast, and inexpensive tool for
the imaging of quorum-sensing recep-
tors by using fluorescently labeled sig-
naling-molecule analogues.
Cell Recognition
J. Gomes, N. Huber, A. Grunau,
L. Eberl,* K. Gademann* . . &&&& — &&&&
Fluorescent Labeling Agents for
Quorum-Sensing Receptors (FLAQS)
in Live Cells
&
6
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!