WO2008030548A2

WO2008030548A2 - Voltage-sensitive fret assay employing pyrene trisulfonic donor

Info

Publication number: WO2008030548A2
Application number: PCT/US2007/019503
Authority: WO
Inventors: Michael Maher
Original assignee: Janssen Pharmaceutica N.V.
Priority date: 2006-09-08
Filing date: 2007-09-06
Publication date: 2008-03-13
Also published as: WO2008030548A3

Abstract

Pyrene trisulfonic donors are used as dyes in FRET-based membrane potential measurements. Such dyes exhibit pH stability, facilitating measurement of membrane potentials during the activation of ion channels at low pH.

Description

VOLTAGE^ENSπiVE FRET ASSAYEMPLOYINGPYRENETRtSULFOlNICDONOR

CROSS-REFERENCE TO RELATED APPLICATION

This application claims priority to United States Provisional Application No. 60/843,275, filed September 8, 2006.

FIELD OF THE INVENTION

The present invention relates to the use of pyrene trisulfonic donors as dyes in assay methods based on FRET-based transmembrane potential measurements. Such methods are useful for determining activities of test compounds against electrogenic targets, e.g., ion channels, transporters, pumps, and ion exchangers.

BACKGROUND OF THE INVENTION

The following background discussion is provided to facilitate a better appreciation of the technology relating to the invention. As statements in this discussion may reflect viewpoints of an inventor, they should not be misconstrued as necessarily corresponding to knowledge in the prior art.

High-throughput screening (HTS) of compound libraries is a common method for discovering novel pharmacological agents against specific protein targets [I]. Compound libraries containing thousands to millions of discrete chemical entities may be tested for binding to or modification of function of a biological target of interest, such as an acid-modulated target, e.g., an acid-sensing ion channel.

Ion channels are recognized as useful drug targets, being involved in many disease states [2, 3]. Exemplary assays for identifying modulators of ion channels are described in International Publication Nos. WO 02/08748, 03/087769, WO 2004/012585, and WO 2006/050257 and U.S. Patent No. 7,002,671. The diverse properties and functions of ion channels present unique challenges for the drug discovery process.

Multiple types of low- to high-throughput assays have been used for identification and characterization of compound-target interactions. Modern approaches for ion channel drug screening include electrophysiology, radioactive flux or tracer assays, binding assays, ligand displacement assays, and optical assays using fluorescent or absorbent ion and transmembrane potential indicators [4]. Although electrophysiological methods are definitive for characterizing channels and modulators, their low throughput typically precludes their use in HTS [5]. Optical methods are well-suited to HTS assays, due to their parallelizabilty, speed, and low cost.

The transmembrane potential of a cell can be detectably altered by the activity of as few as 100 ion channels per cell, so it is a particularly useful measure of ion channel function. Several classes of transmembrane-potential-sensitive indicators have been described [6]. The excited state of the so-called 'fast' or electrochromic dyes involves charge movement through the membrane, causing voltage sensitivity of the emission and/or excitation spectra [7]. 'Slow' or redistribution dyes are charged, generally lipophilic dyes which exhibit aNernstian intracellular concentration [8]. Forster or fluorescence resonance energy transfer (FRET) dyes employ distance- dependent energy transfer between two dyes, at least one of which must be charged and membrane-permeant [9, 10]. Newer methods include using the voltage-dependent motion of a potassium channel gate as a sensor [H].

FRET-based dye systems have several advantages over other transmembrane potential dyes. The response time can be as low as 1 millisecond, which is considerably faster than redistribution dyes and fast enough to capture transient transmembrane potential changes. They are more sensitive (1-2%/mV) than electrochromic dyes (0.01-0.1%/mV). Although electrochromic dyes are faster (response times <1 μsec, cellular responses are rarely fast enough to require this speed. Finally, the FRET dye system is ratiometric, and largely cancels out sources of error and noise, such as addition artifacts and variations in dye loading and cell density.

FRET can occur when a donor fluorophore is in close proximity to a second, acceptor fluorophore with a longer excitation wavelength. When the donor fluorophore is in its excited state, it possesses an oscillating electric dipole that induces dipoles in neighboring molecules. If the acceptor fluorophore is close enough to be affected by the induced dipole, and if its own excited state is close enough in energy to the excited state of the donor, resonance between the excited states occurs. If the transfer rate of the energy between the dipoles is faster than the inverse fluorescence lifetime of the donor, the acceptor can absorb the excitation energy. The acceptor then emits the energy as a photon at its normal emission wavelength. The energy transfer rate is inversely proportional to the 6^th power of the separation between the fluorophores. FRET pairs can be characterized by the distance ro at which 50% energy transfer occurs. This characteristic distance is strongly dependent upon several parameters, including the relationship between the fluorescence spectra of the two fluorophores, the angle between the dipoles of the two fluorophores, the dielectric environment, and the fluorescence lifetime of the donor [12]. Dye pairs in common use typically have ro in the 10-100 A range.

Since the original description of FRET voltage-sensitive dyes [10, 13], little improvement has been made in the method [although see ref.14]. In a widely-used incarnation, a coumarin-linked phospholipid donor (CC2-DMPE; see Figure 1) is bound to the outer face of the cell membrane. An oxonol (DiSB AC_xQ), where x=2,4, or 6) is the mobile acceptor. Hydroxycoumarins are poor choices for FRET donors, due to their strong pH dependence in the physiological range. The peaks of the excitation spectra of the protonated forms (depicted for CC2-DMPE in Figure 1) are shifted 40-50 nm lower relative to the charged forms [15]. Hydroxycoumarins have been used as sensitive probes of local surface potential, which can be altered by lipid composition and polarity [15-18]. Although adding the chloride to the hydroxycoumarin shifts the pK_a from ~7.5 to 6.2 [13], small changes in local pH during the assay can lead to artifacts. Because a drop in overall fluorescence of coumarin is proportionally reflected by a drop in oxonol emission, these artifacts are largely cancelled out by the standard ratiometric analysis. However, the pH dependence precludes the use of assays for which a lower pH is desired, for example, for acid-gated or modulated channels such as the acid-sensing ion channel (ASIC) family and the pro ton- gated channel TRPVl.

FRET has been used in a wide variety of measurement techniques since its discovery in 1965 [19]. See, e.g., U.S. Patent Nos. 5,661,035, 6,107,066, 6,342,379, and 6,596,522. The ability to detect distances between molecules on the order of 50 A provides a powerful tool in examining interactions between molecules [12]. In principle, virtually any fluorophore pair can be used for FRET as long as the emission wavelength of the donor is at or less than the excitation wavelength of the acceptor. Other fluorophores that have been used for FRET include various oxonols, fluoresceins, rhodamines, Texas red, and variants of green fluorescent protein (GFP).

FRET is one of the few techniques capable of probing distances in the 10-100 A range. This corresponds to a molecular density of 1 per 500-5x10⁵ A³, or in terms of concentration, 3 mM - 3 M. In an extrapolation of previous work, Gonzalez and Tsien [10] realized that lipophilic dyes often reach millimolar concentrations in cell membranes. Nernstian redistribution of a charged dye can result in concentration changes by an order of magnitude or more. There remains a need for advantageous dye pairs, such that FRET efficiency is on the order of 50% and very large voltage- dependent signals can be achieved.

Moreover, some targets, such as ASICs, pose challenges for developing suitable assays useful with modern drug discovery techniques and platforms, such as high- throughput systems.

SUMMARY OF THE INVENTION

The invention is directed to the general and preferred embodiments of assay methods defined, respectively, by the independent and dependent claims appended hereto, which for the sake of brevity are incorporated by reference herein.

The detailed description below, taken in conjunction with the appended drawing figures, describes various illustrative and preferred embodiments of the invention.

BRIEF DESCRIPTION OF DRAWINGS

Figure 1 depicts the structures of the following reagents: (A) CC2-DMPE; (B) DiSBAC₂(3); (C) VABSC-I; and (D) PTS_x.

Figure 2 provides contour plots of the (a) normalized ratio response to a high- potassium addition, (b) blue signal/background, and (c) red signal/background for HEK cells stained with DiSBAC₂(3) and either CC2-DMPE (A) or PTSi₈ (B). The high- potassium addition results in a -35 mV depolarization.

Figure 3 shows excitation and emission spectra for cells stained with CC2- DMPE and PTSj₈ (A shows spectra for cells stained with 6 μM CC2-DMPE, excitation peak 415 nm, emission peak 455 nm; B shows spectra for cells stained with 2.5 μM PTSi₈, excitation peak 405 nm, emission peak 435 nm). Dashed lines are the spectra before adding 3 μM oxonol; solid lines are the spectra after adding oxonol. Fluorescence emission was monitored at 450 nm for the excitation scan; the excitation wavelength was 400 nm for the emission scan. The spectra were normalized to the fluorescence intensities before adding oxonol at the peak wavelengths. Excitation and emission spectra for cells stained with PTS]₂, PTSi₄, and PTSi₆ were indistinguishable from those OfPTSi₈, although each dye had a slightly different dependence of FRET efficiency on dye concentration.

Figure 4 depicts fluorescence in saline for CC2-DMPE and PTS_x with 400 nm excitation and 460 nm emission (D CC2-DMPE; O PTSi₈; V PTS,₆; Δ PTS,₂; 0 PTSi₀). The solid line is a fit to the CC2-DMPE data, and indicates that pK_a=5.6. Figure 5 illustrates the behavior of CC2-DMPE and PTS_x loaded into cells, when the pH is changed from 7.4 in the presence of 1 μM gramicidin. The final fluorescence ratio, normalized to the fluorescence ratio at pH 7.4, is plotted as a function of the final pH (D CC2-DMPE; O PTSi₈; V PTSi₆; Δ PTSi₂; 0 PTS₁₀).

Figure 6 is a graph of response as a function of the final potassium reversal potential for HEK cells stained with 3 μM oxonol and either 6 μM CC2-DMPE (D) or 2.5 μM PTSi₈ (O). Solid lines are fits to the data at the three highest reversal potentials. The dashed line indicates the control response when no change in bath solution is made. The intercept of the fits with the control responses indicates the resting membrane potential.

Figure 7 depicts activation of the endogenous ASIC in HEK-293 cells. A. Timecourse of response for HEK-293 cells loaded with PTS- 18 and oxonol. The cells started at pH 7.4, and the pH bathing the cells was changed at t=12 seconds after the start of the assay. Two traces for each of pH 7.4, 6.6, and 6.0 are shown. The scale on the right indicates the transmembrane potential, estimated from the potassium titration experiments. B. Response t=45 seconds after the addition of low pH buffer as a function of final pH using different donor fluorophores. (D CC2-DMPE; O PTS]₈; V

PTS₁₆; A PTS₁₂; O PTS₁₀).

DETAILED DESCRIPTION OF THE INVENTION AND ITS PREFERRED

EMBODIMENTS

The invention may be more fully appreciated by reference to the following detailed description, including the examples. For the sake of brevity, the disclosures of the publications, including patents, cited in this specification are herein incorporated by reference.

The invention is generally directed to assay methods employing pyrene trisulfonic donors, which may be used to screen compounds to determine their activity against biological targets of interest, preferably proteins or polypeptides that are electrogenic (capable of creating or responding to an electric field or potential). For example, one general embodiment comprises: (a) exposing a membrane material to a membrane reagent; (b) exposing the material to a pyrene trisulfonic donor capable of undergoing energy transfer with the membrane reagent; (c) irradiating the material with excitation light; (d) monitoring light intensity from at least one (preferably one or two) fluorophore excitation emissions between the membrane reagent and the pyrene trisulfonic donor as a measure of energy transfer; and (e) relating the monitored light intensity (or measured energy transfer) to a change in transmembrane potential.

In preferred embodiments, such methods are used to assay for activity against an acid-modulated target, such as an ASIC or vanilloid receptor 1 (TRPVl or VRl) or a derivative that is a functional equivalent thereof, e.g., a homologous mutated or truncated polypeptide sequence. For example, the above-described method in a preferred embodiment further comprises contacting a test compound with the membrane material comprising an acid-sensing ion channel target.

The terms "including", "containing" and "comprising" are used herein in their open, non-limiting sense.

A "pyrene trisulfonic donor" refers to a pyrene-based indicator compound having a pyrene trisulfonic acid or trisulfonate core or motif capable of acting as a donor dye or fluorophore. Such compounds may be represented by the following formula:

wherein Ri is H, a hydrophobic moiety, or a charged moiety.

In one preferred embodiment, R] is a charged moiety, which refers to an ionizable group or radical that possesses a charge upon ionization. Exemplary charged moieties include nitrate, sulfate, phosphate, and carboxylate groups. In preferred embodiments of pyrene donors, Ri is a charged moiety selected from sulfonate, phosphonate, carboxylate, and phosphate groups.

In another preferred embodiment, Rj is a hydrophobic moiety selected from groups or radicals having: a partition coefficient between a physiological saline solution (e.g., HBSS) and octanol at least about 50, preferably at least about 1000; and an adsorption coefficient to a phospholipid bilayer (such as for example a membrane derived from a human red blood cell) at least about 100 nm, preferably at least about 300 nm (where the membrane is 3 nm). Methods of determining partition coefficients and adsorption coefficients are available in the art.

Preferred pyrene trisulfonic donors include PTSlO, PTS12, PTS14, PTS16, and PTS 18. PTS 18, i.e., S-octadecyloxypyrene-l^ό-trisulfonic acid, is an especially preferred embodiment of a pyrene trisulfonic donor or dye. Preferred membrane reagents include tetraaryl borate-fluorophore conjugates, such as those of formula: (Ar¹ ^-B-Ar² -Y-F, wherein: Ar¹ is an unsubstituted or substituted aryl group (where "aryl" refers to an aromatic carbocyclic group having a single ring (e.g., phenyl) or multiple condensed rings (e.g., naphthyl, anthracenyl); Ar² is an unsubstituted or substituted bifunctional arylene group; B is boron; Y is oxygen or sulfur; and F is a neutral fluorophore. In another preferred embodiment, the membrane reagent is polymethine oxonol. hi a further preferred embodiment, the membrane reagent comprises a transition metal complex.

In some preferred embodiments, the pyrene trisulfonic donor is connected to the membrane reagent via a linker. A "linker" is a chemically and biologically compatible covalent grouping of atoms that links together the pyrene donor and membrane reagent. In some embodiments, the linkers have from 20 to 40 (in some preferred embodiments from 25 to 30) atoms from end to end and may be branched- or straight-chain or contain rings. The atoms may be connected by carbon-carbon or carbon-heteroatom or heteroatom-heteroatom bonds. The linking group may contain single and/or double bonds, from 0 to 10 heteroatoms (N, O, or S), and/or saturated or aromatic rings. The linking group may contain one or more groups such as esters, ethers, sulfides, disulfides and the like. The linker can be designed to be hydrophobic or hydrophilic. In an exemplary embodiment, the linker is a polyethylene glycol derivative.

The pyrene trisulfonic donor and membrane reagent constitute a donor/acceptor pair that may be used in measuring fluorescence resonance energy transfer (FRET), electron transfer, exchange (Dexter) interaction, paramagnetic quenching, or promoted intersystem crossing. Thus, in some preferred embodiments, the pyrene donor is a FRET donor and the membrane reagent is a FRET acceptor.

Exemplary membrane materials include biological membranes, tissues, and living cells. In one preferred embodiment, the membrane is a living mammalian cell. In preferred embodiments the membrane material comprises an ion channel target, such as a polypeptide selected from ion channels that are known or are identified in the art. Such materials may be used in preferred embodiments to assay test compounds (e.g., selected from library compounds or known ion channel modulators).

In preferred embodiments, the membrane reagent partitions into and remains within the membrane. In other preferred embodiments, the membrane reagent redistributes from an inner face of the membrane to a second inner face of the membrane in response to a transmembrane potential (which may also be referred to as membrane potential). In some preferred embodiments, the pyrene donor is located on the extracellular surface of the membrane and undergoes energy transfer with the membrane reagent located within the membrane. EXAMPLES

Reagents. DiSBAC₂(3), CC2-DMPE, and VABSC-I [20] were obtained from Invitrogen. (The structures of these compounds are shown in Figure 1.) Cell culture reagents were obtained from Gibco. Pyrene donor dyes, Pluronic F- 127, assay buffer ingredients, and gramicidin were obtained from Sigma. The pyrene donors employed are listed in Table 1, and include the following compounds:

PTSio (8-decyloxypyrene-l,3,6-trisulfonic acid)

PTS₁₂ (8-dodecyloxypyrene-l,3,6-trisulfonic acid)

PTSi₄ (8-tetradecyloxypyrene-l,3,6-trisulfonic acid)

PTSi₆ (8-hexadecyloxypyrene-l,3,6-trisulfonic acid)

PTSi₈ (S-octadecyloxypyrene-l^β-trisulfonic acid)

The structure of these dyes are shown in Figure ID as PTS_x, where x indicates the number of carbons in the alkane tail.

Cell culture. HEK-293 cells (ATCC) were maintained in DMEM supplemented with 10% fetal bovine serum in tissue culture flasks at 37⁰C in a 5% CO2-95% air humidified atmosphere. Cells were subcultured every 2-3 days. Freshly dissociated HEK-293 cells were plated at 35,000 cells/well in 100 μL/well of growth medium in 96-well poly-lysine coated plates (Becton-Dickinson BioCoat). For some experiments, 384-well plates were used (15,000 cells in 50 μL/well). No differences were seen in the results obtained from the two different types of plate.

Assay buffer. The assay buffer contained (quantities in mM): 130 NaCl, 4 KCl, 2 CaCl₂, 0.5 MgCl₂, 10 HEPES, and 5 dextrose. The assay buffer was adjusted to 300 mOs and pH 7.4. High-potassium buffer contained: 134 mM KCl, 2 mM CaCl₂, 0.5 mM MgCl₂, 10 mM HEPES, and 5 mM dextrose. Assay buffer and high-potassium buffer were mixed to obtain various potassium concentrations.

Dye preparation. Stock solutions were 10 mM dye in dry DMSO. Pyrene donors were solubilized at 2X final concentration by adding the appropriate volume of stock solution plus an equal volume of 10% Pluronic F 127 in DMSO into a plastic 50 mL centrifuge tube. Assay buffer was added to the dye while vortexing. DiSBAC₂(3) was solubilized at 2X final concentration by adding the appropriate volume of stock solution into a plastic 50 mL centrifuge tube. Assay buffer was added to the dye while vortexing. The blocking dye VABSC-I was added to the oxonol solution at 1 raM. Test compounds were also added at 2X to the oxonol solution where applicable.

Dye loading. Growth medium was replaced with assay buffer using a Biotek ELx405 plate washer. After adding donor dye, the cells were incubated for 30 minutes at room temperature in the dark. The oxonol/test compound solution was added following a second rinse with the plate washer. The cells were then incubated again for 30 minutes at room temperature in the dark.

Optical assays. Static fluorescence intensities and excitation/emission spectra were obtained using a Molecular Probes Gemini plate reader. Dynamic changes in fluorescence intensities were measured using a Voltage-Ion Probe Reader (VIPR™, Aurora Biosciences Corp.)[VIPR™, Aurora Biosciences Corp.; see 21]. The optical filters were 405/BW nm for excitation, and 480/BW nm (blue) and 535/BW nm (red) for emission. Blue and red emission intensities were recorded at 1 Hz. Background signals were obtained using multiwell plates containing the assay buffer only. For potassium addition assays, fluorescence emissions were recorded for 10 seconds to establish a baseline fluorescence ratio. Then, a volume of stimulus buffer equal to the volume already in the well (100 μL for 96-well plates, 50 μL for 384 well plates) was added. The fluorescence intensities 20 seconds after the addition were used for further analysis.

Dye matrix. HEK293 cells were loaded into 96-well plates and stained as above. Blue donor dyes were added as a gradient across the columns of the plate, with concentrations 5 nM to 10 μM. Oxonol was added as a gradient down the rows of the plate, with concentrations 78 nM to 10 μM. All wells had lX=0.5 mM VABSC-I .

VIPR data analysis. The parameter φ(t) is defined as the background-subtracted blue/red ratios normalized to unity at the start of an assay:

Equation 1

•tt-f# R(ή-÷ R₀Y [#B(O)-*B₀

where B(t) and R(t) are the blue and red emission intensities at time t, and Bo and Ro are the background blue and red intensities, φ is directly related to V_1n and is approximately linear in the physiological range [10]. A positive change in transmembrane potential pulls oxonol away from the outer leaflet of the membrane, leading to an increase in blue signal and a decrease in red signal. Therefore, positive φ indicates a depolarization of the transmembrane potential. mRNA expression of human ASICl, 2, 3, and 4 in HEK293. Total RNA was extracted from HEK293 cells using RNeasy kit (Qiagen). First-strand cDNA was synthesized from 5 μg RNA using Superscript II Reverse Transcriptase (Invitrogen) following the manufacturer's protocol. Real-time PCR was performed on Smart Cycler (Cepheid) using LightCycler DNA Master SYBR Green I (Roche). All pairs of gene- specific primers are intron spanned and their sequences are: hASICl : forward: 5'-GCAGATCCTGCTCTGGACTTCC-S' (SEQ ID

NO:1) reverse: 5'-AATGACCTCGTAGGCGTAGTC-B ' (SEQ ID

NO:2) hASIC2: forward: 5'-TCCTCAGAGATGGGCCTCGAC-S' (SEQ ID

NO:3) reverse: 5'-CATTGTGTCACAAGTACTCAC-S' (SEQ ID NO:4) hASIC3: forward: 5'-GTCCCACCTTTGACATGGCG-S' (SEQ ID NO:5) reverse: 5'-CCAGCCCATTGCCCATGCCA-S' (SEQ ID NO:6) hASIC4: forward: 5'-CTACAGTGTGTCTGCCTGCCG-S' (SEQ ID NO:7) reverse: 5'-GGCTGAGCCCCTGTTGGGGA-S' (SEQ ID NO: 8) After initial denaturation for 2 minutes (min) at 95⁰C, thermal cycles consisted of 5 seconds (s) at 95⁰C, 7 s at 6O⁰C and 12 s at 72⁰C for 40 cycles. PCR products were analyzed by electrophoresis on a 1.5% agarose gel and confirmed by sizing and specific restriction patterns.

Cell staining. Fluorescence microscopy revealed that HEK-293 cells stained with CC2-DMPE or PTS_x (x=10,12,14,16,or 18) appeared as bright rings, indicating that only the cell membrane was stained. Although PTS io also initially showed ring staining, this dye came off the membrane with a timescale of -10 minutes. The cell staining gradually grew dimmer as the background fluorescence increased. This effect was not seen with the other pyrene donors over the duration of experiments.

Figure 2 shows the results of dye matrix experiments, where cells were stained with cross gradients of oxonol and either CC2-DMPE or PTSi₈. HEK-293 stained with PTS_x pyrene donors were significantly brighter than CC2-DMPE at equivalent concentrations. Excitation/emission spectra. FRET between two fluorophores was demonstrated by comparing fluorescence scans of the individual dyes with those of the combined dyes. Figure 3 illustrates the excitation and emission spectra for HEK cells stained with a FRET donor (either 6 μM CC2-DMPE or 2.5 μM PTS _{] 8}) with and without 3 μM DiSB AC₂3. The spectra were normalized to the peak intensities in the absence of oxonol. Before adding oxonol, CC2-DMPE (Figure 3A) showed a single broad excitation peak at 418 run ± 33 nm (half- width at half maximum) and a single emission peak at 453 ± 27 nm. Upon adding oxonol, the emission at 453 nm dropped in amplitude by 69%, while a second emission peak grew at 562 ± 17 nm. This second emission peak corresponds to the fluorescence spectrum of DiSBAC₂(3). The shift in emission from the blue (pyrene donor) to the red (oxonol) emission indicates that FRET occurred.

PTSi₈ had a more complex excitation spectrum (Figure 3B), with two peaks at 405 ± 12 and 378 ± 30 nm. The emission had a single peak at 435±25 nm. The emission spectrum of PTS_x is dependent upon the solvent polarity of the pyrene moiety, shifting from 435 nm in water to 420 nm in a lipid environment [22]; these results suggest that the fluorophore is surrounded by saline. Upon adding oxonol, the emission at the 436±25 nm peak dropped in amplitude by 65%, while a second emission peak grew at 562 ± 17 nm. Both excitation peaks were reduced in amplitude uniformly when oxonol was added. The other pyrene donors behaved virtually identical to PTS 18. These spectra and the excitation spectrum for DiSBAC₂(3) (data not shown) were used to calculate a Forster distance of rø=41 A [12], compared to rø=48 A for the CC2-DMPE/DiSBAC₂(3) pair [13]. pH dependence. Dyes in aqueous solution were prepared at 1 μM in assay buffer at various pH values. The assay buffer was supplemented with 10 mM MES (ρK_a 6.1) and 10 mM TAPS (pK_a 8.4) to stabilize the pH above and below the effective range for HEPES (pK_a 7.5). Figure 4 summarizes the fluorescence intensity at 400 nm excitation and 460 nm emission as a function of pH for CC2-DMPE and PTS_x (D CC2- DMPE; O PTSi₈; V PTSi₆; Δ PTSi₂; 0 PTSi₀). While the fluorescence of the pyrene donor was unaffected by changes in pH, CC2-DMPE has pK_a = 5.6 (solid line) and is virtually nonfluorescent below pH 5.

The pH dependence of the dyes loaded into cells was examined in a set of assays in which the pH was changed from pH 7.4 to a test value in while the fluorescence emission at 460 run and 580 nm was monitored in VIPR. To avoid any pH-dependent changes in transmembrane potential, 1 μM gramicidin was included in the assay buffer. The gramicidin essentially locked the transmembrane potential at zero, so that any changes in fluorescence intensities were solely due to the dyes. Figure 5 shows the normalized fluorescence ratio change φ caused by the pH change during the assay. Whereas the pyrene donor dyes show no significant pH dependence, the loss of fluorescence of cells stained with CC2-DMPE at low pH mirrors the effect seen in saline.

Demonstration of voltage sensitivity.

A well-established assay for determining the activity of potassium channels involves engineering a cell line in which the resting potential is set by the target channel. While monitoring the voltage-sensitive fluorescence, a high-potassium buffer is added. If the potassium channel is active, a large transmembrane potential change is seen; if it has been blocked by a test compound, the transmembrane potential change is reduced or absent. This method is extended herein to allow simultaneous, accurate measurement of voltage dye sensitivity in and resting transmembrane potential. This method was used to characterize and compare the voltage sensitivity of the FRET dyes.

HEK293 cells endogenously express several potassium currents, which set the resting potential to approximately -60 mV [23-27]. Multiple transient outward (I_K) and non-inactivating delayed rectifier (I_A) subtypes have been detected. In experiments, whole-cell electrophysiology revealed that the average potassium currents could be well-described by conductances of gκ=1.7 nS(l+exp((-V_m-l)/16)) and g_A=l-5 nS(l+exp((-V_m+8)/8)) (n=8, average membrane capacitance 12 pF). These values are comparable to those previously reported.

To demonstrate the method, a model of the currents in a cell near the resting potential was constructed. Although the Boltzman forms of the potassium conductances above are fairly accurate representations in the voltage range around zero millivolts and above, the resting potential is actually at near -60 mV. In this range, all currents are very small and the channels are nearly completely deactivated. Extrapolating the behavior of the channels to such small currents is inherently inaccurate. So, while the following model cannot be numerically verified, it is useful in illustrating the behavior of currents in setting the resting transmembrane potential. Only a single, 'averaged' potassium current is included for simplicity. The total current near the resting potential V_m is given by:

Equation 2

The above equation must be solved numerically. With normal assay buffer at 4 mM K⁺, and assuming an intracellular concentration of potassium at 130 mM, the reversal potential of potassium is given by the Nernst equation to be Vκ=-S9 mV at room temperature. Using representative values

V_w=\5 mV, the net current is zero at a resting potential of -51 mV. The potassium and leak currents at this potential are only ±3 pA. The solution to Equation 2 is fairly sensitive to changes in the channel parameters, so it cannot be used to predict the resting potential.

In a typical high-potassium addition assay, the voltage-dependent fluorescence is monitored while the extracellular potassium concentration is raised from normal saline (4 mM K⁺) to an elevated level (usually an equal volume mixture of normal saline and 130 mM K⁺). At an extracellular potassium concentration of 69 mM, the potassium reversal potential become -16 mV. In this case, the total current in Equation 2 is zero at V_1n= -16 mV. Once the channels begin to activate, the transmembrane potential is determined solely by the potassium reversal potential.

Consider a high-potassium addition assay in which the final potassium concentration is intermediate between 4 mM and 130 mM. If the potassium reversal potential does not exceed the low-potassium resting potential, very little change in transmembrane potential will be seen. At higher potassium levels, the final transmembrane potential approaches asymptotically the potassium reversal potential. Thus, assuming a linear relationship between the fluorescence change and the transmembrane potential change, the slope of the response versus final potassium reversal potential is a direct measure of the dye sensitivity. Further, the intercept of the response at zero fluorescence change gives the resting transmembrane potential in the starting saline. Figure 6 shows the results of potassium titration experiments for HEK-293 cells stained with 3 μM DiSBAC₂(3) and either 6 μM CC2-DMPE (D) or 2.5 μM PTSi₈ (O). Performing the same experiment with other PTS_x donors gave similar results. Fitting the highest potassium data to linear functions, dye sensitivity and resting transmembrane potential were calculated and are summarized in Table 1.

Table 1 : Voltage sensitivities and resting transmembrane potentials calculated for HEK-293 cells stained with 3 μM DiSBAC₂(3) and various FRET donors.

All donors gave similar values for voltage sensitivity (1.7-1.9%/mV) and for the resting transmembrane potential (-46 to -51 mV). The dye matrix experiments showed that the voltage sensitivity depends quite strongly upon the staining levels of both donor and acceptor, so these calibrations are only strictly valid for a particular experiment. Still, they indicate that the PTS dyes perform remarkably similarly to CC2-DMPE.

Assay for acid-sensing ion channels. In addition to several potassium channel subtypes, HEK-293 cells have been shown to express endogenously the acid-sensing ion channel ASICIa [28]. This amiloride-sensitive, sodium-selective channel was reported to have a pEC₅₀=6.45 with rapid (~1 second) activation and inactivation kinetics. Activation of this channel should lead to a depolarizing response. mRNA for ASICl, 3, and 4 (but not ASIC2) was detected by rtPCR in the HEK-293 cells used in these experiments.

FRET donors were evaluated for their usefulness in an assay for this class of ion channel. Wild-type HEK-293 cells stained with 3 μM oxonol and various FRET donors were challenged in VIPR with low pH. The addition solution was assay buffer supplemented with 20 mMTAPS, MES, or acetic acid, set to various pH values and adjusted to 300 mOs with water. Figure 7A shows the timecourse of the normalized ratio response for HEK-293 cells loaded with PTS- 18 and oxonol. The cells started at pH 7.4, and the pH bathing the cells was changed at t=12 seconds after the start of the assay. Two traces for each of pH 7.4, 6.6, and 6.0 are shown. The scale on the right indicates the transmembrane potential, estimated from the potassium titration experiments. Figure 7B shows the normalized ratio response 45 seconds after the addition of low pH buffer as a function of final pH using different donor fluorophores. (D CC2-DMPE; O PTS₁₈; V PTSi₆; Δ PTSi₂; 0 PTS₁₀).

While the invention has been illustrated by reference to exemplary and preferred embodiments, it will be understood that the invention is intended not to be limited to the foregoing detailed description, but to be defined by the appended claims as properly construed under principles of patent law.

REFERENCES

1. Cox, B., et al., Application of high-throughput screening techniques to drug discovery. Prog Med Chem, 2000. 37: p. 83-133.

2. Ackerman, MJ. and D. E. Clapham, Ion channels—basic science and clinical disease. N Engl J Med, 1997. 336(22): p. 1575-86.

3. Ashcroft, F.M., Ion channels and disease : Channelopathies. 2000, San Diego: Academic Press, xxi, 481.

4. Mattheakis, L.C. and A. Savchenko, Assay technologies for screening ion channel targets. Curr Opin Drug Discov Devel, 2001. 4(1): p. 124-34.

5. Wood, C, C. Williams, and GJ. Waldron, Patch clamping by numbers. Drug Discov Today, 2004. 9(10): p. 434-41.

6. Wolff, C, B. Fuks, and P. Chatelain, Comparative study of membrane potential- sensitive fluorescent probes and their use in ion channel screening assays. J Biomol Screen, 2003. 8(5): p. 533-43.

7. Loew, L.M., G.W. Bonneville, and J. Surow, Charge shift optical probes of membrane potential. Theory. Biochemistry, 1978. 17(19): p. 4065-71.

8. Plasek, J. and K. Sigler, Slow fluorescent indicators of membrane potential: a survey of different approaches to probe response analysis. J Photochem Photobiol B, 1996. 33(2): p. 101-24.

9. Gonzalez, J. E. and M.P. Maher, Cellular fluorescent indicators and voltage/ion probe reader (VIPR) tools for ion channel and receptor drug discovery. Receptors Channels, 2002. 8(5-6): p. 283-95.

10. Gonzalez, J. E. and R. Y. Tsien, Voltage sensing by fluorescence resonance energy-transfer in single cells. Biophysical Journal, 1995. 69: p. 1272-1280.

11. Siegel, M.S. and E. Y. Isacoff, A genetically encoded optical probe of membrane voltage. Neuron, 1997. 19(4): p. 735-41.

12. Wu, P. and L. Brand, Resonance energy transfer: methods and applications. Anal Biochem, 1994. 218(1): p. 1-13.

13. Gonzalez, J.E. and R. Y. Tsien, Improved indicators of cell membrane potential that use fluorescence resonance energy transfer. Chemistry and Biology, 1997. 4: p. 269-77.

14. Farinas, J., A.W. Chow, and H.G. Wada, A microfluidic device for measuring cellular membrane potential. Anal Biochem, 2001. 295(2): p. 138-42. 15. Fromherz, P., Lipid coumarin dye as a probe of interfacial electrical potential in biomembranes. Methods Enzymol, 1989. 171: p. 376-87.

16. Pal, R., et al., Characterization of the fluorophore 4-heptadecyl-7- hydroxy coumarin: a probe for the head-group region of lipid bilayers and biological membranes. Biochemistry, 1985. 24(3): p. 573-81.

17. Kraayenhof, R., GJ. Sterk, and H. W. Sang, Probing biomembrane interfacial potential and pH profiles with a new type of float-like fiuorophores positioned at varying distance from the membrane surface. Biochemistry, 1993. 32(38): p. 10057-66.

18. Mingotaud, C, J.P. Chauvet, and L.K. Patterson, The Behavior ofNitrobenz-2- Oxa-1,3-Diazol and Coumarin Dye Indicators in Lipid Monolayers - a Study of Interfacial Effects on Polarity and Ph. Thin Solid Films, 1994. 242(1-2): p. 243- 248.

19. Forster, T., in Modern Quantum Chemistry, O. Sinanoglu, Editor. 1965, Academic Press, p. 93-137.

20. Hoffman, R., S. Duff, and N. Qadir. Ion Channel Assay Development and High- throughput Screening Applications Using Voltage Sensor Probes, in Society for Biomolecular Screening 9th Annual Conference. 2003. Portland, OR.

21. Gonzalez, J.E., et al., Cell-based assays and instrumentation for screening ion- channel targets. Drug Discovery Today, 1999. 4(9): p. 431-439.

22. Kriechbaum, M., O.S. Wolfbeis, and E. Roller, Evaluation of critical micelle concentrations of non-ionic detergents using new superpolar lipid probes. Chemistry and Physics of Lipids, 1987. 44: p. 19-29.

23. Jiang, B., et al., Endogenous Kv channels in human embryonic kidney (HEK- 293) cells. MoI Cell Biochem, 2002. 238(1-2): p. 69-79.

24. Zhang, Z., Y. Tang, and M.X. Zhu, Increased inwardly rectifying potassium currents in HEK-293 cells expressing murine transient receptor potential 4. Biochem J, 2001. 354(Pt 3): p. 717-25.

25. Avila, G., A. Sandoval, and R. Felix, Intramembrane charge movement associated with endogenous K+ channel activity in HEK-293 cells. Cell MoI Neurobiol, 2004. 24(3): p. 317-30.

26. Yu, S.P. and G.A. Kerchner, Endogenous voltage-gated potassium channels in human embryonic kidney (HEK293) cells. J Neurosci Res, 1998. 52(5): p. 612- 7.

27. Zhu, G., et al., Identification of endogenous outward currents in the human embryonic kidney (HEK 293) cell line. Journal of Neuroscience Methods, 1998. 81(1-2): p. 73-83.

28. Gunthorpe, MJ. , et al., Characterisation of a human acid-sensing ion channel (hASICla) endogenously expressed in HEK293 cells. Pflugers Arch, 2001. 442(5): p. 668-74.

Claims

WHAT IS CLAIMED IS:

1. An assay method comprising:

(a) exposing a membrane material to a membrane reagent;

(b) exposing the membrane material to a pyrene trisulfonic donor capable of undergoing energy transfer with the membrane reagent;

(c) irradiating the membrane material with excitation light;

(d) monitoring light intensity from at least one fluorophore excitation emission between the membrane reagent and the pyrene trisulfonic donor as a measure of energy transfer; and

(e) relating the measured energy transfer to a change in transmembrane potential.

2. A method as defined in claim 1, wherein the pyrene trisulfonic donor is 8-octadecyloxypyrene-l,3,6-trisulfonic acid.

3. A method as defined in claim 1, wherein the monitoring is performed using a fluorescence resonance energy transfer instrument, and the membrane reagent is a FRET acceptor.

4. A method as defined in claim 1 , wherein the monitoring comprises monitoring light intensity due to fluorescence resonance energy transfer, electron transfer, exchange interaction, paramagnetic quenching, or promoted intersystem crossing.

5. A method as defined in claim 1 , wherein the pyrene trisulfonic donor and membrane reagent are connected by a linker.

6. A method as defined in claim 1, wherein the pyrene trisulfonic donor is

7. A method as defined in claim 1, wherein the membrane reagent comprises polymethine oxonol.

8. A method as defined in claim 1, wherein the membrane reagent comprises a tetraaryl borate- fluorophore conjugate.

9. A method as defined in claim 1, wherein the membrane reagent comprises a transition metal complex.

10. A method as defined in claim 1, wherein: the pyrene donor is selected from the group consisting of PTSi o, PTSi₂, PTS_H, PTSi₆, and PTSi₈; and the membrane reagent is selected from the group consisting of polymethine oxonol, tetraaryl borate- fluorophore conjugates, and transition metal complexes.

11. A method as defined in claim 1 , wherein the membrane material is a biological membrane, a tissue, or a living cell.

12. A method as defined in claim 11 , wherein the membrane material comprises an acid-modulated target selected from acid-sensing ion channels and vanilloid receptor 1.

13. A method as defined in claim 1, wherein the membrane material comprises a protein or polypeptide target, and further comprising contacting the membrane material with a test compound prior to said irradiating.

14. A method of assaying for ion channel activity comprising:

(a) exposing a membrane material comprising an ion channel target to a membrane reagent;

(c) exposing the membrane material to a test compound;

(c) irradiating the membrane material with excitation light;

(e) determining ion channel activity by relating the light intensity to a change in transmembrane potential.

15. A method as defined in claim 14, wherein: the pyrene donor is selected from the group consisting of PTSi₀, PTSi₂, PTSi₄, PTSi₆, and PTSi₈.

16. A method as defined in claim 15, wherein the membrane reagent is selected from the group consisting of polymethine oxonol, tetraaryl borate- fluorophore conjugates, and transition metal complexes.

17. A method as defined in claim 16, wherein the membrane is a living mammalian cell.

18. A method as defined in claim 16, the pyrene trisulfonic donor is PTSi₄

19. A method as defined in claim 14, wherein the ion channel is an acid- sensing ion channel.

20. A method as defined in claim 14, wherein the monitoring is performed using a fluorescence resonance energy transfer instrument, and the membrane reagent is a FRET acceptor.