WO2006105985A1

WO2006105985A1 - Reversible sequential binding assays

Info

Publication number: WO2006105985A1
Application number: PCT/EP2006/003201
Authority: WO
Inventors: Christoph Schreiter; Jean-Manuel Segura; Horst Vogel; Rudolf Hovius; Karen Martinez; Marinela Gjoni
Original assignee: Epfl-Swiss Federal Institute Of Technology-Lausanne
Priority date: 2005-04-07
Filing date: 2006-04-07
Publication date: 2006-10-12

Abstract

The present invention relates to a novel method for determining a receptor- ligand interaction which is suitable for ultra sensitive pharmacological receptor profiling on cells, e.g. single living cells based on the repetitive application of reversibly binding fluorescent ligands with fast association- dissociation kinetics. The method allows the determination of single receptor molecules and ligand binding cycles on single cells in microfluidic structures. Thus, the present method is important for high throughput screening of compounds, membrane receptors and tissues. Further, novel fluorescent derivatives of α-conotoxins, particularly of α-conotoxin Gl as described.

Description

Reversible Sequential Binding Assays

Description

The present invention relates to a novel method for determining a receptor- ligand interaction which is suitable for ultra sensitive pharmacological receptor profiling on cells, e.g. single living cells based on the repetitive application of reversibly binding fluorescent ligands with fast association- dissociation kinetics. The method allows the determination of single receptor molecules and ligand binding cycles on single cells in microfluidic structures. Thus, the present method is important for high throughput screening of compounds, membrane receptors and tissues. Further, novel fluorescent derivatives of α-conotoxins, particularly of α-conotoxin Gl are described.

With the advent of combinatorial chemistry, there is a growing need for efficient strategies to screen the pharmacological properties of compounds in the search for novel drugs. Pharmacological investigations are usually performed on solubilized and purified proteins or on cultured cell lines stably overexpressing the receptor used as drug target. However, it is becoming increasingly apparent that pharmacology in such purified systems or in heterologous cell lines might differ from pharmacology in native tissue cells (1). Because primary cells are often very valuable and difficult to obtain and maintain in culture, it is highly desirable to optimize their use by investigating only single cells.

To measure ligand-receptor protein interactions, fluorescence binding assays have the advantage that fluorescent labels provide a much higher sensitivity while being less expensive and easier to handle than commonly used radio ligands (2, 3). In addition, they can be easily applied in vivo.

However, detailed investigations on cells involving repetitive measurements have been considered impossible due to rapid photobleaching of the fluorescent labels and the irreversible binding of previously used labelled ligands. As used herein, fluorescence and fluorescent entail (i) the radiation emitted upon excitation with a proper wavelength, i.e. fluorescence, phosphoresence, and (ii) light absorption and reflection by metal ions.

US patent 4,134,792 discloses a specific binding assay method employing, as a labelling substance, a reversible binding enzyme modulator for the detection of a ligand in a liquid medium. US patent 4,461 ,829 discloses a homogenous specific binding assay element and a method for its use in determining a ligand in or the ligand-binding capacity of a liquid sample. US patent RE34.394 discloses a double receptor, specific binding assay using a receptor complex which is adsorbed onto an insoluble surface. WO 00/075364 and WO 03/081209 describe a method for rapid identification of bi-ligand drug candidates by NMR. WO 00/075657 discloses a method for the screening of target ligand interactions using fluorescent ligands and targets immobilized on solid surfaces.

WO 00/40735 and WO 01/01138 disclose a method for assaying binding between proteins and peptides wherein fluorescence quenching is detected. WO 02/103321 and WO 02/085667 disclose methods of screening for ligands of target molecules using fluorescence resonance energy transfer (FRET). WO 95/15981 discloses compositions and kits for a fluorescence polarisation assay of large molecules. None of the above documents, however, discloses or suggests a fluorescence assay on living cells comprising the determination of reversible analyte binding.

The present invention provides a novel approach to perform fluorescence binding assays, the reversible sequential (ReSeq) binding assay, that solves problems associated with previous measurements of ligand-receptor interactions by applying repetitively-completely reversible and specific, fluorescently labelled ligands with fast association-dissociation kinetics. ReSeq fundamentally differs from existing fluorescent binding assays in that it sequentially performs multiple measurements on the same samples (one or multiple cells) instead of performing one or a small number of measurements on multiple samples as usually done. The new approach overcomes problems related to photobleaching so that cells expressing low amounts of receptors down to the single-molecule level can be investigated. Moreover, ReSeq binding assays can be easily automated and implemented in on-chip analysis.

More particularly, the present invention provides a method for determining a receptor-ligand interaction comprising the steps:

(a) providing a cell comprising a receptor,

(b) providing a ligand of the receptor wherein the ligand has a fluorescence labelling group and binds reversibly to the receptor,

(c) contacting the cell with the ligand under conditions wherein the ligand associates with the receptor and then dissociates therefrom,

(d) monitoring the fluorescence of the cell during step (c) and

(e) repeatedly performing steps (c) and (d).

According to step (a), a cell is provided which comprises a receptor. Preferably, a living cell is provided. The cell may be of any type, e.g. a prokaryotic or eukaryotic cell. Preferably, the cell is a eukaryotic cell such as a mammalian cell, particularly a human cell. The cell may be a cultured cell, for example, a cell from a cultured cell line, e.g. a human embryo kidney (HEK) 293 cell or a CHO cell, or a primary cell, for example a cell derived from an . organism, e.g. body fluids or tissues of an organism. Particularly preferred examples of cells are muscle cells, e.g. C2C12 muscle cells or cells from muscle tissue or neuronal cells. In general, the method can be extended to any cell type comprising any receptor given that a reversible fluorescent ligand for the receptor is available.

The present invention may be carried out on genetically modified cells, e.g. cells which have been transformed or transfected with heterologous nucleic acid molecules encoding the receptor to be analysed. The invention, however, may also be practised on native cells, e.g. cells in which the receptor to be analysed occurs naturally. The sensitivity of the ReSeq binding assay is high enough for determining receptor-ligand interactions on cells comprising about 10⁶ or less, 10⁴ or less or 10³ or less receptor molecules.

The measurement can be carried out on multiple cells, e.g. tissue sections, or on single cells, or on individual cells within a sample. The sensitivity of the method is even high enough for determining the fluorescence of individual receptor-ligand complexes or ensembles thereof.

The method of the present invention is illustrated on the nicotinic acetylcholine receptor (nAChR) which is a pharmacologically important ligand gated ion channel receptor. However, the method can be extended to other receptors, particularly to membrane-bound receptors, e.g. all types of ligand-gated ion channels or G-protein coupled receptors (GPCRs), more generally to all receptors and other proteins on or in a cellular membrane given that a reversible fluorescent ligand is available. Preferably, the receptor is selected from gated ion channels such as the serotonergic 5-HT₃ receptor, the ionotropic γ-amino butyric acid receptor, glutamate receptors, the glycine receptor and the nAChR.

According to step (b) a ligand is provided which has a fluorescence labelling group and binds reversibly to the receptor. The ligand may be any molecule, e.g. an antibody, a polypeptide, a peptide or a non-peptidic compound which can be derivatized with a fluorescence labelling group and which has reversible association-dissociation characteristics under the assay conditions. Preferably, the ligand is a compound which binds specifically to a membrane-bound receptor. The fluorescence labelling group may be selected from known fluorescent molecules such as cyanineδ (Cy5), cyanine3 (Cy3), fluorescent molecules of the Alexa series, fluorescein, rhodamine, phycoerythrine, coumarine, quantum dots, lanthanides or derivatives thereof. Preferably, the fluorescence labelling group is covalently attached to the ligand, e.g. by coupling a reactive derivative of a fluorescence labelling group such as an active ester, for example, an N- hydroxysuccinimide ester to the ligand. Preferably, the ligand is a fluorescently labelled α-conotoxin, particularly an α-conotoxin Gl. α-conotoxin peptides are described in WO02/064740, US200420362 and US2004192610, for example. More preferably, the fluorescently labelled receptor is an α-conotoxin Gl which has covalently attached to an amino group thereof, a fluorescence labelling group.

The fluorescently labelled ligand is capable of binding reversibly to the receptor under assay conditions. In practice, the ligand dissociation lifetime on the receptor is preferably in the range of about 0.5 to 20 min, more preferably in the range of about 1 to about 10 min. Further, the binding constant K_D of the fluorescently labelled receptor to the ligand is preferably in the range of about 10 nM to about 10 μM, more preferably in the range of about 100 nM to about 1 μM. Further, it is preferred that ligands which have a dissociation rate constant kof of about 5 x 10^"2 s^"1 or less are used, more preferably of 1 x 10^"2 s^"1 or less, as determined by fluorescence measurements.

In order to allow the determination of a receptor-ligand interaction, the cell is contacted in step (c) with a fluorescently labelled receptor under conditions wherein the ligand associates with the receptor and subsequently dissociates therefrom. In the course of this association-dissociation cycle, the fluorescence of the cell is measured repeatedly. In a preferred embodiment, the cell is first subjected to conditions wherein the ligand binds to the receptor, e.g. by incubating the cell in a medium which contains the fluorescently labelled ligand in sufficient amounts to obtain the desired ligand-binding to the receptor molecules on the cell. Preferably, the cell is contacted with the fluorescently labelled ligand in a concentration of about 1 nM to about 1 μM. Then, the cell is preferably subjected to conditions wherein the ligand is removed from the receptor, e.g. by incubating the cells in a medium which contains the fluorescently labelled ligand in lower concentrations or does not contain the fluorescently labelled ligand and/or which contains a competitor of the fluorescently labelled ligand. The fluorescence is measured repeatedly, at least during the removal step, in order to determine characteristics of the receptor-ligand interaction under the specific test conditions.

Step (d) of the method of the invention requires monitoring the fluorescence of the cells. Preferably, at least 3 images or measurements, more preferably at least 10 images or measurements and most preferably at least 50 images or measurements, (e.g. Figure 2a shows in each row 4 representative images out of series of 150 images) are carried out preferably at an interval of about 10 s up to 60 min, more preferably in an interval of about 30 s to about 20 min (e.g. Figure 2).

Surprisingly it was found that steps (c) and (d) can be carried out repeatedly on the same cell, i.e. a cell may be subjected to a plurality of ligand association-dissociation cycles. For example, a plurality of subsequent measurements which may be the same or different measurements can be carried out at least twice, e.g. at least three times on the same cell.

During step (e) of the method of the invention, steps (c) and (d) are preferably repeated at least 3 times, more preferably 10 times or more. Repetition can be performed without changing experimental conditions, for instance to obtain better statistics and accuracy, to monitor the evolution of the cells or to measure varying concentrations of receptor at the cell membrane. In this way the ReSeq method allows the measurement of ligand-receptor interactions at low expression levels. Alternatively, the experimental parameters, such as recording frequency or illumination intensity, can be modified, for instance to obtain kinetic information on the ligand-receptor interaction (e.g. Figure 2). Furthermore, the incubation buffer can be changed, for instance by addition of competitive ligands at various concentrations (e.g. Figure 3), preferably antagonists, agonists, or binding modulators. This way, the method allows the determination of the pharmacological properties of unlabelled compounds. In addition, environmental parameters can be varied, such as temperature, pH, ionic strength, and/or the state of the cell can be influenced, for instance by activation of signalling cascades.

The method of the invention allows the accurate determination of receptor- ligand interaction parameters such as association rate (k_on), dissociation rate (ko_ff) and dissociation constant (K_D). The method may be carried out in the presence of further compounds, e.g. unlabelled derivatives of the fluorescently labelled ligands, different known ligands of the receptor or test substances, the binding of which to the receptor is to be tested. For these purposes, different measurement protocols, e.g. kinetic measurements, equilibrium measurements, competition measurements may be carried out. For example, association-dissociation characteristics of the fluorescently labelled ligand in the presence of a further compound which is known or suspected to be a receptor ligand may be determined. In a particularly preferred embodiment, the method is used in a screening procedure for the identification and/or characterisation of pharmaceutical drug candidate molecules.

The method of the present invention can be used to characterize single or multiple cells by determining the presence of receptors or other proteins on the surface of these cells. The high sensitivity of the method allows the detection of little abundant proteins, thus enabling characterisation of native cells.

The method of the present invention can be practised with cells immobilized on a solid surface, or in a matrix, e.g. a gel, or with cells in a liquid medium. Particularly preferred are test formats wherein a plurality of different cells may be tested in parallel. A preferred embodiment involves the measurement of cells immobilized on the surface of chips. Another preferred embodiment involves the measurement of cells present in microfluidic structures, e.g. held by an optical trap or an electrical cage, in a microchannel. These test formats are particularly suitable for high- throughput screening protocols. The fluorescence measurement may be performed according to known methods, wherein the fluorescence labelling group of the ligand is excited by irradiation with light of a suitable excitation wavelength from a light source, e.g. a laser, and detecting the emitted fluorescent light is detected with a suitable detection device, e.g. a photo diode or CCD camera. Generally, the detection device is provided with electronic equipment to allow process control and data acquisition, storage and reproduction.

Furthermore, the present invention relates to a fluorescently labelled α- conotoxin, particularly an α-conotoxin Gl and the use thereof in a method for determining receptor-ligand interactions, particularly in a method as described above. The fluorescently labelled α-conotoxin preferably has a fluorescent group covalently attached to an amino group, particularly the N- terminal amino group thereof. Further, the fluorescent group displays a bright emission with a photobleaching quantum yield of preferably less than

10^'5 at wavelengths preferably longer than 550 nm to allow sensitive measurements on single cells.

Further, the present invention is explained in more detail by the following figures and examples.

Figure Legends

Figure 1: (a) Whole cell current responses of HEK293 cells expressing nAChR. Upper trace: 2 s pulses of ACh were applied-every 60 s (short peaks) and 300 nM α-conotoxin Gl (α-CnTx) was added for 480 s as indicated. The lower trace shows the current response. Peak current responses were constant in absence of antagonist (T = 0 - 100s), while addition of α-CnTx-Cy5 and subsequent washing resulted in a decrease (T = 130 - 600s), respectively increase (T = 600 - 1400s), of the current peaks, yielding rate constants for association and dissociation, (i) and (ii): Magnifications of the current responses in the absence and presence of α- CnTx-Cy5, respectively, showing that only the amplitude but not the shape of the response is modified by the antagonist.

(b) Inhibition of channel currents by α-CnTx and its fluorescent analogous. (+) α-CnTx, (A) α-CnTx-Cy-5, (T) α-CnTx-Cy-3, and (■)<-(- CnTx- Alexa 647. Fits of the inhibition curves yielded IC₅₀ values of 20 nM for the WT (solid curve) and 95 nM (dotted curve), 410 nM (dashed curve), and 159 nM (dotted-dashed curve) for the Cy-5, Cy-3 and Alexa647 toxin derivates, respectively. The Hill coefficient stayed constant

Figure 2: Reversible binding of α-CnTx-Cy5 to nAChRs in HEK293 cells, (a) Fluorescence images in a row from left to right show a single cell during continuous washing after 2 min incubation with 30 nM α-CnTx~Cy5. After complete dissociation, α-CnTx-Cy5 is bound again to the same cell and washing cycles are repeated. Image series are representative images of the fluorescence decay after 1st labelling (1) and after repeated labelling 71 min (2) and 85 min (3) later. Scale bar: 10 μm. (b) Time-traces of the fluorescence intensity integrated over the whole cell during washing. The same absolute fluorescence intensity is reached after each association- dissociation cycle with a deviation of only 6% indicated by the grey bar. (c) Binding kinetic of α-CnTx-Cy5 (30 nM) to the surface of a single cell. The average fluorescence intensity per pixel is plotted versus incubation time and fitted by a single exponential, yielding the association rate constant A_0n 4 1 1

=7 ± 5 -10 MoI^" s . (d) Fluorescence decays recorded at frequencies of 0.2, 0.5, 1, and 2 Hz on the same HEK293 cell expressing nAChR after incubation with 30 nM α-CnTx-Cy5. The apparent fluorescence decay rates k_app were evaluated from corresponding single-exponential fits (solid curves). The fluorescence decay is faster as would be expected from dissociation (dotted line) as a result of photobleaching. Inset: The photobleaching rate varies linearly with the time the cells were exposed to light, i.e. with the number of recorded images per second. By plotting the apparent fluorescence decay rate k_app versus the recording frequency, a value for the dissociation rate constant k_oβ of 1.0 ± 0.6 -10^"3 s^"1 was obtained from the intercept at zero frequency. Figure 3: Site-specific binding of α-CnTx and dTC to the α/δ site of the nAChR on single cells detected by fluorescence competitive assay, (a)

Fluorescence images of a cell incubated with 30 nM α-CnTx-Cy5 mixed with either no (i), 1OnM (ii) or 10OnM (iii) α-CnTx (Scalebar: 10μm). After washing-off the non-labelled α-CnTx, the original initial fluorescence intensity is recovered (iv). (b) Competition binding curve of dTC (■) and α- CnTx (o) compared to patch clamp data for α-CnTx (dashed line). Evaluated

IC₅₀ values were 310 ± 60 nM and 17 ± 4 nM for dTC (solid curve) and WT α-CnTx (dotted line), respectively.

Figure 4: Repetitive and reversible binding of α-CnTx-Cy5 to n ACh Rs on an optically trapped cell inside a microfluidic channel, (a) The microfluidic system comprises two inlet channels for α-CnTx-Cy5 and buffer (arrows), which merge into a central channel where a cell is trapped with optical tweezers. Repetitive binding and washing of α-CnTx-Cy5 was performed by varying the flow speed applied on the two inlet channels, (b) Overlay of two representative fluorescence decay curves (° and +) sequentially recorded on a trapped HEK293 cell during washing with buffer. Fits are shown as solid lines demonstrating that labelling and washing are highly repeatable. (c) Image series showing a trapped cell directly after incubation with 300 nM α-CnTx-Cy5 for 1 min (i) and the subsequent fluorescence decay after 2 s (ii) and 5 s(iii) followed by re-binding α-CnTx- Cy5 to the same cell (iv). Scale bar: 10μm.

Figure 5: Single-molecule images of α-CnTx-Cy5 bound to nAChRs on a HEK293 cell. Single molecules are recognized as diffraction-limited spots. Image series (a) shows the 1^st, the 6^th, and the 17^th image from left to right of a sequence recorded at 4 Hz (scale bar 5μm). (b) After complete photobleaching, the receptor bound toxin could be washed off and nAChRs could be relabelled with fresh α-CnTx-Cy5. (c) Time trace of the fluorescence of the single molecule marked with a white circle in images (a) featuring characteristic single-step photobleaching. (d) Transmission image of the cell. Figure 6: Specific binding of α-CnTx-Cy5 to muscle-type nAChRs.

Confocal microscope images of HEK293 cells expressing either muscle type nAChR coexpressed with cytosolic GFP (green) or neuronal 07/5-HT3 chimera coexpressed with cytosolic CFP (blue), (a) The non-specific antagonist α-BgTx-Alexa647 (red) binds to both receptor types, (b) α-CnTx- Cy5 (red) binds exclusively to cells expressing muscle type nAChRs. Scale bar: 25 μm.

Example

1. Introduction

To illustrate the novel approach of the present invention, the nicotinic acetylcholine receptor (nAChR) was chosen as the prototype of a pharmacologically important family of ligand gated ion channels comprising the serotonergic 5-HT₃ receptor, the ionotropic y-aminobutyric acid and glycine receptors. Different subtypes of nAChRs are found in the postsynaptic membrane of muscle and nerve cells. While muscle type receptors are composed of 4 different subunits (2 x α1 , β1 , Y / ε, δ), neuronal receptors can be homopentamers (α7 / αδ) or consist of two different subunits (2 x α, 3 x β) (4, 5).

Ligand binding to the nAChRs is usually investigated by competition assays using either expensive radiolabeled α-bungarotoxin, or a fluorescently labeled α-bungarotoxin derivate. Both labels are based on the snake toxin α- bungarotoxin (α-BgTx) and bind irreversibly to all types of nAChRs. So far, only few site-specific reversible weak fluorescent ligands for nAChRs have been published (6-8). Toxins from marine snails have recently gained an increasing interest for novel medical applications because of their outstanding receptor-ligand specificity (9, 10). The venom of marine snails contains different α-conotoxins, which target specifically either neuronal or muscle type nAChR as reviewed in (H). Thus, they can serve as tools to elucidate structure and function of nAChRs (12). It has been shown that fluorescein labelled α-CnTxs bind specifically to purified Torpedo nAChRs (13, 14). Here, WΘ describe novel fluorescent derivates of α-conotoxin Gl (α- CnTx), a small 13 amino acid peptide from the snail conus geographicus (15), which bind with high affinity, specificity for the α/δ site and strong selectivity for the muscle-type nAChR (16).

Using the reversible binding with fast association-dissociation rates of the novel fluorescent α-CnTx analogues, kinetic and competitive reversible binding assays were performed on single cells and binding constants of the unlabelled compounds d-tubocurarine (dTC) and α-CnTx were measured.

2. Materials and Methods

2.1. Labelling α-conotoxin Gl α-conotoxin Gl ((α-CnTx) Bachem, Bubendorf, CH) was labelled with N- hydroxysuccinimide (NHS) esters of Alexa 647 (Molecular Probes, Eugene, OR, USA), Cy5 and Cy3 (Amersham, Buckinghamshire, UK) to yield α- CnTx-A647, α-CnTx-Cy5 and α-CnTx-Cy3, respectively. Typically, 1 mg α- CnTx dissolved in 100 μl 100 mM NaHCO₃ at pH 8.5 was added to 70 μl of a 10 mg/ml solution of fluorophore-NHS-ester in DMF and incubated overnight at room temperature in the dark. After evaporating the solvent, the residue was dissolved in a minimal amount of methanol, applied to a silica gel G60 preparative thin layer chromatography plate (0.25 mm, Merck, Darmstadt, D), dried thoroughly and developed in MeOH/25% NH₄OH (95/5, v/v). The product was extracted with methanol and its purity demonstrated by thin layer chromatography. Identity of the product was established by electron spray (ESl) or matrix-assisted laser desorbtion ionisation (MALDI) mass spectrometry: α-CnTx-Cy5 (calculated: m/z = 2075.8, measured

MALDI: (m+3H⁺)/z =2078.8, measured ESI: (m+3H⁺)/2z =1038.3); α-CnTx-

Cy3 (calculated: m/z = 2049.7, measured ESI: (m+3H⁺)/2z =1025.3); α- CnTx-A647 (calculated: m/z = chemical formula unknown, measured ESI: m/nz = 1051.7).

2.2 Transient transfection ofHEK293 cells Human embryonic kidney (HEK293) cells were cultured in Dulbecco's modified Eagle medium supplemented with 2.2 % fetal calf serum in a humidified 5% CO₂ atmosphere at 37 ⁰C. HEK293 cells (60-80% confluent), growing either on 25 mm glass slides in 6-well plates or 35 mm cell culture

dishes, were transfected using Effectene lipofection according to the protocol of the manufacturer (Qiagen, Hilden, D) with 0.2 μg of plasmids containing the coding region of human muscle type nAChR subunits α (0.08 μg), β (0.04 μg), δ (0.04 μg), and ε (0.04 μg) (4, 5). 0.2 μg DNA of cytosolic green fluorescent protein (GFP; Clontech, Palo Alto, CA, USA) was cotransfected to identify transfected cells.

Experiments were performed 24-55 hours after transfection. Expression levels were typically around 1000 active receptors per cell as obtained from whole-cell patch-clamp currents. The cell culture medium was replaced prior to fluorescence and patch clamp experiments by buffer containing 147 mM NaCI₁ 12 mM glucose, 10 mM HEPES, 2 mM KCI, 1 mM MgCI₂ adjusted to pH 7.4 (NaOH). This buffer was used to dissolve the ligands and to continuously perfuse the cells.

2.3 Electrophysiology Standard patch-clamp measurements were done in whole cell configuration employing an EPC-9 patch-clamp amplifier (HEKA Elektronik Dr. Schulze GmbH, Lambrecht, D). For data acquisition and storage the software PULSE 8.3 (HEKA) was used. Borosilicate glass pipettes (resistances of 2-5 MΩ) were filled with 140 mM NaCI, 10 mM EGTA, 10 mM HEPES adjusted to pH 7.4 (NaOH). The ground electrode was connected to the bath via a 1 M KCI agar bridge. All experiments were performed at room temperature and the membrane potential was kept at -60 mV. Recorded inward currents are displayed downwards. Ligands and buffer were applied with a software controlled RSC-200 perfusion system (Bio-Logic, Claix, F). 2 or 3 s pulses of 100 μM acetylcholine (ACh) were repetitively applied every 60 s while α- CnTx or its fluorescent analogues were added to various concentrations.

The change in peak current response was evaluated as a function of time and α-CnTx concentration to extract rate constants of the inhibition kinetics, /fei, on and fa, off _T from exponential fits yielding

_on- [α-CnTx] +

_ofi) and /c_βι, o_ff, respectively. Antagonist binding curves were computed from the equilibrium peak responses for each concentration and fitted to the equation:

/ = /₀/{1+(IC₅₀/[antagonist]r^π} (Eq.1) using the Levenberg-Marquardt algorithm of Igor Pro (Wavemetrics Inc., Lake Oswego, OR, USA) where / is the peak current at a particular antagonist concentration, /₀ the peak current in absence of any antagonist, IC₅₀ the half maximal inhibitory concentration, and n the Hill coefficient.

2.4 Fluorescence Experiments

Cells grown and transfected on 0.17 mm thick glass coverslips were mounted on a modified epiluminescence wide-field microscope (Axiovert 200, Zeiss, Feldbach, CH). To investigate Cy5 and Alexa 647 fluorescence, circularly polarized light of the 632.8 nm line of a HeNe laser (Coherent, Auburn, CA, USA) was directed by a dichroic mirror (Q645LP, Chroma Corp., Rockingham, VT, USA) into a microscope objective (C-Apochromat 63x W Korr, 1.2 NA, Zeiss) to illuminate a 22 μm diameter region of the sample. Fluorescence was collected by the same objective, passed through a filter (HQ710/100, Chroma) and imaged on an intensified CCD camera (I- Pentamax 512 EFT, Roper Scientific, USA). GFP fluorescence was excited with the 488 nm line of an Ar⁺ laser (Innova Sabre, Coherent, USA). To minimize photobleaching, cells were only illuminated for 50 ms per image using a shutter (LS3T2, Vincent Associates, Rochester, USA). Illuminated cells were continuously perfused using a VC-77SP fast step perfusion system (Warner Instruments Corp, Hamden, CT, USA). 2.5 a-CnTx-Cy5 binding kinetics

The association of α-CnTx-Cy5 to the nAChR was measured repetitively on the same cell by recording fluorescence images upon incubation with 30 nM α-CnTx-Cy5 from 1 to 25 min and a rapid wash (10 s) to remove free fluorescent ligand from solution. The time dependency of average fluorescence intensity of the whole cell was fitted by single-exponentials to yield (ko_n [α-CnTx-Cy5] + kotr). After complete removal of the ligand (total wash time: ≥20 min) the measurement was repeated.

To determine the dissociation rate constant, cells were first incubated for 2 min., unless stated otherwise, with 30 nM α-CnTx-Cy5 followed by continuous washing with buffer. A series of fluorescence images was taken with a delay of 15 s after the start of the wash to ensure that no free fluorescent ligand remained in solution. After complete removal of the ligand (total wash time: ≥20 min) the measurement was repeated. All experiments were performed on at least three different cells from independent cultures. The rate constant of dissociation of α-CnTx-Cy5 from nAChRs was evaluated from the decay of fluorescence intensity during washing. This decrease in fluorescence intensity is a superposition of ligand dissociation and photobleaching of the fluorophore and can be described by:

t) = F₀ exp (~k_app ^■ t) (Eq. 2) where F is the fluorescence intensity at time t, FQ the fluorescence intensity of the first image,

the photobleaching rate, f_exp the exposure time per image, and v the frequency at which images were recorded. /c_off is obtained from the intercept of a single linear fit of the apparent fluorescence decay rate k_app = (Zc₀If +

^■ t_exp ^■ v) against v. The k_on was evaluated from image series taken at frequencies from 0.2 Hz to 8 Hz on at least three different cells from independent preparations.

2.6 Fluorescent single cell binding assay for unlabelled ligands

The affinity of unlabelled α-CnTx and δ-tubocurarine (dTC) was measured in fluorescence competition binding assays. Cells were repetitively incubated with increasing concentrations of the competitor for 8 min (control experiments at longer incubation times showed this was sufficient to reach equilibrium) followed by a incubation with a mixture of the competitor and 30 nM α-CnTx-Cy5. The initial fluorescence intensities evaluated by averaging the 2^nd image over the whole cell were fitted to:

F - F₀/{1+(IC₅₀/[compemoή)^'n} (Eq. 3) where F and F₀ are the fluorescence intensity measured in presence and absence of competitor, /C₅₀ the half maximal inhibitory concentration and n the Hill coefficient.

2.7 Single-Molecule Imaging

Single-molecule measurements were performed on the same epiluminescence microscope as for bulk fluorescence. Incubation times with 3 -30 nM α-CnTx-Cy5 were reduced to some seconds so that only about 1-5 % of all nAChR expressed on the cell surface did bind a fluorescent ligand. Single-molecules images were recorded at a frequency of 4 Hz illuminating the cells for 50 ms with excitation intensities around 0.5 kW/cm². Labeling was found to be specific as no single molecules were observed on non- transfected cells, i.e. cells not expressing nAChRs.

2.8 Discrimination between nAChR subtypes

Separate cell cultures were transfected as described above with nAChR and cytosolic GFP or with DNA encoding the ctf/δ-HT_β chimera (21) and cytosolic cyan fluorescent protein (Clontech). 24 h after transfection, cells from both cultures were mixed and co-cultured on 25 mm cover glass slides. The nAChRs were probed sequentially by incubation first for 20 min with 100 nM α-CnTx-Cy5 and then for 90 min with 10 nM of α-BgTx-Alexa647 (Molecular Probes). Toxin binding was imaged after incubation with ligand and a rapid wash with buffer to remove free fluorescent iigand. Fluorescence images were acquired on a laser scanning confocal microscope (LSM 510, Zeiss), using appropriate laser and filter settings.

2.9 Single cell fluorescent binding assay in microfluidic structures Binding experiments inside a glass microstructure were carried out using an optical trapping system. The beam of the 1064 nm light of a Nd³⁺: YVO₄ laser

(J-20;BL10-106Q, Spectra Physics, CA, USA) was expanded to slightly overfill the back aperture of a high NA microscope objective (Plan- Apochromat 63x 1.4 Oil DIC, Zeiss) on an inverted microscope (Axiovert 100TV, Zeiss) to form a diffraction-limited 50 mW optical trap in the sample plane. The optical trap was positioned in the centre of the microfluidic channel inside the microstructure (Fig.6) using steering mirrors.

For fluorescence experiments, a 30 μm region of the sample was illuminated with the 632.8 nm light of a HeNe laser (05-HLP-171 , Melles Griot, Carlsbad,

CA, USA) at an averaged power of 4 mW. The emitted fluorescence passed through a set of filters (Z488/633/1064RPC and Z488/633M, Chroma) to reject back-scattered excitation and trapping light and was imaged on a CCD camera (Pixelfly, PCO, Kelheim, D) at a frequency of 0.5 Hz with 50 ms acquisition time.

The microfluidic circuit consisted of two syringe pumps (Versaδ, Kloehn, Bonaduz, CH), two multi-position valves (EMHMA-CE and E60-CE, VICI₁USA) to handle liquids, and a microfabricated glass Pyrex 7740 microchip with 50 μm high and 110 μm wide channels (kind gift of Laura Ceriotti and Elisabeth Verpoorte, IMT, Neuchatel, CH). The microfluidic system comprised two inlet channels, one for the ligand and one for the washing buffer, which merged into a central channel. Medium around the cells was flowing from only one inlet channel at a time and was changed by switching the flow from one channel to the other. Measurements were performed by first trapping a cell and then repetitively alternate between 1 min. incubations with 300 nM α-CnTx-Cy5 and washing with buffer. 3. Results

3.1 Electrophysiological characterization of fluorescent α-CnTx analogues Binding properties of α-CnTx and its fluorescent analogues to nAChR were quantified by whole cell patch-clamp measurements. Addition of α-CnTx or its fluorescent analogues to cells expressing nAChR gradually decreased the maximal ACh induced current response to reach a final stable level (Fig.1a). Both kinetics and final level depended on α-CnTx concentration. The antagonist could bind and be washed off completely within minutes allowing repetitive measurements on the same cell, only limited by the stability of the patch-clamp seal. The rate constants

and Aτ_ei, off of the inhibition kinetics were calculated from single-exponential fits to the time-dependence of the maximal current responses. Control experiments with lower pulse rates (4 or 8 min between ACh pulses) yielded identical data indicating that binding kinetics were not affected by the repetitive ACh pulses. Dissociation kinetics of all fluorescent α-CnTx analogues were comparable to WT α-CnTx (/c_ei. off =

5 ± 2 -10^'3 s^'1), whereas association was 3 to 4-fold slower than WT α-ChTx (/fei, on = 12 ± 5 "10⁴ M^"1s^"1). Accordingly, IC₅O values evaluated for the Cy5, Cy3, and Alexa647 analogues showed 5, 20, and 8 fold-lower affinities compared to WT α-CnTx (IC₅₀ = 20 ± 2 nM), respectively (Fig. 1b). The Hill coefficients for all toxins were close to unity, indicating that the same number of bound toxin molecules is needed to block channel activation. The maximum peak currents in the absence of any α-CnTx were usually several nA corresponding to only about 10⁴ activated channels in the cell membrane. Such low concentrations are usually difficult to observe in fluorescence experiments and make standard fluorescence binding assays impossible.

3.2 Single cell fluorescent ligand binding experiments

Fluorescence experiments were performed using α-CnTx-Cy5 because of its high affinity to the nAChR and its emission wavelength in the red favourable for cellular investigations. Cells expressing nAChRs were first perfused with α-CnTx-Cy5 for several minutes; during subsequent washing with buffer, a series of images was recorded (Fig. 2a). After complete wash-off of the α- CnTx-Cy5, the measurement could be repeated many times over hours on the same cell reaching every time the same initial fluorescence intensity within 6 % deviation (Fig. 2a,b). The ligand interaction was specific to the nAChR as no binding of α-CnTx-Cy5 to cells not expressing nAChR could be observed.

Ligand binding kinetics and binding constants were evaluated from the fluorescence intensities of the image series with a precision comparable to patch clamp experiments. The association rate constant was determined from the repetitive recording of the initial fluorescence intensity after increasing incubation times, yielding Zc_0n = 7 ± 5 -10⁴ MoI^"1 s^"1 (Fig. 2c), in good agreement with patch-clamp experiments. Ligand dissociation was measured by monitoring the fluorescence decay during washing. To distinguish the fluorescence intensity decrease due to ligand dissociation from that of fluorescence bleaching, repetitive measurements were taken at various image recording frequencies as described in materials and methods.

The resulting k_ofS - 1.0 ± 0.6-10^"3 s^"1 was slightly slower (Fig. 2d) than the value observed with patch clamp. . Standard attempts to gain association or dissociation rates from the gradual increase or decrease, respectively, of the continuously monitored fluorescence intensity fail due to photobleaching caused by the high excitation intensity needed at the very low expression level of nAChR (around 1000 receptors per cell). Such low receptor concentrations are typical for native cells or tissues. The complete reversibility of α-CnTx binding allows rebinding of fresh α-CnTx-Cy5 over and over again; thus many measurements can be performed on the same cell, either under identical conditions to get better statistics or under varying conditions to get insight in e.g. dynamic properties of the system.

3.3. Compound screening on single cells The fast and reversible binding of α-CnTx-Cy5 can be used to rapidly and repetitively measure the binding constants of unlabelled substances in a competition assay. In these experiments we monitored the reduction of initial fluorescence intensity when a single cell was incubated with increasing concentrations of a competing non fluorescent ligand before applying α- CnTx-Cy5 as illustrated in Fig. 3a. Competition with non-labelled α-CnTx yielded an /CgQ(O-CnTx) = 17 ± 4 nM in excellent agreement with IC₅O = 20 ±

2 nM measured with the patch clamp technique. This shows that α-CnTx and its Cy5-analogue compete for the same binding site. Competition experiments with the toxin dTC gave a value of /C₅₀(dTC) = 310 ± 60 nM, which corresponds to the affinity of dTC for the α/δ site of the muscle type nAChR (17). This proves that both, α-CnTx-Cy5 and α-CnTx specifically bind to the α/δ site of the nAChR and hat Cy5 labelling did not alter the pharmacological properties of the antagonist.

3.4 Ligand binding on single cells in microfluidic structures Assay miniaturization combined with the use of high sensitive detectors opens the possibility to investigate large libraries of receptors and active compounds with high efficiency at massively reduced recording time and sample consumption. Here, we realized an important step towards this goal demonstrating repetitive ligand binding on a single cell in microfluidic structures. First, individual HEK293 cells expressing the nAChR were trapped by optical tweezers inside a microchannel. Individual trapped cell could be released and new cells trapped from a small volume of cell suspension (<100 nl) temporarily pumped across the focus of the laser tweezers. In the next step, α-CnTx-Cy5 was reversibly bound to the trapped cell (Fig. 4) in the same manner as described above. Only 100 nl of 300 nM α-CnTx-Cy5 were required per cycle of binding and washing.

3.5 Imaging single receptor molecules

When cells were incubated with reduced α-CnTx-Cy5 concentrations for shorter time periods, only about 1 to 5 % of the nAChR on the cell surface were labelled with the toxin. Under these conditions, individual α-CnTx-Cy5- receptor complexes could be observed as diffraction-limited spots showing the characteristic single-step photobleaching(18, 20) (Fig. 5). Typically, several tens of frames could be recorded before a fluorophore photobleached. Because α-CnTx-Cy5 could be washed off completely, fresh α-CnTx-Cy5 could be added again. This made possible to repeat single- molecule experiments more than 30 times on a single cell and thus to collect data with excellent statistics. No fluorescence was observed on cells not expressing nAChRs showing that labelling is specific. Hence, α-CnTx-Cy5 can be used as excellent specific label for the nAChR.

3.6 nAChR subtype profiling high content assay

As α-CnTx WT is known to bind specifically to muscle type but not to neuronal nAChRs (16), we investigated whether this specificity was conserved for α-CnTx-Cy5. HEK293 cells expressing the muscle type nAChR, coexpressed with cytosolic GFP, and cells expressing the neuronal α7/5-HT₃ chimera (21, 22), coexpressed with cytosolic CFP, were co- cultivated on microscope slides. While the non-specific antagonist α-BgT- Alexa647 did bind to both receptor types (Fig. 6a), α-CnTx-Cy5 did exclusively bind to cells expressing muscle type nAChRs (Fig. 6b).

4. Discussion

The above results describe the general application of fluorescent receptor ligands with fast association-dissociation kinetics in pharmacological investigations of membrane receptors. The measurement concept preferably consists of cycles of incubation-acquisition enabling to carry out numerous experiments on the same single cell. A key benefit of this approach is that it can be applied even in the case of strong photobleaching opening the possibility to investigate single cells with low receptor concentrations comparable to native conditions in tissue cells. Furthermore, the influence of stimuli, such as ligands, can be rapidly and repetitively investigated using our reversible binding assay.

This approach was illustrated using novel fluorescent α-conotoxin derivatives that bind with fast kinetics to nAChRs. The addition of the fluorophore resulted in only a minimal loss of affinity, in particular for the Cy5 derivative. α-CnTx-Cy5 conserved the attractive properties of conotoxins: High affinity, site specificity, in this case for the α/δ site, and strong selectivity for muscle-type nAChR as compared to neuronal-type acetylcholine receptors. Thus, that these novel fluorescent antagonists might find a broad application in nAChR research, for instance in investigations on cells that express multiple variants of acetylcholine receptors and in pathology. Our ReSeq binding assay can be particularly useful for medical diagnostics in tissue cells of patients suffering from congenital diseases such as myasthenic syndromes where it has been shown that alterations of the nAChR binding properties are a major source of malfunction (23).

The present approach can be extended to other membrane receptors: First, conotoxins are a large family of antagonists that are specific for a broad panel of receptors and can be labelled in a similar way as α-CnTx-Cy5.

Second, preferred requirements on the ligand properties are relatively modest as dissociation lifetimes in the range of one to ten minutes usually correspond to K_D in the range of 100 nM to 1 μM. Synthesizing fluorescent ligands exhibiting such affinities should prove easier as in standard fluorescence applications where K₀ in the nM range are required,

A key advantage of the measurement approach is that pharmacological investigations can be performed on single cells expressing the natural amount of receptor, which usually is very difficult due to photobleaching. In conventional measurements, receptor overexpression is often used to circumvent this problem at the risk of inducing effects that are absent at lower expression levels similar to native cells. For instance, there are indications that the function of recombinant G-protein coupled receptors are modified when expressed in non-native cells (1 ). In this context, the present approach to investigate receptor function in primary cell lines will be of importance. Using repetitive measurements, binding parameters can be extracted and investigations can be performed even when photobleaching becomes extremely strong as it is the case at lowest expression levels. Ultimately, single-molecule sensitivity can be achieved as illustrated on the nAChR. This opens the possibility to perform pharmacological investigations on tissue cells expressing minimal amounts of membrane receptors. In general, repetitive reversible labelling is an ideal approach to monitor single molecules where photobleaching usually is the limiting factor. By performing repetitive measurements, tenth of image sequences can be acquired to collect extensive statistics within a single experiment.

Repetitive competition experiments allow the rapid measurement of binding curves of ligands on single cells. This methodology might prove very useful in pharmacological investigations on rare tissue cells. In particular, the repetitive measurement procedure opens the way to automation and miniaturization, for instance using microfluidic structures. High-throughput screening of potential drugs can be performed on a single cell with a rate only depending on the association-dissociation cycle time, consuming only minimal amount of testing components. For example, in the case of α-CnTx- Cy5, a measurement currently requires only 30 fmol of toxin and takes about 20 minutes. One can expect that the use of ligands with faster kinetics could reduce this time down to 5 minutes. In addition, the use of more complex microstructures enables complete binding assays on a single cell or increasing statistics by parallel approaches using only some hundred nl of cell suspension. This way, screening more than 250 ligands per day using one single cell extracted from an animal tissue or obtained from a body fluid is a realistic goal. References

1. Ellis, C. The state of GPCR research in 2004. Nat Rev Drug Discov 3, 577-626 (2004). 2. Zemanova, L., Schenk, A., Valler, MJ. , Nienhaus, G.U. & Heilker, R. Confocal optics microscopy for biochemical and cellular high-throughput screening. Drug Discov Today 8, 1085-1093 (2003).

3. Sundberg, SA High-throughput and ultra-high-throughput screening: solution- and cell-based approaches. Curr Opin Biotechnol 11 , 47-53 (2000). 4. Corringer, PJ. , Le Novere, N. & Changeux, J.P. Nicotinic receptors at the amino acid level. Annu Rev Pharmacol Toxicol 40, 431-458 (2000).

5. Karlin, A. Emerging structure of the nicotinic acetylcholine receptors. Nat Rev Neurosci 3, 102-114. (2002).

6. Waksman, G., Changeux, J.P. & Roques, B.P. Structural requirements for agonist and noncompetitive blocking action of acylcholine derivatives on

Electrophorus electricus electroplaque. MoI Pharmacol 18, 20-27 (1980).

7. Meyers, H.W. et al. Synthesis and properties of NBD-n-acylcholines, fluorescent analogs of acetylcholine. EurJ Biochem 137, 399-404. (1983).

8. Martinez, K.L., Corringer, PJ., Edelstein, SJ., Changeux, J.P. & F, M. Structural differences in the two agonist binding sites of the Torpedo nicotinic acetylcholine receptor revealed by time-resolved fluorescence spectroscopy. Biochemistry 39, 6979-6990. (2000).

9. Nelson, L. Venomous snails: one slip, and you're dead. Nature 429, 798- 799 (2004). 10. Tsetlin, V.I. & Hucho, F. Snake and snail toxins acting on nicotinic acetylcholine receptors: fundamental aspects and medical applications. FEBS Lett 557, 9-13 (2004).

11. Arias, H. R. & Blanton, M.P. Alpha-conotoxins. lnt J Biochem Cell Biol 32, 1017-1028. (2000). 12. Nicke, A., Wonnacott, S. & Lewis, RJ. Alpha-conotoxins as tools for the elucidation of structure and function of neuronal nicotinic acetylcholine receptor subtypes. EurJ Biochem 271 , 2305-2319 (2004).

13. Rogers, K.R., Eldefrawi, M. E., Menking, D.E., Thompson, R.G. & Valdes, JJ. Pharmacological specificity of a nicotinic acetylcholine receptor optical sensor. Biosens Bioelectron 6, 507-516 (1991 ).

14. Ashcom, J. D. & Stiles, B.G. Characterization of alpha-conotoxin interactions with the nicotinic acetylcholine receptor and monoclonal antibodies. Biochem J 328 (Pt 1), 245-250 (1997).

15. Gray, W.R., Luque, A., Olivera, B.M., Barrett, J. & Cruz, LJ. Peptide toxins from Conus geographus venom. J Biol Chem 256, 4734-4740 (1981 ).

16. Quiram, P.A. & Sine, S.M. Structural Elements in alpha -Conotoxin ImI Essential for Binding to Neuronal alpha 7 Receptors. J. Biol. Chem. 273, 11007-11011 (1998).

17. Blount, P. & Merlie, J. P. Molecular basis of the two nonequivalent ligand binding sites of the muscle nicotinic acetylcholine receptor. Neuron 3, 349-

357 (1989).

18. Baumle, M., Stamou, D., Segura, J. M., Hovius, R. & Vogel, H. Highly fluorescent streptavidin-coated CdSe nanoparticles: Preparation in water, characterization, and micropatterning. Langmuir 20, 3828-3831 (2004). 19. Weiss, S. Fluorescence Spectroscopy of Single Biomolecules. Science 283, 1676-1683 (1999).

20. Xie, X.S. & Trautman, J.K. OPTICAL STUDIES OF SINGLE MOLECULES AT ROOM TEMPERATURE. Annu. Rev. Phys. Chem. 49, 441-480 (1998). 21. Eisele, J. L. et al. Chimeric Nicotinic Serotonergic Receptor Combines Distinct Ligand-Binding and Channel Specificities. Nature 366, 479-483 (1993).

22. Corringer, PJ. et al. Identification of a new component of the agonist binding site of the nicotinic alpha 7 homooligomeric receptor. J. Biol. Chem. 270, 11749-11752 (1995). 23. Engel, A.G., Ohno, K. & Sine, S.M. Sleuthing molecular targets for neurological diseases at the neuromuscular junction. Nat Rev Neurosci 4, 339-352 (2003).

Claims

1. A method for determining a receptor-ligand interaction comprising the steps:

(a) providing a cell comprising a receptor,

(d) monitoring the fluorescence of the cell during step (c) and

(e) repeatedly performing steps (c) and (d).

2. The method of claim 1, wherein the receptor is a membrane-bound receptor.

3. The method of claims 1 or 2, wherein the receptor is a gated ion-channel, particularly a nicotinic acetylcholine receptor (nAChR).

4. The method of any one of claims 1 to 3, wherein the ligand has a dissociation lifetime in the range of about 0.5 to about 20 min.

5. The method of any one of claims 1 to 4, wherein the ligand is selected from fluorescently labelled peptides, polypeptides and non-peptidic compounds which bind specifically to a membrane-bound receptor.

6. The method of any one of claims 1 to 5, wherein the ligand is a fluorescently labelled α-conotoxin, particularly an α-conotoxin Gl.

7. The method of any one of claims 1 to 6, wherein the cell is a single cell or an individual cell within a sample.

8. The method of any one of claims 1 to 7, wherein the cell comprises about 10⁶ receptor molecules or less, particularly 10⁴ receptor molecules or less.

9. The method of any one of claims 1 to 8, wherein the fluorescence of individual ensembles of receptor-ligand complexes is determined.

10. The method of any one of claims 1 to 9, wherein step (c) comprises: (ci) subjecting the cell to conditions wherein the ligand binds to the receptor and

(cii) subjecting the cell to conditions wherein the ligand is removed from the receptor.

11. The method of any one of claims 1 to 10, wherein step (d) comprises at least 3 measurements.

12. The method of any one of claims 1 to 11, wherein the steps (c) and (d) are repeatedly carried out on the same cell or cells.

13. The method of any one of claims 1 to 12 for determining the binding of a test substance to the receptor.

14. The method of any one of claims 1 to 13 which is carried out in a microfluidic structure or on a chip. .

15. Use of the method of any one of claims 1 to 14 in a screening procedure for the identification and/or characterisation of pharmaceutical drug candidate molecules.

16. A fluorescently labelled α-conotoxin, particularly an α-conotoxin Gl, which has a bright emission with a photobleaching quantum yield of preferably less than 10^'5 at wavelengths of preferably longer than 550 nm to allow sensitive measurements on single cells.

17. Use of a fluorescently labelled α-conotoxin in a method for determining a receptor-ligand interaction which has a bright emission with a photobleaching quantum yield of preferably less than 10^'5 at wavelengths of preferably longer than 550 nm to allow sensitive measurements on single cells.