WO1984002193A1 - Chromogenic support immunoassay - Google Patents

Abstract

An immunoassay, utilizing an active support capable of concentrating a detectable signal resulting from the interaction of a product of a reaction between an enzyme-labelled immune complex and enzyme substrates with the support. The immune complex is either formed or collected on the active support during the assay. The assay is useful for the detection and measurement of analytes in biological samples.

Description

TITLE CHROMOGENIC SUPPORT IMMUNOASSAY TECHNICAL FIELD This invention relates to immunoassay techniques for the detection and measurement of soluble organic antigens and haptens, cells, microorganisms, and their products and, more specifically, to immunoassay techniques in which the signal produced by the label component or by the interaction of a product of label component of a labeled immune complex on or with an active support is concentrated on the support material by virtue of its chemical, physical, or ionic interaction with the support.

BACKGROUND ART Clinical laboratory chemical diagnostic tests are an important component of health care delivery. In recent years, a large number of immunoassay techniques have been described which take advantage of the specificity of antigen-antibody reactions for the measurement of soluble organic antigenic or haptenic compounds, including drugs, hormones, vitamins, metabolites, etc., in body fluids.

Numerous label components for producing a detectable signal in immunoassays have been described in the patent and nonpatent literature, including radioisotopes, enzymes, fluorophores, enzyme inhibitors, chemi- or bioluminescent materials, etc. See. for example. Methods in Enzymology, Volume 70 (1980), H. van Vunakis and J. Langone, ed., New York, Academic Press and references contained therein; and Methods in Enzymology, Volume 73 (1981), H. van Vunakis and J. Langone, ed.. New York, Academic Press and references contained therein.

Some immunoassays require a separation step in order to distinguish the bound from the free label component. Such assays are said to be heterogeneous and include such well-known techniques as the radioimmunoassay (RIA) and the enzyme-linked imraunosorbent assay (ELISA) . A variety of means of separation is known in the art, such as centrifugation, filtration, and chromatography.

Other immunoassays do not require a separation step and are said to be homogeneous. Homogeneous assays, while faster and simpler to perform, are frequently less sensitive than heterogeneous assays and can be more prone to interferences.

In conventional immunoassays of either the homogeneous or heterogeneous type, the reaction product (s) of the label components is free to diffuse into the test medium. This facilitates spectrophotometric detection since the products can be uniformly distributed throughout the test volume. However, dilution of the reaction product(s) limits sensitivity and makes direct visualization difficult. Multilayer test devices have been described; see for example. Pierce et al., U.S. Patent 4,258,001, issued March 24, 1981, and U.S. Patent 3,723,064, issued March 27, 1973 to Liotta. The fabrication of such devices is complex, requiring multiple reagent coatings or impregnations and lamination of different materials. Interface irregularities can cause nonuniform fluid migration within the test device. Reverse fluid migration can lead to products in the lower layers migrating back up into the top layers, resulting in chemical interferences and diminished test response. Furthermore, large molecules and cellular debris contained in many clinical samples create particular difficulty in the design and development of multilayer test devices, since these substances can clog the device and impede fluid flow. One of the most important constraints of all is that analysis is limited to the free label fraction only. Direct measurement of the bound fraction requires additional washing and elution steps, usually too cumbersome and time consuming to perform in this type of test device. While immunoassays have been widely used for the measurement of soluble antigens and haptens, traditional diagnostic methods for infectious diseases have involved the cultivation of the infectious agent in vitro or in laboratory animals. This approach suffers from several disadvantages. First, some medically important viruses cannot be cultured in commonly available tissue culture or animal systems. These include rotavirus, hepatitis A and B viruses, and Epstein-Barr virus. Second, many infectious agents require long periods of cultivation before identification can be accomplished. In some cases this period may be so long that the results are not useful to the clinician. Third, traditional cultivation methods can be inadequate when specimens are obtained from individuals who have already received antibiotic therapy.

There is a rapidly expanding market for assays capable of detecting infectious agents directly in clinical samples. Such assays can be based on antigen-antibody reactions or on measurement of microbial enzyme activity. It should be noted that assays based on antigen or enzyme detection do not necessarily measure live or infectious organisms. For any such assay to be practically useful, it is essential that the assay have a rapid turnaround time, be specific for the organism or class of organisms of interest, be sensitive to a relatively small number of organisms, and be relatively free of interferences from components encountered in body fluids. A number of immunoassays for microbial and viral antigens have been described; see R. H. Yolken. Rev. Infect. Dis., Volume 4. 35 (1982), for a review. Despite their advantages over traditional methods of cultivation, many of these assays are still comparatively time-consuming (1-3 hours) and labor intensive (requiring multiple reagent additions and washing steps). Because the generated signal is measured in solution, inadequate sensitivity is also often a problem.

U.S. Patent 3.856,628, issued December 24, 1974 to Sbarra et al., discloses a method for the identification of microorganisms which utilizes a flat sheet of absorbent material having a plurality of separate test zones, each zone including a substrate and an indicator. The reaction between the products of the microorganism of interest and the substrate and indicator is a fermentation reaction. The absorbent sheet material does not display any particular affinity for either the substrate or the indicator, allowing both to diffuse into the surrounding medium.

U.S. Patent 4,327,073, issued April 27, 1982 to Huang, discloses a method for the quantitative analysis of biological fluids for a plurality of substances, each of which undergoes at least one reaction with a respective cognate compound. The substrate carrier which is used for immobilizing the functional reactive compound (antibody or nonantibody binding protein) in the invention can have a functional group for covalent bonding of the reactive compound or ion exchange sites for ionic bonding of the functional reactant. The substrate carrier can also be an adsorbent in which the reactive compound is attracted by Van der Waals forces. Enzymes are used as the tracers and are measured by bathing the carrier in a chromo genie substrate for the enzyme; the resultant product is a precipitated colored solid.

U.S. Patent 4,011,308, issued March 8, 1977 to Giaever, discloses a method for the detection of bacteria, viruses, and other cells which makes use of an immune reaction on a substrate between the particle to be detected and its tagged antibody. The tag is usually a radioisotope and the substrate is usually a metal or metallized glass slide. The substrate as described is inert and does not participate in the development of the tag.

U.S. Patent 4,322,495, issued March 30, 1980 to Kato, discloses an immunoassay for antibodies to a microorganism which utilizes a nonporous ceramic or polymer support on which the organism of interest is immobilized. The support is incubated with a sample suspected of containing antibody to the organism, washed by flooding to remove unbound components, and incubated with an enzyme-labeled anti-antibody. After another wash, an indicator capable of reacting with the enzyme label is applied to produce a color. The colored product does not interact physically or chemically with the support.

U.S. Patent 3,187,075, issued February 5, 1980 to Noller, describes an immunoassay for hepatitis B virus which utilizes an antibody-coated, inert, solid support to capture virus from the sample. After washing, the bound virus is reacted with a second enzymelabeled antibody and a substrate for the enzyme label. The colored product of the reaction precipitates on the nylon support and is eluted with a mixture of saline and acetone for measurement.

German Patent Application 2,923,735, published

December 13, 1979, discloses a process for measuring antigen receptors on cells using a purified, labeled antigen. Cells are collected by a filtration process and the amount of labeled antigen which is bound is measured by. for example, gamma-counting. Retention of label on the filter is based on binding to the cell and is solely a function of molecular size versus pore size.

Canadian Patent 1,104,927, issued July 14, 1981 to Cole et al., discloses a method for the detection of a component of an immunochemical reaction in which one component is immobilized on a microporous membrane, it is reacted with sample containing the second component of an immunochemical reaction, followed by an enzyme-labeled immunochemical reactant for the second component. The amount of color change on the surface of the membrane is determined following reaction of the enzyme label with a chromogen. The chromogen can precipitate on the membrane when acted on by the enzyme label or it can bind preferentially to the membrane. Precipitation on or preferential binding of the chromogen to the support is largely fortuitous.

U.S. Patent 4.200,690, issued April 29. 1980 to Root et al., discloses an immunoassay for an intestinal parasite in which the test result is measured by the color which forms on a support. The chromogenic substance used is poorly soluble in the reaction solvent or is preferentially entrapped in the filter matrix in the colored state. Again, precipitation on or preferential binding of the chromogen to the support is largely fortuitous.

U. S. 4,340,670, issued July 20, 1982 to Mennen, discloses a swab kit for the detection of Neisseria gonorrhea in which a solution of tetramethyl chromogen reacts with an enzyme produced by organisms present in patient sample to form a purple color on the same swab which is used for sampling. This is a non-immunochemical, microbial enzyme-based assay.

R. J. Doyle, in an article in E. Y. Laboratories, Inc. 's "Biotech", Summer '82-1, entitled "Lectins in Diagnostic Microbiology", reviews the use of lectins to selectively interact with cellular carbohydrates. In addition to their carbohydrate binding sites, lectins also possess hydrophobic binding sites and can be utilized for specifically and selectively binding various microorganisms through the differing surface components of those microorganisms.

U.S. Patent 4,299,916, issued November 10, 1981 to Litman et al., discloses an immunoassay method utilizing a surface to which there is conjugated a member of an immunological pair and a signal producing system; the latter comprising a catalyst conjugated to a member of an immunological pair and a solute capable of undergoing a reaction catalyzed by the catalyst. As a result of insolubilization of a signal generating compound, a measurable signal is generated on the surface.

There is a need for an immunoassay method for antigens, especially of viral or bacterial origin, which is rapid, sensitive to the small numbers of organisms present in biological samples in the early stages of infection, and is potentially quantitative. DISCLOSURE OF THE INVENTION The chromogenic support immunoassay of this invention for the detection or measurement of analytes such as soluble organic antigens and haptens, cells, microorganisms, and their products in biological samples utilizes a system in which the analyte is ultimately contacted with an enzyme-labeled antibody and in which the signal generated by the reaction of the enzyme with its substrate is concentrated on an active support and comprises the following steps: (A) collecting or forming on the active support an immune complex having an enzyme component comprising an analyte immunochemically bound to at least one antibody; (B) contacting the label component such as an enzyme of the immune complex on the support with signal generating reagents whereby a product is formed, provided that when said product is a chromophore, it is soluble in the assay medium; and (C) allowing the product to interact with the support which interaction results in the concentration of a detectable signal thereon.

The labeled immune complex can be formed either in solution or directly on the active solid support. When formed in solution by pre-incubation of the analyte with an enzyme-labeled antibody or with an antibody first followed by an enzyme-labeled antibody, it is subsequently collected on a support. Direct formation of the complex on the support can involve immunochemical or nonimmunochemical capture of the analyte (followed by a reaction with an enzyme-labeled antibody) or an enzyme-labeled antibody complex thereof by an antibody or lectin bound to the support or it can involve immunochemical capture of an antibodyenzyme conjugate by analyte which is bound to the support through other than the intermediacy of a specific binding reaction. In this latter case, the analyte can be bound to the support either covalently or noncovalently. The chromatic signal, the electromagnetic radiation of a signal-emitting product, can be emitted directly by the product of the label (enzyme) catalyzed breakdown of substrate. In this case, the signalemitting product is a chromophore, is soluble in the assay medium and interacts with the active support resulting in its concentration thereon. Alternatively, the signal can result from the interaction of a product of the label (enzyme) catalyzed reaction with the support itself. In this case, the signal-emitting product, if existed independently of the support, would be soluble in the assay medium.

DESCRIPTION OF THE INVENTION The chromogenic support immunoassay method of this invention is applicable to the detection and measurement of a variety of analytes such as soluble organic antigens and haptens, cells, microorganisms, and their products. Soluble organic antigens and haptens include synthetic antigenic peptides, drugs, hormones, vitamins, food additives and pesticides. The microorganisms include bacteria, mycoplasma, rickettsia, chlamydia. and viruses. The products can be synthetic such as enzymes, toxins, colicins, or cell surface antigens or breakdown products such as antigenic fragments of viral coat proteins. Some bacteria of interest are staphylococci, streptococci, neisseria, and the Enterobacteriaciae. Some viruses of interest are Herpes simplex virus, Cytomegalovirus. myxoviruses, adenoviruses, and hepatitis viruses. When the analyte is an antigen or enzyme produced by a cell or microorganism, it can be found on the surface of living or dead cells or organisms, in extracts of cells or organisms, or as cellular debris. The presence of analyte in patient sample, therefore, does not necessarily correlate with infectivity. The assay of this invention utilizes an active support designed to fulfill the functions of concentrating the chromatic signal generated by the label component. In addition, the support can also serve as a carrier for signal generating reagents and for reagents that can capture the analyte immunochemically or nonimmunochemically. By active support is meant a support capable of binding to or associating with the reaction components of the immunoassay of this invention and with the products of the reaction between the enzyme label and the signal generating reagents.

Concentration of the signal by the support makes possible its visualization which, were it uniformly distributed in solution, would be substantially indistinguishable over background because of the high dilution of the signal in the assay medium. Direct visual readout of a qualitative test result is thus possible. The results can also be read quantitatively for example by densitometry or semi-quantitatively by visual comparison to a series of standards of varying color intensities.

In addition to the intensity of color concentrated on the support, the localization of color can also be informative. For example, when a patient sample suspected of containing the analyte of interest is applied to a support and an antibody-enzyme conjugate is added, the localization of signal on the support corresponds to the localization of analyte. This phenomenon can be exploited for comparative purposes to relate the number of live or infectious organisms, determined by traditional cultivation methods, to the presence of antigenic material contained in the sample.

The support can be fabricated from a variety of porous materials. Among suitable supports are included polyamides such as nylon; starch granules; starch-coated materials such as cotton or filter paper; cellulose nitrate; ethyl cellulose; granules based on sodium starch glycollate, polyvinyl acetate and partially hydrolyzed polyvinyl acetate, and poly(acrylamide) /starch; other synthetic or natural polymers; modified polyamides; and supports modified with 5,5 '-dithiobis(2-nitrobenzoic acid) (DTNB). Nylon mesh discs can be hydrolyzed with, for example. HCl, to yield functional groups, for example, -NH₂ and -COOH, for possible further reactions. Nylon can be further treated with a crosslinking reagent such as XAMA-7, a trifunctional aziridine compound available from Cordova Chemical Co., or hexanediamine. DTNB can be attached to supports through, for example, ion exchange agents. The support can be fabricted from nonporous materials also such as ceramic beads, dipsticks, test tubes and granular materials where porosity of the granules would not be required for carrying out the immunoassay. The support can also contain other materials capable of binding to or associating with antigens or their antibodies. It is sometimes advantageous to coat the above materials with latex polymers to which proteins can be attached covalently or noncovalently. It is further possible to prepare the active support by mixing the above porous materials with other carriers capable of binding to or associating with antigens or their antibodies. These carriers include glass beads, gel filtration beads such as Sephadex, Sepharose, Bio-Gel, Bio-Beads and Zorbax^® packing materials (a trademark of E. I. du Pont de Nemours and Company) .

The support itself can have any appropriate physical form on which to perform an assay. The form is dictated by the circumstances of the assay and can be disc-like such as filter paper, elongated such as a swab or a dipstick, columnar, rectangular, etc. In general, a porous or microporous support is preferred since the kinetics and efficiency of capture of the reactive components and of washing are maximized. While macroporous supports can be suitable in some circumstances, microporous supports (average pore size <lμ diameter) are generally preferred because of the increased surface area available for reaction. Pore size can also be dictated by the dimensions of the various immune complexes formed during the assay. For general use, it is preferred that the pore size of the support be less than 10 nm and the interstitial space between particles in the support or the pore size of the support be greater than 10 μM. The choice of the smaller pore size would eliminate any intraparticle entanglement of viral antigens present in biological samples (most viruses are approximately 18-250 nm in size). The choice of the larger pore size would not lead to size-related fractionation of even bacteria (average size of the generally rod-shaped organisms is 0.5 μM x 2 μM) by the particles. As a general requirement, the porosity of the support material should be such as to avoid fractionation of the antibody-enzyme conjugates (approximate molecular weight 10⁴-10⁶ daltons). Support characteristics can also be dictated by a need to capture effectively the analyte from the biological sample.

The assay of this invention can be performed on a disc-shaped porous support, in a column packed with an active support, or on a swab made of a suitable support material. Each of the above can be adapted to test for the presence of a single analyte or can be subdivided into multiple areas, each adapted to test for the presence of a different analyte. Where the assay is performed on a swab, the swab on which the assay is performed can be the same one used in obtaining patient sample. Where the assay is performed on a disc or in a column, patient sample is customarily collected in a separate vessel and an aliquot withdrawn for assay. Alternatively, when the column is in the form of a syringe, sample can be drawn directly into the syringe.

The assay of this invention can also be performed on filter such as a starch coated glass fiber filter. In this case the patient sample can be applied directly to the filter to which antibodies have been previously attached.

Nonspecific binding to the support can be reduced or eliminated by pretreating the support with a dilute protein solution (e.g., albumin) which saturates the nonspecific attachment sites or by including in the various reagent and wash solutions a low concentration of a nonionic surfactant. The interaction of the support with the ρroduct(s) generated by the reaction of the label (enzyme) component with its substrate must occur rapidly and selectively. The concentration of the chromatic signal on the support can occur through the intervention of physical forces or chemical bonding. The bonding can be ionic or covalent.

Physical forces such as hydrogen bonding, van der Waals forces, and hydrophobiσ interactions can participate in the attachment of the reaction products between the enzyme label and signal generating reagents to various support materials. Ionic properties of the support can be tailored to complement the ionic properties of the reaction product(s). For example, cationic supports can be employed to affix anionic products, and vice versa. Some examples of acidic polymers are polystyrene sulfonic acid, carboxymethyl cellulose, and copolymers of acrylic, methacryliσ and maleic acids. Some examples of basic polymers are polypeptide, copolymers of amino-functional monomers, and diethylaminoethyl cellulose. Covalent attachment of the reaction product(s) to the support can be achieved in different ways. An advantage of this approach is enhanced color permanence, since the products become an integral part of the support.

The support can be prepared to contain functional groups for the covalent attachment of the reaction products. These groups are parts of color couplers which are known in the photographic arts. Examples include azomethine. azo-, leuco-, indazolin-, indoaniline- and pyrazalone dyes.

In order to prepare supports with color couplers, these need to have low mobility. They can be made nondiffusing by increasing their molecular weight through the addition of hydrophobic and/or ionic groups. This results in surfactant-like structures which can be incorporated directly into a polymeric support. The couplers can also be immobilized on the support by covalently bonding the couplers to the support.

In some cases, it is desirable to coat the support with anti-analyte antibodies or lectins in order to specifically capture analyte from the sample. In general, it is preferable to attach the antibody or lectin covalently to the support, either directly through a functional group on the antibody or indirectly through a spacer arm such as a protein, polyamino acid, or synthetic linking group. Absorption of the antibody or lectin can sometimes afford sufficient stability during the assay, although covalent attachment is thought to be desirable from the standpoint of stability. The method of attachment should be chosen so as to preserve the binding activity to the highest degree possible. When anti-analyte antibodies are utilized in this manner, the antibody utilized in the enzyme-labeled antibody conjugate can be the same as that used to coat the support or it can be an antibody directed to different antigenic determinant on the analyte. In contrast to the common practice of immunochemically capturing analytes, which step requires an antibody of the analyte already on the support, the assay of this invention utilizes a nonimmunochemical capture of the analyte. This capture can occur through the physical forces described above or can be mediated by, for example, lectins on the support.

In some other cases, where the labeled immune complex is formed in the assay medium, such as by pre-incubation of the sample containing the analyte with an enzyme-labeled antibody or first with an antigen-antibody complex or with an antibody and then with an enzyme labeled antibody, the support must be of such pore size to retain the immune complex formed but to permit excess reagents to be passed through. Commercially available filter papers or porous membranes allow the convenient selection of a variety of suitable pore sizes. One particularly suitable porous membrane has a nominal pore size of 0.45 μ.

The analyte-binding protein(s) is usually an antibody but it can also be a nonantibody protein such as protein A, lectins, cellular receptors and serum transport proteins.

Because of the specificity lectins show for different cellular carbohydrates, one can utilize different lectins to selectively bind a large variety of microorganisms found in biological samples to the active support utilized in the immunoassay of this invention. Among the many lectins available, a few are shown in the Table below together with the microorganism(s) they are capable of capturing:

As can be appreciated from the above, the lectins can act in at least two different functions. In one mode of operation, a lectin can capture all of the microorganisms from a biological sample for which the lectin has specificity. In such a situation the antianalyte antibody-enzyme conjugate acts to identify the microorganism sought. Alternatively, the lectin itself can serve to capture selectively the microorganism of interest (see footnote 2 above). The microorganism so immobilized on the active support can then be detected by the immunoassay of this invention.

In some other cases polyacrylamide can be utilized for the non-immunochemical capture of various analytes as can be fibronectin which can capture various bacteria.

In yet some other cases, it is desirable to coat the support with erythrocytes or erythrocyte ghost membranes from different mammalian species. These erythrocytes or erythrocyte membranes selectively bind to glycoprotein hemagglutinin determinant found as surface projections on different groups of viruses. The following table lists the erythrocyte and the viruses they bind:

Antibody can be produced by methods well known in the art, including animal immunization and cell fusion (hybridoma) techniques. It can be used as whole serum, ascites or tissue culture fluid, but it is generally preferred to use a purified or partially purified preparation. An immunoglobulin fraction can be prepared by ammonium sulfate precipitation; an IgG fraction can be obtained by ion-exchange chromatography or gel filtration. Antibodies can also be affinity purified by elution from an antigen-coated support.

In some cases, it can be preferable to use fragments of antibodies, such as Fab, Fab', F(ab')₂ or haIf-molecules, instead of whole antibodies. Possible advantages can be related to the valence of the fragments or their lack of an Fc region. The label in the labeled antibody conjugate can be chosen from a variety of substances known in the art such as radioisotopes, fluorophores, enzyme inhibitors and chemi- or bioluminescent materials. Most often the label is an enzyme. The enzyme is chosen for its stability, turnover number, ease of purification, ease of measurement, and freedom from interfering substances often found in biological samples. When the analyte is an enzyme, it can act both as a label and as a member of a specific binding pair.

The conjugation of the enzyme label to an antibody can be carried out by known methods and in such a way as to minimize both the loss of immunoreactivity of the antibody and the loss of activity of the label. Attachment can be direct or indirect, via a spacer arm. Usually, it is necessary to purify the conjugate extensively before use in the assay in order to remove any unreacted label, unlabeled antibody, and nonimmunoreactive labeled antibody. This can be accomplished by a combination of gel filtration and affinity chromatography steps. The conjugate can be an enzyme-labeled anti-analyte antibody or an enzyme-labeled antibody which is an antibody to an anti-analyte antibody. In general, it is desirable to have as many enzyme molecules per antibody as possible and in no case should the ratio of enzyme to antibody be less than 1:1. The number of enzyme label molecules is limited by the overriding need to maintain both immunoreactivity of the antibody and enzymatic rectivity of the label. In some cases, it can be advantageous to link the antibody and the label each to a common carrier rather than to each other. For example, one antibody and several labels can be conjugated to a high molecular weight protein or polysaccharide, such as dextran. The reaction product(s) of the enzyme component of the immune complex with signal generating reagents is soluble and preferentially interacts with and binds to the support. This concentration of the signal on the support is a consequence of careful matching of the support material and the product (s) and makes the assay of this invention uniquely useful for the rapid determination of agents causing infectious diseases. The signal generating reagents are those components of the immunoassay of this invention which are necessary to produce a detectable signal upon interaction with the enzyme label component of the immune complex. These reagents include substrates for the enzyme label or a combination of materials capable of ultimately producing a substrate for the enzyme label such as in a coupled enzyme reaction.

In some cases, the product itself is a colored substance; for example, chromogenic substrates such as para-nitrophenyl phosphate and ortho-nitrophenyl galactoside are known for many common enzymes. Chromogen/substrate pairs are also known, for exmple, peroxides and ortho-phenylenediamine for horseradish peroxidase. Alternatively, the product can be less visible or substantially colorless but can become more visible or colored upon interaction with the support. For example, the support can be a starch-coated material which can complex with iodine generated as the product of an enzyme-catalyzed reaction to yield a characteristic blue color on the support. In another embodiment, the porous structure of the support can create a micro pH environment different from the pH of the solution, the reaction product being colored in the environment of the support but not in the solution. For example, the hydrolysis of para-nitrophenyl phosphate by acid phosphatase led to the formation of a yellow support even though the pH of the assay medium was acidic. It is also possible that the support can function to bind an ionized, colored form of product which is in equilibrium with an unionized, colorless form thus shifting the equilibrium.

The assay of this invention can be carried out by first mixing patient sample suspected of containing analyte with a conjugate of anti-analyte and enzyme. Patient sample can include whole blood, blood serum, blood plasma, urine, σerebrospinal fluid, a throat swab, lesion exudate, an aqueous dispersion of feces, and discharges, scrapings or lavages. Sample and antibody-enzyme conjugate are usually incubated for one minute to one hour, most often five minutes to fifteen minutes, in the temperature range of 4-45ºC, most often 23-37ºC. A molar excess of antibody-enzyme conjugate over the analyte is used to insure adequate antigen-antibody reaction in the short time allowed. Advantageously, the largest possible excess is utilized without leading to soluble complex formation. Reaction stoichiometry can be determined empirically. After preincubation, the antigen-antibody complexes formed are collected by filtration on a suitable support, for example, a starch-coated microporous filter. The complexes are retained on and within the porous support while other substances present in the sample, as well as uncomplexed conjugate, are washed through. Washing can be done with an aqueous solution buffered to a pH between 4 and 10. The buffer can also contain surfactant such as nonionic surfactant, protein(s) such as albumin, and electrolytes. Washing can be accomplished by flowing buffer through the support or, in some cases, it can be sufficient to wash the support by repeated immersions in fresh buffer. The former approach is usually faster and more efficient and is preferred. After washing, the support is contacted with signal generating reagent. When the enzyme label is horseradish peroxidase (HRP), for example, the signal generating reagent consists of glucose oxidase, glucose, and sodium iodide in a suitably buffered, aqueous medium. The signal generating reagent can be flowed through the support or the support can be immersed in the solution. H₂O₂ generated by the action of glucose oxidase on glucose serves as a substrate for HRP, generating a hydroperoxide complex. This, in turn, oxidizes iodide ion to iodine which forms a complex with the amylose in starch to form a deep blue color. This color, the intensity of which is directly proportional to the concentration of analyte in the sample, is concentrated on the support. Preferably, the color formed is stable and can be kept as a permanent record of the test result.

Alternatively, patient sample can be applied to the support first, followed by addition of antibodyenzyme conjugate. Sample can be deposited on the support by filtration, spotting or swabbing. Free conjugate passes through the support while conjugate bound to analyte is retained on the support and is available for subsequent reaction with the signal generating reagent.

In another embodiment, the support can be coated with anti-analyte. Filtration of the sample results in specific capture of analyte. After washing, a molar excess of a conjugate of anti-analyte antibody and an enzyme is filtered through the support. The amount of conjugate bound is a function of the amount of analyte bound in the first step and is quantified by addition of the signal generating reagents.

The assay of this invention can also be carried out by contacting the support with an analyte-anti analyte antibody complex which can be captured nonimmunochemically. The enzyme conjugate in this case can contain the same or a different anti-analyte antibody or an antibody to the first anti-analyte antibody. The support can then be washed with an aqueous solution buffered to a pH between 4 and 10. The buffer can also contain surfactant such as nonionic surfactant, protein(s) such as albumin, and electrolytes. Washing can be accomplished by flowing buffer through the support or, in some cases, it can be sufficient to wash the support by repeated immersions in fresh buffer. The former approach is usually faster and more efficient and is preferred.

EXAMPLE 1 ASSAY FOR h-IgG

This example illustrates the concentration of color on the support produced by the reaction of an enzyme-labeled conjugate on the support and signal generating reagents. Polyamide mesh discs (3/4" x .002") were treated first with aqueous methanol containing 10% w/v CaCl₂ and then treated with 3.65 M HCl for 30 minutes at 45ºC. This hydrolyzed the polymer backbone and provided functional (COOH, NH₂) groups on the film surface for protein attachment. The film surface was then activated with a freshly prepared 5% aqueous solution of the trifunctional aziridine crosslinking reagent XAMA-7 (Cordova Chemical Co.) for 25 minutes at room temperature. Excess XAMA-7 reagent was removed by successive washes with purified water.

Protein attachment was then achieved by equilibrating the activated nylon discs in a saline (0.85%) solution containing 2 mg/mL human IgG (h-IgG. Miles Laboratories). [Rabbit IgG (r-IgG) was used as control]. The films were then washed free of excess protein in cold phosphate-buffered saline (PBS) buffer, pH 7.5, and stored at 4ºC in PBS buffer until ready for use.

The assay was performed by equilibrating (2 hours) the test films with a solution of anti-h-IgGHRP (horseradish peroxidase) enzyme conjugate (0.1 mg/mL, available from Polysciences, Inc.) in 0.1 M PBS buffer, pH 7.8, containing 1% BSA (bovine serum albumin), washed free of excess conjugate by three successive washes with cold PBS buffer, and then added to the signal generating reagents (2 mL/disc). The signal generating reagents were prepared immediately before use by dissolving 35 mg of o-phenylenediamine dihydrochloride and 25 mg of urea peroxide in 50 mL of pH 5.0 buffer. Those test films which had h-IgG attached to them showed a positive color response within minutes while those without h-IgG and, therefore, having no HRP-conjugate, exhibited no color response. The color oh the support was permanently bound as indicated by stability to subsequent washings and there was no color visible in the test solution.

EXAMPLE 2 ASSAY FOR PAP This Example is a special case wherein, since the analyte is an enzyme and therefore can act as a label, the immune complex of the inventions is simply the antibody bound to the analyte/enzyme.

Discs of approximately 1 cm diameter were punched from a sheet of woven nylon fabric (CMN-5 monofilament nylon mesh, 5 micron mesh opening, available from Small Parts, Inc., Miami, FL) . Half of the discs were treated by soaking for 1 hour at room temperature in 0.1 M 1,6-hexanediamine in water. The remainder were untreated. All were washed by immersion four times in 0.1 M Tris buffer, pH 7.0. Discs were stored in 0.1 M Tris, pH 7.0, until used. Treated and untreated discs were soaked in 0.1 M Tris, pH 7.0, containing 0.108 mg/mL monoclonal antiprostatic acid phosphatase (PAP) antibody (IgG, 1 mL/ disc) for one hour at room temperature. The IgG fraction was obtained from a 50% ammonium sulfate precipitation of mouse ascites fluid by ion exchange chromatography on DEAE-cellulose in 0.02 M phosphate buffer, pH 8.0. according to standard procedures. Control discs were soaked in buffer only, without antibody. After soaking, discs were washed by immersion twice in 0.1 M Tris, pH 7.0, and twice in 0.1 M borate buffer, pH 8.0. (Previous experiments had indicated that even untreated discs have antibody physically adsorbed on them.) PAP (human, 70 units activity/mg protein, obtained from Calbiochem, Inc.) was dissolved in 0.1 M borate buffer. pH 8.0, to provide a concentration 5 μg/mL. Discs prepared as above were soaked for 15 minutes at room temperature in this PAP solution (1 mL/disc). They were then washed twice by immersion in 0.1 M borate buffer, pH 8.0, containing 0.1% Tween 80 nonionic surfactant. Control discs were prepared by omitting the PAP soak.

Discs were placed in 0.25 mL of a 4 mg/mL aqueous solution of p-nitrophenyl phosphate, disodium salt, and 0.25 mL of 0.09 M citric acid, 0.01 M sodium chloride buffer, pH 4.8. After 15 minutes at room temperature, the discs were washed twice by immersion in 0.1 M borate buffer, 0.1% Tween 80, pH 8.0. In this qualitative assay for the presence or absence of PAP, hexanediamine-treated discs exposed to both antibody and enzyme were yellow in color indicating that the signal, the product of PAP-catalyzed hydrolysis of p-nitrophenyl phosphate, was being concentrated on the support. There was no color concentrated on the untreated discs. Control discs lacking either antibody or enzyme, which is the analyte in this Example, were colorless.

EXAMPLE 3 ASSAY FOR STREPTOCOCCUS A

A. Affinity Purification of Antibody to Streptococcus

Streptococcus Group A antiserum was prepared in rabbits according to the protocol established by the Center for Disease Control. Antisera from the third bleed were pooled and utilized in the purification step. The antiserum so obtained reacted positively with streptococcal group antigens A, C, E, F, and G and did not react with B and D when tested by the Lancefield Capillary Precipitin Test. The streptococcal group antigens utilized were Difco Bacto-Streptoccal Antigen Set. Catalogue No. 2368-32, Control No. 683957.

Thirty mL of the antiserum was placed in a 0.7 cm X 17 cm column (column bed volume approximately 10.8 mL) containing N-acetyl glucosamine agarose (obtained from the Sigma Chemical Company). After removal of the unbound serum components by washing with 120 mL of PBS (prepared from 8.0 g NaCl, 0.2 M KH₂PO₄. 2.9 g Na₂HPO₄.12 H₂O, and 0.2 g KCl) . The bound antibody was eluted with 90 mL of 3 M NH₄SCN, pH 7.4. The fractions were collected, pooled, dialyzed extensively against PBS, and concentrated to 8.0 mL using an Amicon Model 202 Ultrafiltration cell with a YM10 membrane having a nominal molecular weight exclusion of 10.000 daltons. The final yield of immunoglobulin, assuming that a 1 mg/mL solution of IgG has an absorbance at 280 nm of 1.4. was 12.8 mg. This affinity purified anti-streptococcal antibody (APSA) reacted positively only with Streptococcus A antigen in the Lancefield Precipitin Test. As control, APSA was premixed with an equal volume of N-acetylglucosamine; no reactivity was seen with any of the Strep group antigens. When APSA was mixed with an equal volume of PBS, reactivity was seen with Strep A. B. Preparation of APSA-HRP Conjugate

8.0 mg of the APSA prepared in (A) above was conjugated with horseradish peroxidase (HRP) by the following procedure: 4 mg HRP (RZ 3.0) in 1.0 mL of water was mixed with 0.2 mL of freshly prepared 0.1 M NalO₄. The mixture was stirred for 20 minutes at room temperature and then dialyzed against 1 mM sodium acetate buffer, pH 4.4, overnight at 4ºC. To the retentate was added 20 μL of 0.2 M carbonate, buffer, pH 9.5, and 8 mg of Ig dissolved in 1.0 mL of 0.01 M carbonate buffer, pH 9.5, and stirred for 2 hours at room temperature. After this time period, 0.1 mL of freshly prepared sodium borohydride solution (4 mg/mL in water) was added and the mixture was allowed to stand for 2 hours at 4ºC. This was followed by the addition of an equal volume of saturated ammonium sulfate solution, the mixture was then centrifuged, the precipitate was washed twice with 50% saturated ammonium sulfate solution and then dialyzed extensively against PBS. Bovine serum albumin was then added to a concentration of 1% and the conjugate was filtered through a Millipore filter (0.22 μm) . To the filtrate was added an equal amount of glycerol. The conjugate was stored at -20°C. C. Assay APSA-HRP conjugate prepared in (B) above was diluted in a PBS/Tween/BSA buffer to a dilution of 1:20. One hundred μL of the conjugate and 100 μL of heat killed Strep A bacterial antigen were mixed and incubated for 10 minutes at room temperature. FiftyμL aliquots of this material were pipetted onto pre washed filters on a vacuum filter apparatus . The starch coated filter was prepared by spray coating filters (Catalogue No. HAWPO4700, 0.45 μ pore size, 47 mm diameter, available from Millipore Corp.) with Easy-On spray starch. The filters were saturated and the excess starch removed by drying at 45°C for 2 hours. The filters were then soaked in approximately 10 mL of a carbonate-BSA buffer for 1 minute, then washed with an additional 20 mL of the same buffer, and finally with approximately 20 mL of a PBS-Tween buffer and held under vacuum until all excess buffer was removed. The filters were then transferred to a Petri dish.

Approximately 1 mL of the signal-generating reagent was then utilized to flood the spotted filters. The signal-generating reagent was prepared from 0.40 mL of 2% aqueous sodium iodide, 0.50 mL of 4% aqueous ß-D-glucose solution, 0.1 mL of a citratephosphate buffer, pH 6.0, and 0.05 mL of a 1 mg/mL aqueous solution of glucose oxidase (Obtained from Sigma Chemical Co.). The citrate-phosphate buffer was prepared from 7.74 g citric acid and 17.93 g of Na₂HPO₄/l L of H₂O. A positive reaction was indicated by the formation of dark blue spots in approximately 30 seconds. The data below show that as low as 22 colony forming units per sample could be detected by this assay. The color intensity of the test results were graded from 0 to +4, +4 being the darkest color, 0 being no color: CFU/Sample Color Intensity

2.2 0

22 1+

220 1+

2,200 2+

22,000 3+

220,000 4+ EXAMPLE 4 ASSAY FOR HERPES SIMPLEX VIRUS TYPE 1 HSV antigen was purchased from MA Bioproducts (Walkerville, MD) together with negative controls. The Herpes antibody was Dako anti-herpes simplex virus type 1 rabbit antibody, conjugated to horseradish peroxidase. The conjugate was diluted 1:50 in PBS-TweenBSA. The antigen was diluted 1:100. 1:1000, and 1:10,000 in PBS-Tween-BSA buffer. Fifty-μL aliquots of the HSV dilutions above were mixed with 50 μL of the conjugate. After 30 seconds, 50 μL of a PEG solution was added (20% w/v solution of polyethylene glycol 8000 in PBS) and mixed thoroughly. After 10 minutes of incubation at room temperature the solutions were pipetted onto pre-washed starch coated filters as described above. The filters were washed with a PBS-Tween buffer and then flooded with the signal-generating reagent prepared from 0.1 mL sodium citrate-phosphate buffer (prepared from 7.74 grams of citric acid, 17.93 grams of NA-HPO. per 100 mL of water, pH 6.0) adjusted to pH 5, 0.45 mL of 2% aqueous Nal and 0.45 mL of 0.03% H₂O₂. The assay for HSV antigen showed positive results down to the 1:1000 dilution. The negative control gave a faint spot, readily distinguishable from positive spots at the same dilution.

EXAMPLE 5 ASSAY FOR ENTEROTOXIN Enterotoxin antigen (purified porcine heatlabile E. coli enterotoxin) and goat antiserum to purified porcine heat labile E. coli enterotoxin were both obtained commercially. Swine anti-goat IgG-peroxidase conjugate was purchased from Tago, Inc. (catalog no. 6401). Enterotoxin was diluted with a Tween-1% BSA buffer to the following concentrations: 100 μg/mL, 10 μg/mL, 1 μg/mL. 0.1 μg/mL. 0.01 μg/mL. and 0.001 μg/mL. It is known that agarose is capable of binding enterotoxin.

One hundred-μL portions of agarose preparation (Sepharose 4B. available from Pharmacia, 40-190 μ diameter beads suspended to a packed volume of 50% in the Tween-BSA buffer) were mixed with 100-μL portions of the enterotoxin samples and incubated for 15 minutes at room temperature. After incubation, 100 μL of goat antiserum (1:100 dilution in Tween-BSA buffer) was added to each of the samples and incubated for a further 15 minutes at room temperature. Then 100 μL of the conjugate (diluted 1:10 in Tween-BSA) was added to each of the samples and the final mixture was again incubated for 15 minutes at room temperature. Each sample was then applied in spots to starch-coated filter mounted on a Millipore filter unit (Catalogue No, XX10 04700) and the liquid portion of the mixture was pulled through the filter by vacuum. The beads remaining on the filter were washed with the Tween-BSA buffer. A wash volume of approximately 10 mL per spot was used. The filter was then removed and was flooded with a mixture containing 0.40 mL of 2% Nal. 0.50 mL of 4% ß-D-glucose solution, 0.1 mL of a citrate-phosphate buffer, pH 6 (as described above), and 0.05 mL of a 1 mg/mL solution of glucose oxidase. After 2 minutes, the color developed at each of the spots was graded using the 0 to 4+ system. Positive results were obtained down to the 1 μg/mL dilution of the toxin, corresponding to 100 ng per sample.

The following control reactions were negative: Buffer, agarose, antibody, and conjugate, no toxin; toxin, buffer, antibody, conjugate, no agarose; and toxin, agarose, buffer, conjugate; no antibody. EXAMPLE 6 SWAB ASSAY FOR HERPES VIRUS Various concentrations of infectious HSV-1 (F-strain). grown in Hep-2 cells, were prepared by diluting in a PBS/0.5% Tween 20/1% BSA buffer and placed in glass tubes. Starch coated cotton swabs were prepared by spraying with Easy-On spray starch. 1-2 sprays, until the swabs were soaked, followed by drying at 45°C for 2 hours in an inverted position. A swab was placed into each of the tubes containing virus and allowed to absorb the sample liquid for approximately 20 seconds. The swabs were then removed and placed into a second set of tubes each one containing 150 μL of a rabbit anti-HSV-1/HRP conjugate (available from Dako; diluted 1:100 in the above PBS/Tween/BSA buffer).

After a 5-minute incubation time the swabs were removed and washed three times by immersion in a PBS/0.5% Tween 20 buffer. After washing, the swabs were transferred to fresh tubes each one containing 0.2 mL of freshly prepared signal-generating reagent (prepared from 0.1 mL of citrate-phosphate buffer, 0.45 mL of 2% by weight sodium iodide solution, and 0.45 mL of a 0.3% (v/v) solution of hydrogen peroxide in distilled water). Color development was observed within 10 seconds. After 60 seconds, each of the swabs was washed under tap water to stop the reaction. The various color intensities corresponding to the different HSV sample concentrations are shown below: HSV Dilution Color Intensity

Undiluted 1-1/2+

1:10 2+

1:100 3+

1:1000 3+

1:10,000 1-1/2+

1:100,000 1/2+

Control +-

The nonlinearity of color intensity observed was due to the prozone effect, that is, only a single dilution of antibody was utilized in the assay. The control for this assay was performed as above but without HSV antigen.

EXAMPLE 7 SWAB ASSAY FOR STREPTOCOCCUS A The procedure in this assay was the same as described for the swab Herpes virus assay, utilizing an affinity-purified anti-strep/HRP conjugate (APSAHRP) described in Example 3 above and Strep A bacteria.

The antigen was diluted 1:10, 1:100 and 1:1000. The control contained no antigen. Strep A Dilution Color Intensity

1:10 2+ 1:100 3+ 1:1000 3+ Control +-

The nonlinearity of color intensity observed was due to the prozone effect. EXAMPLE 8 DETECTION OF HUMAN IgE Rabbit anti-human IgE was prepared by standard techniques, injecting the Fc fragment of human IgE myeloma protein into rabbits and then bleeding approximately 1-3 months later. Sheep anti-rabbit immunoglobulin, having specificity for the total immunoglobulin fraction of rabbit serum, was obtained commercially. Decreasing amounts of sheep anti-rabbit Ig (stock solution diluted 1:5, 1:15, 1:45, 1:135,

1:405. and 1:1215) were added to a constant amount of rabbit anti-human IgE (stock solution diluted 1:15). After 18 hours at 4ºC, each solution was inspected visually to determine the relative proportions of the two reagents at which soluble complexes were formed. The 1:135 dilution was the first one producing a clear solution with the rabbit anti-IgE.

For assay purposes, four five-fold dilutions of human IgE, covering a range of concentrations from 80 ng/mL to 10 μg/mL, were prepared. The source of IgE was a human myeloma obtained from patient PS. A 100-μL portion of each of the five dilutions was mixed with 100 μL of goat anti-IgE-HRP conjugate solution (obtained from Pelfreez Biologicals, lot number 282D, 10 mg/mL concentration diluted 1:500 in a borate buffer containing 0.005M boric acid, 0.0015M sodium tetraborate. 0.14 M NaCl, and 1 mg/mL gelatin. pH 8.3), and incubated for 1 minute at room temperature. 100 μL of the soluble complex prepared above was then added to each of the mixtures followed by an amount of a 16% (w/w) solution of 5 mM phosphate-buffered PEG-6000 sufficient to bring the final concentration of PEG in the reaction mixture to 4%. The mixture was incubated for 10 minutes at room temperature and then a 3-μL aliquot of the resulting suspension was placed onto a Millipore HA filter having a pore size of 0.45 μ. As soon as the fluid was absorbed into the filter the filter was placed into a Swinney filter holder and the filter was washed using 2 mL of PBS containing 0.05% Triton X-100. After washing, the filter was removed and placed into a signal generating reagent solution having a final concentrations of hydrogen peroxide of 0.02% and of ortho-phenylenediamine (OPD) of 0.02% in a citrate buffer. 0.05 M citric acid, 0.065 M K₂HPO₄, pH 6. The presence of IgE could be detected at levels as low as 80 ng/mL in less than 3 minutes by generating a yellowish brown spot on the filter paper. Controls for this assay consisted of utilizing the soluble complex and the anti-IgE-HRP conjugate in presence of PEG but without IgE and resulted in very faint yellow spots only.

When OPD was replaced by 3.3'-diaminobenzidine. the assay did not work satisfactorily; only faint brown spots were formed even at the 10 μg/mL concentration level (slightly darker than the controls described above, all other concentrations were indistinguishable from the controls). The oxidation product of this benzidine substrate is insoluble and is, therefore, outside of the scope of this invention. EXAMPLE 9

ASSAY FOR STREPTOCOCCUS A

A. Preparation of Crosslinked Amylose-APSA Column

A 30-gram portion of amylose (from potato starch, containing some insoluble matter and approximately 5% amylopectin) was treated with 100 mL of ice-cold 5 N sodium hydroxide and stirred until a uniformly grainy slush was obtained. Stirring was continued for 30 additional minutes at which time 15 g of epichlorohydrin was added with stirring and the reaction was allowed to continue for 25 minutes in an ice bath. The reaction was continued for 1 additional hour at 45-50°C. After this time period the reaction mixture was transferred into a larger container and 1 L of purified water was added. The reaction mixture was allowed to stand overnight at between 2-8ºC. Suspended fines were removed by twice decanting the supernatant liquid. The remaining gel was filtered through a coarse glass filter, washed 3-times with purified water, twice with ethanol, and twice with acetone. After washing, the gel was suctioned dry and dried in an oven at 40-50°C overnight, followed by vacuum drying at room temperature for two hours. The dried gel was screened and the 100-200 mesh size material was then utilized for activation and attachment of the antibody.

A 0.5-g quantity of the crosslinked amylose prepared above was transferred to a glass counting vial. 10 mL of acetone and 0.30 gram of 1,1' -carbonyldiimidazole were placed into the vial. The vial was capped and shaken at room temperature for 20 minutes. The contents were then filtered using a coarse sintered glass funnel and washed extensively with water and then 0.1 M carbonate/bicarbonate buffer, pH 9.6. The washed gel was transferred to a clean glass counting vial. Then 5 mg of APSA (described above) was diluted to 5 mL in 0.1 M carbonate/bicarbonate buffer. The pH was adjusted as necessary to pH 9.6. This solution was transferred to a 15-mL polystyrene centrifuge tube and the washed gel added. The tube was capped and rotated for 22 hours at room temperature. This slurry was placed into a column and washed extensively with water and then with PBS/0.05% Tween-20. B. Assay for Streptococcal Group A Bacteria

Seven columns (internal diameter approximately 0.5 cm) were filled with the suspension prepared above to afford a column height of 2 mm. Each column was washed with a total volume of 20-25 mL of water and three 1-mL portions of a carbonate/bicarbonate buffer, pH 9.6, containing 1% BSA. To the columns were added various concentrations of killed Strep Group A bacteria in a total volume of 0.1 mL in PBS containing 1%

BSA, 0.05% Tween 20, pH 7.4. The Strep Group A concentrations ranged from 9 x 10⁷ colony forming units (CFU)/mL to 3 x 10⁵ CFU/mL. The seventh column served as a blank, that is, PBS buffer as described above without Strep A. The columns were allowed to equilibrate for 15 minutes and then washed with three 1-mL portions of the above PBS/BSA/Tween 20 buffer. After this time, 0.1 mL of a 1:20 dilution of the APSA-HRP enzyme conjugate, as prepared above, in the above PBS/BSA/Tween buffer was added to each of the columns and allowed to equilibrate for 15 minutes. Each of the columns was washed with three 1-mL portions of the same buffer. After washing, 0.1 mL of the signal generating reagent, prepared from 0.5 mL of 2% sodium iodide, 0.4 mL of a 4% solution of ß-D-glucose, 0.1 mL of a citrate/phosphate buffer, pH 6.0, and 0.05 mL of the 1 mg/mL solution of glucose oxidase was added. After 80 seconds, the columns were washed with water to stop the reaction. The blue color intensity of each of the columns is tabulated below: Strep A Color Intensity (CFU)

9 x 10⁷ 4+

3 x 10⁷ 3+

9 X 10⁶ 2+

3 X 10⁶ 2+

9 X 10⁵ 1+

3 X 10⁵ 1+ 0 +-

EXAMPLE 10 ASSAY FOR CYTOMEGALOVIRUS (CMV) A. Preparation of Antibody-Virus Complex CMV antigen (complement fixation titer of 1:8, obtained from MA Bioproducts, Inc.) was diluted in PBS/0.05% Tween 20 buffer, containing 1% bovine serum albumin, pH 6.5, to various concentrations ranging from undiluted to 1:256. Goat anti-CMV serum (IgG fraction, obtained from Dynatech Diagnostics) was diluted 1:200 in the above mentioned buffer. Soluble antigen/antibody complexes were formed by mixing 50 μL each of CMV antigen and the above antiserum for 15 minutes at 22ºC. A Millipore HA filter (0.45 μ) coated with Easy-On spray starch was prepared as described previously and washed with a 0.05 M carbonate buffer at pH 9.6 containing 1% BSA. Fifty μL of antigen/antibody complex was added to the starch coated filter while under vacuum. After addition, the filter was rinsed with a PBS/Tween buffer. pH 6.5.

Then 1 mL of rabbit anti-goat IgG/horseradish peroxidase conjugate (Cappel Laboratories, diluted 1:300 in PBS/Tween/BSA) was applied to the filter and incubated for 5 minutes at room temperature. The filter was then washed with a PBS/0.05% Tween 20 buffer, pH 6.5. The filter was removed from the suction apparatus and flooded with 1 mL of a signal generating reagent (prepared from 500 μL of 2% aqueous sodium iodide, 400 μL of 4% aqueous ß-D-glucose solution. 100 μL of a phosphate/citrate buffer, and 50 μL of a 1 mg/mL glucose oxidase solution). Readings were made 2 minutes later at which time the filter was washed with distilled water and then read again after a total of 5 minutes. Reactions were graded on a scale of 1+ to 4+ as shown below; the data represent those obtained after 2 minutes, no further color changes were observed with time:

CMV Dilution Color Intensity

0 3 +

1 : 2 3 +

1 : 4 3 +

1 : 8 2+

1 : 16 1+

1 : 32 +-

1 : 64 +-

1 : 128 -

1 : 256 -

Utilizing a mock infected cell line (MA Byproducts, Inc.) as a negative control, the undiluted sample showed +-, while all other dilutions gave negative results. EXAMPLE 11

STARCH GRANULE LATEX-ANTIBODY COLUMN ASSAY FOR STREPTOCOCCUS GROUP A

A. Preparation of Crosslinked Amylose

Latex-Antibody Column

Dry crosslinked amylose (60-100 mesh size) was prepared as described in Example 1. A 10% suspension of the crosslinked amylose was prepared in 0.1 M borate buffer. pH 8.5, and allowed to rehydrate for 30 minutes at room temperature. After rehydration, 200 μL of the above suspension was layered into plastic columns (inside diameter 0.5 cm. fitted with a fritted glass filter to prevent washout) and washed 3 times with 1 mL of 0.1 M carbonate buffer, pH 9.6, containing 1% BSA.

Styrene-divinylbenzene copolyraer latex particles of 25.7 μra average diameter were obtained from Sigma at 10% solids content. Affinity purified streptococcal antibody (APSA, prepared as in Example 1) was adsorbed to the latex particles as follows. One mL of the latex was mixed with 1 mL APSA and 1 mL 0.1 M carbonate buffer, pH 9.6, and incubated 2 hours at 37°C with intermittent mixing. One hundred μL of the above suspension was then layered on the amylose layer in the columns and a fritted glass filter was placed on top to prevent mixing. The column was then washed three times with 1 mL of 0.1 M carbonate buffer, pH 9.6, containing 1% BSA to remove free APSA.

B. Assay for Streptococcus Group A

Streptococcus was grown at 37°C for 24 hours in Todd-Hewitt broth culture and infectivity was assayed by standard colony counting methods on blood agar plates. Cells were heat-killed (60°C, 1 hour) and washed once in PBS, pH 7.4, prior to use. Antigen was serially diluted ten-fold in high salt acidic buffer (containing 9.0 g NaCl, 6.97 g Na₂HPO₄, 0.36 g KH₂PO₄, and 0.19 g KCl per liter), pH 6.5. containing 0.05% Tween 20 and 1% BSA. One hundred-μL portions of the different dilutions of the antigen were added to each of seven columns and allowed either to flow through the columns within 1 minute or incubate for 15 minutes at room temperature with the bottoms sealed. The control contained buffer without antigen. The columns were then flushed 3 times with the high salt acidic buffer. A 100-μL aliquot of APSA-HRP conjugate (as prepared in Example 1). diluted 1:20 in high salt acidic buffer, was added to each column and allowed to either flow through within one minute or incubate for 15 minutes at room temperature. After incubation, the columns were flushed three times with 1 mL high salt acidic buffer followed by the addition of 100 μL of signal generating reagent. The signal generating reagent was prepared from 0.40 mL of 2% aqueous sodium iodide, 0.50 mL of 4% ß-D-glucose solution, 0.1 mL of a citrate-phosphate buffer, pH 6.0, and 0.05 mL of a 1 mg/mL aqueous solution of glucose oxidase (obtained from Sigma). The intensity of blue color was graded and recorded 2 minutes after the addition of the signal generating reagent (0-4+ basis):

Color Intensity

CFU/Sample 15-Min Incubation Flow Through

10⁶ 4+ 4+

10⁵ 4+ 4+

10⁴ 4+ 3+

10³ 3+ 2+

10² 3+ 2+

10¹ 2+ 1+

10° 1+ 1+ control ± ± EXAMPLE 12 ASSAY FOR HERPES SIMPLEX VIRUS TYPE 1 A. Preparation of PolyacrylamideCyanoethylated Starch Particles Polyacrylamide-cyanoethylated starch (PACS) beads were produced by a variation of a procedure of K. Sutoh. et al., Anal. Biochem., Volume 79, 329 (1977). A mixture of 6 g acrylamide, 0.6 g N.N'methylene-bisacrylamide (BIS), 0.25 g σyanoethylated starch (available from Polysciences, Inc.), and 40 mg ammonium persulfate in 50 mL deionized water was deaerated for 5 minutes with frequent swirling to keep the starch in suspension. Deaeration was continued for an additional minute following the addition 20 μL N.N.N' ,N'-tetramethylethylenediamine (TEMED) . The mixture was poured slowly into 100 mL peanut oil under an atmosphere of nitrogen with constant stirring. Stirring was continued under nitrogen for two hours as the acrylamide solution polymerized into bead form. The suspended starch was physically entrapped in the beads as polymerization occurred. All manipulations were performed at room temperature.

The peanut oil was separated from the PACS beads by centrifugation at 200 x g (1000 rpm in an IEC PR-J centrifuge) for 5 minutes. The pellet was resuspended and stirred in 100 mL toluene overnight. The toluene was removed by filtration in a Buchner funnel, and the beads were suspended in 400 mL acetone. After ten minutes they were isolated again by filtration and allowed to air dry. Fines were removed by sieving the dry material with a 60-mesh screen. Final dry bead diameter was approximately 0.3-0.8 mm. B. Direct Binding Immunoassay for HSV Antigen

Dry PACS beads (0.1 g) were added to plastic columns (internal diameter approximately 0.5 cm) the bottoms of which were sealed with a cap. One hundred fifty-μL portions of HSV antigen, at dilutions of 1:25. 1:75, 1:225, 1:675 and 1:2,025 (antigen purchased from MA Bioproducts, Walkersville, MD, diluted in PBS, pH 7.4, containing 0.5% Tween 20 and 1.0% BSA) were added to the columns allowing the contents of the columns to rehydrate (approximately 1-2 minutes). After rehydration. 50 μL of rabbit anti-HSV-1-HRP conjugate (DAKO), diluted 1:100 in PBS, pH 7.2, containing Tween 20 and BSA as above, was added and allowed to incubate 5 minutes. The columns were then opened by removal of the cap and washed with 3 mL PBS, pH 7.2, containing 0.05% Tween 20 (column void volume approximately 200 μL) . At this stage the following enzyme detection system was used to detect the amount of conjugate bound to antigen immobilized on PACS beads: 200 μL of signal generating reagent, prepared from 0.5 mL of 2% sodium iodide. 0.4 mL of a 4% solution of ß-D-glucose. 0.1 mL of citrate-phosphate buffer, pH 6.0, and 0.05 mL of a 1 mg/mL solution of glucose oxidase, was added to the columns. After 60 seconds the blue color intensity of each of the columns was graded from 0 to 4+, 4+ being the darkest color and 0 being no color.

HSV Dilution Column Color Intensity

1:25 4+

1:75 3-4+

1:225 3+

1:675 2+

1:2,025 2+ negative control

1:75 2+ diluent alone 1-2+ EXAMPLE 13 ASSAY FOR HERPES SIMPLEX VIRUS TYPE 1 The immunoassay of this invention can be carried out utilizing cotton swabs, to which a lectin has been attached, coated with starch. The attachment of lectin is a two-step process. First, the swabs are activated with cyanogen bromide [J. Porath. et al., J. of Chromatography, Volume 86. 53 (1973)]. In the second step, the lectin is attached to the CNBr-activated swab ("Cell Affinity Chromatography Principles and Methods". Pharmacia Fine Chemicals. Sweden, 1979). The starch coating of the swabs and the assay protocol are described in Example 1 above.

A. Attachment of Lectin to Swabs Swabs (composed of polysaccharide fibers such as cotton, flax, hemp, jute, and rayon) are washed with water and then air dried. They are placed in 5 M potassium phosphate buffer. pH 11.9 (25 mL buffer per 1 g of cotton fiber), to which 2 mL of a cyanogen bromide solution (100 mg/mL) is added in small aliquots with gentle stirring over a 2-minute period at 5-10°C. The temperature is maintained for an additional 8 minutes at 5-10°C after which the swabs are washed thoroughly with water and allowed to air dry. CNBr-activated swabs expected to be obtained by the above procedure are washed with 200 mL of 1 mM HCl in several aliquots followed by washing with 5 mL coupling buffer (0.1 M NaHCO₃, 0.5 M NaCl, pH 8.3). The swabs are immediately transferred to 30 mL of coupling buffer containing 18-35 mg of purified snail (Helix pomatia) agglutinin (HPA. available from E. Y. Laboratories. Inc., San Mateo, CA) and allowed to agitate gently (no magnetic stirrer) for 2 hours at room temperature or 4°C overnight. The swabs are then transferred to 30 mL of blocking buffer (0.2 M gly cine, pH 8.0) and allowed to agitate gently for 2 hours at room temperature or 4°C overnight. Excess lectin is removed by washing with coupling buffer, followed by acetate buffer (0.1 M, pH 4. containing 0.5 M NaCl) followed by coupling buffer. The lectin coated swabs can be stored in coupling buffer at 4-8ºC with a bacteriostatic agent such as Merthiolate or allowed to air dry.

B. Starch-Coating of Lectin-Swabs Lectin-swabs (prepared in A above) can be starch-coated by spraying with Easy-On spray starch. 1-2 sprays, until the swabs are soaked, followed by drying at 45°C for 2 hours in an inverted position as described in Example 1. C. Assay

The assay can be carried out as in Example 1 above. Various concentrations of infectious HSV-1 (F-strain), grown in Hep-2 cells, are prepared by diluting in PBS/O.5% Tween 20/1% BSA buffer, pH 7.4, and 100 μl of each dilution placed in glass tubes. A starch-coated lectin-swab is placed into each of the tubes containing virus and allowed to absorb the sample liquid. HPA is expected to bind the following microorganisms (if present in the clinical specimen): Herpes simplex virus types 1 or 2 [(through the surface glycoprotein gC; S. Oloffson. et al., J. of Virology, Volume 38, 564 (1981)], Escherichia, Salmonella, and Streptococcus. After a 5-minute incubation period at room temperature, the swabs are removed and washed three times by immersion in a PBS/0.05% Tween 20 buffer, pH 7.4, to eliminate normal flora, host cell debris, and other potentially interfering substances. The swabs are then transferred to a second set of tubes containing 150 μL of a rabbit anti-HSV-1/HRP conjugate (available from DAKO ; diluted 1 : 100 in the above PBS/Tween/BSA buffer) . After a 5-minute incubation period, the swabs are removed and washed three times in buffer as before. After washing, the swabs are transferred to fresh tubes each containing 200 μL of signal-generating reagent (prepared from 0.5 mL of 2% sodium iodide, 0,4 mL of a 4% solution of ß-D-glucose, 0.1 mL of a citrate/phosphate buffer, pH 6.0, and 0,05 mL of a 1 mg/mL solution of glucose oxidase). After 60 seconds, the swabs are washed under tap water to stop the reactions expected to be occurring. The color intensities are expected to correspond to the different HSV sample concentations; a predicted set of results are shown below:

HSV Dilution Color Intensity undiluted 4+

1:10 4+

1:100 3+

1:1,000 3+ 1:10,000 1 1/2+

1:100,000 1/2+

1:1,000,000 +/- negative control⁽¹⁾ +/- diluent buffer alone +/-

(1)Mock-infected Hep-2 cell lysate

Nonlinearity of color intensity, attributable to prozone effects, are not expected since several antibody dilutions can be utilized. The positive color response is indicative of the presence of HSV.

Claims

CLAIMS :

1. An immunoassay for the detection of analytes in biological samples comprising the steps of:

(A) collecting or forming on an active support an immune complex having an enzyme component comprising the analyte immunocheraically bound to at least one antibody;

(B) contacting the enzyme component of the immune complex on the support with signal generating reagents whereby a product is formed, provided that when said product is a chromophore, it is soluble in the assay medium; and

(C) allowing the product to interact with the support which interaction results in the concentration of a detectable signal thereon.

2. The immunoassay of Claim 1 wherein said immune complex is formed in solution by pre-incubation of the analyte with an enzyme-labeled anti-analyte antibody and is subsequently collected on the support.

3. The immunoassay of Claim 1 wherein the formation of said immune complex on the support comprises the immunochemical capture of the analyte by an anti-analyte antibody bound to the support followed by a reaction with an enzyme-labeled anti-analyte antibody.

4. The immunoassay of Claim 1 wherein the formation of said immune complex on the support comprises the immunochemical capture of an enzyme-labeled anti-analyte antibody by the analyte covalently bound to the support.

5. The immunoassay of Claim 1 wherein the formation of said immune complex on the support comprises the immunochemical capture of a complex of an enzyme labeled anti-analyte antibody and the analyte by an anti-analyte antibody bound to the support.

6. The immunoassay of Claim 1 wherein the product formed in step (B) is soluble in the assay medium and is capable of emitting a detectable signal.

7. The immunoassay of Claim 1 wherein the analyte is selected from the group consisting of soluble organic antigens and haptens. cells, microorganisms and their products.

8. The immunoassay of Claim 1 wherein the support comprises nylon, cellulose nitrate, ethyl cellulose or modified polyamide.

9. The immunoassay of Claim 1 wherein the porous support is disc-like.

10. The immunoassay of Claim 1 wherein the product formed in step (B) is covalently attached to the support through functional groups on the support.

11. The immunoassay of Claim 10 wherein the functional groups are parts of color couplers selected from the group consisting of azomethine, azo-, leuco-, iηdazolin-, indoanalin- and pyrazolone-dyes.

12. The immunoassay of Claim 1 wherein the formation of said immune complex on the support comprises the nonimmunochemical capture of the analyte.

13. The immunoassay of Claim 12 wherein the analyte is captured by lectin.