EP0733209A1

EP0733209A1 - Composition and method for detection of analytes

Info

Publication number: EP0733209A1
Application number: EP95904843A
Authority: EP
Inventors: Stephen J. Kennedy
Original assignee: BIOSYSTEMS TECHNOLOGY Corp
Current assignee: BIOSYSTEMS TECHNOLOGY Corp
Priority date: 1993-12-06
Filing date: 1994-12-06
Publication date: 1996-09-25
Also published as: WO1995016206A1; JPH09509481A

Abstract

The present invention relates to a biosensor composition comprising a binding element, associated with a lipid membrane, in combination with a signaling element to qualitatively or quantitatively detect an analyte in a mixture. The composition is arranged so that binding of an analyte by the binding element produces a measurable change in the signaling element. The composition may optionally contain an intermediary element to couple the response of the binding element to the signaling element through intermediate chemical reactions.

Description

"COMPOSΓΠON AND METHOD FOR DETECTION OF ANALYTES"

Technical Field The present invention relates to a composition and method for detecting components in a fluid. More particularly, the present invention relates to a biosensor composition comprising a binding element, associated with a lipid membrane, in combination with a signaling element to qualitatively or quantitatively detect a component of a mixture. The composition is arranged so that binding of an analyte to the binding element produces a measurable change in the signaling element. Background of the Invention

There are many techniques known in the art for detecting a component in a mixture and/or measuring its amount. Immunoassay s, which employ antibodies that bind specifically to the compound of interest, are one of the better known measurement techniques. Classical methods involve reacting a sample containing the analyte with a known excess amount of antibody specific for the analyte, separating bound from free antibody, and determining the amount of one or the other. Often the antibody is labeled with a reporter group to aid in the determination of the amount of bound analyte. The reporter group or "label" is commonly a fluorescent or radioactive group or an enzyme.

Many early forms of immunoassay employed radioactive analyte in a competitive reaction with unlabeled analyte from the mixture. Both labeled and unlabeled analytes can be bound by the antibody. The amount of radioactivity bound was used to calculate the concentration of the unlabeled analyte in the mixture.

There are disadvantages to using radioactive labels in detection systems. The radioactive labels undergo spontaneous decay which limits their shelf-life and requires recalibration of the equipment. There are environmental and safety concerns associated with the use of radioactive materials. Tests using radioactive labels require sophisticated instrumentation and trained personnel to run the assays. These inherent problems lead to higher costs and restrict the locations where the assay can be performed. For these reasons, radioimmunoassay techniques are being replaced by immunoassay systems employing fluorescent or enzyme labels. Another form of immunoassay, called a "sandwich immunoassay" employs two antibodies specific for the analyte, one of which is labeled with an appropriate reporter group. The first antibody, which is usually bound to a solid support, binds the analyte from the mixture, forming an antibody-analyte complex. The labeled second antibody is added and binds to this complex.

The support is washed to remove unbound labeled antibody, and the amount of bound labeled antibody is determined. If the label on the second antibody is an enzyme, the amount of analyte is quantitated by determining the amount of enzyme activity present. For example, some enzymes catalyze a reaction in which a colored product is formed. The color change can be monitored by spectrophotometry. If the label on the second antibody is fluorescent or luminescent, the amount of bound analyte is calculated based on the amount of fluorescence or luminescence measured.

Attempts at improvements in immunoassay technology revolve around the development of more sensitive labels and simplification of the assay procedure. Homogeneous immunoassays are those which do not require the separation of bound from free. These assays detect the complexation of analyte and antibody in various ways. Fluorescent polarization measurements, for example, can detect bound labeled antibody in the presence of free labeled antibody. The drawback to this technique is that the analyte must be a molecule large enough to significantly affect the rotational diffusion rate of the antibody when it is bound.

Biosensors are a relatively recent development in immunoassay technology that detect the complexation of antibody and analyte using a variety of physical methods. Several kinds of biosensor use antibodies or other binding agents bound to a surface and measure the change in surface charge that occurs when analyte is bound. In Janata et al., U.S. Patent No. 3,966,580, a hydrophobic polymer membrane bearing attached binding agents on its surface is used to coat an electrode used in the measurement of surface charge. In McConnell, U.S. Patent No. 4,490,216, a bilayer membrane with attached binding agents is used to coat an electrode or semiconductor device. In Taniguchi, et al., U.S. Patent No. 4,839,017, an electrically conductive polymer film with attached antibodies or antigens is used to coat an electrode. In Cheung et al., U.S. Patent No. 5,074,977, binding agents are attached directly to the gates of field-effect transistors. In Hafeman et al., U.S. Patent No. 5,164,319, an insulating layer bearing attached binding agents is used to coat a semiconductor electrode. All of these devices sense changes in surface charge, and all suffer the serious limitation that they are sensitive to changes in surface charge produced by any material binding to the surface, regardless of whether it is specifically bound to the binding agents provided. Biological sample fluids such as whole blood, serum, or plasma contain a variety of surface-active agents that adhere non-specifically to any surface in contact with the fluid. These substances are often present in far greater concentration than the analyte being tested, and can greatly interfere with the ability of these sensors to obtain accurate readings.

Other biosensor techniques use binding agents attached to a surface and measure a change in a physical property of the support, other than surface charge, upon binding of analyte. For example, in Malmros, U.S. Patent No. 4,444,892, a semiconductive polymer support is used and changes in the resistance of the support upon binding of analyte are measured. In Reading e t al., U.S. Patent No. 4,927,502, a galvanic electrode with attached binding agents is used to conduct a current. Binding of analyte causes a decrease in the current by interfering with the operation of an electrode. In Ribi, U.S. Patent No. 5,156,810, a conductive polymer layer with attached binding agents is provided. Binding of analyte causes a change in an electrical, optical, or structural property of the conductive polymer layer. Again these techniques are limited in that non-specific binding to the surface is detected as well as specific binding to the binding agents attached to the surface.

Other biosensor techniques use a specific property of a labeled binding agent or antigen to produce a measurable change. Mroczkowski, U.S. Patent No. 4,794,089, uses electrically conductive metal sol particles, such as colloidal gold, attached to a binding agent, to detect complexation of analyte and binding agent. Complexation occurs on a non-conductive surface between two electrodes used to measure changes in conductance upon analyte binding. In Forrest et al., U.S. Patent No.

4,945,045, binding agents are labeled with an electron transfer mediator, such as ferrocene. Complexation enhances or interferes with a redox reaction taking place at the surface of the electrode. In Issichar, U.S. Patent No. 5,156,972, binding agents and analyte analogs are connected together and bound to a sensing surface.

The components are labeled with different groups that are sensitive to the state of complexation of the components, as in the case of a fluorophore and a fluorescence quencher. In the inner complex between binding agent and analyte analog, fluorescence is quenched, and no signal is produced. Upon addition of sample, analyte displaces the analog from the complex, and fluorescence increases. The sensing surface may be a waveguide employing evanescent wave phenomena to detect fluorescence or luminescence, or it may be a piezoelectric device, in which case the components are not labeled, and increases in mass bound to the sensing surface are measured.

Lipid or polymer membranes have been employed in sensor devices where changes in conductivity are detected. In Rechnitz et al., U.S. Patent No. 4,402,819, a hydrophobic polymer membrane, incorporating an antigen bonded to an ion carrier, is used to detect the presence of specific antibody to the antigen by measuring the change in ion conductance that occurs upon antibody binding. In Krull et al., U.S. Patent Nos. 4,637,861, 4,661,235, and 4,849,343 a lipid membrane, incorporating a complexing agent for the analyte, is attached to a support and increases in transmembrane ion movement are detected upon analyte complexation. The complexing agent may be an enzyme, an antibody or a biological receptor. Taylor et al., U.S. Patent No. 5,001,048 provide a film, coated onto a transducer, and polymerized from a mixture of biological receptors, base proteins and stabilizers. Binding of analyte to the receptors produces a change in the electrical characteristics of the film, which is measured. In Fare et al., U.S. Patent No. 5,111,221, a semiconductor substrate is coated with a receptor film specific for the analyte. Binding of analyte causes an increase in ionic current through the film which is measured by the semiconductor and associated circuitry. In Yager, U.S. Statutory Invention Registration H201, membrane proteins specific for a given analyte are extracted from biological cells, purified, and reincorporated into a lipid bilayer where binding of the analyte produces a change in transmembrane voltage or conductance.

In some parts of the prior art, the term "receptor" is used to encompass any kind of binding agent including antibodies, enzymes and lectins; see for example Krull et al., U.S.

Patent Nos. 4,661,235, and 4,849,343, and Ribi, U.S. Patent No. 5,156,810. Such usage is counter to the generally accepted definition of the term. Here the word receptor will be used to refer to an integral membrane protein or glycoprotein having binding activity toward a specific ligand, and capable of transducing the binding of that ligand into a measurable response.

Some of the above mentioned membrane and receptor devices are limited by the use of non-membrane proteins to increase transmembrane conductance. For example, the two Krull patents cited above describe the formation of an antigen- antibody complex used to "perturb" the structure of a lipid bilayer, thereby increasing transmembrane ion flow. Such water-soluble protein complexes are not optimal for promoting ion flow across the hydrophobic interior of a lipid bilayer membrane. Ion channels are proteins appropriate for such use. Ion channels are integral membrane proteins that span the lipid bilayer and provide a conduction pathway for ions. They have the disadvantage that only a few types of compounds can interact with them to affect transmembrane ion flow. All of the above mentioned membrane and receptor devices are limited in that they must use naturally occurring (biological) receptors. Biological receptors are important in cell to cell signaling functions, and receptors typically bind to hormones, growth factors and neurotransmitters, such as insulin, glucagon, steroid hormones, acetyl choline, glutamate, serotonin, cytokines, and peptide neurotransmitters. While some of these compounds are of interest from a diagnostic standpoint, the range of compounds that can be detected using receptors is limited, and comprises those compounds that react with the natural receptor, i.e. the natural ligands and their analogs. The scope and potential utility of these biological receptors is much less than that of other binding agents, such as antibodies, which can be produced to bind to almost any substance. A further limitation of these membrane and receptor devices is the requirement for receptors that produce an increase in ionic flux. Only a few types of receptors are also transmembrane ion channels and are therefore able to transduce the binding of ligand into an increase in transmembrane ion flow. These receptors are members of the ligand-gated ion channel family such as the receptors for acetylcholine, glutamate and gamma amino butyric acid (GAB A). Most receptors transduce the binding of specific ligand into an increase in enzyme activity or the activation of a secondary molecule such as a G protein. What is needed is a means to provide membrane bound receptors with functional binding diversity as great as that of antibodies, and means for coupling the response of receptors to changes in the activity of other agents to produce a measurable signal.

Also needed is a means to provide ion channels for use in modifying transmembrane conductance, and means for gating ion channels, and means for coupling the activity of ion channels to the response of receptors.

Summary of the Present Invention

The present invention is a composition for measuring an analyte in a fluid comprising a binding element, associated with a membrane, the binding element being capable of modulating the activity of a signaling element in the presence of the analyte. One or more intermediary elements may be included in the composition to couple the response of the binding element to the signaling element through intermediate chemical reactions.

Because the activity of the signaling element reflects the state of complexation of the binding element, the composition may be used in a homogeneous immunoassay system.

In some embodiments, the binding element can modulate the activity of a plurality of signaling agents, providing a means for amplification of the signal associated with analyte binding. In one embodiment of the present invention, the binding element is a naturally occurring protein selected from members of the immunoglobin superfamily, such as antibodies, cellular adhesion molecules, the T-cell receptor, T-cell accessory molecules such as CD2, CD3, CD4 and CD8, Fc receptors, certain receptor protein tyrosine kinases, and many others.

The binding element can also be a genetically engineered form of a member of the immunoglobin superfamily. In creating the binding element by genetic engineering, changes can be introduced into the molecule such as an alteration in the specificity of binding, a change in the response of the binding element to ligand binding, addition or deletion of a means of membrane attachment, or a change in the means thereof.

The binding element can also be a naturally occurring protein that functions as a receptor for molecules such as hormones, growth factors, cytokines, neurotransmitters, odorants, vitamins, antibodies, bacteria, viruses, serum lipoproteins, other proteins, toxins, or binding agent-ligand pairs such as antibody-antigen complexes, carbohydrate-lectin complexes, and complexes of metal-binding molecules with metal ions.

The binding element can also be a reassortment of binding domains from naturally occurring proteins to create a new binding specificity. The binding element can also be an artificially created membrane-bound form of a receptor normally found in the cytosolic fraction of a cell. The receptor can be membrane bound by either genetically engineering a membrane component for the receptor or the membrane component can be covalently attached to the receptor.

In another embodiment ofthe present invention, the signaling element is an enzyme. The signaling element can also be a genetically engineered enzyme that can catalyze reactions of new substrates, exhibit altered reaction kinetics of normal substrates, exhibit altered binding of coenzymes, or require different coenzymes or factors.

The signaling element can also be a member of the naturally occurring group of voltage-gated ion channel proteins or a member of the naturally occurring group of ligand-gated ion channel proteins.

The signaling element can also be a genetically engineered member ofthe superfamily of voltage-gated ion channel proteins, or the superfamily of ligand-gated ion channel proteins. The genetically engineered feature of the channel is an alteration of the naturally occurring form. This may be a change in the type of ion which traverses the channel, a change in the voltage- dependence or gating kinetics of the channel, the addition, deletion, or modification of a channel blocking region or domain, a change in the number of protein domains or subunits associated to form the channel, a change in the types of protein domains or subunits associated to form the channel, or a change in other aspects of the protein domains or subunits associated to form the channel. In another embodiment of the present invention, the membrane is selected from a group consisting of lipid bilayers and lipid monolayers. The lipid membrane can optionally be crosslinked, and may contain molecules other than lipids.

In another embodiment of the present invention, one of the above combinations of binding elements and signal elements are associated with one of the possible configurations of the lipid membrane and an electrical parameter of the composition is measured. In another embodiment, one of the above combinations of binding elements and signal elements are associated with one of the possible configurations of the lipid membrane and an optical parameter of the composition is measured.

One object of the present invention is to provide membrane bound receptors with functional binding diversity as great as that of antibodies.

Another object of the present invention is to provide means for coupling the response of receptors to changes in the activity of other agents to produce a measurable signal. Another object of the present invention is to provide ion channels for use in modifying transmembrane conductance, and means for gating ion channels, and means for coupling the activity of ion channels to the response of receptors. A further object of the present invention is to provide means for producing, handling, manipulating and arranging receptors, ion channels and other agents.

Yet another object of this invention is to provide a homogeneous immunoassay system for the detection of a great multitude and variety of compounds, and to provide for the detection of a multiplicity of compounds at one time.

The composition of the present invention can be used to provide an assay system to accurately measure analytes in a fluid. The fluid can be a mixture such as blood, serum or other bodily fluids.

The present invention can be used to monitor changes in the concentrations of analytes in a sample as a diagnostic determinant, over the course of a disease, during surgical procedures, or over the course of treatment. The present invention can be used to detect the presence and/or determine the amount of infectious agents, hormones, drugs, metabolites, or other chemicals in diagnostic samples from humans, animals, or other organisms. The present invention can be used to screen candidate pharmaceutical compounds in the search for new drugs and therapeutic agents.

The present invention can also be used to detect the presence of environmental toxins or contaminants in samples from animals, plants, soil, air or water.

These and other objects, features and advantages of the present invention will become apparent after a review of the following detailed description of the disclosed embodiment and the appended claims.

Brief Description of the Drawings

Figure 1 is a schematic representation of combined binding and signaling elements in one membrane protein and also dimerization of the combined-form membrane protein with multivalent binding element membrane protein.

Figure 2 is a schematic cross sectional representation of a tyrosine kinase functioning as a binding element and an intermediary element.

Figure 3 is a schematic cross sectional representation of a typical voltage-gated ion channel.

Figure 4 is a schematic representation of univalent binding elements interacting with the specific analyte to form a dimer and trigger the signal element. Figure 5 is a schematic representation ofthe normal functioning of a voltage-gated ion channel. The open, closed and inactivated stages are illustrated.

Figure 6 is a schematic representation of binding of an analyte to inhibit the inactivation stage of the voltage-gated ion channel.

Figure 7 is a schematic representation of the inhibition ofthe functioning of the ion channel by an attached toxin molecule. With the addition of a specific binding element, the presence of an analyte prevents the inhibition by the attached toxin molecule.

Figure 8 is a schematic representation of the support structure for the lipid membrane.

Figure 9 is a schematic representation of the molecule which anchors the lipid membrane to the support structure.

Detailed Description

The present invention provides a composition for detecting the presence of an analyte in a mixture. The composition comprises a binding element, such as an antibody or receptor, associated with a membrane, and combined with a signaling element, such as an enzyme or ion channel. Binding of an analyte by the binding element produces a measurable change in the activity of the signaling element. Intermediary elements may be included in the composition to couple the response of the binding element to the signaling element through intermediate chemical reactions.

Binding Elements

The binding elements of the present invention may be selected from the immunoglobin family of proteins or other naturally occurring binding proteins or receptors. The binding elements may also be portions of, combinations of, or altered forms of these proteins or entirely new forms of proteins developed by genetic engineering techniques which are well known in the art.

The immunoglobin superfamily includes, but is not limited to, antibodies including all classes of antibodies including, but not limited to, IgG, IgE, IgM and IgA, cellular adhesion molecules, the T-cell receptor, T-cell accessory molecules such as CD2, CD3, CD4 and CD8, the B-cell receptor, Fc receptors, certain receptor tyrosine kinases, certain receptor tyrosine phosphatases, and others. The distinguishing feature of these molecules is a characteristic domain structure known as the immunoglobin fold. The three dimensional structure of the immunoglobin fold is well characterized and genetic engineering techniques can be used to modify antibody binding properties. The binding sites of these proteins are formed by polypeptide strands known as Complementarity Determining Regions or CDRs. CDRs can be transferred from one member of the superfamily to another using standard genetic engineering techniques. In addition, the domains that contain the CDRs and determine specificity of binding, the Fv domains, may be transferred from one member to another. For example, the CDRs or Fv regions of an antibody with known binding specificity may be transferred to a membrane- bound member of the superfamily to form a binding element having the specificity of the parent antibody. Larger segments or multiple domains may also be transferred. Techniques for modifying antibody molecules and for producing single-chain polypeptides that contain antibody variable regions are described in U.S. Pat. No. 4,946,778, issued to Ladner et al., which is hereby incorporated by reference.

The response of the binding element to binding of the analyte is dependent on the type of binding element.

Some binding elements may activate guanosine nucleotide-binding proteins (G proteins) which in turn activate or inhibit adenylate cyclase. Adenylate cyclase causes the production of the "second messenger" cyclic AMP, which activates a protein kinase cascade, resulting in the phosphorylation of a variety of proteins.

Other binding elements may activate a G protein that activates phospholipase C (PLC). This results in the production of two second messengers: diacylglycerol (DAG) and inositol 1,4,5-triphosphate (IP₃). Diacylglycerol activates protein kinase C (PKC) to cause phosphorylation of a variety of proteins, and IP₃ causes the release of calcium ions from internal stores by opening (or gating) a calcium ion channel.

Still other binding elements may activate protein kinases, phosphatases, cyclases or other enzymes or effectors. Compounds activated directly by the binding elements of the present invention may be utilized as signaling elements in various embodiments, or they may be used as intermediary elements to couple the response of the binding element to the signaling element via intermediate chemical reactions. One or more intermediary elements may be included in the composition of the present invention.

In Figure 1, a combined form of binding element 4 and signaling element 5 e.g., a receptor tyrosine kinase (RTK) is shown. The membrane-spanning portion 6 of the composite molecule prevents the molecule from dissociating from the lipid membrane 1. Binding of analyte, 7, causes an increase in kinase activity. In Figure 2, the receptor tyrosine kinase functions as a binding element and an intermediary element is shown. Binding of analyte, 7, causes the kinase domain to phosphorylate the signaling element, in this case a tyrosine phosphatase domain, 8, with associated membrane-spanning portion, 9. Phosphorylation produces a change in the activity of the phosphatase, which is measured.

Multiple binding elements may also be used in certain embodiments of the present invention. An example is the combination of an antibody and an Fc receptor specific for the antibody. Binding of analyte to the antibody triggers a response of the Fc receptor.

Intermediary Elements The intermediary elements are optional components of the present invention. The intermediary elements may be naturally occurring proteins or chemicals or proteins modified by genetic engineering. Examples of intermediary proteins are G proteins, calmodulin, kinases, phosphatases, cyclases, and lipases. Examples of intermediary chemicals are cyclic nucleotides, diacylglycerol, inositol triphosphate, and calcium ions.

Intermediary proteins may also be altered by genetic engineering techniques. For example, some G Proteins are soluble cytosolic proteins and therefore do not reside in or on membranes. A cytosolic G protein can be made membrane bound by the addition of a hydrophobic region. Portions or domains of intermediary proteins may also be transferred from one type of intermediary protein to another. For example, the tyrosine kinase domain of an RTK can be replaced with a serine/threonine kinase domain to change the substrate specificity of the parent RTK.

In embodiments of the invention that use intermediary elements, the signaling element can be phosphorylated or dephosphorylated to produce a measurable change in its activity. Also, the signaling element can respond to cyclic AMP or an activated G protein or a second messenger such as IP₃ or calcium ion to produce a measurable change.

Signaling Elements

The signaling elements ofthe present invention may be selected from a group consisting of enzymes, the superfamily of voltage-gated ion channel proteins, and the superfamily of ligand-gated ion channel proteins. The signaling elements may also be portions of, combinations of, or altered forms of these proteins or entirely new forms of proteins developed by genetic engineering techniques.

The superfamily of voltage-gated ion channel proteins is exemplified by the Shaker potassium channel from Drosophila. This family of proteins includes sodium, potassium and calcium channels, cyclic nucleotide-gated channels, and calcium-activated potassium channels. These proteins are distinguished by a characteristic domain structure consisting of six putative transmembrane helical regions, and a pore region that defines channel selectivity.

The normal functioning of a typical voltage-gated ion channel is shown in Figure 3, in cross-section. The channel protein 10 spans the lipid membrane 1. In the closed state of the channel, the channel domains or subunits 11 do not permit ions to flow through the channel. Depolarization of the membrane favors a conformational change in the subunits to produce the open state, which provides a pathway, 12, for transmembrane ion flow through the channel protein. Subsequent to channel opening, the ion pathway is blocked by a protein domain, 13, associated with the channel, to produce the inactivated state. Upon repolarization of the membrane, the channel returns to the closed state and is ready to be activated again.

The superfamily of ligand-gated ion channel proteins is exemplified by the nicotinic acetylcholine receptor. This family of proteins includes receptors for acetylcholine, glycine, and GABA, and is characterized by a domain structure containing four putative transmembrane helical regions.

The structure of ion channels is well defined and genetic engineering techniques can be used to modify the properties of ion channels. For example, the region of the channel protein that forms the lining of the "aqueous pore" can be transplanted from one channel type to another, thus changing the ion selectivity of the channel. Individual amino acid residues in the pore region may also be modified, and a particular channel may be made more selective or less selective for a given ionic species. The activation and inactivation properties of the channel can be modified by deleting the region of the channel that is responsible for channel inactivation, or by transplanting the "voltage sensor" region (helix S4 and associated residues) from one channel species to another. Phosphorylation sites may be introduced at various positions in the channel protein, and existing phosphorylation sites may be removed or modified.

The multimeric structure of a channel may be modified, for instance, by ligating two (or four) potassium channel polypeptides to form a covalently linked dimer (or tetramer), or by separating the four domains of the sodium channel into non- covalently linked polypeptides.

Voltage-gated ion channels are so named because they are opened and closed by changes in the transmembrane potential. However, a variety of second messengers and effector proteins also affect or modify the function of voltage-gated ion channels. Calcium-activated potassium channels (K(Ca) channels) are opened by increases in calcium ion concentration. Some types of K(Ca) channel are opened synergistically by voltage and calcium ion, while others are only slightly voltage-dependent. Cyclic nucleotide-gated channels are opened by cyclic GMP and/or cyclic AMP. A type of calcium channel is gated by IP₃ as mentioned above, and a related channel is gated by IP₄. Finally, there are examples of ion channels directly activated by G proteins as well as channels modified through phosphorylation by protein kinases.

Any of these kinds of ion channel may be used as the signaling element in the practice of the present invention. A particular signaling element may require that the appropriate intermediary element(s) be included in the composition, so that binding of an analyte to the binding element produces a measurable change in the activity of the signaling element.

The activity of ion channels can be measured electrically, and techniques for these measurements are well known in the art. Electrical measurement techniques can detect the activity of a single ion channel protein. Usually, the transmembrane voltage is controlled ("voltage clamp" conditions) and transmembrane current is measured. This current is the sum of the current through all of the ion channels present plus a background or "leakage" current component. However, open circuit voltage measurements of transmembrane potential can also reflect the activity of ion channels. Also, ion channel activity may be measured indirectly, for example, by means of potential sensitive fluorescent dyes or by measuring a secondary effect of the channel activity, as in the case of a calcium channel current flow causing the opening of K(Ca) channels.

Enzymes may be used as signaling elements in the present invention, and the kinds of enzymes that may be used include kinases, phosphatases, cyclases, and lipases. Various techniques for measurement of enzymatic activity are known in the art. Spectrophotometric techniques measure a change in the absorb ance or fluorescence of an appropriate enzyme substrate upon exposure to the enzyme, and may be miniaturized by the use of optical fibers to conduct the incident and emitted or transmitted light beams. These and other similar techniques may be used in the practice of the present invention where the signaling element is an enzyme.

Binding Elements Linked to Signaling Elements.

Receptor tyrosine kinases (RTKs) are a diverse family of membrane receptors that transduce the extracellular binding of analyte into phosphorylation of tyrosine residues on the interior side of the cell membrane. The first event that takes place after analyte binding is autophosphorylation of certain receptor tyrosine residues. These phosphorylated tyrosine residues serve as binding sites for various intermediary proteins that bind to the RTK and are then activated by tyrosine phosphorylation. Activation of these intermediary proteins results, either directly or indirectly, in the production of second messengers such as cAMP, IP₃, calcium, or diacylglycerol. Some intermediary proteins are themselves kinases, and their activation results in a cascade of phosphorylation events that activates (or inactivates) a variety of proteins and enzymes throughout the cell.

Many RTKs are members of the immunoglobin superfamily of proteins since they contain extracellular binding domains that are homologous to Ig domains. These regions may be modified using genetic engineering techniques as described above to alter binding specificity. In this manner, RTKs may be produced that display selectivity for essentially any analyte. Since RTKs contain both a binding domain and a measurable enzyme activity that is induced upon analyte binding, some embodiments of the present invention may use a single RTK to provide both a binding element and a signaling element. The kinase activity of a particular RTK may be measured by various means. One method involves measuring the incorporation of radioactive phosphate into substrate proteins. Another procedure is to measure the increase or decrease in absorbance or fluorescence of a dye that is able to be phosphorylated by the kinase domain.

Analyte binding by RTKs has been shown to result in receptor dimerization in some cases, and it is believed that this may be a necessary part of signal transduction. In certain types of RTKs, binding of a monovalent ligand produces a conformational change in the receptor that results in dimerization. In other types of RTKs, dimerization is produced by the binding of a multivalent ligand to two receptors. In some cases, two identical univalent receptor molecules bind to one molecule of ligand to form a homodimeric receptor-analyte complex. Some multivalent ligands bind to two different receptor molecules to form a heterodimeric receptor-analyte complex.

In the practice of the present invention, such dimerization may be provided by the binding of multivalent analytes, or, when it is desirable to detect monovalent analytes, dimerization may be produced by the use of a type of RTK that is activated by monovalent ligands. Alternatively, inclusion of a second, multivalent, binding component specific for the analyte of interest may be used to induce dimerization. The multivalent binding component should be designed to bind to an epitope on the analyte that is distinct from the epitope bound by the first binding component. In Figure 4, a multivalent analyte 14 is bound at one epitope by a combined-form of binding element 4 and signal element 5 in one molecule, and at another epitope by a multivalent binding element 15. The binding of a second molecule of analyte 14 causes aggregation of the binding elements, autophosphorylation and activation of the signal elements, 5.

Another means of providing analyte-induced receptor dimerization is to incorporate a portion of the binding element into each of two separate proteins. This concept is also illustrated in Figure 4 where a univalent binding element 16 is composed of two separate domains, 17 and 18, each connected to a membrane spanning portion, 6, and a signal element 5, in this case a tyrosine kinase. The binding domains are selected, or designed, so that the affinity of each domain for the other is insufficient to produce dimerization in the absence of analyte.

Binding of analyte 19 induces the formation of a dimer 20. Dimerization causes autophosphorylation of the tyrosine kinase domains and activates the tyrosine kinase activity of the signaling elements. In other embodiments, the tyrosine kinase domain may function as an intermediary element to couple analyte binding to the signaling element or to other intermediary elements. The enzyme phospholipase C-gamma (PLC-g) is activated by phosphorylation by a tyrosine kinase associated with the receptor for platelet derived growth factor (PDGF). One embodiment of the present invention which uses a plurality of intermediary elements is the combination of an RTK, the enzyme PLC-g, phosphatidylinositol 4,5-bisphosphate (PDP₂), and an IP₃-gated calcium channel. In this embodiment, binding of analyte activates the RTK kinase activity, which activates PLC-g, which cleaves PIP₂ into DAG and IP₃, and IP₃ opens the calcium channel. The calcium channel in this embodiment may act as the signaling element, and its activity may be measured by various methods known in the art. Alternatively, the calcium ions conducted by the channel may be used as another intermediate element to activate a signaling element such as a luminescent protein like aequorin, or a calcium-sensitive fluorescent dye.

In the practice of these embodiments, the genetic code for any proteins forming the binding element, intermediary elements, and signaling element may be combined in the genome of an appropriate cloning or expression vector and expressed in a host cell line transformed with such vectors. Protein Engineering of Ion Channels.

In their natural state, voltage-activated ion channels are opened by depolarization of the cell membrane. See Figure 3. Following a change in membrane potential, voltage activated channels typically open, conduct a flow of ions for a time, and then close by a mechanism known as inactivation. Inactivation is a separate process from activation, and involves a conformational change in the channel protein that results in the physical obstruction of the intracellular mouth of the channel. Upon restoration of the normal membrane potential, inactivated channels return to the closed state.

Ion channels may be modified to respond to the presence of an external analyte by providing means for analyte- dependent alteration of channel function. As shown in Figure 5, a binding element 4 which specifically binds an analyte 22 can be introduced into a voltage-gated ion channel protein to produce a hybrid channel that is also ligand-gated. In the absence, 21, of analyte, the channel subunits 11 and the inactivation particle 13 function normally. The binding of analyte 22 by the binding element 4 prevents the inactivation of the channel by the inactivation structure 13.

Ion channels are also blocked by some toxins in a manner reminiscent of channel blockade at the internal mouth by an inactivation particle. Charybdotoxin (CTX) is a 37 amino acid peptide, isolated from scorpion venom, which binds to and occludes the external mouth of the Shaker potassium channel. Figure 6 shows an ion channel protein 10, without an inactivation domain, and with an introduced binding element, 4, and toxin molecule 23. The toxin molecule may be attached to the channel, as shown in Figure 6, or it may be free in solution. In the absence, 24, of analyte, the channel is blocked by the toxin molecule. When analyte 25 is bound by the binding element 4, the toxin molecule 23 is prevented from inhibiting the functioning of the ion channel. Another means of producing analyte-dependent modulation of ion channel function is shown in Figure 7. An ion channel protein 10 with associated binding element 4 also has associated an analyte analog 26 and blocking region 27. The analyte analog and attached blocking region may be attached to the channel, as shown in Figure 7, or they may be free in solution.

In the absence of analyte, the analog is bound to the binding element, holding blocking region 27 in proximity to the channel opening and preventing ion flow. In the presence of analyte 28, the analyte analog and blocking region are displaced, permitting ions to flow through the channel.

It will be recognized that the important feature of these embodiments is that there be a measurable difference in channel function in the presence and absence of analyte. Insertion of the binding element may alter any number of channel properties compared to the unmodified or "native" channel. Binding of analyte may not completely abolish inactivation or toxin binding or analyte analog binding. These effects are unimportant as long as the presence of analyte produces a measurable change in channel function. Members of the voltage-dependent ion channel superfamily of proteins are either oligomeric, or have multiple homologous domains. Consequently, binding elements may be introduced at multiple sites in the channel. This would be advantageous where improved sensitivity or specificity of binding is desired. Two binding elements with specificity for distinct epitopes on the binding partner can be arranged so that simultaneous binding to both elements is required before channel function is modified. This arrangement provides enhanced sensitivity and enhanced specificity compared to the situation where there is only one binding element per channel. Since potassium channels are formed by a tetrameric arrangement of protein monomers, insertion of a binding element into the monomer produces a channel with four identical binding sites. Alternatively, two or four monomer proteins may be joined and binding elements inserted so that channels are formed with two or four binding sites.

This arrangement would provide enhanced sensitivity and specificity in the situation where a multivalent analyte is to be detected.

Obviously, the techniques used in these embodiments of the invention may be applied to other types of channels besides voltage-gated channels. Blocking proteins or toxins are known for many types of ion channels, and new blocking proteins may be produced for a given ion channel by genetic engineering techniques. One method of producing new blocking proteins or peptides for a particular ion channel is to use the channel to screen a library of random sequence polypeptides displayed on the surface of bacteriophage. Phage that bind to the channel may be isolated and cloned, and the sequences of the binding polypeptides determined. A second screening may be used to determine which peptides block the channel as well as bind.

Multiple cycles of this process may be used to fine tune the size of the blocking particle or its affinity for the channel. The same process can be used to produce analyte analogs that bind to a particular binding element. In these embodiments, the ion channel and its associated blocking protein functions as the signaling element and the channel contains an inserted domain that functions as the binding element. In the practice of these embodiments, the genetic code for the binding element and signaling element may be combined in the genome of an appropriate expression vector and used to transform an appropriate host cell line.

Membrane Supports

In the practice of the present invention, binding elements and signaling elements are combined and associated with a lipid membrane. This combination and association can be accomplished, in certain embodiments, through the use of genetic engineering techniques to cause the binding and signaling elements to be expressed in an appropriate host cell line. Cell lines and expression vectors appropriate for use with the elements of the present invention are well known in the art. The lipid membrane employed can have a bilayer or monolayer structure, and may include molecules other than lipids, where necessary, as stabilizers or enhancers of membrane structure. The membrane can be derived from the cell membrane of an expression vector, and may contain any of the molecules present in the membrane of the parent cell. The membrane may be attached to, or associated with a solid support.

The present invention may provide cells or vesicles containing the binding elements and signaling elements embedded in the membrane surrounding these cells or vesicles, so that exposure of the external side of the cells or vesicles to the analyte in the mixture produces the change in the signaling element.

Alternatively, the protein-containing lipid membranes of the subject invention may be incorporated onto a support for use in analysis and binding applications. To form the membranes, the proteins of the subject invention can be extracted from the membranes of the cell in which they were produced and incorporated into preformed lipid membranes. The proteins may be incorporated into lipid vesicles for ease of handling and enhanced protein stability. These vesicles may be fused with planar lipid membranes, or vesicles can be attached directly to a support and converted into planar format. The planar format has the advantage that both sides of the membrane are accessible, and the support can be designed so that the sample mixture may be applied to the side of the membrane that was originally intracellular.

Planar membranes can be provided by attaching vesicles or cells to a support over or within an annular opening, and breaking open said vesicles or cells. Figure 8 shows a support useful for the practice of converting vesicles or cells into planar membranes. Cylindrical apertures 29 in a thin electrical insulator 30, are surrounded, on one surface, by an annular ring 31 containing on its surface, attached hydrophobic molecules 32 capable of penetrating into or through the lipid membrane of the cells or vesicles. Photolithography techniques are used to define the annular ring as well as adjacent areas containing deposited metal such as silver or platinum electrodes 33. Upon exposure of the support to a suspension of vesicles or cells, said vesicles or cells become affixed to the annular ring, and centered over the cylindrical opening.

The hydrophobic molecules 32 are attached to the annular ring through reactive groups 34 located at one end of the molecule. As shown in Figure 9, the hydrophobic portion 35 of the hydrophobic molecule 32 is connected to a polar, photolabile protective group 36 on the distal end of hydrophobic portion 35. The polar group 36 is designed to prevent the penetration of the hydrophobic molecule into the membrane of the cell or vesicle. In such case, photolithography techniques can be used to break the photolabile linkage 37 and "deprotect" a portion of the support so that vesicles or cells can be affixed to only that portion of the support. Subsequent exposure of other sections and fixation of different vesicles or cells provides means for producing a sensor capable of detecting multiple analytes simultaneously. In addition, the hydrophobic portion of the molecule may be connected to nonpolar cleavable groups 38 that pass through the membrane and reveal polar groups upon cleavage. The linkages 39 connecting these groups may be cleaved by enzymes or other agents within the vesicles or cells and the revealed polar groups, 40 prevent the withdrawal of the hydrophobic molecule once it has penetrated the vesicle or cell.

The hydrophobic portion 35 may be a hydrocarbon, a peptide, or other compound of appropriate size and hydrophobicity. The reactive groups 34 may be sulfhydryl, amino or other groups suitable for covalent reaction with the surface of the annular ring. The photolabile linkage 37 may be an ortho-nitro ester or the like. Polar groups 36 and 40 may be carboxylate, sulphate, phosphate, carbohydrate, and the like, and combinations thereof. Nonpolar cleavable groups 38 may be aliphatic or aromatic esters or the like. Vesicles or cells can also be fixed within the cylindrical apertures through the use of hydrostatic pressure, or hydrophobic molecules as described above.

After attachment to the support, the vesicles may be broken open to provide a planar membrane. The breakage may be accomplished electrically, by a method similar to electroporation, where the magnitude and duration of the applied voltage are made sufficiently large to prevent resealing of the vesicles or cells. The electrodes disclosed earlier can be used to apply the voltage pulse for disruption of the vesicles or cells as well as to monitor the current through ion channels contained in the resulting planar membrane. In the photolithography process, electrodes are also provided on the lower surface of the thin support so that the planar membrane has electrodes on each side. These electrodes are used to apply a transmembrane voltage and measure the resulting transmembrane current. Alternatively, the breakage of the cells or vesicles of the subject invention can be effected through the use of ultrasound, shear forces, or physical contact. In the latter case, cells or vesicles are attached to two opposed surfaces using the membrane penetrating molecules described above. The two surfaces are then pulled apart, leaving behind planar membranes attached to each surface.

This invention is further illustrated by the following examples, which are not to be construed in any way as imposing limitations upon the scope thereof. On the contrary, it is to be clearly understood that resort may be had to various other embodiments, modifications, and equivalents thereof which, after reading the description herein, may suggest themselves to those skilled in the art without departing from the spirit of the present invention and/or the scope of the appended claims.

Example I

Cloning, .Modification and Expression of an Ion Channel in Oocytes. cDNA containing the 1.8 kb Shaker K⁺ channel cDNA (ShaK), or a mutant channel with amino acid residues 6 to

46 deleted to remove inactivation (ShaKΔI) is obtained using standard polymerase chain reaction (PCR) techniques, and inserted into a commercial vector (pBluescript, Stratagene). Insertion of a 17 amino acid peptide epitope into the Shaker K⁺ channel open reading frame is made as follows. Briefly, primers are designed flanking the start and stop codons of the Shaker open reading frame, as well as within the S1-S2, S2-S3 and S3-S4 loops. The primer pairs within each of the loops contain an extension on their 5' ends corresponding to the coding region for the 17 amino acid peptide. The initial PCR reactions using the cloned Shaker cDNA as a template, generates products corresponding to the NH₂- and COOH-terminal sides of the open reading frame, each with the antigen insert appended to their respective ends. The secondary rounds of PCR using the NH₂- and COOH-terminal primers and the two initial PCR products as templates result in a fusion of the two halves of the Shaker cDNA construct with the antigen tag spliced into the joining region. Primers used in these constructs are obtained commercially (Operon Technologies) and extended using Pfu polymerase (Stratagene) in a commercial PCR thermal cycle instrument (Perkin Elmer).

The primary PCR products are desalted over a Sephadex G-50 column to remove the first round primers and quantitated by ethidium bromide staining on an agarose gel. Equal amounts of the two initial PCR products are then mixed and subjected to the secondary round of amplification. The final PCR product is cloned into the TA-cloning vector (pCR II, Invitrogeπ) as described by the manufacturer. The final constructs are sequenced as double stranded templates using the "dideoxy" chain termination method and synthetic oligonucleotide primers to confirm that no errors have been into the coding region during the PCR process. The recombinant cDNA plasmids containing the modified K⁺ channel cDNAs are linearized by digestion with Hind III, extracted with 1:1 pheno chloroform, made 0.3 M in sodium acetate, and precipitated with 2.5 volumes of ethanol. Linearized template DNA is quantitated by spectrophotometry at 260 nm.

Transcription into mRNA is done at 37 ^*C in 40 mM Tris (pH 7.4) containing 10 mM NaCl, 10 mM dithiothreitol, 6 mM MgCl₂, 2 mM spermidine, 0.5 mM diguanosine 5 '-5' triphosphate, 0.5 mM of each ribonucleoside triphosphate, 1-3 mg of template cDNA, 140 units of human placental RNase inhibitor, and 60 units of T7 RNA polymerase (Promega). Following in vitro synthesis of RNA, cDNA templates are destroyed by a 15 minute incubation with 5 units of DNase I (Promega) at 37'C.

RNA is extracted with 1:1 pheno chloroform, made 0.3 M in sodium acetate, and precipitated with 2.5 volumes of ethanol. Following centrifugation, RNA pellets are washed with 70% ethanol and resuspended in distilled water at approximately 200 ng/μL. The transcript is quantitated by agarose gel electrophoresis followed by ethidium bromide staining.

Fresh Xenopus oocytes (stage V and VI) are obtained commercially (Xenopus One) and defolliculated with 2 mg/mL collagenase (Typel, Sigma) in a solution containing 100 mM NaCl, 2 mM KC1, 1 mM MgCl₂, and 5 mM HEPES, pH 7.5.

Treated oocytes are stored in ND96 (96 mM NaCl, 2 mM KC1, 1.8 mM CaCl₂, 1 mM MgCl₂ and 5 mM HEPES, pH 7.5) supplemented with penicillin (100 units/mL), streptomycin (100 μg/mL) and 2.5 mM sodium pynivate. Within 24 hours, the oocytes are injected with 50 nL RNA (approximate concentration

200 ng/μL) using a micropipettor and incubated in the above solution for 2-3 days at 20 ^*C to express potassium channels. Example II

Insertion of an Analyte Binding Region into an Ion Channel, and Modulation of Channel Function in the Presence of Analyte. cDNA coding for a single-chain binding protein specific for bovine growth hormone (bGH) is prepared according to Example 2 in Ladner, et al., U.S. Patent No. 4,946,778. This cDNA construct is inserted in-frame into the region of the ShaK cDNA corresponding to the S2-S3 loop of the channel, as described in Example I, above. The final PCR product is purified, transcribed into mRNA and expressed in Xenopus oocytes as described above. Channel function is assayed using standard patch clamp techniques as described in Smith, et al., 1985. In excised patches, addition of bGH to the bath produces a marked slowing of the rate of channel inactivation.

It should be understood that the foregoing relates only to a preferred embodiment of the present invention and that numerous modifications or alterations may be made therein without departing from the spirit and the scope of the invention as set forth in the appended claims.

References

Darnell, J., Lodish, H. and Baltimore, D. Molecular Cell Biology

W.H. Freeman & Co., New York. 1990.

Watson, J.D., Gilman, M., Witkowski, J. and Zoller, M.

Recombinant DNA. W.H. Freeman & Co., New York. 1992.

Kuby, J. Immunology. W.H. Freeman & Co., New York. 1992.

Miller, C. Ion Channel Re constitution. Plenum Press, New York.

1986.

Roberts, G, Ed. Langmuir-Blodgett Films. Plenum Press, New

York. 1990.

Smith, T.G., Lecar, H., Redman, S.J. and Gage, P.W., Eds.

Voltage and Patch Clamping with Microelectrodes. Williams and

Wilkins, Baltimore, MD. 1985.

Williams, A.F. and B arclay, A.N. (1988). The Immunoglobin Superfamily - Domains for Cell Surface Recognition. Ann. Rev. Immunol. 6: 381-405.

Young, P.R. (1992). Protein Hormones and Their Receptors.

Curr. Opinion in Biotechnology. 3: 408-421.

Jorgensen, J.L., Reay, P. A., Ehrlich, E.W. and Davis, M.M.

(1992). Molecular Components of T Cell Recognition Ann. Rev.

Immunol. 10: 835-873.

Reth, M. (1992). Antigen Receptors on B Lymphocytes. Ann. Rev.

Immunol. 10: 97-121. Ravetch, J.V. and Kinet, J.-P. (1991). Fc Receptors. Ann. Rev. Immunol. 9: 457-492.

Fantl, W.J., Johnson, D.E. and Williams, L.T. (1993). Signaling by Receptor Tyrosine Kinases. Ann. Rev. Biochem. 62: 453-481. Gilman, A.E. (1987). G Proteins: Transducers of Receptor- Generated Signals. Ann. Rev. Biochem. 56: 615-649. Majerus, P.W. Inositol Phosphate Biochemistry. (1992). Ann. Rev. Biochem. 61: 225-250.

Pongs, O. (1992). Molecular Biology of Voltage-Dependent Potassium Channels. Phiosiol. Rev. 72: S69-S88.

Claims

1. A composition for measuring an analyte in a fluid comprising: a. a binding element associated with a membrane, the binding element being capable of binding to the analyte; and b. a signaling element associated with the binding element so that the activity of the signaling element is modulated in the presence of the analyte.

2. The composition of Claim 1, further comprising an intermediary element that is functionally associated with the binding element and the signaling element.

3. The composition of Claim 1, wherein the binding element is selected from the group consisting of immunoglobulins, cellular adhesion molecules, T-cell receptors, T- cell receptor accessory molecules, Fc receptors, and receptor tyrosine kinases.

4. The composition of Claim 1, wherein the signaling element is selected from the group consisting of voltage- gated ion channel proteins, and ligand-gated ion channel proteins.

5. The composition of Claim 2, wherein the intermediary element is selected from the group consisting of G proteins, kinases, phosphatases, cyclases, and lipases.

6. The intermediary element of Claim 5, wherein the intermediary element comprises cyclic nucleotides, diacylglycerol, inositol triphosphate, and calcium ions.