A fluorescent sensor for GABA and synthetic GABAB receptor ligands

Article abstract of DOI:10.1021/ja306320s

While γ-aminobutyric acid (GABA) is the main inhibitory neurotransmitter, suitable tools to measure its concentration in living cells with high spatiotemporal resolution are missing. Herein, we describe the first ratiometric fluorescent sensor for GABA, dubbed GABA-Snifit, which senses GABA with high specificity and spatiotemporal resolution on the surface of living mammalian cells. GABA-Snifit is a semisynthetic fusion protein containing the GABA_B receptor, SNAP- and CLIP-tag, a synthetic fluorophore and a fluorescent GABA_B receptor antagonist. When assembled on cell surfaces, GABA-Snifit displays a GABA-dependent fluorescence emission spectrum in the range of 500-700 nm that permits sensing micromolar to millimolar GABA concentrations. The ratiometric change of the sensor on living cells is 1.8. Furthermore, GABA-Snifit can be utilized to quantify the relative binding affinities of GABA_B receptor agonists, antagonists and the effect of allosteric modulators. These properties make GABA-Snifit a valuable tool to investigate the role of GABA and GABA_B in biological systems.

Full text of DOI:10.1021/ja306320s

Article
pubs.acs.org/JACS
A Fluorescent Sensor for GABA and Synthetic GABAB Receptor
Ligands
Anastasiya Masharina, Luc Reymond, Damien Maurel,† Keitaro Umezawa,‡ and Kai Johnsson*
Institute of Chemical Sciences and Engineering (ISIC), Institute of Bioengineering, NCCR in Chemical Biology, Ecole Polytechnique
Federale de Lausanne (EPFL), 1015 Lausanne, Switzerland
́ ́
S
* Supporting Information
ABSTRACT: While γ-aminobutyric acid (GABA) is the main
inhibitory neurotransmitter, suitable tools to measure its
concentration in living cells with high spatiotemporal
resolution are missing. Herein, we describe the first ratiometric
fluorescent sensor for GABA, dubbed GABA-Snifit, which
senses GABA with high specificity and spatiotemporal
resolution on the surface of living mammalian cells. GABA-
Snifit is a semisynthetic fusion protein containing the GABAB
receptor, SNAP- and CLIP-tag, a synthetic fluorophore and a
fluorescent GABAB receptor antagonist. When assembled on
cell surfaces, GABA-Snifit displays a GABA-dependent
fluorescence emission spectrum in the range of 500−700 nm
that permits sensing micromolar to millimolar GABA
concentrations. The ratiometric change of the sensor on living cells is 1.8. Furthermore, GABA-Snifit can be utilized to
quantify the relative binding affinities of GABAB receptor agonists, antagonists and the effect of allosteric modulators. These
properties make GABA-Snifit a valuable tool to investigate the role of GABA and GABAB in biological systems.
glutamate,12 carbohydrates,13 cyclic nucleotides,14,15 ATP, and
INTRODUCTION
■
others,16−18 no FRET biosensors for GABA have been reported
so far. Using traditional methods for the generation of FRET
biosensors we would first need to identify an appropriate
GABA binding protein which undergoes a significant conforma-
tional change upon GABA binding − so far none has been
found. Recently, we reported an approach for the creation of
FRET-based biosensors that circumvents the need for a
conformational change of an analyte binding protein.19−21
These sensors, termed Snifits (SNAP-tag based indicator with a
fluorescent intramolecular tether), are fusion proteins consist-
ing of SNAP-tag, CLIP-tag, and a receptor protein (Figure 1A).
SNAP- and CLIP-tag, two self-labeling protein tags,22,23 are
labeled with a synthetic fluorophore tethered to a receptor
protein ligand and a second synthetic fluorophore, respectively.
The ligand binds to the receptor protein and maintains the
Snifit in a closed conformation. Free analyte competes for
binding to the receptor protein and can shift the equilibrium
toward an open conformation. The conformational change
results in reduced proximity of the two fluorophores and hence
a change in their FRET efficiency.
γ-Aminobutyric acid (GABA) is the main inhibitory neuro-
transmitter in the mammalian nervous system.1 Perturbations
in GABAergic inhibition are responsible for neurological
disorders such as epilepsy, anxiety disorders, and schizophre-
nia.2 In addition to its role in neuronal communication, GABA
plays a role in embryonic development and adult neurogenesis.3
Furthermore, GABA participates in intercellular communica-
tion outside the nervous system.4 A detailed characterization of
the role of GABA in these different biological processes will
require the measurement of GABA concentrations with high
temporal and spatial resolution. Reported noninvasive methods
for analysis of GABA concentrations in living systems are based
on in vivo magnetic resonance spectroscopy (MRS)5,6 and in
vivo microdialysis techniques.7,8 Furthermore, electrochemical9
and piezoelectric10 GABA sensors were developed to detect
GABA ex vivo. However, these methods lack the required
sensitivity or spatiotemporal resolution to measure GABA
concentrations on the level of a single cell required for further
elucidation of the role of GABA. Fluorescent biosensors based
on Forster resonance energy transfer (FRET) are well suited
̈
Here we introduce a Snifit for GABA that is based on the
metabotropic GABAB receptor as a receptor protein and a
fluorescent antagonist as a synthetic ligand (Figure 1B). GABA-
Snifit represents the first biosensor that permits the measure-
for the measurement of small-molecule concentrations in cells
and in vivo.11 Such FRET biosensors are usually based on a
receptor protein that undergoes a conformational change upon
binding to the analyte of interest. Expressing such a receptor
protein as a fusion protein with two autofluorescent proteins
can then result in the generation of a FRET sensor. Despite
successful examples of FRET sensors for the neurotransmitter
Received: June 28, 2012
Published: October 24, 2012
© 2012 American Chemical Society
19026
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
change in the relative orientation of the two subunits and to G
protein activation by the 7TM of GB2.27 It has been shown that
SNAP-tag fusion proteins of the GABAB receptor are functional
and have been used for FRET studies on receptor dimerization
in living cells.28 For the generation of GABA-Snifit we focused
on GB1a and generated a construct for the expression of a
SNAP-CLIP-GB1a fusion protein (Figure 1B). Functional
GABA-Snifit then is based on the coexpression of SNAP-
CLIP-GB1a and GB2, and the labeling of the sensor with a
fluorescent ligand that competes with GABA for binding.
Various GABAB receptor agonists and antagonists have been
reported,29 and some of these were derivatized with synthetic
probes without abolishing their binding affinity to the
receptor.29−31 For the labeling of GABA-Snifit, antagonists
are preferable over agonists as they keep the G protein-coupled
receptor (GPCR) in an inactive state. We therefore chose the
GABAB receptor antagonist CGP 51783 1 ((3-[[4-
chlorophenyl)methyl]amino-propyl]-(P-diethoxymethyl)-phos-
phinic acid) as a starting point for the generation of a
fluorescent ligand (Figure 2A).32 The affinity of CGP 51783
has been reported to be 1 μM. Furthermore, it has been shown
that coupling of a similar antagonist to a spacer did not affect its
affinity to GABAB.31 This suggests that tethering of compound
1 to SNAP-tag should not disrupt its binding affinity. To
transform CGP 51783 1 into an appropriate substrate for
GABA-Snifit, we synthesized derivative 2 (Figure 2A) to which
fluorophores and the SNAP-tag substrate benzylguanine (BG)
were coupled via a short flexible linker (Figure 2B). The
resulting probes BG-Cy5-CGP 3, BG-TMR-CGP 4, and BG-
AlexaFluor594-CGP 5 contained a polyethylene glycol (PEG)
linker between BG and the fluorophore to facilitate binding of
the antagonist to the receptor. For control experiments we
prepared the identical SNAP-tag substrates but lacking the
GABAB receptor antagonist (BG-Cy5 6, BG-TMR 7, and BG-
AlexaFluor594 8; Figure S1(Supporting Information [SI]). A
detailed outline of the synthesis of all derivatives is provided in
the SI.
Characterization of GABA-Snifit. To confirm the cell
surface expression of GABA-Snifit, human embryonic kidney
293 cells (HEK 293 cells) transiently expressing SNAP-CLIP-
GB1a and GB2 were labeled with fluorescent substrates of
SNAP- and CLIP-tag and subsequently analyzed by confocal
fluorescence microscopy (Figure 3A). These data demonstrate
that GABA-Snifit is well expressed and can be specifically
labeled via SNAP-tag and CLIP-tag. Furthermore, the cell lysate
of cells expressing SNAP-CLIP-GB1a and GB2 was analyzed by
sodium dodecyl sulfate polyacrylamide gel electrophoresis
(SDS-PAGE) and subsequent in-gel fluorescence scanning,
demonstrating the presence of doubly labeled sensor proteins
of the correct size (Figure 3B).
We then investigated if GABA-Snifit retains the affinity of
native GABAB receptor for GABA and is capable of GPCR
signaling. To detect GABA-binding and receptor activation of
GABA-Snifit, we coexpressed SNAP-CLIP-GB1a, GB2, and the
chimeric G protein Gqi9 in HEK 293 cells and analyzed the
nonlabeled GABA-Snifit. It has been previously shown that
Gqi9 permits activation of phospholipase C (PLC) by the
GABAB receptor in HEK 293 cells.33 Active PLC stimulates
inositol 1,4,5-trisphosphate (IP3) formation which triggers
calcium release from the endoplasmic reticulum (ER) into the
cytosol. Upon addition of different GABA concentrations we
measured an increase in cytosolic calcium concentrations
utilizing a calcium-sensitive fluorescent dye. We analyzed
Figure 1. Snifits. (A) Snifit principle. A Snifit is composed of a fusion
protein between SNAP-tag, CLIP-tag, and a receptor protein (RP) for
the analyte of interest. SNAP-tag is labeled with a molecule containing
a fluorophore (red star) and a ligand (gray ball) that binds to the RP;
CLIP-tag is labeled with a second fluorophore (green star). In the
absence of analyte (pink ball), the intramolecular ligand keeps the
Snifit in a closed conformation. In the presence of sufficient
concentrations of analyte, the Snifit is shifted toward an open
conformation. Opening and closing of the Snifit are detected through
changes in FRET efficiency between the two fluorophores. (B) GABA-
Snifit. The metabotropic GABAB receptor composed of the GB1a
(light gray; containing the sushi domains (orange)) and the GB2
subunit (dark gray) is used as RP and a GABAB receptor antagonist
(gray ball) as intramolecular ligand. GABA-Snifit detects GABA or
synthetic GABAB ligands through displacement of the intramolecular
antagonist and a resulting change in FRET efficiency. As GABAB is
known to form oligomers on cell surfaces, it should be noted that an
intermolecular binding of the tethered antagonist cannot be excluded.
ment of GABA concentrations on the surface of mammalian
cells with high temporal and spatial resolution.
RESULTS
■
Design of GABA-Snifit. The design of a Snifit for GABA
requires a suitable GABA-binding protein as well as a synthetic
ligand that competes with GABA for binding to the receptor.
For the GABA-binding protein we chose the metabotropic G
protein-coupled GABAB receptor. The GABAB receptor is an
obligatory heterodimeric transmembrane receptor composed of
the homologous subunits GABAB1 (GB1) and GABAB2 (GB2).
Furthermore, two GB1 receptor isoforms resulting from
alternative splicing exist, GB1a and GB1b. GB1a contains two
N-terminal complement control protein (CCP) modules, the
so-called sushi domains (SDs) that are lacking in GB1b.24 The
SDs are conserved protein interaction motifs which are believed
to be important for axonal targeting of the receptor.25 Both
subunits GB1 and GB2 possess a large extracellular domain, the
so-called venus flytrap domain (VFT) which is linked to the N-
terminus of a heptahelical transmembrane domain (7TM). The
VFT of GB1 binds orthosteric ligands such as GABA and other
agonists or antagonists, whereas the VFT of GB2 does not bind
any orthosteric GABAB receptor ligand.26 Upon agonist
binding, the two lobes of GB1 VFT close. This results in a
19027
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
(FDY‑547/FCy5) (Figure 4B). Control experiments of cells
expressing GABA-Snifit and labeled with BG-Cy5 6 lacking
the GABAB receptor antagonist did not result in a detectable
intensity ratio change upon GABA addition (Figure S4, SI).
Furthermore, mock-transfected HEK 293 cells which were
incubated with BG-Cy5-CGP 3 (2 μM) and CLIP-Surface 547
25 (10 μM) did not show any unspecific fluorescent labeling or
background fluorescence (Figure S5, SI). In further control
experiments GABA-Snifit was perfused with GABAB-unspecific
ligands such as glutamate, glycine, aspartate, and others,
resulting in no intensity ratio change (Figure S6, SI). This
indicates that the observed intensity ratio change of GABA-
Snifit labeled with BG-Cy5-CGP 3 is indeed due to a specific
replacement of the GABAB receptor antagonist by GABA and a
concomitant switching of GABA-Snifit to an open conforma-
tion. The fact that FRET efficiency increases upon sensor
opening is noteworthy as previously described Snifits showed a
decrease in FRET efficiency upon sensor opening. However,
the GABAB receptor is a heterodimeric transmembrane
receptor that is known to form stable dimers of heterodimers.28
The dimerization was shown to occur at the level of GB1,34
raising the possibility that the covalently bound antagonist
binds to the GB1a of the other heterodimer. Another unique
property of GABA-Snifit is that the SNAP-CLIP fusion is
separated from the GABA-binding domain by an additional
domain, the so-called Sushi domains (Figure 1B), whereas in
previously described Snifits SNAP-CLIP was fused directly to
the ligand-binding domain. To gain more insights into the
sensing mechanism of GABA-Snifit, we compared its FRET
efficiency in the open state with those of two other Snifits
previously utilized on cell surfaces. These were a Snifit for
sulfonamides based on human carbonic anhydrase (HCA-
Snifit)20 and a Snifit for glutamate based on the ionotropic
glutamate receptor iGluR5 (Snifit-iGluR5).21 All three Snifits
were expressed in HEK 293 cells and labeled with BG-Cy5 6
and BC-DY-547 25, and the fluorescence emission intensity
ratios were measured for all three Snifits under identical
instrument settings. Snifits labeled with fluorophores not
containing a ligand for the receptor proteins mimic the open
state of the sensor. We found that the intensity ratio is similar
for all three Snifits (Figure S7, SI). This demonstrates that the
dimerization of GABAB does not influence the FRET efficiency
in the open Snifit state. However, the absence of structural
information on (dimeric) GABAB receptor prevents a
clarification if sensor closing is due to an inter- or intra-
molecular interaction. Regardless, our results demonstrate that
the GABA-dependent emission ratio change is due to a
displacement of the fluorescent antagonist from the GABA
binding site.
Figure 2. Molecules used for labeling of GABA-Snifit. (A) GABAB
receptor antagonist CGP 51783 1 and its derivative 2 used for
generation of tethered fluorescent ligands. (B) Synthetic probes for
semisynthesis of GABA-Snifit; these probes contain the SNAP-tag
substrate benzylguanine (BG) (blue), the intramolecular ligand CGP
(pink), and different fluorophores. Top: BG-Cy5-CGP 3 (red).
Middle: BG-TMR-CGP 4 (green). Bottom: BG-AlexaFluor594-CGP 5
(violet).
PLC activation as a function of [GABA] and obtained a half
maximal effective concentration (EC50) of 56
19 nM for
GABA-Snifit (Figure 3C). This value describes the GABA
concentration that results in a half maximal calcium response in
cells transiently expressing GABA-Snifit. These data demon-
strate that GABA-Snifit binds GABA and is functional with
respect to GPCR signaling.
Measuring GABA with GABA-Snifit. To test the
suitability of GABA-Snifit for GABA sensing, HEK 293 cells
coexpressing SNAP-CLIP-GB1a and GB2 were labeled for 20
min at 37 °C with BG-Cy5-CGP 3 (2 μM) and CLIP-Surface
547 25 (10 μM) (Figure 4A). CLIP-Surface 547 contains the
fluorophore DY-547, a good FRET donor for Cy5 (Figure S2,
SI). It should be noted that the use of membrane impermeable
substrates for the labeling of GABA-Snifit allows us to assemble
the sensor exclusively on the cell surface (Figure S3, SI).
Perfusion of the cells with buffer containing 1 mM GABA
resulted in a decrease in DY-547 fluorescence, an increase in
Cy5 fluorescence, and a resulting decrease in the emission ratio
To determine the concentration range in which GABA-Snifit
can be utilized to measure changes in [GABA], we perfused
cells expressing GABA-Snifit with different GABA concen-
trations (Figure 5A) and fitted the corresponding ratio changes
to a single-site binding isotherm (eq 4, SI; Figure 5B). These
comp,GABA
experiments yielded a Kd
of 100
10 μM; this
constant represents the concentration at which GABA has
displaced half of the antagonist from the binding site. It should
be noted that the affinity of GABA is lower than its efficacy at
GABAB35 and that in labeled GABA-Snifit GABA competes for
binding with tethered GABAB antagonist. For these reasons,
higher GABA concentrations are needed to open the GABA-
Snifit than to induce an intracellular calcium response. Different
reports indicate that GABA concentrations in brain rise to the
19028
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
Figure 3. Characterization of GABA-Snifit. (A) Expression of GABA-Snifit on the surface of mammalian cells. SNAP-CLIP-GB1a was labeled on the
surface of HEK 293 cells with the dyes Alexa Fluor 488 and DY-647 via their corresponding benzylcytosine (BC)- and benzylguanine (BG)-
derivatives. Images were taken using a confocal Zeiss LSM 700 microscope. Scale bar 10 μm. (B) SDS-PAGE and in-gel fluorescence scanning of the
labeled sensor protein SNAP-CLIP-GB1a on HEK 293 cell surface and Coomassie staining. Lane 1: MW, lane 2: GABA-Snifit (150 kDa). (C)
Calcium dose response generated by increasing concentrations of GABA in HEK 293 cells expressing SNAP-CLIP-GB1a together with GB2 and the
chimeric G protein Gqi9. In this experiment GABA-Snifit was not labeled. Data are means s.d. of triplicate determinations and representative of
three independent experiments.
Figure 4. Perfusion experiments for GABA detection on the surface of HEK 293 cells. (A) Donor channel (DY-547), FRET channel (Cy5), and
transmission channel of the GABA-Snifit on HEK 293 cells labeled with BC-DY-547 25 and BG-Cy5-CGP 3. Scale bar 50 μm. (B) Time course of
the intensity ratio of donor emission vs acceptor emission (top), of the donor channel (middle), and the acceptor channel (bottom) upon addition
and removal of 1 mM GABA. The red bar indicates the time span of GABA perfusion.
millimolar range and that GABA’s half-maximal response
present and at saturating GABA concentration (eq 2, SI) we
obtained a maximum ratio change of ΔRmax = 1.5 0.08 for the
DY-547−Cy5 pair. We have observed for other Snifits that the
quantum yield (QY: the ratio of photons absorbed to photons
emitted through fluorescence) of fluorophores bound to SNAP-
tag can slightly decrease in the closed state of the sensor and
(EC50) in brain stem slices of rats is about 20−70 μM.36 The
comp,GABA
here measured Kd
demonstrates that GABA-Snifit
detects GABA in a concentration-dependent manner and that it
should be well suited for GABA detection under physiological
conditions. By dividing the intensity ratio when no GABA is
19029
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
Figure 5. Calibration of GABA-Snifit. (A) Perfusion of GABA-Snifit with increasing GABA concentrations. GABA-Snifit is labeled with BC-DY-547
25 and BG-Cy5-CGP 3 on the surface of HEK 293 cells and perfused with GABA (a = 1 μM, b = 10 μM, c = 20 μM, d = 40 μM, e = 100 μM, f = 1
mM, g = 10 mM). Shown is the ratio of donor (DY-547) and acceptor (Cy5) emission. (B) GABA titration curve of GABA-Snifit on the surface of
HEK 293 cells. Shown is the intensity ratio change ΔR = rzeroDY‑547/Cy5/rGABADY‑547/Cy5 for different GABA concentrations. Data are means s.d. of
three independent experiments; (n = 15).
that this can contribute to the observed ratio change.19 To
investigate if ΔRmax is due to a FRET effect alone or if other
factors contribute, we labeled GABA-Snifit with BG-Cy5-CGP
3 but not with CLIP-Surface 547 25. We perfused the cells with
1 mM GABA and recorded the emission of Cy5 (Figure S8A,B
SI). We observed a decrease in Cy5 fluorescence when the
fluorescent antagonist was displaced from the GABA-binding
site of GABA-Snifit (Rnon‑FRETCy5 = 1.07 0.03; eq 3, SI). This
environmental effect decreases the total intensity ratio change.
To improve the observed ratio change of GABA-Snifit, we
labeled the sensor with BG-TMR-CGP 4 as donor fluorophore
and CLIP-Surface 647 24 containing DY-647 as acceptor
fluorophore. In this arrangement, a reduced QY of SNAP-
bound FRET donor in the closed state should increase the
intensity ratio change. As predicted, we found a maximum ratio
change of ΔRmax = 1.6 0.08 (eq 2, SI) for the TMR/Cy5 pair
(Figure S9A, SI and Table 1). This increased maximum ratio
Control experiments for each fluorophore pair confirmed
that the observed ratio change is dependent on the presence of
covalently attached GABAB receptor antagonist (Figure S10,
SI). The intensity ratio changes for all five different fluorophore
pairs tested with GABA-Snifit are summarized in Table 1. The
ΔRmax varied between 1.3 and 1.8, and the Alexa488/Cy5 pair
showed the highest ratio change.
The use of membrane-impermeable dyes results in the
exclusive semisynthesis of GABA-Snifit on the cell surface. This
prevents interference from sensor protein not fully transported
to the cell surface when measuring extracellular GABA
concentrations. To visualize the intracellular protein pool of
GABA-Snifit we incubated HEK 293 cells expressing GABA-
Snifit with a cell-permeable fluorescent SNAP-tag substrate,
revealing the presence of a significant intracellular pool of
SNAP-CLIP-GB1a (Figure S11, SI). It should be noted that the
cell-permeable substrate does not contain the GABAB
antagonist, and thus the intracellular pool of GABA-Snifit
does not allow for GABA sensing. In contrast to Snifits, the
intra- and extracellular sensor populations of FRET sensors
based on autofluorescent proteins can only be discriminated
through confocal microscopy, complicating the use of these
sensors.
Table 1. Fluorophore Pairs Utilized with GABA-Snifit
maximum ratio
change
environmental
SNAP-tag
CLIP-tag
effect
BG-Cy5-CGP
BC-DY-547
BC-Alexa488
BC-DY-647
BC-Alexa488
BC-Alexa488
1.5 0.08
1.8 0.06
1.6 0.04
1.4 0.02
1.3 0.04
−
−
+
BG-Cy5-CGP
BG-TMR-CGP
BG-TMR-CGP
BG-Alexa594-CGP
Signal transduction of GABAB might interfere with the use of
GABA-Snifit and/or might result in the internalization of
GABA-Snifit. To prevent GPCR signaling after GABAB
receptor activation, we replaced wt-GABAB with the GABAB
receptor chimera GB1/237 (Figure S12A, SI). GB1/2 contains
the VFT of GB1a and the 7-TM of GB2. It is able to bind
GABAB receptor ligands but is not able to couple to G proteins.
SNAP-CLIP-GB1/2 fusion protein was expressed in HEK 293
cells, and its cell surface expression was confirmed by confocal
microscopy (Figure S12B, SI). When SNAP-CLIP-GB1/2 was
labeled with BG-Cy5-CGP 3 and CLIP-Surface 547 25, we
could detect GABA by a decrease in the emission ratio with a
ΔRmax of 1.4 (Figure S12C, SI). GB2 VFT was reported to
−
−
change relative to the DY-547/Cy5 pair is due to the
TMR
environmental effect on the TMR fluorophore (Rnon‑FRET
= 1.08 0.04; eq 3, SI) (Figure S6C,D). Furthermore, control
experiments with a derivative of BG-TMR-CGP 4 lacking the
GABAB antagonist (i.e., BG-TMR 7) did not result in an
intensity ratio change upon GABA addition (Figure S10A, SI).
To further demonstrate the ease with which different FRET
pairs can be utilized for GABA-Snifit we labeled it with three
additional FRET fluorophore pairs: Alexa Fluor 488/Cy5, Alexa
Fluor 488/TMR, and Alexa Fluor 488/Alexa Fluor 594. We
found a GABA-dependent intensity ratio change for all
fluorophore pairs upon perfusion (Figure S9, SI).
increase GB1 VFT affinity for GABA.38 As this GABAB domain
comp,GABA
is missing in the new GABA-Snifit, we observed Kd
of
420 50 μM upon perfusion with different [GABA] (Figure
19030
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
Figure 6. Titration curves of SNAP-CLIP-GB1a on the surface of HEK 293 cells with different GABAB receptor ligands. GABA-Snifit is labeled with
BC-DY-547 25 and BG-Cy5-CGP 3 on the surface of HEK 293 cells and perfused with increasing concentrations of ligands. (A) Intensity ratio
change ΔR = rzeroDY‑547/Cy5/rligandDY‑547/Cy5 for different concentrations of the agonists R,S-baclofen (black squares), GABA (cyan triangles), 3-APPA
DY‑547/Cy5
(red circles), and for the antagonist CGP 52432 (blue triangles). (B) Intensity ratio change ΔR = rzeroDY‑547/Cy5/rGABA
for different GABA
concentrations in the absence of PAMs (black squares) and the presence of the PAMs CGP 7930 (blue triangles) or rac-BHFF (red circles). Data are
means s.d. of three independent experiments (n ≥ 18).
comp,GABA
S12D SI) compared to the previously described Kd
100 10 μM.
of
currently in clinical use is the agonist Baclofen (4-amino-3-(4-
chlorophenyl)butanoic acid - Lioresal). GABA-Snifit could be
employed in drug research for the characterization of the
affinity of synthetic GABAB receptor agonists and antagonists as
the binding of such ligands to the orthosteric site of GABAB
receptor should switch GABA-Snifit to its open conformation.
We characterized the interaction of GABA-Snifit with the
known agonists R,S-baclofen42 and 3-APPA (3-aminopropyl-
phosphinic acid)43 as well as the antagonist CGP 52432 ([3-
[[(3,4-dichlorophenyl)methyl]amino]propyl](diethoxy-
methyl)phosphinic acid).44 All three ligands were able to open
the sensor (Figure S15, SI), and by titrating the sensor with
different ligand concentrations and fitting the data to a single-
site binding isotherm (SI eq 4) we obtained Kdcomp,baclofen = 300
Functional tests with SNAP-CLIP-GB1/2 and the chimeric
G protein Gqi9 showed no increase in cytosolic calcium
concentrations upon addition of GABA (Figure S12E, SI). This
demonstrates that SNAP-CLIP-GB1/2 is suitable as GABA
Snifit. It senses GABA on the cell surface in a higher
concentration range and is nonfunctional with respect to G
protein coupling. GABA-Snifit based on GB1/2 is therefore
suitable for sensor applications where cell signaling has to be
avoided.
The release of GABA from neurons takes place over a period
of ∼200 ms,39 and the temporal resolution of any GABA sensor
ideally should be in a similar range. Our previous experiments
with other Snifits19−21 suggest that the kinetics of the opening
and closing of GABA-Snifit should be determined by the
dissociation rate constant (koff) of the tethered fluorescent
antagonist and GABA, respectively. A koff of 5.8 s−1 has been
reported for the dissociation of GABA from GABAB.40 We
attempted to determine the rate of sensor opening and closing
through GABA perfusion experiments. The measured opening
rate of the sensor (kopening = 0.68 0.17 s−1; obtained by fitting
the change in emission ratio to a single exponential function;
SI, eq 5 and Figure S13A) was similar to the speed with which
the bath solution inside the perfusion chamber could be
60 μM, Kdcomp,APPA = 20 1.7 μM, and Kdcomp,CGP52432 = 280
comp,ligand
20 nM (Figure 6A). Kd
reflects the affinity of each
comp
ligand for the GABAB receptor. The Kd
values of R,S-
baclofen and 3-APPA are in agreement with the reported 15-
fold difference between their GABAB receptor affinities (35 nM
for R,S-baclofen and 2.4 nM for 3-APPA; inhibition of binding
of [3H]-baclofen to cat cerebellum).45 In contrast, according to
comp
our Kd
values, the GABAB affinity of the antagonist CGP
52432 is 70-fold higher than that of 3-APPA, whereas its
reported affinity on brain membranes is 11 times lower (55 nM
for CGP 52432 and 5 nM for 3-APPA; inhibition of binding of
[3H]-APPA to rat cerebral cortex membranes).29 We speculate
that this discrepancy is due to the different assay systems
employed: GABA-Snifit measures the affinity of a ligand for the
heterologously expressed GABAB receptor consisting of GB1a
and GB2 in HEK 293 cells, whereas in previous reports the
affinity of a ligand for the ensemble of GABAB receptors was
measured in preparations of brain membranes. It was shown for
3-APPA that agonist affinities to native GABAB receptors differ
from those measured for recombinant GABAB.46 Furthermore
GABAB receptor isoforms may be pharmacologically hetero-
genic in vivo, and some show insensitivity toward CGP
52432.47 Thus, GABA-Snifit can be utilized in drug discovery to
characterize the specific binding affinity of GABAB agonists or
antagonists, but the measured affinity does not necessarily
reflect the affinities of such ligands in vivo.
exchanged (kperfusion = 0.77
0.04 s−1; Figure S14, SI). As
observed for sensor opening, the measured rate for sensor
closing upon GABA removal was at least partially determined
by the rate of GABA removal from the perfusion chamber
(kclosing = 0.35
0.14 s−1; obtained by fitting the change in
emission ratio to a single exponential function; SI eq 6 and
Figure S13B). In conclusion, the temporal resolution of GABA-
Snifit is in the range of seconds, but more sophisticated
experiments are needed to determine if GABA concentration
changes in the range of milliseconds can be measured.
Characterization of GABAB Receptor Agonists, Anta-
gonists, and Allosteric Modulators through GABA-Snifit.
The pharmacological manipulation of GABAB receptors is of
interest for the treatment of a wide variety of neurological
diseases and psychiatric disorders.41 The only synthetic ligand
19031
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
Allosteric modulators of GABAB receptors bind to GABAB
outside its orthosteric site, thereby altering the binding affinity
and biological activity of agonists or antagonists.48 To test if
GABA-Snifit can sense the activity of allosteric modulators of
the GABAB receptor we used the positive allosteric modulators
(PAMs) CGP 7930 (2,6-di-tert-butyl-4-(3-hydroxy-2,2-dimeth-
yl-propyl)-phenol)49 and rac-BHFF ((R,S)-5,7-di-tert-butyl-3-
hydroxy-3-trifluoromethyl-3H-benzofuran-2-one).50 CGP 7930
and rac-BHFF are known to increase the affinity of GABAB
receptor agonists but not antagonists, presumably by binding to
the heptahelical domain of GB2.51 The addition of CGP 7930
GABAB in a defined conformation through a tethered ligand
might facilitate structural studies. A synthetic agonist covalently
tethered to β(2) adrenergic receptor was shown to stabilize the
receptor in an active conformation and facilitate its
crystallization.52
Another important aspect of this work is the demonstration
that GPCRs can act as binding proteins to generate fluorescent
ratiometric sensors for applications on cell surfaces on the basis
of the Snifit concept. Hundreds of different GPCRs have been
identified, and their ligands include neurotransmitters,
hormones, odorants, and pheromones. GABA-Snifit might
therefore be the first example of a family of GPCR-based Snifits
for biologically important analytes. It has been shown for more
than 20 GPCRs that they can be functionally expressed as N-
terminal SNAP-tag fusion proteins,53−55 and synthetic ligands
for numerous GPCRs have been described that could serve as
starting points for the design of tethered fluorescent
ligands.56,57
(30 μM) or rac-BHFF (10 μM) in GABA titration experiments
comp,GABA
of GABA-Snifit reduced Kd
from 90 μM 10 μM to
30 μM 3 μM with CGP 7930 and to 10 μM 1 μM with rac-
BHFF (Figure 6B). The observed 3-fold increase in agonist
affinity is in agreement with previously published data on the
CGP 7930-dependence of [3H]3-APPA binding to GABAB
receptors in rat cortical membranes.49 For rac-BHFF a
nonselective GABA-binding assay is reported where rac-
BHFF enhanced [3H]-GABA binding in rat cerebral cortex
by 2-fold.50 However, rac-BHFF is 20-fold more potent than
CGP 7930 in increasing GABA efficacy at GABAB receptors.48
Our data confirm that rac-BHFF is a more potent PAM than
CGP 7930. In conclusion, GABA-Snifit is capable of sensing the
activity of positive allosteric modulators of GABAB, broadening
its use for applications in drug discovery. These results also
demonstrate that the interaction of ligands other than GABA
with GABA-Snifit can interfere with GABA sensing.
CONCLUSION
■
We have introduced the first fluorescent sensor for the
important neurotransmitter GABA based on the GABAB
receptor. The sensor should become a powerful tool for
studying the role of GABA in biology and for quantifying the
relative binding affinities of GABAB receptor agonists,
antagonists and the effect of allosteric modulators. Further-
more, GABA-Snifit should serve as blueprint for the generation
of other GPCR-based fluorescent sensors for biologically
relevant extracellular metabolites.
DISCUSSION
■
GABA-Snifit is the first fluorescent ratiometric sensor for
measuring GABA concentrations on the surface of live cells.
GABA-Snifit senses GABA in a physiologically relevant
concentration range from low μM to mM with high
spatiotemporal resolution. Considering (i) the ubiquitous role
of GABA in neurobiology, embryonic development and adult
neurogenesis, and (ii) the absence of other suitable sensors for
measuring GABA concentrations with appropriate spatiotem-
poral resolution, GABA-Snifit should become a valuable
research tool in biology. An important feature of GABA-Snifit
is its modular design which facilitates an adaption of the sensor
toward a particular application. For example, by exchanging its
underlying fluorophores different FRET pairs can be readily
generated. This flexibility is important when GABA-Snifit will
be used simultaneously with other, less flexible fluorescent
sensors in multiplexing experiments. Furthermore, previous
studies on the structure−activity relationship (SAR) of GABAB
antagonist CGP 51783 suggest how to modify the tethered
ligand to either improve or decrease its affinity.32 Changing the
affinity of the tethered ligand of GABA-Snifit would shift the
concentration range at which the sensor could be used.
Another application of GABA-Snifit is to sense the affinity of
synthetic ligands of GABAB. Such ligands can be agonists,
antagonists, or allosteric modulators, all of which are of
importance in drug research.
METHODS
■
Synthesis. The detailed synthetic procedures and compound
characterizations are described in the SI.
Cell Culture and Transfection. HEK 293 cells were grown on
polyornithine-coated glass coverslips (Ø 15 mm) in DMEM Glutamax
medium (Lonza) supplemented with 10% fetal bovine serum (Lonza).
Transient transfection was performed using Lipofectamine 2000
(Invitrogen) and a total of 0.75 μg DNA (SNAP-CLIP-GB1a/GB2 =
1:1) and a DNA/lipofectamine ratio of 0.4.
SNAP- and CLIP-Tag Labeling on the Surface of Live
Mammalian Cells. At 24−48 h after transfection HEK 293 cells
were labeled by incubation with a solution of 2 μM of the fluorescent
O6-benzylguanine (BG) and 10 μM of the fluorescent O2-
benzylcytosine (BC) derivatives in Hank’s buffered salt solution
(HBSS; Lonza) complemented with 10 mg/mL BSA (Sigma-Aldrich)
for 20 min at 37 °C followed by four washing steps with HBSS.
Confocal Microscopy. Twenty-four hours after transfection HEK
293 cells were labeled with BC-Alexa Fluor 488 23 and SNAP-Surface
647 (New England Biolabs) and imaged using a Zeiss LSM 700
confocal microscope equipped with a 40 × plan Apochromat 1.3
numerical aperture (NA) oil immersion objective lens. The imaging
was performed using a 488 nm laser line for excitation of Alexa Fluor
488 and a 639 nm laser line for excitation of DY-647. Fluorescence was
collected at 510−560 nm and at 650−700 nm for BC-Alexa Fluor 488
and DY-647, respectively. The settings for scanning were: × 2 zoom,
image format 1024 × 1024 pixels, pinhole 1 Airy unit (AU), average 16
frames.
Wide-Field Microscopy and Live Cell FRET Imaging. Perfusion
experiments were performed by transferring the glass coverslips with
the labeled HEK 293 cells to a Warner imaging chamber (RC-20) and
perfusing it with HBSS or HBSS containing different GABAB receptor
ligands (GABA (Sigma-Aldrich); R,S-Baclofen, 3-APPA, CGP 52432
and CGP 7930 (Abcam)) gravity-fed at a flow rate of 0.5 mL/min.
Imaging of sensors time-course experiments was performed using a
Leica LAS AF 7000 wide-field microscope equipped with a 40 × plan
Apochromat 1.25 NA oil immersion objective lens and a xenon arc
GABA-Snifit might also be used for studying the function of
GABAB. The tethering of an antagonist to GABA-Snifit locks
the receptor in an inactive conformation until GABA
concentrations rise to μmolar levels. As the activation of
GABA-Snifit by GABA or other agonists results in an emission
ratio change, a simultaneous visualization of receptor activation
and internal signaling events for which appropriate fluorescent
sensors are available should be possible. Furthermore, locking
19032
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
lamp. For recording each frame the donor and the FRET channel were
measured consecutively with an interval between 10 and 0.7 s,
depending on the experiment. The image size was 293 μm × 293 μm if
not stated otherwise. The following filter sets were used for the FRET
ratio imaging: for DY-547/Cy5, excitation at 530 nm (bandwidth 35
nm), emission at 580 nm (bandwidth 40 nm) (DY-547) and at 700
nm (bandwidth 72 nm) (Cy5); for Alexa Fluor 488/Cy5, excitation at
470 nm (bandwidth 40 nm), emission at 580 nm (bandwidth 40 nm)
(Alexa Fluor 488) and at 700 nm (bandwidth 72 nm) (Cy5); for Alexa
Fluor 488/TMR, excitation at 470 nm (bandwidth 40 nm), emission at
520 nm (bandwidth 40 nm) (Alexa Fluor 488) and at 605 nm
(bandwidth 70 nm) (TMR); for Alexa Fluor 488/Alexa Fluor 594,
excitation at 470 nm (bandwidth 40 nm), emission at 520 nm
(bandwidth 40 nm) (Alexa Fluor 488) and at 632 nm (bandwidth 60
nm) (Alexa Fluor 594).
SDS-PAGE Analysis. After 72 h of transient protein expression
transfected HEK 293 cells were labeled with 2 μM SNAP-Surface 647
and 10 μM CLIP-Surface 547 (New England Biolabs) for 20 min at 37
°C. The labeled cells were lysed by rocking them for 45 min at 4 °C
with RIPA buffer (50 mM TrisHCl, pH 7.4, 150 mM NaCl, 1% Igepal
CA-630, 0.5% sodium deoxycholate, 0.1% SDS, protease inhibitor
cocktail tablets (Roche) in presence of 100 μM SNAP-Cell Block and
CLIP-Cell Block (New England Biolabs). The supernatant was cleared
by spinning the cells at 10000 rpm at 4 °C for 30 min, incubated for 1
h at RT with SDS loading buffer and analyzed on a 7.5% SDS-
polyacrylamide gel by electrophoresis (SDS-PAGE) and in-gel
fluorescence scanning.
REFERENCES
■
(1) Roberts, E. GABA and Benzodiazepine Receptors; CRC Press:
Boca Raton, FL, 1988.
(2) Paredes, R. G.; Agmo, A. Neurosci. Biobehav. Rev. 1992, 16, 145−
170.
(3) Owens, D. F.; Kriegstein, A. R. Nat. Rev. Neurosci. 2002, 3, 715−
727.
(4) Erdo, S. L.; Wolff, J. R. J. Neurochem. 1990, 54, 363−372.
(5) Rothman, D. L.; Petroff, O. A. C.; Behar, K. L.; Mattson, R. H.
Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 5662−5666.
(6) Yoon, J. H.; Maddock, R. J.; Rokem, A.; Silver, M. A.;
Minzenberg, M. J.; Ragland, J. D.; Carter, C. S. J. Neurosci. 2010, 30,
3777−3781.
(7) Kehr, J. J. Chromatogr., B. 1998, 708, 49−54.
(8) Buck, K.; Voehringer, P.; Ferger, B. J. Neurosci. Methods 2009,
182, 78−84.
(9) Niwa, O.; Kurita, R.; Horiuchi, T.; Torimitsu, K. Anal. Chem.
1998, 70, 89−93.
(10) Wang, T. T.; Muthuswamy, J. Anal. Chem. 2008, 80, 8576−
8582.
(11) Frommer, W. B.; Davidson, M. W.; Campbell, R. E. Chem. Soc.
Rev. 2009, 38, 2833−2841.
(12) Okumoto, S.; Looger, L. L.; Micheva, K. D.; Reimer, R. J.;
Smith, S. J.; Frommer, W. B. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
8740−8745.
(13) Fehr, M.; Okumoto, S.; Deuschle, K.; Lager, I.; Looger, L. L.;
Persson, J.; Kozhukh, L.; Lalonde, S.; Frommer, W. B. Biochem. Soc.
Trans. 2005, 33, 287−290.
Intracellular Calcium Measurements. HEK 293 cells were
reversely transfected with plasmids encoding SNAP-CLIP-GB1a
together with GB2 and a chimeric protein Gqi9 using Lipofectamine
2000 (Invitrogen) seeding them in 96-well clear-bottom black-well
plates (Corning) pretreated with poly ornithin (Sigma-Aldrich) at
100,000 cells/well. Forty-eight hours after transfection intracellular
calcium measurements were performed as previously described28 after
addition of various GABA concentrations.
(14) Honda, A.; Adams, S. R.; Sawyer, C. L.; Lev-Ram, V.; Tsien, R.
Y.; Dostmann, W. R. G. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 2437−
2442.
(15) Nagai, Y.; Miyazaki, M.; Aoki, R.; Zama, T.; Inouye, S.; Hirose,
K.; Iino, M.; Hagiwara, M. Nat. Biotechnol. 2000, 18, 313−316.
(16) Romoser, V. A.; Hinkle, P. M.; Persechini, A. J. Biol. Chem.
1997, 272, 13270−13274.
(17) Cicchetti, G.; Biernacki, M.; Farquharson, J.; Allen, P. G.
Biochemistry 2004, 43, 1939−1949.
(18) Bogner, M.; Ludewig, U. J. Fluoresc. 2007, 17, 350−360.
(19) Brun, M. A.; Tan, K. T.; Nakata, E.; Hinner, M. J.; Johnsson, K.
J. Am. Chem. Soc. 2009, 131, 5873−5884.
(20) Brun, M. A.; Griss, R.; Reymond, L.; Tan, K. T.; Piguet, J.;
Peters, R. J.; Vogel, H.; Johnsson, K. J. Am. Chem. Soc. 2011, 16235−
16242.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthetic procedures, protein sequences, additional methods,
and additional figures for the characterization of GABA-Snifit
on the cell surface. This material is available free of charge via
the Internet at http://pubs.acs.org.
(21) Brun, M. A.; Tan, K. T.; Griss, R.; Kielkowska, A.; Reymond, L.;
Johnsson, K. J. Am. Chem. Soc. 2012, 134, 7676−7678.
(22) Keppler, A.; Gendreizig, S.; Gronemeyer, T.; Pick, H.; Vogel,
H.; Johnsson, K. Nat. Biotechnol. 2003, 21, 86−89.
(23) Gautier, A.; Juillerat, A.; Heinis, C.; Correa, I. R.; Kindermann,
M.; Beaufils, F.; Johnsson, K. Chem. Biol. 2008, 15, 128−136.
(24) Blein, S.; Ginham, R.; Uhrin, D.; Smith, B. O.; Soares, D. C.;
Veltel, S.; McIlhinney, R. A.; White, J. H.; Barlow, P. N. J. Biol. Chem.
2004, 279, 48292−48306.
AUTHOR INFORMATION
Corresponding Author
kai.johnsson@epfl.ch
■
Present Addresses
†Plateforme ARPEGE, Institut de Gen
́
omique Fonctionnelle,
IGF NORD, Bureau 225, 141, rue de la Cardonille, 34094
Montpellier, France.
‡Laboratory of Chemical Biology and Molecular Imaging,
University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-
0033, Japan.
(25) Biermann, B.; Ivankova-Susankova, K.; Bradaia, A.; Aziz, S. A.;
Besseyrias, V.; Kapfhammer, J. P.; Missler, M.; Gassmann, M.; Bettler,
B. J. Neurosci. 2010, 30, 1385−1394.
(26) Jones, K. A.; Borowsky, B.; Tamm, J. A.; Craig, D. A.; Durkin,
M. M.; Dai, M.; Yao, W. J.; Johnson, M.; Gunwaldsen, C.; Huang, L.
Y.; Tang, C.; Shen, Q.; Salon, J. A.; Morse, K.; Laz, T.; Smith, K. E.;
Nagarathnam, D.; Noble, S. A.; Branchek, T. A.; Gerald, C. Nature
1998, 396, 674−679.
(27) Duthey, B.; Caudron, S.; Perroy, J.; Bettler, B.; Fagni, L.; Pin, J.
P.; Prezeau, L. J. Biol. Chem. 2002, 277, 3236−3241.
(28) Maurel, D.; Comps-Agrar, L.; Brock, C.; Rives, M. L.; Bourrier,
E.; Ayoub, M. A.; Bazin, H.; Tinel, N.; Durroux, T.; Prezeau, L.;
Trinquet, E.; Pin, J. P. Nat. Methods 2008, 5, 561−567.
(29) Froestl, W. Adv. Pharmacol. 2010, 58, 19−62.
(30) Kaupmann, K.; Huggel, K.; Heid, J.; Flor, P. J.; Bischoff, S.;
Mickel, S. J.; McMaster, G.; Angst, C.; Bittiger, H.; Froestl, W.; Bettler,
B. Nature 1997, 386, 239−246.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Swiss National Science
Foundation, the European Research Training Network ENEFP
(A.M.), and a JSPS Research Fellowship for Research Abroad
(K.U.). We thank Dr. W. Froestl, Dr. G. Lukinavicius, Dr. M.
Brun, and R. Griss for valuable discussions, Dr. J. P. Pin for the
GABAB receptor and the Gqi9 plasmids, Dr. C. Trefzer for
technical assistance, and N. Tappin for corrections of the
manuscript.
19033
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034

Journal of the American Chemical Society
Article
(31) Froestl, W.; Bettler, B.; Bittiger, H.; Heid, J.; Kaupmann, K.;
Mickel, S. J.; Strub, D. Neuropharmacology 1999, 38, 1641−1646.
(32) Froestl, W.; Mickel, S. J.; von Sprecher, G.; Bittiger, H.; Olpe,
H.-R. Pharmacol. Commun. 1992, 2, 52−56.
(33) Franek, M.; Pagano, A.; Kaupmann, K.; Bettler, B.; Pin, J. P.;
Blahos, J. Neuropharmacology 1999, 38, 1657−1666.
(34) Comps-Agrar, L.; Kniazeff, J.; Norskov-Lauritsen, L.; Maurel,
D.; Gassmann, M.; Gregor, N.; Prezeau, L.; Bettler, B.; Durroux, T.;
Trinquet, E.; Pin, J. P. EMBO J. 2011, 30, 2336−2349.
(35) Kniazeff, J.; Galvez, T.; Labesse, G.; Pin, J. P. J. Neurosci. 2002,
22, 7352−7361.
(36) Grabauskas, G. Neurosci. Lett. 2005, 373, 10−15.
(37) Galvez, T.; Duthey, B.; Kniazeff, J.; Blahos, J.; Rovelli, G.;
Bettler, B.; Prezeau, L.; Pin, J. P. EMBO J. 2001, 20, 2152−2159.
(38) Liu, J. F.; Maurel, D.; Etzol, S.; Brabet, I.; Ansanay, H.; Pin, J. P.;
Rondard, P. J. Biol. Chem. 2004, 279, 15824−15830.
(39) Connors, B. W.; Malenka, R. C.; Silva, L. R. J. Physiol. 1988, 406,
443−468.
(40) Chu, D. C.; Albin, R. L.; Young, A. B.; Penney, J. B. Neuroscience
1990, 34, 341−357.
(41) Marshall, F. H. J. Mol. Neurosci. 2005, 26, 169−176.
(42) Hill, D. R.; Bowery, N. G. Nature 1981, 290, 149−152.
(43) Ong, J.; Harrison, N. L.; Hall, R. G.; Barker, J. L.; Johnston, G.
A. R.; Kerr, D. I. B. Brain Res. 1990, 526, 138−142.
(44) Lanza, M.; Fassio, A.; Gemignani, A.; Bonanno, G.; Raiteri, M.
Eur. J. Pharmacol. 1993, 237, 191−195.
(45) Froestl, W.; Mickel, S. J.; Hall, R. G.; von Sprecher, G.; Strub,
D.; Baumann, P. A.; Brugger, F.; Gentsch, C.; Jaekel, J.; Olpe, H. R.;
et al. J. Med. Chem. 1995, 38, 3297−3312.
(46) Hirst, W. D.; Babbs, A. J.; Green, A.; Minton, J. A.; Shaw, T. E.;
Wise, A.; Rice, S. Q.; Pangalos, M. N.; Price, G. W. Biochem.
Pharmacol. 2003, 65, 1103−1113.
(47) Bonanno, G.; Fassio, A.; Schmid, G.; Severi, P.; Sala, R.; Raiteri,
M. Br. J. Pharmacol. 1997, 120, 60−64.
(48) Urwyler, S. Pharmacol. Rev. 2011, 63, 59−126.
(49) Urwyler, S.; Mosbacher, J.; Lingenhoehl, K.; Heid, J.; Hofstetter,
K.; Froestl, W.; Bettler, B.; Kaupmann, K. Mol. Pharmacol. 2001, 60,
963−971.
(50) Malherbe, P.; Masciadri, R.; Norcross, R. D.; Knoflach, F.;
Kratzeisen, C.; Zenner, M. T.; Kolb, Y.; Marcuz, A.; Huwyler, J.;
Nakagawa, T.; Porter, R. H.; Thomas, A. W.; Wettstein, J. G.; Sleight,
A. J.; Spooren, W.; Prinssen, E. P. Br. J. Pharmacol. 2008, 154, 797−
811.
(51) Binet, V.; Brajon, C.; Le Corre, L.; Acher, F.; Pin, J. P.; Prezeau,
L. J. Biol. Chem. 2004, 279, 29085−29091.
(52) Rosenbaum, D. M.; Zhang, C.; Lyons, J. A.; Holl, R.; Aragao, D.;
Arlow, D. H.; Rasmussen, S. G.; Choi, H. J.; Devree, B. T.; Sunahara,
R. K.; Chae, P. S.; Gellman, S. H.; Dror, R. O.; Shaw, D. E.; Weis, W.
I.; Caffrey, M.; Gmeiner, P.; Kobilka, B. K. Nature 2011, 469, 236−
240.
(53) Zwier, J. M.; Roux, T.; Cottet, M.; Durroux, T.; Douzon, S.;
Bdioui, S.; Gregor, N.; Bourrier, E.; Oueslati, N.; Nicolas, L.; Tinel, N.;
Boisseau, C.; Yverneau, P.; Charrier-Savournin, F.; Fink, M.; Trinquet,
E. J. Biomol. Screen. 2010, 15, 1248−1259.
(54) Delille, H. K.; Becker, J. M.; Burkhardt, S.; Bleher, B.;
Terstappen, G. C.; Schmidt, M.; Meyer, A. H.; Unger, L.; Marek, G. J.;
Mezler, M. Neuropharmacology 2012, 62, 2184−2191.
(55) Leyris, J. P.; Roux, T.; Trinquet, E.; Verdie, P.; Fehrentz, J. A.;
Oueslati, N.; Douzon, S.; Bourrier, E.; Lamarque, L.; Gagne, D.;
Galleyrand, J. C.; M’Kadmi, C.; Martinez, J.; Mary, S.; Baneres, J. L.;
Marie, J. Anal. Biochem. 2011, 408, 253−262.
(56) Middleton, R. J.; Kellam, B. Curr. Opin. Chem. Biol. 2005, 9,
517−525.
(57) Kuder, K.; Kiec-Kononowicz, K. Curr. Med. Chem. 2008, 15,
2132−2143.
19034
dx.doi.org/10.1021/ja306320s | J. Am. Chem. Soc. 2012, 134, 19026−19034