Live cell labeling of native intracellular bacterial receptors using aniline-catalyzed oxime ligation

Article abstract of DOI:10.1021/ja200455d

Live cell fluorescent labeling of proteins has become a seminal tool in biology and has led to hallmark discoveries in diverse research areas such as protein trafficking, cell-to-cell interactions, and intracellular network dynamics. One of the main chall

Full text of DOI:10.1021/ja200455d

ARTICLE
pubs.acs.org/JACS
Live Cell Labeling of Native Intracellular Bacterial Receptors Using
Aniline-Catalyzed Oxime Ligation
Josep Rayo, Neri Amara, Pnina Krief, and Michael M. Meijler*
Department of Chemistry and National Institute for Biotechnology in the Negev, Ben-Gurion University of the Negev,
Be’er-Sheva 84105, Israel
S Supporting Information
b
ABSTRACT: Live cell fluorescent labeling of proteins has
become a seminal tool in biology and has led to hallmark
discoveries in diverse research areas such as protein trafficking,
cell-to-cell interactions, and intracellular network dynamics.
One of the main challenges, however, remains the ability to
label intracellular proteins using fluorescent ligands with high
specificity, all the while retaining viability of the targeted cells.
Here, we present the first example of live cell labeling and
imaging of an intracellular bacterial receptor involved in cell-to-
cell communication (i.e., quorum sensing), using a novel two-
step approach involving covalent attachment of a reactive mimic
of the primary endogenous Pseudomonas aeruginosa quorum-
sensing signal to its receptor, LasR, followed by aniline-catalyzed oxime formation between the modified receptor and a fluorescent
BODIPY derivative. Our results indicate that LasR is not distributed evenly throughout the cytoplasmic membrane but is instead
concentrated at the poles of the P. aeruginosa cell.
’ INTRODUCTION
It is known that small changes in the structure of the LasR
cognate ligand, 3-oxo-C12-HSL, may lead to a significant de-
crease in ligand affinity for its receptor. However, by covalently
labeling LasR with a reactive 3-oxo-C12-HSL analogue contain-
ing a keto group, we enabled the exploration of bio-orthogonal
ligation reactions on the protein.
Cell-to-cell communication is employed by single-cell organ-
isms to sense their population density and coordinate gene
expression in a manner that benefits their ability to compete
and adapt to changing environments. This form of interbacterial
chemical communication is known as “quorum sensing” (QS).1,2
Examples of QS-controlled group behavior include the formation
of biofilms, the expression of virulence factors, antibiotic produc-
tion and bioluminescence. Each of these processes is beneficial to
a bacterial population only when carried out in a synchronized
manner, at high cell density. While significant progress has been
made in understanding the mechanisms behind QS in recent
years, many important molecular details remain to be elucidated.
One such point in question addresses the regulation of QS
in Pseudomonas aeruginosa, an organism often resistant to com-
mon antibiotic agents and commonly associated with severe
hospital-acquired infections, especially in immuno-compromised
individuals.3 The primary QS signal in this bacterium is 3-oxo-
dodecanoyl homoserine lactone (3-oxo-C12-HSL, 1, Figure 1a),
which binds to the transcriptional activator, LasR, once thresh-
olds of cell density and concentration have been reached. Correct
folding of LasR, which is bound to the cytoplasmic membrane,
has been suggested to occur only upon ligand binding, enabling
the formation of a homodimer that is able to bind to its target
DNA promoter, leading to gene expression.4 It is less clear,
however, if LasR remains associated with the membrane or
whether it is evenly distributed throughout the cell.5
Only limited examples for the specific intracellular labeling of
proteins with fluorescent ligands have been reported.6,7 The
reason for this is that the use of a conventional approach, in which
a ligand is labeled with a fluorescent tag, often results in a
significant decrease in affinity of the modified ligand for its target
protein, particularly when the ligand is small.8 In the case of larger
ligands, such as peptides or proteins, the main obstacle for
fluorescent labeling remains penetration of modified ligands into
the cell such that membrane structure and polarity are not
compromised. Recent progress in the labeling of proteins in live
cells by fluorescent probes has been achieved through the
development of powerful bio-orthogonal chemistry, such as
Cu(I)-catalyzed azideÀalkyne cycloaddition (CuAAC) assisted
by a tris(triazolylmethyl)amine-based ligand,9 copper-free reac-
tions between azides and strained cyclo-octynes,10 Staudinger
ligation between azides and phosphines,11 and thiolÀene
reactions.12
One of the few reactions considered to be bio-orthogonal and
involving carbonyl moieties is the formation of oximes/hydrazines
Received: January 17, 2011
Published: April 22, 2011
r
2011 American Chemical Society
7469
dx.doi.org/10.1021/ja200455d J. Am. Chem. Soc. 2011, 133, 7469–7475
|

Journal of the American Chemical Society
ARTICLE
Figure 1. QS receptor labeling using aniline-catalyzed oxime chemistry. (a) Condensation reaction between 3-oxo-C12-HSL 1 and methoxyamine. (b)
Selective in vivo and in vitro labeling of ITC-12 covalently bound to a quorum sensing receptor, LasR-LBD, by aminooxy-BODIPY. (c) Deconvoluted
ESI mass spectra of the in vitro reaction (pH 5, 1 mM aniline, 100 μM aminooxy-BODIPY, and 1 μM LasR-ITC-12) at t = 0 h (top) and 12 h (bottom).
The 22 430 Da product corresponds to unreacted LasR-LBD.
upon reaction with an appropriate amine. The reaction of amino-
oxy compounds or hydrazines with carbonyl moieties usually
takes several days for completion at neutral pH, while shorter
reaction times can be achieved by significantly lowering the pH
(pH < 4.5). Such considerations render the reaction not useful
for in vivo studies. Elegant studies by Dawson and co-workers,
however, led to the discovery that aniline catalyzes the reaction
between aldehydes and oxyamines at neutral pH, resulting in
swift oxime formation with good reaction rates.13À15 Accord-
ingly, we have adopted this strategy to label LasR in two steps in
live P. aeruginosa cells, after covalent modification with a reactive
3-oxo-C12-HSL analogue, isothiocyanate ITC-12 (Figure 1b),
followed by reaction with a fluorescent aminooxy-tag (BODIPY-
ONH2). This novel and highly specific live cell dual tagging
approach has enabled the first fluorescent labeling of a native
intracellular bacterial receptor.
reaction product, 3-methoxyimino-C12-HSL (2), was isolated
and characterized by electrospray ionization mass spectrometry
(ESI-MS) and NMR spectroscopy (SI, Figure S1). Full conver-
sion was achieved within 10 h at room temperature, which is
faster than a similar reaction reported by Fleissner et al.16 but
slower than reported reactions with aldehydes.13,15 At neutral
pH, a moderate conversion (60%, 8 h) was obtained using a 100-
fold excess of the aminooxy reagent (SI, Table S1). Once the
optimal conditions for this model reaction were determined, we
next assessed whether the aminooxy reagent could also specifi-
cally react with the keto moiety of a 3-oxo-C12-HSL analogue,
ITC-12,17 covalently bound to the ligand-binding domain
(LBD) of a bacterial quorum-sensing regulator, LasR. We
previously established that this covalent binder reacts specifically
with Cys79 in the LasR-LBD binding pocket (Figure 1b).17
Indeed, upon applying the procedure used in the reaction
between methoxyamine and 3-oxo-C12-HSL, we were able to
observe efficient labeling of LasR-LBD-ITC-12 with BODIPY-
ONH2 (SI, Figure S2). Briefly, the reaction performed at pH 5 in
the presence of 100 μM of BODIPY-ONH2 and 1 mM aniline
yielded a single product with a mass of 23 179 Da after 12 h, as
revealed by ESI-MS. This value corresponds to the expected
mass of the conjugated product (Figure 1c).
’ RESULTS AND DISCUSSION
In Vitro Ligation. To explore the suitability of this oxime
formation-based methodology for in vivo imaging in P. aerugi-
nosa, we began our study by characterizing the product resulting
from a reaction between a model aminooxy reagent, methoxy-
amine, and the P. aeruginosa cognate ligand, 3-oxo-C12-HSL
(1). This reaction proceeds smoothly under acidic conditions
(pH 5, Supporting Information (SI), Table S1). The single
Next, we set out to examine whether this methodology could
be applied to label the ITC-12-LasR complex in vitro, in the
7470
dx.doi.org/10.1021/ja200455d |J. Am. Chem. Soc. 2011, 133, 7469–7475

Journal of the American Chemical Society
ARTICLE
washed with PBS containing 0.01% pluronic F-127 to remove
unreacted ITC-12. In each experiment, cell density was normal-
ized to an OD600 of 2, and ligations were carried out at pH 6.6.
After incubation with the membrane-permeable aminooxy probe
for 12 h at 8 °C, the cells were pelleted, washed with PBS
containing 0.01% pluronic F-127, resuspended in SDS sample
buffer, sonicated, and the proteome was resolved on 12%
SDSÀPAGE. In agreement with Zhang et al,22 strong labeling
of the protein target was only observed upon introduction of a
reactive ketone into the protein (Figure 3a, lane 5). To further
explore the specificity of the ligation reaction, we performed a
competition experiment with 1. As expected, due to the higher
affinity of 1 for the receptor, the fluorescent band corresponding
to the labeled LasR-ITC-12 complex disappeared upon preincu-
bation with an equimolar amount (10 μM) of the cognate LasR
ligand (Figure 3a, lane 4). As mentioned, we ultimately aimed to
label the bacterial virulence regulator, LasR, in its endogenous
environment. Toward this end, we applied the same conditions
used in the E. coli ligations in repeating the approach with live
P. aeruginosa cultures, taking into account the increased challenge
to success due to the presence of lower levels of LasR expression.
Encouragingly, as seen in the in vitro studies, incubation of intact
cells with 3 showed specific labeling of a protein with an apparent
molecular weight of approximately 30 kDa (Figure 3b; full length
BODIPY-labeled LasR-ITC-12 has a mass of 28.5 kDa). Low-
ering the concentration of 3 from 100 to 10 μM decreased the
intensity of this band. No labeling was detected in the absence of
the probe. Finally, the fluorescent band was excised, digested
with trypsin, and its identity was assessed by mass spectrometry
using a linear trap quadrupole (LTQ) Orbitrap apparatus
(Figure 3c). SEQUEST searches of the resulting LC-MS/MS
data set identified three LasR-derived peptides, corresponding to
a total sequence coverage of 32%.
To further corroborate these results, the ligation reaction was
analyzed by flow cytometry. Pseudomonas cells were treated with
ITC-12 and incubated with the BODIPY-aminooxy probe using
the same methodology as employed for in vivo labeling. In
preliminary experiments in which the cells were washed exten-
sively with PBS buffer, a relatively high background signal was
observed for cells that were not labeled with ITC-12 (SI, Figure
S4). However, through use of a modified procedure described by
Aharoni et al.,23 remaining unreacted probe was easily removed.
Briefly, cells were subjected to two 5 min cycles of incubation
with fresh LB media. With this washing procedure, it is likely that
products of the reaction of compound 3 with small metabolites
are released from the cells. As shown in Figure 4, the negative
control displays only a slight increase in mean fluorescence over
the background. Significantly, when cells were pretreated with
ITC-12, a concentration-dependent increase in the fluorescent
readout was detected (Figure 4a). Furthermore, these results are
in good agreement with those obtained in the gel fluorescent
readout (Figure 3b). Additionally, when cells were preincubated
with increasing concentrations of 3-oxo-C12-HSL, a marked
decrease in fluorescence was observed (SI, Figure S5). Finally,
we set out to determine the effect of cell density on the extent of
LasR labeling. As several studies have shown a clear relation-
ship between the expression of LasR and cell density,18,24 we
expected an increase in overall fluorescence with increasing cell
density at the time of labeling. To determine whether this was
indeed the case, bacteria were grown to different cell densities
(OD600 = 0.4À6) in the presence of ITC-12 (10 and 100 μM),
followed by incubation with the aminooxy probe (cell densities
Figure 2. In vitro labeling of LasR in a whole cell lysate of Pseudomonas
aeruginosa. Ligation reactions were carried out after incubation of
P. aeruginosa lasI-rhlI double mutant (PAO-JP2) with different concen-
trations of ITC-12. Cell lysates were treated with 100 μM of aminooxy-
BODIPY and 1 mM aniline for 12 h at pH 5.1. The labeled proteome was
resolved by SDSÀPAGE (12%) and analyzed by in-gel fluorescence
scanning. See SI, Figure S3 for the corresponding Coomassie blue-
stained gel. *Nonspecific labeling. The positions of molecular weight
markers are indicated.
context of the complete P. aeruginosa proteome. This is not a
trivial matter when one considers the relatively low solubility,
stability, and expression levels of LasR.18 However, if both
reactions (i.e., the initial reaction of the affinity-based covalent
probe (ITC-12) with LasR followed by reacting bound ITC-12
with BODIPY-ONH2) are efficient and highly specific, we would
expect to observe a single band in the fluorescent readout of the
proteome following separation by SDSÀPAGE. To determine
whether this was the case, bacteria were grown in the presence or
absence of ITC-12, after which the cells were washed and lysed,
and the supernatant was incubated at pH 5.1 in the presence of
100 μM BODIPY-ONH2 (3) and 1 mM aniline for 12 h. When
the samples were analyzed by SDSÀPAGE (as shown in
Figure 2), robust labeling of a protein with an apparent molecular
weight of approximately 30 kDa (calculated molecular weight for
ligated full length LasR is 28.5 kDa) was observed when the
bacteria were incubated in the presence of 10 and 100 μM ITC-
12 (lanes 4 and 5, respectively). In contrast, when the concen-
tration of ITC-12 was reduced to 1 μM (lane 3), no labeling was
observed. Furthermore, no labeling of this protein was observed
when the bacteria were grown in the presence of 1 (lane 2),
although a small degree of nonspecific labeling of an unknown
protein with smaller molecular weight was observed, reflecting
the nearly complete chemoselectivity of the ligation.
In Vivo Labeling of LasR in E. coli and P. aeruginosa.
Although oxime/hydrazine ligation has already been employed
for labeling aldehydes on the cell surface,13,19,20 in which one
study used aniline mediated conjugation,13 it has been previously
suggested that this mode of ligation is not suitable for in vivo
intracellular imaging due to competition with endogenous
aldehydes and ketones.21 Zhang et al., however, reported
in vivo intracellular labeling of an unnatural aminoacid, m-acetyl-
phenylalanine, with fluorescein hydrazide,22 suggesting a method
we could adopt to label LasR intracellularly with a fluorescent
aminooxy tag. To validate this approach, we first performed
the conjugation in E. coli overexpressing LasR-LBD. Cells were
incubated in the presence or absence of ITC-12 for 12 h, and then
7471
dx.doi.org/10.1021/ja200455d |J. Am. Chem. Soc. 2011, 133, 7469–7475

Journal of the American Chemical Society
ARTICLE
Figure 3. In vivo labeling of LasR by aminooxy-BODIPY. Ligation reactions were carried out at pH 6.6 with 100 μM of aminooxy-BODIPY and 1 mM
aniline for 12 h at 8 °C. A total cell protein (TCP) extract was separated by SDSÀPAGE (12%). (a) TCP of E. coli BL21 (DE3) overexpressing LasR-
LBD, in the presence (lanes 4 and 5) or the absence (lane 3) of ITC-12. The fluorescent readout (top) and Coomassie-stained SDSÀPAGE (bottom)
gels are shown. Pretreatment of cells with 3-oxo-C12-HSL (lane 4) diminished the labeling. Lane 1, molecular weight markers; lane 2, empty. (b) Specific
labeling of LasR in P. aeruginosa in the presence (10 and 100 μM) or absence (negative control) of ITC-12. See SI, Figure S3 for the corresponding
Coomassie blue-stained gel. *Nonspecific labeling. (c) High-resolution mass spectrometry analysis of the 30 kDa band after digestion with trypsin (see
Supporting Information for details). The primary sequence LasR (top) and peptides identified by LC-MS/MS (ESI) and the SEQUEST search
algorithm (bottom) are shown. The sequences covered by the identified peptides are marked in red.
were normalized to 2 in all labeling experiments). As shown in
Figure 4b, an increase in fluorescence correlates with increased
cell density, and thus with an increase in LasR concentration.
However, at high cell densities (OD600 = 6) the fluorescence is
diminished. A possible explanation for this phenomenon is that
upon reaching the threshold cell density, i.e. when the quorum
has been reached and sensed, any further increase in cell density
is accompanied by degradation of both quorum signals and
receptors, as prolonged sensing of the population size may not
be needed and would likely even be wasteful. Collectively, these
two experiments performed in the presence or absence of ITC-
12 and at different LasR expression levels proves that ligation
involves covalently bound ITC-12 and not any free affinity probe
that might remain in the cell.
In Vivo Imaging of LasR. The current model describing LasR
activity in P. aeruginosa QS is based on the assumption that the
N-terminal region of LasR is inserted in the cytoplasmic mem-
brane and that binding of its cognate ligand, 3-oxo-C12-HSL,
results in structural changes that lead to a conformation that
allows LasR to bind its target DNA. It is not clear whether the
interaction of LasR with the membrane is retained upon such
structural change. A second question regarding the localization of
LasR asks whether this QS receptor is distributed evenly through-
out the cell, or instead, whether a certain degree of compartmen-
talization occurs. To address these points, P. aeruginosa cells were
treated with 3 using the same methodology as described vide supra
for the in vivo labeling and flow cytometry experiments above
(Figure 4). Corresponding with the observations made upon
analysis of the SDSÀPAGE fluorescent readout and flow cyto-
metry measurements, a concentration-dependent increase in LasR
labeling was observed, proving the suitability of the aniline-
catalyzed oxime chemistry methodology for in vivo labeling of
this bacterial virulence regulator. When cells were pretreated with
ITC-12 (Figure 5a), the resulting fluorescent staining was not
consistent over the entire bacterial surface; instead, it appeared to
be localized at the poles of the bacteria. This effect was enhanced
when a higher concentration (100 μM) of ITC-12 was used. It is
worth mentioning that a similar labeling profile was obtained by
Ventre et al.25 by expressing RhlR (another LuxR family type
protein in P. aeruginosa) as an EGFP fusion protein.
Figure 5b shows a clear association of the labeled protein with
the cell membrane, while in cells that were treated with a higher
7472
dx.doi.org/10.1021/ja200455d |J. Am. Chem. Soc. 2011, 133, 7469–7475

Journal of the American Chemical Society
ARTICLE
Figure 4. Analysis of in vivo QS receptor labeling by flow cytometry. (a) Pseudomonas aeruginosa cells were incubated in the absence (blue) or presence
of 10 (yellow), 100 (green), or 250 μM (red) ITC-12. After removal of most of the unreacted ITC-12, the cell density in all samples was normalized to an
OD600 of 2 and treated at 8 °C and pH 6.6 with 100 μM aminooxy-BODIPY (3) and 1 mM aniline for 12 h. As negative control (black), absence of 3. (b)
Mean fluorescence as function of cell density after incubation with ITC-12 ((•) 10 μM, (9) 100 μM) and reaction with probe 3. Before the ligation
reaction with 3, cell densities were normalized to an OD600 of 2.
Figure 5. Imaging of aniline-catalyzed aminooxy ligation of QS receptors on live cells. Pseudomonas aeruginosa cells were incubated in the presence of
100 μM (a), in the presence of 10 μM (b), or in the absence (c) of ITC-12. Cells were washed and subsequently labeled at pH 6.6 with 100 μM of
aminooxy-BODIPY and 1 mM aniline for 12 h. Left, fluorescence channel. Right, bright field. Middle, overlay of BODIPY fluorescence over bright field.
The inset is an expanded section of the fluorescence image. The scale bar in each panel corresponds to 3 μm.
concentration of ITC-12 (Figure 5a), this association did not
appear as strong. This may indicate that LasR indeed dissociates
from the membrane upon activation and dimerization. Addi-
tional studies including higher resolution fluorescence micro-
scopy, however, will be required to further examine this
hypothesis.
LasR Localization. The observation that LasR localizes at the
poles of the cell is significant. This finding may have important
implications for our understanding of the overall mechanism of
quorum sensing signal generation and collection. We propose
the following model to explain the observed localization of LasR:
Several studies have shown that bacteria distribute proteins to
7473
dx.doi.org/10.1021/ja200455d |J. Am. Chem. Soc. 2011, 133, 7469–7475

Journal of the American Chemical Society
ARTICLE
distinct locations in the cell.26 Here, the localization of LasR
observed by fluorescence microscopy (Figure 5) can be ex-
plained by considering reports that newly synthesized autoindu-
cers undergo active secretion from P. aeruginosa to the
environment via the MexAB-OprM efflux pump.27 The precise
location of this efflux pump in bacteria is still unknown, although
recent studies indicate that a similar efflux pump, Opr86, is
mainly localized at the central part of the cell.28 LuxR-type family
proteins, such as LasR, are membrane-associated monomers,
which in the absence of their cognate ligand misfold and are
rapidly degraded by endogenous proteases.18 Is it likely that
through their association with the membrane, these proteins
allow bacteria to specifically detect extracellular autoinducers
that diffuse passively into the cell, especially if the autoinducers
are secreted at a location in the cell that is far removed from the
poles, where the receptors appear to be located. In a previous
study that strengthens this hypothesis, Zhu and Winans deter-
mined that Clp proteases are the primary proteins responsible for
TraR (a LuxR-type protein in Agrobacterium tumefaciens)
turnover.29 In addition, it has been proposed that Clp proteases
are deployed mainly at the bacterial poles, and that LasR under-
goes degradation at low cell densities.30,31 These findings, in
combination with the potential benefit conferred to bacteria in
terms of noise reduction, by the physical separation of the cellular
machineries responsible for autoinducer secretion and sensing
functions, suggest that LasR is most abundant at the cell poles
(see SI, Figure S6 for a schematic description of this model). The
benefit for bacteria to separate localization of autoinducer
secretion and sensing would be in reducing the chance that
secreted autoinducers bind receptors on the cell that secretes
them, thereby increasing sensitivity to gauge the presence and
amount of other cells.
buffer, pH 7, 5% glycerol, 250 mM NaCl, 0.3 mg/mL lysozyme, and
protease inhibitor cocktail (Merck). The lysate was centrifuged and
subjected to buffer exchange with 100 mM sodium acetate (pH 5.1) by
repeated concentration and dilution on a centricon YM-10 (Millipore).
Ligations (200 μL total reaction volume) were performed with 0.1
mg mLÀ1 of proteome in the presence of aniline (1 mM) and 3 (100
3
μM). After 12 h at room temperature reactions were diluted with 2X
SDS sample buffer (without bromophenol blue), heated for 10 min at
70 °C, and subjected to 12% SDSÀPAGE. Fluorescence readouts were
obtained using a Fuji LAS-3000 scanner.
rLasR-LBD in Vivo Ligation. E. coli BL21-DE3 strain, harboring a
pETM-11 vector encoding for a shortened, His6-tagged LasR construct,
LasR-LBD (ligand-binding domain), was inoculated into 10 mL of LB
media containing kanamycin (50 μg/mL) and 3-oxo-C12-HSL (10 μM)
and/or ITC-12 (0, 10 μM). Cells were grown at 21 °C for 12 h, after
which the cells were harvested, washed twice with cold PBS containing
0.01% of pluronic F-127. The cells were then resuspended to an OD600
of 2 in 1 mL phosphate buffer (100 mM) containing aniline (1 mM)
and 3 (100 μM). After 12 h at 8 °C, cells were harvested and washed
twice with cold LB media containing 0.01% of pluronic F-127.
Finally, reactions were diluted with 2X SDS sample buffer (without
Bromophenol blue), heated for 10 min at 70 °C, and subjected to a 12%
SDSÀPAGE, and the fluorescence readout was obtained using a Fuji
LAS-3000 scanner.
Native LasR in Vivo Ligation. P. aeruginosa strain JP2 was grown
in the presence and absence of ITC-12 as described above and washed
with PBS containing 0.01% of pluronic F-127. Cells were then resus-
pended to an OD600 of 2 in 1 mL of phosphate buffer (100 mM)
containing aniline (1 mM) and 3 (100 μM). After 12 h at 8 °C, cells were
harvested and washed twice with cold LB media containing 0.01% of
pluronic F-127.
Fluorography. Treated bacteria (P. aeruginosa JP2 or E. coli) were
resuspended in 200 μL of 1X SDS sample buffer. Samples were
sonicated, heated for 10 min at 70 °C, and subjected to 12%
SDSÀPAGE. Labeled proteins were visualized by in-gel scanning using
a Fuji LAS-3000 scanner and excised from the gel for trypsin digestion.
The resulting peptides were analyzed on a LTQ-Orbitrap ion trap mass
spectrometer (Thermo-Fisher Scientific). Samples were loaded on
custom-made C-18 reverse phase columns (Jupiter 5 μm, 300 Å) and
eluted at a flow rate of 300 nL/min (linear gradient of CH3CN/H2O).
SEQUEST analyses from two independent experiments identified
peptides from a P. aeruginosa PAO1 genomic V2 database (www.
pseudomonas.com).
Microscopy. Labeled (or control) P. aeruginosa (strain JP2) bacteria
(2 μL of culture with an OD600 of 0.1 in PBS) were immobilized on
agarose-coated (2%) microscope slides. Images were obtained using an
Olympus FV1000 laser-scanning confocal microscope, with a 100Â oil-
immersion lens. Ligated BODIPY was excited by a 488 nm argon laser
line, and fluorescence was detected at 520 nm. Images were processed
using ImageJ software (National Institutes of Health).
’ CONCLUSIONS
In summary, our data demonstrate that a combination of
isothiocyanate and oxime chemistry can be successfully em-
ployed for the visualization of specific native proteins in live cells
through a two-step labeling strategy. We have shown, using
several different analytical methods, that we can bio-orthogonally
label QS receptors. In spite of the propensity of an aminooxy
probe to react with endogenous carbonyl moieties (i.e., pyruvic
acid, glucose, etc.), we have shown that this problem can be
overcome through optimization of the aniline-catalyzed oxime
reaction followed by appropriate washing of the cells. Relying on
this strategy, we report the first demonstration of activity-based
fluorescent labeling and visualization of a quorum sensing
receptor in living bacterial cells. This methodology could, more-
over, be applied for the intracellular ligation of other types of
proteins in bacterial and/or eukaryotic cells in their natural
settings, offering the possibility of imaging, mapping and quanti-
fying specific proteins in vivo.
Flow Cytometry. Treated bacteria were diluted to an OD600 of 0.1
in PBS and analyzed with a FACSCalibur flow cytometer (Becton
Dickinson) or alternatively with an Eclipse Analyzer (icyt). The average
sorting rate was ∼1000 events per second. FACS data were processed
using the FCS Express (De Novo Software).
’ EXPERIMENTAL SECTION
In Vitro Proteome Ligation. P. aeruginosa strain JP2 (lasI/rhlI-
deleted), harboring plasmid pKD201 carrying a LasI reporter coupled to
the luxCDABE luminescence system,32 was inoculated into 10 mL of LB
media (with 2.5 g/L of NaCl) containing 300 μg/mL of trimethoprim
and 3-oxo-C12-HSL (10 μM) or ITC-12 (0, 1, 10, 100 μM). Cells were
grown at 37 °C to an optical density (OD600) of 2, after which the cells
were harvested, washed twice with cold PBS containing 0.01% of
pluronic F-127, frozen at À20 °C, and lysed in 50 mM phosphate
’ ASSOCIATED CONTENT
S
Supporting Information. Synthetic protocols and char-
b
acterization of compounds reported herein; optimization of
conditions for the model reactions; Coomassie blue-stained gels
for the in vivo and in vitro labeling; additional flow cytometry
analysis; and proposed model for noise control in P. aeruginosa.
7474
dx.doi.org/10.1021/ja200455d |J. Am. Chem. Soc. 2011, 133, 7469–7475

Journal of the American Chemical Society
ARTICLE
This material is available free of charge via the Internet at http://
pubs.acs.org.
(24) Schuster, M.; Greenberg, E. P. BMC Genomics 2007, 8, 287.
(25) Ventre, I.; Ledgham, F.; Prima, V.; Lazdunski, A.; Foglino, M.;
Sturgis, J. N. Mol. Microbiol. 2003, 48, 187–198.
(26) Shapiro, L.; McAdams, H. H.; Losick, R. Science 2009,
326, 1225–1228.
’ AUTHOR INFORMATION
(27) Pearson, J. P.; Van Delden, C.; Iglewski, B. H. J. Bacteriol. 1999,
181, 1203–1210.
(28) Tashiro, Y.; Nomura, N.; Nakao, R.; Senpuku, H.; Kariyama, R.;
Kumon, H.; Kosono, S.; Watanabe, H.; Nakajima, T.; Uchiyama, H.
J. Bacteriol. 2008, 190, 3969–3978.
Corresponding Author
meijler@bgu.ac.il
’ ACKNOWLEDGMENT
(29) Zhu, J.; Winans, S. C. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
The authors thank A. Brik, I. Fishov, and I. Khalaila (Ben-
Gurion University of the Negev) and P. E. Dawson (The Scripps
Research Institute) for helpful suggestions and discussions,
A. Aharoni (Ben-Gurion University of the Negev) for valuable
advice and assistance with FACS experiments, and M. G. Surette
(University of Calgary) and M. J. Bottomley (IRBM) for providing
us with bacterial strains. This work was supported by the European
Research Council (Starting Grant 240356, M.M.M.).
1507–1512.
(30) Kain, J.; He, G. G.; Losick, R. J. Bacteriol. 2008, 190, 6749–6757.
(31) McGrath, P. T.; Iniesta, A. A.; Ryan, K. R.; Shapiro, L.;
McAdams, H. H. Cell 2006, 124, 535–547.
(32) Duan, K.; Surette, M. G. J. Bacteriol. 2007, 189, 4827–4836.
’ REFERENCES
(1) Engebrecht, J.; Nealson, K.; Silverman, M. Cell 1983, 32, 773–781.
(2) Fuqua, W. C.; Winans, S. C.; Greenberg, E. P. J. Bacteriol. 1994,
176, 269–275.
(3) Arias, C. A.; Murray, B. E. N. Engl. J. Med. 2009, 360, 439–443.
(4) Kiratisin, P.; Tucker, K. D.; Passador, L. J. Bacteriol. 2002,
184, 4912–4919.
(5) Pappas, K. M.; Weingart, C. L.; Winans, S. C. Mol. Microbiol.
2004, 53, 755–769.
(6) Keppler, A.; Pick, H.; Arrivoli, C.; Vogel, H.; Johnsson, K. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 9955–9959.
(7) Griffin, B. A.; Adams, S. R.; Tsien, R. Y. Science 1998,
281, 269–272.
(8) Speers, A. E.; Adam, G. C.; Cravatt, B. F. J. Am. Chem. Soc. 2003,
125, 4686–4687.
(9) Soriano del Amo, D.; Wang, W.; Jiang, H.; Besanceney, C.; Yan,
A. C.; Levy, M.; Liu, Y.; Marlow, F. L.; Wu, P. J. Am. Chem. Soc. 2010,
132, 16893–16899.
(10) Jewett, J. C.; Sletten, E. M.; Bertozzi, C. R. J. Am. Chem. Soc.
2000, 132, 3688–3690.
(11) Saxon, E.; Bertozzi, C. R. Science 2000, 287, 2007–2010.
(12) Dondoni, A. Angew. Chem., Int. Ed. 2008, 47, 8995–8997.
(13) Zeng, Y.; Ramya, T. N. C.; Dirksen, A.; Dawson, P. E.; Paulson,
J. C. Nat. Methods 2009, 6, 207–209.
(14) Dirksen, A.; Dawson, P. E. Bioconjugate Chem. 2008, 19,
2543–2548.
(15) Dirksen, A.; Hackeng, T. M.; Dawson, P. E. Angew. Chem., Int.
Ed. 2006, 45, 7581–7584.
(16) Fleissner, M. R.; Brustad, E. M.; Kꢀalai, T.; Altenbach, C.;
Cascio, D.; Peters, F. B.; Hideg, K.; Peuker, S.; Schultz, P. G.; Hubbell,
W. L. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 21637–21642.
(17) Amara, N.; Mashiach, R.; Amar, D.; Krief, P.; Spieser, S. A. H.;
Bottomley, M. J.; Aharoni, A.; Meijler, M. M. J. Am. Chem. Soc. 2009,
131, 10610–10619.
(18) Siehnel, R.; Traxler, B.; An, D. D.; Parsek, M. R.; Schaefer, A. L.;
Singh, P. K. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 7916–7921.
(19) Mahal, L. K.; Yarema, K. J.; Bertozzi, C. R. Science 1997,
276, 1125–1128.
(20) Baskin, J. M.; Dehnert, K. W.; Laughlin, S. T.; Amacher, S. L.;
Bertozzi, C. R. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 10360–10365.
(21) Sletten, E. M.; Bertozzi, C. R. Angew. Chem., Int. Ed. 2009, 48,
6974–6998.
(22) Zhang, Z.; Smith, B. A. C.; Wang, L.; Brock, A.; Cho, C.;
Schultz, P. G. Biochemistry 2003, 42, 6735–6746.
(23) Aharoni, A.; Thieme, K.; Chiu, C. P. C.; Buchini, S.; Lairson,
L. L.; Chen, H.; Strynadka, N. C. J.; Wakarchuk, W. W.; Withers, S. G.
Nat. Methods 2006, 3, 609–614.
7475
dx.doi.org/10.1021/ja200455d |J. Am. Chem. Soc. 2011, 133, 7469–7475