Capacitive Sensor for Chemical Analysis and
Measurement Corresponding to U.S. Patent Application Serial No. 799,761
Filed: November 19, 1986
Be it known that Arnold L. Newman, a citizen of the United States of America, and resident of Maryland, has invented a certain new and useful apparatus for detecting the concentration of an analyte in a fluid medium, of which the following is a specification: STATEMENT OF GOVERNMENTAL INTEREST
The Government has rights in this invention pursuant to Contract No. N00024-85-C-5301, awarded by the Department of the Navy.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an apparatus for determining the concentration of an analyte in a fluid medium. More particularly, the invention relates to a capacitive sensor which is uniquely designed to detect a change in the dielectric constant caused by biospecific binding of an analyte with a biochemical binding system. The biochemical binding system is selected to have specific affinity to the particular analyte or group of analytes under test.
2. Description of the Prior Art
Various prior art techniques have attempted to measure the concentration of an analyte in a fluid medium using a binding substance having specific affinity for the analyte. Immunoassays are used to identify analytes, such as haptens, antigens and antibodies in a fluid medium. These immunoassays are based on biospecific binding between components of a reaction pair, such as the biospecific
binding between an antigen and an antibody. Tagging one of the components of the binding pair enables more detailed quantification. For example, radioimmunoassay uses a radioisotope as a label for one of the components of the biospecific binding pair. Similarly, fluorescent labels have been used with fluorescent i munoassay.
More recently, attempts have been made to develop an electrochemical sensor which can directly measure analyte concentration. Such sensors would greatly simplify and speed up immunoassay laboratory procedures and provide greater accuracy. These sensors generally detect a change in the physical, electrical or optical properties as one of the binding pairs (generally an antibody) biospecifically binds to its mated pair (generally an antigen). U.S. Patent 4,314,821, issued to Thomas K. Rice detects the change in resonance frequency of a piezoelectric oscillator as antibodies bind to the oscillator. The change in resonant frequency is proportional to the build-up of bound complexes on the oscillator surface (i.e., the build-up of the antibody-antigen complex physically changes the resonance of the oscillator). In U.S. Patent 4,238,757, issued to John F. Schenck, an antigen in a fluid medium is brought into contact with a protein surface layer and alters the charge of the surface layer through an antigen- antibody biospecific binding reaction. A field effect transistor is used to detect this change in charge. Similarly, U.S. Patents 4,444,892 and 4,334,880 detects a change in charge which occurs with certain biospecific binding reactions by using a polyacetylene semiconductive device.
U.S. Patent 4,219,335, issued to Richard C. Ebersole, teaches the use of immune reagents labeled with reactance tags. These tags can be detected electrically since they alter the dielectric, conductive or magnetic properties of the test surface. The patent teaches binding a receptor agent to a test surface. The patient's body fluid containing a certain antibody is added to the test area and
the antibody complexes with the receptor agents." In a second step, the test area is exposed to a second immune reagent that is bonded to a reactance tag. This immune tag complexes with the receptor agent-patient antibody complex, if present, on the test surface. The reactance tag containing a metal or metal-oxide is then detected by electrical means.
U.S. Patent 4,054,646, issued to Ivan Giaever, teaches a method for determining, by electrical means, whether an antigen-antibody reaction produces a monomolecular layer or a biomolecular layer. An antigen is used to coat a metal substrate. The coated substrate is then brought into contact with the fluid suspected to contain a certain antibody. If the antibody is present it adheres to the antigen layer forming a biomolecular layer. If the antibody is not present, a monomolecular layer remains. The next step is to place a mercury drop on the upper layer and measure the capacitance between the mercury drop and the metal substrate. Since the distance between the mercury drop and the metal substrate changes for the biomolecular layer as compared to the monomolecular layer, the measured capacitance also changes. U.S. Patent 4,072,576, issued to Hans Arwin et al, teaches measuring the alternating voltage impedance between two platinum electrode plates immersed in a fluid medium. A biochemical substance, is adsorbed onto the metallic surface. If the fluid under test contains an analyte biospecific to the adsorbed substance binding will occur. For example, an antigen may be absorbed directly on the metal electrodes and a specific antibody in the test fluid may bind to it forming a complex which remains on the surface of the metal electrodes. The capacitance changes depending on whether the surface is coated with a mololayer of the antigen or whether a biomolecular layer, composed of antigen and antibody layers, are absorbed onto the surface. SUMMARY OF THE INVENTION
The present invention represents a new type of electrochemical sensor for determining the concentration of
ananalyte in a fluid medium. The invention has increased speed and accuracy compared to prior art methods.
The invention utilizes an "open" capacitor which produces a higher electric field in a first volumetric region VI and a lower electric field in a second volumetric region V2. A change in the dielectric constant within the first region VI will have a greater effect on the measured capacitance than a change in the dielectric constant within the second region V2. Biospecific binding reactions are used to draw into or release large biochemical molecules from a surface located within the first region VI. Movement of these large molecules displaces molecules of the fluid medium which has a higher dielectric constant. The region VI is specifically designed so that the large molecules released from the binding surface can rapidly diffuse from region VI thereby allowing the sensor to respond relatively rapidly.
The sensor has two general embodiments. In the first embodiment, referred to as the direct binding configuration, a surface in region VI is coated with-a layer of immobilized binding agent molecules. The binding agent molecules, may be antibodies immobilized on the substrate surface. The binding agent molecules are biospecific with a particular analyte, such as a virus, bacteria or large molecule. As fluid containing the analyte is introduced onto the sensor and approaches the surface, the analyte binds to the immobilized binding agent. As the analyte binds to the surface, fluid molecules are displaced from region VI changing the dielectric constant of the "open" capacitor.
The second embodiment, referred to as the competitive binding embodiment, uses a more elaborate biochemical binding system. This method is preferred when the analyte molecules are relatively small. The biochemical binding system has a first layer of the analyte or analyte-analog immobilized on the substrate surface. A second layer of a binding agent, biospecific to the analyte, is bound onto the immobilized analyte layer. The binding agent molecules
are larger molecules and have a lower dielectric constant than the fluid medium. When free analyte molecules in the fluid medium are introduced onto the sensor, they compete with the immobilized analyte molecules to bind with the binding agent molecules. This competitive binding results in a certain amount of the binding agent molecules forming a complex with the free analyte molecules. The free analyte-binding agent complex then diffuses from region VI allowing the higher dielectric fluid molecules to enter region VI, and increase the measured capacitance.
The invention also teaches combining the invented analyte affinity capacitor with at least one reference capacitor to form a differential affinity sensor. The reference capacitor is used to compensate for non-analyte effects. These non-analyte effects include changes in the dielectric constant of the fluid medium caused by a change in temperature, ionic concentration, pH, composition and physical state of the fluid medium, as well as non-specific binding of other proteins contained within the fluid medium.
The invented capacitive sensor can be used to measure the concentration of specific analytes in body fluids and can function as either an in vivo or in vitro sensor. The capacitor sensor can also be used to detect specific substances in the environment. The use of the reference capacitor allows the sensor to continuously measure analyte concentration even though the physical and chemical characteristics of the fluid medium containing the analyte may change. The capacitance affinity sensor can be used to detect a broad range of analytes including: bacteria, viruses, antibodies, large protein molecules, antigens, haptens, polysaccharides, glycoproteins, glycolipids, enzyme inhibitors, enzyme substrates, neurotransmitters and hormones. BRIEF DESCRIPTION OF THE DRAWINGS
Figures la and b are schematic cross-sectional views of the direct binding configuration with Figure la showing the structure of the capacitive sensor, and Figure lb
illustrating the operation of the capacitive sensor' to detect the presence of an analyte in a fluid medium.
Figures 2a and 2b are schematic cross-sectional views of the competitive binding configuration with Figure 2a showing the structure of the capacitive sensor and Figure
2b illustrating the operation-of the capacitive sensor to detect the presence of an analyte of in a fluid medium.
Figure 3 is a perspective view of an "open" capacitor that uses a plurality of interdigited fingers. Figure 4 is a top view of an "open" capacitor which uses interleaved conductors.
Figure 5 is a perspective view of an "open" capacitor that uses two parallel conductive wires positioned in an insulator. Figures 6a, b and c are schematic cross-sectional views showing various embodiments of the reference capacitor with Figure 6a showing a reference capacitor which does not contain the biochemical binding system.
Figure 6b showing a reference capacitor that uses a "dummy" binding agent for the binding system, and Figure 6c showing a reference capacitor using a binding system composed of a
"dummy" analyte and binding agent pair.
Figures 7a and b are schematic cross-sectional views of the differential affinity sensor using a molecular sieve, with Figure 7a showing a single molecular sieve associated with both the affinity and reference capacitors and Figure 7b showing a first molecular sieve associated with the affinity capacitor and a second molecular sieve associated with the reference capacitor. Figure 8 is an embodiment of the differential affinity sensor having an affinity capacitor and a reference capacitor located side-by-side.
Figure 9 is an embodiment of the differential affinity sensor having the affinity capacitor and the reference capacitor located back-to-back.
Figure 10 is a schematic diagram of the circuit to detect the phase difference between the affinity capacitor and the reference capacitor.
Figure 11 is a„schematic diagram of a microprocessor system for use with a differential affinity sensor that has an affinity capacitor and at least one reference capacitor. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
The Capacitive Chemical Sensor can be made chemically sensitive to an analyte by any of a variety of biospecific chemical binding methods. These biospecific binding methods fallinto two general categories: (1) competitive binding configuration, and (2) direct binding configuration. As used herein, the term "analyte" means the species to be analyzed.
Direct Binding Embodiment
Figure la is a schematic cross-sectional view showing the first general configuration of the sensor, referred to as the direct binding configuration. A first conductor 10 is positioned on the surface of an insulating material or substrate 12; and, a second conductor 14 is also positioned on substrate 12 and disposed a distance from the first conductor 10 creating a channel between the two conductors. The two conductors 10, 14 are coated with a thin electrically insulating layer 16, and the resulting structure forms an "open" capacitor. When a direct alternating voltage is applied across the conductors, an electric field is generated having electric lines of flux 18. As seen generally in Figure la, the electric field has a higher field intensity within the volumetric region VI and a lower field intensity within volumetric region V2. Molecules of a binding agent 20 are immobilized on a surface in the volumetric region VI. In Figure la, the binding agent is immobilized within the channel formed between the two conductors; however, a layer of the immobilized binding agent may coat the entire surface covering the insulated conductors as well as the top surface of the substrate. The techniques for immobilizing the binding agent on the surface are known in the art and will be discussed later in this specification. The binding
agent is an affinity ligand that will bind specifically to the analyte, such as an antibody binds specifically to a particular virus. Alternatively, the affinity ligand may bind to a specific group of analytes, such as nucleotide analogs and lectins bind to certain groups of biochemical analytes.
In Figure lb, a fluid medium to be tested for a particular analyte is introduced onto the "open" capacitor. The sensor may be immersed into the fluid as in the case of an in vivo medical sensor or an environmental sensor; or, a small volume of the fluid medium may be poured onto the sensor. The fluid medium, shown in Figure lb, is composed of molecules of fluid 22 and molecules of analyte 24. The fluid medium fills the sensor volume VT which is composed of volumetric regions VI and V2. The fluid medium may be body fluids such as blood, urine, tears, saliva, semen or it may be other buffered solutions containing the analyte. The fluid molecules 22 will generally include water molecules and small amounts of protein molecules, ionic substances, etc. The dielectric constant of the analyte species must be lower than the dielectric constant of the dominant fluid molecule, generally the water molecule.
In operation, when an analyte species in the fluid medium enters the "open" capacitor sensor and approaches the surface, it binds to the immobilized binding agent
(i.e., the ligand layer). This binding will occur until equilibrium is reached between the binding agent, the analyte, and the binding agent-analyte complex (i.e., the ligand-analyte bound species). This equilibrium relationship can be related by the following equation:
(A) + (B) =___: (A'B), where A = Analyte, B= Binding Agent and (A-C) is the Bound Complex.
As the analyte species binds to the surface, fluid molecules from region VI are displaced and the resulting dielectric constant in a region VI will decrease. This change in the dielectric constant will be proportional to the analyte species concentration as related by the
following equations :
( 1 ) [A- B] = K [A ] [B ]
( 2 ) TA = [A ] + [A- B ]
( 3 ) TB = [B ] + [A- B]
where, [A] = free analyte concentration [B] = binding agent (ligand) concentration [A-B] = bound analyte-ligand complex
TA = total analyte concentration
Tβ = total binding agent (ligand) concentration
It is to be understood that the above equilibrium equations are only approximations and are used only to illustrate the general functioning of the sensor. The quantity TQ, the number of immobilized binding agent molecules, is know; the quantity K is known or can be determined by experimentation; the concentration [A-B] is measured by the change in the dielectric constant of the "open" capacitor; and, the total concentration of the analyte in the test fluid (TA) is what one wants to determine. For these equations to be generally representative, there should not be a large concentration gradient of the free analyte molecules in region VI. This concentration gradient can be reduced by thermal diffusion over a small volume. Therefore, the "open" face capacitor is specifically designed so that region VI, having the highest electric field flux, is small and there is a short diffusion distance for analyte molecules released from the binding surface 20 to migrate from region VI. It is also within the inventor's contemplation to measure the sensor
response during non-equilibrium conditions. The use of kinetic rate equations or empirical data can relate non- equilibrium measurements to total analyte concentration.
Usually, but not exclusively, the analyte species for the direct binding configuration will be large molecules (generally larger than 150,000 daltons) such as bacteria, viruses, other antibodies, or protein molecules. The larger the analyte molecule and the lower its dielectric properties, the greater will be the change in the bulk dielectric constant of region VI as the analyte binds to surface 20. Table I contains anon-limiting example of the type of binding agents (ligands) and analytes that can be used with the direct binding configuration of the sensor:
TABLE I immobilized analyte binding agent
bio-specific antibody bacteria bio-specific antibody viruses bio-specific antibody a second antibody bio-specific antibody large molecule analytes such as protein mole¬ cules
Competitive Binding Embodiment
The second general embodiment of the present invention is shown in the schematic cross-sectional view of Figure 2a. This embodiment is referred to as the competitive binding configuration of the sensor and is particularly useful in sensing analytes that are "small" molecules. In this case, small is defined as significantly smaller in molecular weight than 150,000 daltons (1 dalton = 1 atomic mass unit), the approximate atomic mass of antibodies. A first conductor 26 is positioned on the surface of an insulating material or substrate 27; and, a second conductor 28 also positioned on substrate 27 is disposed a
distance from first conductor 26, creating a channel between the two conductors. The two conductors 26, 28 are coated with a thin electrically insulating layer 30, and the resulting structure forms an "open" capacitor, similar to that used in the first direct binding embodiment. As with the first embodiment, when a direct or alternating voltage is applied across the conductors, an electric field is generated having electric lines of flux 32. As seen generally in Figure 2a, the electric field has a higher field intensity within the region of VI, and a lower field intensity within region V2.
The essential difference between the direct and competitive binding embodiments is that a two-layer biochemical binding system is used in the latter. A first layer 34 is made from molecules of the analyte or an analog of the analyte that is immobilized on a surface in the volumetric region VI. A second layer 36 is made from molecules of a binding agent that are biospecific with the analyte. The second layer 36 binds to the immobilized analyte layer 34. The molecules of the binding agent are generally large compared to the analyte molecules. Figure 2a shows the two-layer binding system positioned within the channel formed between the two conductors; however, the two-layer binding system may coat the entire surface covering the insulated conductors as well as the top surface of the substrate.
In Figure 2b, the fluid medium to be tested for a particular analyte is introduced onto the "open" capacitor, as was done with the direct binding embodiment. The fluid medium that can comprise body fluids or a fluid buffer, is composed of fluid molecules 38 and analyte molecules 40. The fluid molecules 38 will generally include water molecules, as well as small amounts of protein molecules, ionic substances, etc. The binding agent is selected to have a dielectric constant lower than the dielectric constant of the dominant fluid molecule, generally the water molecule; and, the binding agent molecule is selected to be substantially larger than the dominant fluid
molecule.
In operation, when analyte species in the fluid medium enters the "open" capacitor sensor and approaches the two- layer biochemical binding system, it competes with the immobilized analyte 34 to bind with binding agent molecules 36. Since the binding agent molecules are in dynamic equilibrium, there is always a small fraction of these molecules not bound to the immobilized analyte. When free analyte enters into the system, some of these unbound binding agent molecules bind to the free analyte. This results in an overall loss of the binding agent molecules from the surface of the biochemical binding system as equilibrium is restored. The binding agent-free analyte complex diffuses from the binding system to region V2, allowing higher dielectric fluid molecules to enter the higher intensity electric field region VI. The result is an increase in the dielectric constant of the capacitor. This change in the dielectric constant will be proportional to the concentration of the analyte species as related by the following equations:
(4) [A-C] = Ki [A] [C]
(5) [A-B] = K2 [A] [B]
(6) TA = [A] + [A-C] + [A-B]
(7) TB = [B] + [A-B]
(8) Tc = [C] + [A-C]
where [A] = binding agent concentration [B] = free analyte concentration [C] = immobilized analyte concentration
[A-B] = free analyte-binding agent complex [A-C] = immobilized analyte-binding agent complex
A = total binding agent concentration β = total free analyte concentration Tc = total immobilized analyte concentration It is again to be understood that the above equilibrium equations are only approximations and used only to illustrate the general functioning of the sensor. For these equations, the quantity TA, the number of binding agent molecules, is known; the quantities K and K2 are known or can be determined by expe imentation; the concentration [A-C] is measured by the change in the dielectric constant of the "open" capacitor; the quantity Tc, the number of immobilized analyte molecules, is known; and, the total concentration of the analyte in the test fluid (TA) is what one wants to determine. For these equations to be generally representative there (1) should not be a large concentration gradient of the free analyte molecules in region VI; and (2) the free analyte-binding agent complex (A:B) should diffuse rapidly from region VI. This concentration gradient can be reduced by thermal diffusion over a small volume. Therefore, the "open" capacitor is specifically designed so that region VI is small and there is a short diffusion distance allowing free analyte-binding agent complexes to move from the surface of the two-layer biochemical binding system and out of region VI, and the concentration gradient of free analyte in region VI is thereby reduced. Applicant envisions that the use of additional thermal energy or fluid agitation may increase the mobility of the free analyte molecules as well as the free analyte-binding agent complex molecules. It is also within the the Applicant's contemplation to measure the sensor response during nonequilibrium conditions. The use of kinetic rate equations or empirical testing can relate nonequilib ium measurements to total analyte concentration. The binding agent that forms the second layer of the biochemical binding system can be selected from general or specific affinity ligands and may include, but is not limited to, antibodies, lectins, enzymes and receptors.
The immobilized analyte which forms the first layer of the biochemical binding system may be the same molecular substance as the analyte under test, or it may be an analog of the analyte that is biospecific to the binding agent. The immobilized analyte may, for example, be an antigen, a hapten, a polysaccharide, a glycoprotein, a glycolipid, an enzyme inhibitor, an enzyme substrate, a neurotransmitter, a hormone, etc. The immobilized analyte is covalantly bound to the substrate surface. Table II contains non- limiting examples of the biochemical binding system used in a competitive binding embodiment to test for particular analytes . TABLE II biochemical binding system analyte class of sensor
immobilized binding agent analyte
antigen antibody antigen
hapten antibody hapten
polysaccharides lectin polysaccharides B
glycoproteins lectin glycoproteins B
glycolipids lectin glycolipids
enzyme enzyme enzyme inhibitor inhibitor
enzyme enzyme enzyme substrate substrate
enzyme enzyme enzyme inhibitor substrate
095
1 5 neurotrans¬ neural neurotrans¬ mitters receptor mitters
hormones neural hormones receptor
As can be seen from Table II, there are four classes of the competitive binding sensor. In class A the binding agent is an antibody specific to the analyte. The analyte may be an antigen or hapten. The biochemical binding system comprises a first immobilized layer of the antigen or hapten analyte with a second layer of the biospecific antibody biochemically bound to the immobilized analyte in the first layer.
In class B, the binding agent is a lectin, which is a general ligand specific to a group of analytes. A lectin- based sensor can be made more specific by an appropriate molecular sieve membrane that excludes larger molecules in the general analyte group from reacting with the biochemical binding system. In this class, for example, the binding system could have a first immobilized layer of a polysaccharide or a membrane protein containing sugar residues of certain configurations and a second layer of the general lectin bound to the first layer.
In class C, the binding agent is an enzyme reactive with an enzyme inhibitor or enzyme substrate. In this class, for example, the binding system could have an inhibitor for a particular enzyme immobilized on the sensor surface and a second layer containing the enzyme bound to the inhibitor in the first layer. With a particular enzyme substrate in the test fluid, the enzyme binding agent will be drawn from the surface of the binding system.
In class D, the binding agents are neuroreceptors. The neuroreceptor has its function greatly altered by various neurotoxins and other agents. The binding system can have a layer of succinylcholine immobilized on the sensor surface with a second layer of acetylcholine
receptor molecules bound to the first layer. If a neurotoxin, for example, is present in the test fluid, the receptor binding behavior will be altered and it will be released from the binding system surface, thereby altering the dielectric properties of the sensor. It is of course to be understood that these are merely examples of the biochemical binding systems that can be used with the competitive binding embodiment of the present invention.
"Open" Capacitor Structures Figures 3, 4 and 5 show various embodiments of the
"open" capacitor structure that can be used for either the direct or competitive binding embodiment of the sensor. Each of these alternative structures of the "open" capacitor contain similar features: (1) the electrical field intensity of the capacitor is higher in a first region VI than a second region V2; (2) the biochemical binding system is located on a surface area in the first region VI; and, (3) molecules released from the binding system have a short diffusion distance to migrate from the region VI into region V2.
Figure 3 is a perspective view of an "open" capacitor that uses a plurality of interdigitated fingers. Metallic conductors 42 and 44 are positioned on an insulating substrate 46. Each conductor has a plurality of fingers that are disposed in an interdigitated manner relative to the fingers of the other conductors. The interdigitated fingers from both conductors form a plurality of channels that comprise a significant portion of the higher electric field region VI, as seen in Figures la and lb. Known photolithographic etching techniques are used to form the interdigitated fingers on the substrate. The substrate can be made from insulating materials such as Corning 7059 glass or alumina wafers. The interdigitated fingers can be made of copper and gold. Applicant selected 2 mil wide fingers that are approximately 1 mil high and separated by 3 mil spaces, although other dimensions may be used. The interdigitated fingers are covered with an insulating layer
48. Applicant made the insulating layer 48 with a 1-2.5 micron coating of parylene polymer deposited using known deposition processes and a 0.3 micron of SiO deposited using vapor vacuum evaporation deposition; however, alternative electrically insulating material can be used. In the direct binding configuration, a layer of the binding agent is immobilized onto the insulated layer 48. (see, generally Figure la). In the competitive binding configuration, the first layer of the two-layer biochemical binding system is immobilized onto the insulated layer 48 (see, generally Figure 2a). Fluid to be tested for a particular analyte is brought into contact with the "open" capacitor as discussed earlier.
Figure 4 is a top view of an "open" capacitor that uses two interleaved conductors covered with an electrically insulated layer. Interleaved metallic conductors 50 and 52 are deposited on insulating substrate 54 using the same technique and materials discussed above. Each conductor is approximately 2 mil by 2 mil with a 2 mil spacing between the interleved conductors; although, other dimensions may be used. The binding agent, for the direct configuration, and the biochemical binding system, for the competitive embodiment, is immobilized on the surface of the insulated conductor and in the channels between the conductors.
It is to be understood that the interdigitated and interleaved configurations of the two conductors are not limiting examples, and that other geometries can provide the desired features of the "open" capacitor. For example, in Figure 5 an embodiment of the "open" capacitor is shown that uses two parallel conductive wires 56, 58 embedded in a molded insulator 60. The molded insulator 60 is shaped to provide two channels positioned between and running parallel with the conductive wires. If a direct or alternating voltage were applied across conductors 56 and 58, electrical lines of flux 62 would be generated. The volume generally within the two channels will have a higher electric field intensity (similar to region VI in Figures
la or 2a) than the region displaced further radially (similar to region V2 and Figures la and 2a). The binding agent, for the direct binding embodiment, and the biochemical binding system, for the competitive binding embodiment, are immobilized onto the surfaces 64 of the molded insulator. As with the interdigitated and interleaved embodiments, the following occurs: (1) the field intensity of the capacitor is higher within the two channels (region VI) than in the radially extended regions (region V2); (2) the biochemical binding system or binding agent is immobilized within the area (VI) having the higher electric field intensity; and (3) molecules removed from the binding system have a short diffusion distance to migrate from the region of the two channels (the region of higherns electrical field intensity) the radially extending regions having lower field intensity. This embodiment of the "open" capacitor can be placed in a 1 millimeter dialysis tube 66 which acts as a molecular sieve and the entire sensor can be inserted into a patient's vein or artery to measure the concentration of a particular analyte in the patient's blood. As an alternative to this embodiment, conductive wires 56, 58 are twisted around a center line. This embodiment may provide additional noise immunity. Further, in each of the embodiments in Figures 3, 4 or 5, the surface area of the binding agent or biochemical binding system can be increased by adding a plurality of ridges, corrugations, or protrusions in region VI. These ridges, corrugations or protrusions are positioned within the channels formed in region VI are be coated with the immobilized binding agent or biochemical binding system.
Differential Capacitive Sensor
The accuracy of both the direct binding and competitive binding embodiments of the present invention is increased if differential sensing is employed. The differential capacitive sensor uses an analyte affinity sensor (i.e.,the direct binding capacitive sensor or the
competitive binding capacitive sensor discussed above) and at least one reference capacitor to compensate for non- analyte effects. The reference capacitor compensates for changes in dielectric constant of the fluid medium caused by changes in temperature, ionic concentration, pH, composition and physical and chemical state of the fluid medium, as well as non-specific binding of proteins that may be in the fluid medium. Figures 6a, b, and c, show various embodiments of the reference capacitor. Each reference capacitor has a first and second conductor 68, 70 positioned on a substrate to form the "open" capacitor as described above. In Figure 6a, a reference capacitor that can be used with both the direct and competitive binding embodiments is shown. This reference capacitor has no protein coat, i.e., it does not have the immobilized binding agent or binding system. In Figure 6b, a reference capacitor for use with the direct binding embodiment is shown. This reference capacitor contains an immobilized layer of a "dummy" binding agent 72. The "dummy" binding agent is selected from the same class as the analyte sensitive binding agent but it is made biospecific to a molecule not found in the test environment. Alternatively, if the reference capacitor uses the same binding agent as the affinity capacitor, a molecular sieve would be used to prevent the analyte from entering the reference capacitor. In Figure 6c, a reference capacitor for use with the competitive binding embodiment is shown. This reference capacitor contains a "dummy" biochemical binding system. The "dummy" binding system uses an immobilized "dummy" analyte 74 specifically reactive with a "dummy" binding agent 76. The "dummy" analyte and binding agent are chosen to have an affinity constant and other physical characteristics that closely match the real analyte and real binding agent. If an antigen-antibody pair are chosen for the binding system of the affinity capacitor, the
"dummy" antibody would be selected from the same class of antibodies and from the same type of animal, but would not be biospecific with the analyte antigen. The reference
capacitor may use only the immobilized "dummy" analyte layer, and not the "dummy" binding layer. Alternatively, the reference capacitor may use the same antigen-antibody pair as the affinity capacitor but a molecular sieve would be used to prevent the analyte from entering the reference capacitor. Each of the different types of reference capacitors outlined above compensates for non-analyte changes in the fluid medium. However, a multiplicity of reference capacitors could be used with one affinity capacitor. These reference capacitors would identify the end points and/or other specific points of the dose/response curve. The analyte concentration would be determined by the dielectric change in the analyte affinity capacitor as compared to the boundary values provided by the reference capacitors.
The molecular sieves shown in Figures 7a7 b enable the invented affinity sensor to be immersed in the test fluid. The molecular sieve provides two functions: (1) it retains the binding agent molecules in the sensor; and, (2) it selectively screens certain larger molecules from entering the "open" capacitor sensor. Figure 7a is a schematic drawing of a competitive binding differential sensor having an analyte affinity capacitor 78 and reference capacitor 80 (for simplicity the biochemical binding system is not shown in Figure 7a). Fluid molecules flowing into or from the analyte and reference capacitors must pass through molecular sieve 82. The molecular sieve is of a known construction having a pore size that can easily pass the fluid and analyte molecules but will not allow the larger binding agent molecules to escape from the sensor. The pore size for an antigen-antibody binding system would be less than 150,000 daltons to keep the antibody within the sensor. Molecular sieves are particularly useful when the sensor is an in vivo sensor implanted, for example, in a patient's blood stream. The molecular sieve prevents the binding agent molecules released by the binding system from being removed by the blood flow from the sensor.
Figure 7b is a schematic drawing of a competitive
binding differential sensor in which the analyte capacitor
78 and the reference capacitor 80 have separate molecular sieves 84 and 86. In this case, molecular sieve 84 prevents the binding agent molecules from leaving the affinity capacitor. A separate molecular sieve 86 is used with the reference capacitor if the reference capacitor does not use a "dummy" binding system but uses the same binding system as the affinity capacitor. In this case, the molecular sieve 86 provides the following two functions: (1) preventing the binding agent molecules from leaving the reference capacitor and, (2) preventing the analyte molecules from entering the reference capacitor. This form of reference capacitor would be particularly sensitive to changes in the affinity constant of the binding agent-immobilized analyte complex caused by temperature changes. It is to be further understood that a molecular sieve of this nature can be used to filter unwanted larger molecules from interacting with the biochemical binding system. In that case, the pore size of the molecular sieve would be such that fluid and analyte molecules could pass through whereas larger unwanted molecules would be blocked by the molecular sieve. The construction, fabrication and choice of materials for these types of molecular sieves are known in the art. Figures 8 and 9 show various embodiments for a differential sensor that includes an affinity capacitor and a reference capacitor. Figure 8 is a top view of an affinity capacitor 88 and a reference capacitor 90 located side by side on the same substrate. Figure 9 is a cross- sectional view of an affinity capacitor 92 and a reference capacitor 94 located back-to-back. A metal shield 96 located between the capacitors isolates the electrical field generated by each capacitor. For both the side-by- side and back-to-back embodiments, the fluid medium under test would be simultaneously introduced onto both the affinity and reference capacitors. It is also to be understood that a molecular sieve could be used to encompass either or both the reference capacitor and the
affinity capacitor.
The following non-limiting examples, describe several specific embodiments of the differential sensor:
Example 1. Competitive binding embodiment. The analyte or analyte analog is immobilized on the dielectric surface forming the first layer of the biochemical binding system. An analyte specific antibody is conjugated to the immobilized analyte species and forms the second layer of the biochemical binding system. The sensor is enclosed by a molecular sieve membrane with pores large enough to be permeable to the analyte but small enough to confine antibodies on or close to the sensor. This example is appropriate for small and medium molecular weight analytes compared to antibodies, which have molecular weights of approximately 150,000 daltons. With this example, the most appropriate, but not exclusive, reference capacitor is made exactly the same way as the analyte sensitive side, except that a "dummy" analyte and its associated specific "dummy" antibody is used. The "dummy" analyte and its specific antibody are chosen to have an affinity constant and other physical characteristics that closely match the analyte and analyte specific antibody characteristics. The reference capacitor is also enclosed by a molecular sieve. A second reference capacitor configuration with no bound "dummy" antibody may also be used.
Example 2. Direct Binding Embodiment. An antibody specific to particular cells, such as bacteria or to viruses, or to large molecules, is immobilized on the surface of the "open" capacitor, forming the binding agent molecules. A large molecule, bacterium, or virus, when bound to this antibody, will displace a significant amount of the fluid molecules, (predominantly water molecules) from the higher density electric field volume VI, and thus cause a detectable change in capacitance. In this case, a molecular sieve membrane would not be required. However, it would be useful to cover the surface with a mesh. The reference side of this sensor consists of a capacitor with a "dummy" antibody immobilized on the insulating substrate.
This antibody is of the same class as the analyte sensitive antibody, but is made specific to a molecule not found in the test environment.
Example 3. Competitive Binding Equipment. This sensor is analogous to Example 1, but uses a receptor in place of an antibody as the second layer of the biochemical binding system. A generic sensor for neurotoxins can be configured using acetylcholine receptors. A substrate, such as succinylcholine, for which the receptor has affinity, is immobilized on the dielectric substrate, forming the first layer of the biochemical binding system. Receptor molecules are then conjugated to the substrate forming the second layer of the biochemical binding system. The receptor molecules are confined within the sensor by the use of a molecular sieve. When a neurotoxin permeates, the receptor is pulled off the surface, and capacitance changes. A reference capacitor is made identical to the analyte sensitive side except that the molecule chosen for surface immobilization is one with an affinity so large that substances of interest will not pull the receptor off the immobilized layer.
The above three examples show models that can be used for a large number of possible sensor configurations. It is to be understood that other binding agents and biochemical binding systems than those shown above are within the scope of this invention.
Figures 10 and 11 are schematic diagrams which illustrate two possible circuits to be used with the differential sensor as taught by' the present invention. Figure 10 is a schematic diagram of a circuit to detect the phase difference between the affinity and reference capacitors. A stable oscillator 98 supplies an alternating signal to the affinity capacitor 102 and the reference capacitor 104. These capacitors are placed in parallel with trimmer capacitors 104 and 106. Phase detector 108 detects the phase angle shift between the affinity capacitor 102 and the reference capacitor 104. The phase difference is functionally related to the analyte
concentration in the fluid medium.
Figure 11 is a schematic diagram of a microcomputer system for use with the differential sensor. The system contains an analyte affinity capacitor 110 and a plurality of reference capacitors 112 and 114 (although, a single reference capacitor may be used). The affinity and reference capacitors (110, 112, 114) are brought into contact the fluid under test. Each capacitor is connected to an oscillator (116, 118, 120) and a change in the capacitance will alter the frequency of oscillation of its associated oscillator. The output frequency of each - oscillator (116, 118, 120) is fed to an associated counter (122, 124, 126) which sends_ the frequency count in digital form via bus 128 to microcomputer 130. A look-up Table or Equations similar to Equations (1) through (8) are stored in the microcomputer and a determination of the concentration of the analyte in the fluid medium is made. This value is displayed on output display 132. It is to be understood, that other circuits can also be envisioned once one understands the differential change in capacitance between the analyte affinity capacitor and the reference capacitor as taught by the present invention.
Binding Systems
As described earlier, for the direct binding embodiment, molecules of a binding agent are immobilized on the substrate surface; and, for the competitive binding configuration, a layer of the analyte or analyte-analog is immobilized on the substrate surface to form the first layer of the biochemical binding system. As used herein, immobilized means attaching a molecule by one or more covalent bonds, or other biochemical bonds. Various immobilization techniques are known in the art. The attachment site on the molecule is chosen so that functional groups of the molecule have no interference. For example, in the direct binding embodiment, an antibody (the binding agent) is immobilized on the substrate so that its analyte recognizing and binding site or sites are free
to function. For binding proteins, most reactions are nucleophilic with the nucleophitic group most often NH , OH or SH. Specific examples of biochemical binding systems are found in the art of affinity chromatography and are listed in Table II of Waters, R. , "Affinity
Chromatography", Analytical Chemistry, Volume 57, No. 11, pp. 1099A-1114A and listed in the figures on pages 19, 21 and 22 of Parikh, I., and P. Cuatrecasas, "Affinity Chromatography", Chemical and Engineering News, August 26, 1985, pp. 17-32 (these articles being incorporated here and by reference). Attachment reactions include the use of Cyanogen Bromide, Active Esters, Epoxide, Tresyl Chloride, Carbonyldiimidazole, Thiol and Diazonium reagents.
By way of illustration, the following experimental example performed by the Applicant shows covalent attachment of the biochemical binding system to the "open" affinity capacitor. The example is a sensor to detect the Trichothecene mycotoxin T-2, which is found in the environment and is produced by the fungal species Fuarium. Trichothecene mycotoxin is an agricultural toxin causing the loss of grain yield on various food crops. It has been implicated in human and animal mucotoxicoses. Experimental Example
1. The "open" capacitor is coated with a 0.3 micron thick layer of SiO. Without care to prevent hydration of the surface (dry vacuum), the surface becomes composed of silanol groups:
OH OH i t
Si Si The surface will have approximately 10 silanols per m^.
2. Amino groups are attached to the SiO surface for later attachment of proteins, using the following steps: a. - soak substrate in 10% Y -aminopropyl- triethoxysilane [(EtO)3-Si-(CH2.3-NH2I and dry toluene overnight at room temperature. b. wash with dry toluene; and, c. dry at 60 degrees C for two hours. The aminosilanized surface will be:
3. The surface is now ready for introduction of the Trichothecene (T-2) groups. a. The T-2 molecule is converted to a hemisuccinate derivative by heating it in the presence of Pyridine and Succinicynhydride. This derivization was necessary in this example, but some hemisuccinates can be bought off the shelf. For example, in making a hydrocortisone sensor, hydrocortisone hemisuccinate can be purchased directly from Sigma Chemical Co., and others. b. The hemisuccinate derivative of the analyte is then conjugated to the γ - amino function of the silanized surface, using a water soluble carbodiimide as a catalyst. The T-2 analyte is now immobilized on the surface of the "open" capacitor and the surface appears as follows:
4. The second layer of the biochemical, binding system is produced by adding the anti T-2 toxin antibody to fluid bathing the surface of the open face capacitor. The antibodies will bind with an affinity similar to that in the standard immunoassay (5.28 x 10? liters/mol) . The resulting biochemical binding system has a first layer of the T-2 analyte immobilized on the surface and a second layer of the anti T-2 toxin antibody specifically bound to the immobilized layer. Since the anti T-2 antibodies and the immobilized T-2 toxin are in dynamic equilibrium, an influx of free T-2
toxin molecules would perturb the equilibrium and draw the antibodies from the immobilized surface forming free analyte-antibody complexes. Removal of the free analyte- antibody complexes from the region of the capacitor sensor having higher field intensity, region VI, causes a change in the capacitance that is a direct indication of the concentration of free T-2 molecules in the fluid medium. Obviously many modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims the invention may be practiced otherwise than as specifically described.