CA2131054A1 - Macroporous hydrophobic cloth-based affinity-concentration device for microbiological and chemical analyses - Google Patents

Abstract

A novel affinity concentration device is provided herein for use in the analysis of target cells or other organic chemical analytes in dilute suspension. Such device comprises a macroporous, hydrophobic synthetic polymer cloth coated with a suitable, affinity ligand, e.g., a target-cell-specific capturing agent, or a suitable analyte-specific capturing agent.

Description

This invention relates to a macroporous hydrophobic cloth-based affinity concentration device for microbiologi-cal and chemical analyses.
The detection of pathogens and other analytes by means S of diagnostic tests claims a large share of the health care market and the agri-food industry. The definitive identi-fication of microbial pathogens in agricultural commodities (i.e., foods) and clinical specimens requires the demon-stration of the infectious agents or their components.
Traditional culture methods for the detection of pathogens are slow, expensive and of uncertain sensitivity, and require extensive laboratory personnel and facilities.
To overcome some of these disadvantages, specific binding assay techniques have provided analytical methods for determining various organic substances of diagnostic, medical, environmental and agricultural importance which appear in liquid media at very low concentrations. Speci-fic binding assays are based on the specific interaction between the ligand, i.e., the bindable analyte under deter-mination, and a binding partner for such analyte. In addi-tion, several immunological tests are now commercially available which exploit the specificity of the antibody-antigen (or hapten) reaction, namely: agglutination tests;
immunofluorescence tests; and immunoassays (e.g., enzyme immunoassay and radioimmunoassay).
Radioimmunoassay employs a radioactive isotope as the label. Because of the inconvenience and difficulty of handling radioactive materials and the necessity of a Z131~54 _ 2 separation step, assay systems have been devised using materials other than radioisotopes as the label component, including enzymes, bacteriophages, metals and organo-metallic complexes, coenzymes, enzyme substrates, enzyme activators and inhibitors, cycling reactants, organic and inorganic catalysts, prosthetic groups, chemiluminescent reactants, and fluorescent molecules. Such homogenous specific binding assay systems provide a detectable response, e.g., an electromagnetic radiation signal, e.g., chemiluminescence, fluorescence emission, or colour change, related to the presence of amount of the ligand under assay in the liquid sample.
Enzyme immunoassays use enzyme-labelled immunoreagents (antibodies or antigens) for the detection of antigens or antibodies captured on a solid phase. Adsorption onto an easily recoverable solid phase is a simple and rapid means of immobilization of immunoreactants for the subsequent capture of antigens or antibodies from a test sample.
Since antibodies and many antigens contain hydrophobic regions in their structures, they bind readily to hydro-phobic surfaces. Most commonly used enzyme immunoassays depend on the adsorption of immunoreactants onto either a flat solid surface or a microporous membrane. Solid phases, e.g., microtiter plates, tubes or beads, and plastics, e.g., polystyrene, polyvinyl chloride, nylon, and polymethacrylate have commonly been used. Although micro-porous membranes (e.g., nylon and nitrocellulose) have been used to adsorb antigens as well as antibodies, these are . ~ 2131054 thin and can only accommodate a small volume of test sample per unit area which has a limited contact with the surface area. Furthermore, since their pore sizes are small (e.g., microporous), their effective washing between each step in the assay requires a vacuum suction apparatus.
Biotinylated antibodies and avidin-(or streptavidin-) enzyme conjugates have also become a popular combination for the detection of antigens captured by immobilized anti-bodies in enzyme immunoassay since it provides greater sensitivity than direct antibody-enzyme conjugates. There-fore, a variety of avidin-(or streptavidin-) enzyme conju-gates are commercially available. The preparation of bio-tinylated antibodies involves the purification of the desired antibodies, which are then biotinylated in free solution. The entire procedure commonly requires a few days to complete.
In addition to the above techniques, several methods that use a variety of different principles have been developed. These include assay of microbial ATP, direct counting of single cells or microcolonies by fluorescent microscopy or flow, incorporating fluorescent antibodies, DNA or rRNA probes and the polymerase chain reaction or specific phages containing cloned genes for bioluminescence or ice nucleation.
Recently, the polymerase chain reaction (PCR) tech-nique for the amplification at unique deoxyribonucleic acid (DNA) sequences of bacteria and other target ce~ls has been developed for the detection of specific pathogens (e.g., _ 4 food-borne disease causing bacteria). This powerful tech-nique, which uses oligonucleotide primers targeting speci-fic DNA sequences, can achieve tremendous amplification of very low numbers of target DNA to levels which can be easily visualized by agarose gel electrophoresis analysis.
The techniques, above-described, can be extremely effective with pure cultures but when they are applied to chemically-complex food or clinical samples their sensi-tivity may be reduced or be inadequate to detect the numbers of organisms of concern. Enrichment methods which involve pre-growth of the target microorganisms in nutrient-rich media are usually applied but they lengthen the overall time needed for any particular assay. On the other hand, particulate matter can interfere with direct counting methods, food constituents interfere with assays of ATP and nucleic acid amplification methods, although this interference can partially be alleviated by elaborate pretreatment procedures.
The presence of certain exogenous chemical residues -20 (e.g., antibiotics, pesticides, etc.) in foods and other materials destined for human or animal consumption is a major issue which must also be addressed by government regulators and industry. For example, antibiotics (e.g., chloramphenicol) used in the treatment of bacterial infec-tions in animal husbandry present a public health hazard,and animal-derived commodities, e.g., milk must be regu-larly monitored to ensure their absence ,from these products.

_ 5 Several chromatographic methods have been developed for the detection of a variety of chemical residues (including antibiotics) in foods and other samples, including gas chromatography and high performance liquid chromatography (HPLC). HPLC is very commonly used in the analysis of antibiotics, e.g., chloramphenicol in foods.
In the chromatographic methods (gas chromatography, HPLC), components of a mixture are carried through a stationary phase (walls of a tubular column or porous solid packed within a tubular column) by the flow of a mobile phase (gas or liquid), and separations are based on differences in the migration rates among the sample components. Capillary electrophoresis has recently been introduced for the rapid electrophoretic separation and characterization of chemical and biological components in complex samples. In the most common application of the capillary electrophoresis method, a small volume of sample is introduced into a very narrow-bore charged-wall capillary filled with a suitable buffer and then subjected to a high voltage, resulting in the rapid and high-efficiency separation of the sample com-ponents based on their electrophoretic mobilities and the electroosmotic flow of the buffer.
Among the important disadvantages of most analytical chemistry techniques, e.g., HPLC and capillary electro-phoresis, are the limited volume of sample which can beintroduced into the instrument (e.g., a few nanoliters to several microliters), and the requirement for a relatively "clean" preparation of the analyte which must be free of `. ~ 213~054 _ 6 extraneous sample components that might interfere with the analysis. The limited sample volume which can be intro-duced into the instrument seriously limits the effective detectability (minimum concentration of analyte that can be detected) of the method, since the analyte is normally present in very small quantities (e.g., parts-per-billion concentrations) in a large volume of sample. Furthermore, the elaborate sample clean-up procedures (e.g., extraction in organic solvents, acid precipitations, etc., required to extract the analyte from the test sample (e.g., food mat-rix) and remove extraneous ample components add consider-able labour, time and cost to the analysis.
For the HPLC analysis of antibiotics, e.g., chloram-phenicol in foods, an on-line immunoaffinity sample clean-up procedure has been developed in which a microporousactivated tresyl support is used to covalently immobilize an anti-chloramphenicol antibody and then packed into a column which is placed in-line with and upstream from the analytical HPLC column. This requires a physical adapta-tion of the HLPC instrument to accommodate such an on-line column and may not be convenient for all users. The sample to be tested (e.g., milk) must be extensively treated in a series of steps involving acid precipitation of extraneous components, centrifugation, neutralization and then filtration, before it can be injected into the immunoaffinity clean-up column. This extensive sample preparation is required to remove food solids and 2~3~054 particulate matter which might clog the microporous packing of the column.
It has been suggested that the application of many assay methods could be improved if the analyte (e.g., bacteria or the chemical residue) could be separated from the bulk of the sample solution and concentrated from the food or clinical sample matrix before analysis. Indeed, several rapid methods for the detection of bacteria include - microporous membrane filtration or centrifugation steps either to capture or to pellet cells enhancing the sensi-tivity of the methods when applied to foods due to concen-tration of the cells from larger volume of sample. When they are used alone, however, these separation steps may have limited application because the filters may be blocked by food particles/components or organisms partitioned at interfaces during centrifugation. Other approaches, e.g., ion-exchange resins or electropositive filters have had limited success.
Antibodies immobilized on magnetic microspheres (beads) have recently been used to separate and concentrate bacteria from food and clinical samples but their routine use may be limited by their specificity and cost and the complex manipulations required to handle the beads.
Methods for the rapid detection and/or counting of bacteria can be highly sensitive when applied to aqueous suspensions, but were of restricted value, however, when applied directly to samples, e.g., foods. For example, the assay of microbial metabolites (e.g., ATP) is subject to ` 2131054 _ 8 high background levels of non-microbial origin. Particle-counting methods based on microscopy or flow cytometry can similarly suffer from interference by particulate matter.
Many of these problems might be reduced if the bacteria S were separated and concentrated from the sample before analysis. Food samples can be complex and varied matrices and simple centrifugation and filtration methods cannot generally be successfully applied. As alternative approaches, the adsorption of bacteria to ion exchange resins or electropositive filters have been examined.
However, another possibility is to exploit the specific binding activities of biological molecules, e.g., specific antibodies recognizing bacterial cell surface antigens.
The use of antibodies immobilized on magnetic beads, mentioned hereinabove, would be an example of such a possi-bility.
An alternative approach might be to exploit the sugar-binding specificities of lectins, which can be isolated from a wide range of biological materials. Lectins have been used for tissue typing and the separation of popu-lations of thymocytes and lymphocytes and are known to bind selectively to specific sugar residues present on the sur-face of most cells, including bacteria.
Lectins are proteins with specific carbohydrate binding activities. The selective agglutination of immature thymocytes by peanut lectin allows their separa-tion from mature thymocytes, and affinity chromatographic t~hn;ques have been used to separate different human g lymphocytes. Some lectins have been shown to bind to a number of bacteria and a range of bacterial components, e.g., teichoic acids, lipopolysaccharides and other bacterial polysaccharides.
It has previously been found that lectins immobilized on magnetic microspheres can be used successfully to separate and concentrate potential pathogens from cultures and from foods. Thus, M.J. Payne, Shona Campbell, R.A.
Patchett and R.G. Kroll, AFRC Institute of Food Research, Reading Laboratory, Shinfield, Reading, UK, in a paper entitled "The use of Immobilized Lectins in the Separation of Staphylococcus Aureus, Escherichi Coli, Listeria and Salmonella spp. from Pure Cultures and Foods", Journal of Applied Bacteriology, 1992, 73, 41-52, reported the inter-action of four lectins with a number of strains of food-borne pathogens, reported methods for lectin immobilization on particulate solid phases (e.g., beads) and reported the ability of three of these lectins to separate and con-centrate these organisms from pure and mixed bacterial cultures and from food samples.
R.A. Patchett, Allison F. Kelly & R.G. Kroll, AFRC
Institute of Food Research, Reading Laboratory, Shinfield, Reading, UK, in a paper entitled "The adsorption of Bacteria to Immobilized Lectins, Journal of Applied Bacteriology 1991, 71, 277-284, reported the agglutination of some food-borne bacteria by a selection of lectins and the application of lectin affinity methods to,adsorb some of these bacteria from aqueous suspension.

-M.A.B. Starbuck, P.J. Hill & G.S.A.B. Stewart, University of Nottingham, Faculty of Agriculture and Food Sciences, Department of Applied Biochemistry and Food Science, Sutton Bonington, Loughborough, Leics LEI12 5RD, UK, in a paper entitled "Ultra Sensitive Detection of Listeria Monocytogenes in Milk by the Polymerase Chain Reaction (PCR), published in Letters in Applied Microbio-logy 1992, 15, 248-252, reported that the polymerase chain reaction (PCR) had great potential for the detection of micro-organisms in environmental samples, allowing rapid and specific identification of pathogenic, spoilage and indicator organisms.
Asim K. Bej, Meena H. Mahbubani, Joseph L. Dicesare, and Ronald M. Atlas, Department of Biology, University of 15Louisville, Louisville, Kentucky, 40292, and Perkin-Elmer Corporation, Norwalk, Connecticut, 06859, in a paper entitled "Polymerase Chain Reaction-Gene Probe Detection of Microorganisms by Using Filter-Concentrated Samples", published in Applied and Environmental Microbiology, Dec.
201991, p. 3529-3534, 0099-22409/123529-06502.00/0, Vol. 57, No. 12, taught the detection of low levels of microor-ganisms in environmental samples by using polymerase chain reaction (PCR)-gene probe detection, by concentrating samples by filtration on microporous membranes.
25A.C. Fluit, R. Torensma, M.J.C. Visser, C.J.M.
Aarsman, M.J.J.G. Poppelier, B.H.I. Keller, P. Klapwijk, and J. Verhoef, Eijkman-Winkler Laboratory ,for Medical Microbiology, University Hospital Utrecht, P.O. Box 85500, HP G04-515, 3508 GA Utrecht, U-Gene Research B.V., Utrecht, and Section of Microbiology and Hygienic Processing, Unilever Research Laboratorium, Vlaardingen, The Nether-lands, in a paper entitled "Detection of Listeria Monocy-togenes in Cheese With the Magnetic Immuno-Polymerase Chain Reaction Assay, published in Applied and Environmental Microbiology, May 1993, p. 1289-1293, Vol. 59, No. 5, reported a detection system, the magnetic immuno-polymerase chain reaction (PCR) assay (MIPA) which was developed to detect Listeria monocytogenes in food. This method separates Listeria cells from PCR-inhibitory factors present in enrichment broths containing food samples by using magnetic beads coated with specific monoclonal antibodies (MAbs). The separated bacteria were lysed, and the supernatant containing the bacterial DNA was subjected to the PCR. Detection of _. monocytogenes in three naturally contaminated cheese samples with two different MAbs and PCR primers specific for the gene encoding the delayed-hypersensitivity factor showed that with MAb 55 all three samples were positive whereas with MAb A two samples were positive.
These papers taught that lectins [e.g., concanavalin A (con A)] or antibodies immobilized on various surfaces can bind to various bacterial cells. Particularly, it was found that con A could bind to L. monocytoqenes. Applica-tions for PCR using con A were not demonstrated, nor did they seem to be anticipated. Binding cells to a surface did not guarantee that the approach would be suitable for 213105~

PCR, since the DNA of the cells may be impossible to release from the surface (i.e., the DNA would not be available for PCR detection). Furthermore, solid phase material itself when included into the PCR may be inhibi-tory to the reaction. They also described methods forcapturing cells on a microporous polysulphone membrane, followed by PCR detection of the captured cells. They also described the use of magnetic beads with an immobilized Listeria-specific antibody for the capture of target cells and their assay by PCR.
Accordingly, the PCR technique has the sensitivity and specificity required to achieve the detection limits needed for bacterial pathogens in food. However, it is the isola-tion and harvesting procedures for bacterial target DNA
that are the limiting factors to the sensitivity of the assay. Moreover, the published techniques suffer from the disadvantages of: clogging of microporous filters by sample particles (e.g., food); complicated chemical extraction procedures required to remove captured cells for PCR;
expensive, since in the case of immuno-magnetic capture techniques, the complicated wash requires a magnetic separator to recover beads after mixing with the sample;
the beads themselves are also expensive; and the volume of sample that can be pressed is limited by capacity of the separator.
Thus, the polymerase chain reaction (PCR) technique for the enzymatic amplification of specific DNA sequences has been applied to the rapid and sensitive detection of a ;~.~J~

_ 13 large number of microbial pathogens in foods and clinical samples. Although very sensitive PCR systems have been developed for some food-borne pathogens, making it possible to detect less than 10 cells per reaction in pure cultures, the application of PCR to the detection of pathogens in actual food and environmental samples has not always enjoyed such exquisite levels of detectability, due to the paucity of practical and efficient methods to recover the cells from the sample prior to performing the PCR analysis.
Thus, there is a need for rapid, simple and inexpen-sive methods for concentrating cells (e.g., bacteria) and washing away possible interfering substances originating from the sample in order to increase the reliability and effective detectability of assays, e.g., the PCR technique.
Furthermore, there is great need for simple, inexpen-sive and user-friendly sample preparation ("clean-up" and concentration) methods for use in the analysis of organic chemical analytes (e.g., antibiotics or pesticides) present in complex samples (e.g., foods) by analytical chemistry techniques, e.g., HPLC and capillary electrophoresis.
Accordingly, those concerned with the development and use of analytical techni~ues for microbiology and chemistry and related devices have recognized the desirability for further improvements. It would be extremely useful to have an affinity clean-up system based on the use of antibodies or other specific ligands (e.g., lectins), which is simple, inexpensive, easy to use and minimizes or eli,minates the need for prior sample treatment.

~ Z131054 _ 14 It is therefore an object of one aspect of the present invention, to provide a simple, rapid and efficient method for the affinity-concentration and clean-up of microbial and chemical analytes present in complex samples (e.g., foods) for their subsequent analyses by relevant analytical methods, e.g., PCR, chromatographic or electrophoretic techniques.
An object of a further aspect of the present invention is the provision of relatively simple, yet highly effective and sensitive, diagnostic tests for the detection of speci-fic bacterial pathogens and chemical contaminants.
An object of yet another aspect of this invention is the use of macroporous hydrophobic cloths to make affinity-concentration and sample clean-up procedures rapid and simple.
An object of yet another aspect of this invention is to enable the use of macroporous hydrophobic cloths coated with antibodies or other suitable specific ligands in the analysis of antigens and haptens.
According to a broad aspect of this invention, a simple method is provided for the affinity concentration and clean-up of bacterial cells based on the use of a macroporous, hydrophobic, synthetic polymer cloth coated with specific antibodies or other suitable specific ligands (e.g., the lectin concanavalin A) as an inexpensive high surface area adsorbent.
By another aspect, the present invention,provides a - simple method for the affinity concentration and clean-up -of small organic chemical analytes (e.g., proteins, antibiotics, drugs, etc.) based on the use of a macroporous, hydrophobic, synthetic polymer cloth coated with specific antibodies as an inexpensive high surface area adsorbent.
Therefore, by a broad aspect of the present invention, an improvement in an affinity concentration and sample clean-up procedure for subsequent analysis, is provided by the following three main steps: capturing analytes (bacteria or chemical residues) on affinity ligand coated cloth by filtration; desorbing the captured analytes from the cloth (e.g., by lysis of bacterial cells on the cloth or by exposure to chemical agents, e.g., acids); and then analyzing the desorbed analytes by suitable means (e.g., - 15 PCR for bacterial cells; or high performance ligand chromatography or capillary electrophoresis for chemical analytes, e.g., antibiotics).
The present invention in one aspect thereof thus provides a macroporous, hydrophobic, synthetic polymer cloth, e.g., a polyester cloth mini-column device, for the affinity concentration of bacterial cells and their subsequent detection by the polymerase chain reaction.
The present invention in another aspect also provides a macroporous, hydrophobic, synthetic polymer cloth, e.g., polyester cloth mini-column device, for the affinity con-centration of antibiotics (e.g., chloramphenicol) and the subsequent detection by analytical chemistry~ techniques (e.g., capillary electrophoresis).

213~054 -In more general terms, the present invention in an general aspect thereof, provides a simple and inexpensive affinity concentration device for use in the analysis of target cells and organic chemical analytes in dilute suspensions by microbial assay and analytical chemistry techniques. In one embodiment thereof, the device consists of a one ml disposable pipet tip packed with small segments of a macroporous hydrophobic polyester cloth coated with a suitable target cell-specific capturing agent (e.g., lectin, antibody or other suitable ligand). Target cells (e.g., bacteria, virus particles) are captured on the cloth surface by passage of the sample suspension (e.g., enrichment broth or body fluids) through such cloth-packed mini-column. Captured cells are then lysed on the cloth by briefly heating in a detergent solution, and the lysate is subjected to PCR analysis. Because of the macroporosity of the cloth, clogging of the mini-columns by colloidal and particulate samples is minimized, and it is possible to concentrate relatively large volumes (e.g., several milli-liters) of sample using gravity to draw the liquid through.
In another embodiment thereof, the device of an aspectof this invention, consists of a one ml disposable pipet tip packed with a suitable organic chemical analyte-speci-fic antibody (e.g., anti-chloramphenicol). The analytes (e.g., antibiotic residues) are captured on the cloth sur-face by passage of the sample suspension (e.g., milk or other liquid food sample) through such cloth-p,acked mini-column. The captured analyte is then eluted by exposure to _ 17 heat or a mild acid, and the eluate is subjected to analy-sis using standard analytical chemistry techniques, (e.g., capillary electrophoresis).
The device of yet another aspect of the present inven-tion utilizes a macroporous, hydrophobic, synthetic polymercloth, e.g., a polyester cloth, as the solid phase for the immobilization of analyte-specific antibodies (e.g., anti-bacterial or anti-chloramphenicol antibody or other suit-able affinity ligand), which can be packed into a column (e.g., disposable pipette tip) to form an inexpensive immunoaffinity concentration column.
While it is believed that the terms "macroporous" and "hydrophobic" are well known to those skilled in the art, the following definitions are hereby provided.
The term "macroporous" as applied to cloths when used herein is intended to mean textiles composed of hydrophobic synthetic polymeric fibres, which are either woven or non-woven into a physically structurally stable cloth, (e.g., polyester cloth), of more than 200 ~m thickness, such that the pores (i.e., spaces between the fibres) exceed 20 ~m in diameter.
The term "hydrophobic" as applied to cloths when used herein is intended to mean that the cloths repel water, the degree of repelling being dependent on the pore size and the inherent polymeric properties.
Because of its macroporosity, hydrophobic, synthetic polymer cloth, e.g., a polyester or nylon clo~th exhibits better flow characteristics for liquid samples than micro-porous membranes or beads, thus allowing efficient contact with large volumes of sample by filtration.
While it is not desired to be limited in any way, one suitable macroporous polyester cloth has the following phy-sical properties:
The cloth has a Frazier Air Permeability in CFM/ft2 at 0.5"H20 of 215 for a cloth of thickness 40 mils, the cloth thereby having such porosity that it can accommodate a large volume of liquid per surface area thereof, that it has a large surface are for binding to the affinity ligand and that it has minimum flow resistance.
The following examples show the concentration and washing of several different bacterial pathogens as an example of the application of a specific macroporous hydro-phobic synthetic polymer cloth, e.g., a polyester cloth coated with an affinity ligand, te.g., con A or anti-bodies), in the detection of bacterial cells by PCR.
The following examples also show the concentration and clean-up of the antibiotic chloramphenicol (tagged with a fluorescent marker) as an example of the application of antibody-coated polyester cloth in the detection of chlor-amphenicol by capillary electrophoresis.
Before describing the Examples the following descrip-tion of materials and methods are provided.
Bacteria used in these Examples include two Listeria monocytogenes reference strains (ATCC 15313 and ATCC 43256) and 3 L. monocytoqenes isolates from cheese (~2 isolates) and environmental (1 isolate) samples collected by Canadian Government inspection staff and submitted for routine microbiological analysis by Laboratory Services Division, Agriculture Canada. Unless otherwise stated, the Examples were routinely carried out using one of the cheese L. mono-cytoqenes isolates. Other Listeria sPP. examined includedtwo strains of L. innocua, two strains of L. ivanovii, and one strain each of L. seeliqeri, L. welshimeri, L. graYi and L. murrayi. All these bacteria were routinely grown by inoculating single colones into Trypticase Soy Broth (TSB) (BHD) Inc., No. 5459) and shaking for 24 h at 30C. Viable counts were obtained by plating serial dilutions of the broth cultures on Brain Heart Infusion (BHI) (Difco Labor-atories, No. 0418-01-5) agar. Other media P~m;ned were Listeria Enrichment Broth (LEB) (Difco Laboratories, No.
0223-17-2), BHI broth, and Fraser broth (FB) Oxoid, No.
CM895) with Fraser Selective Supplement (Oxoid, No.
SR156E).
A S. tYphimurium strain (ATCC 14028) was grown by shaking in Buffered Peptone Water (BPW) at 37C for 16-20 h. Viable counts were determined by plating serial dilu-tions of the suspension on Nutrient Agar plates and incu-bating as above.
A virulent Y. enterocolitica human isolate (serotype 0:3) was a gift from G. Kapperud, Turku University, Finland. It was grown in Trypticase Soy broth (TSB) at 25C for 16-20 h. Viable counts of the organism were determined by plating serial dilutions of a suspension on Nutrient Agar and incubating as above.

-The following were obtained from Sigma Chemical Co.:
chloramphenicol-fluorescein conjugate, rabbit anti-chlor-amphenicol antiserum, and concanavalin A (con A). The antiserum was supplied as a lyophilized preparation con-5 taining 0.1 mg of rabbit serum proteins as stabilizers and an unspecified quantity of anti-chloramphenicol antibodies, and was rehydrated in 1 ml of 0.01 M phosphate-buffered (pH
7.2)/0.85% NaCl (PBS). The chloramphenicol fluorescein conjugate was supplied as an 18.6 picomol/ml stock in 0.1 M phosphate buffer (pH 7.4) containing 0.2 mg rabbit IgG as a stabilizer. These solutions were stored at 4C. Anti-Salmonella serogroup B antibodies were purchased from Difco. Anti-Y. enterocolitica serotype 0:3 antiserum was a kind gift from G. Wauters, Université Catholique de Lauvain, Belgium. All other chemicals used were of the analytical reagent grade.
(i) CapillarY ElectroPhoresis Equipment A P/ACE System 2200 equipped with an on-line laser-induced fluorescence detector (air-cooled Ar ion laser, excitation: 488 nm; emission; 560 nm, laser module 488) (Beckman Instruments, Inc.). Post-run data analysis was performed on System Gold software (Beckman Instruments, Inc.). Separations were carried out in a 57 cm length (50 cm to detector window) X 75 ~m i.d. fused-silica capillary (Beckman Instruments, Inc., no. 338454).
(ii) Preparation of Bacterial Lysates for the PCR
Unless otherwise stated, bacteria in br~th cultures were lysed as follows: Briefly, 100 ~l of broth culture 1 0~-~

were mixed with an equal volume of 2% (w/v) TRITONTM X-100 and heated at 100C for 10 min. The samples were then cooled to room temperature and used immediately in the polymerase chain reaction (PCR).
(iii) PreParation of Concanavalin A-Coated PolYester Cloth (Con A-Cloth) Non-woven polyester cloth (SONTARA 81O0TM, a trademark of DuPont) was cut into 6 mm squares and washed by briefly rinsing with 95% ethanol followed by distilled H2O on a filter under vacuum. The cloth squares were blotted and then transferred to a 0.1 mg/ml solution of concanavalin A
(con A) (Sigma Chemical Co., No. L-7647) in 0.01 M phos-phate-buffered/0. 85% NaCl (pH 7.2) (PBS) containing 0.05%
glutaraldehyde (ca. 30 cloth squares per ml), and incubated for 16-20 h at room temperature. The cloth squares were then washed with PBS on a filter under vacuum, and the unreacted glutaraldehyde sites were blocked by incubating the cloths in 0.1 M glycine in PBS for 1 h at room tem-perature. The con A-coated cloth squares (con A-cloth) were then washed as above, and stored in PBS at 4C until use. The con A-cloth was stable for at least two months.
(iv) PreParation of Anti-Chloramphenicol AntibodY-Coated PolYester Cloth and Mini-Columns Polyester cloth (SONTARA 81O0TM) was cut into 5 mm squares and thoroughly washed with PBS on a filter under vacuum. The cloth squares were then incubated with the antichloramphenicol antiserum diluted 1:10 in PBS (1 ml of antibody solution per 20 cloth squares) for 20 h at 37C.

The antibody-coated polyester cloth (antibody-cloth) was then thoroughly washed with PBS as before, and then stored in PBS at 4C until use.
Antibody-cloth-packed mini-columns were prepared by inserting 5 antibody-cloth squares in a disposable 1 ml pipette tip (VERI-TIPSTM, no. B10201VT)_and packing firmly into the tip using a narrow rod. These antibody-cloth-- packed mini-columns were used immediately for the immuno-affinity concentration procedure.
(v) PreParation of Anti-Salmonella Antibody-Coated Polyester Cloth (Antibody-Cloth) Non-woven polyester cloth (SONTARA 8100TM) was cut into 6 mm squares, which were incubated overnight in a solution of 1:40 diluted anti-serogroup B antiserum (Difco Labora-tories) in PBS. Cloths were then washed with PBST and stored at 4C moistened with PBS.
(vi) PreParation of Anti-Yersinia Enterocolitica Antibody-Coated Polyester Cloth tAntibodY-Cloth) Non-woven polyester cloth (SONTARA 8100~) was cut into 6 mm squares, which were incubated overnight in a solution of 1:10 diluted rabbit antiserum to serotype 0:3 (gift from G. Wauters, Université Catholique de Louvain, Belgium) in PBS. Cloths were then washed with PBST and stored at 4C
moistened with PBS.
(vii) Polymerase Chain Reaction Primers for the L. monocytoqenes PCR were selected from the published nucleotide sequence of the L. monocY-toqenes hly A gene. For the amplification of a 730 base-pair (bp) fragment spanning nucleotides 680 to 1411, a 21-oligomer forward primer 5'CATTAGTGGAAAGATGGAATG-3' and a 20-mer reverse primer 5'-GTATCCTCCAGAGTGATCGA-3' were used.
Oligonucleotides were synthesized on a DNA synthesizer (Applied Biosystems, Model 391, PCR-Mate-EP), using phos-phoramidite chemistry (Applied Biosystems) according to the manufacturer's instructions. For the PCR reaction, 10 ~l of bacterial lysate were added to 89.5 ~l of PCR mixture containing 0.22 mM of each dNTP, 1.1 ~M each of the forward and reverse primers, 2.2 mM MgCl2, 55 mM KCl, 11 mM Tris-HCl (pH 8.3) and 0.11% (w/v) TRITONTM X-100. The mixtures were then overlaid with mineral oil, placed in a thermal cycler (Perkin-Elmer Cetus, Model TC 480) and held at 80C for 10 min before adding 0.5 ~l of Taq DNA polymerase (2 units) (Promega, No. 1861). The reaction mixture was then sub-jected to d30 cycles of denaturation at 94C for 1 min, primer annealing at 55C for 1 min, and primer extension at 72C for 2 min. An additional 2 min was given for the com-pletion of primer extension after the last cycle. Ampli-cons were then analyzed by electrophoresis of 10 ~l of PCRproduct in a 1.2% agarose gel at 100 V for about 1.5 h, followed by staining for 20 min in 10 ~g/ml ethidium bromide solution. DNA on the gels was visualized by fluor-escence under ultraviolet light and photographed on Polaroid 667 film. The size of the amplicon was determined by including a sample of 123 bp ladder DNA molecular weight marker (Gibco BRL, No. 5613SA) in each gel.

Primers for the Salmonella PCR were selected from the published nucleotide sequence of the Salmonella invA gene.
Conditions for the PCR were precisely as described above with the exception that the PCR mixture was used 2 x con-centrated.
Primers for the Y. enterocolitica PCR were selected from the published nucleotide sequence of the YoPA gene.
For the generation of a 238 bp amplicon starting at posi-tion 1862 and terminating at position 2100, the following pair of primers were used:
5'-CAGTATTGACCAAAACCAGGC-3' (forward primer); and 5'-TGTCGAGGTTACAAGTC-3' (reverse primer).
The conditions for the PCR were precisely as described above.
(viii) Affinity Concentration of L. monocytogenes on Con A-Cloth and PCR Detection Affinity concentration mini-columns were prepared by inserting two con A-cloth squares in a 1 ml disposable pipette tip (VERI-TIPS~, No. B10201VT). The con A-cloth squares were packed in the pipette tip using a narrow rod to firmly push the squares into the tip. L. monocytogenes cells were concentrated on the con A-cloths by passing 1-4 ml samples of enrichment broth through the mini-column by gravity flow. In this manner, 1 ml of sample could be passed through the mini-column within about 30 min, at room temperature. Upon passage of the sample, the con A-cloths were washed 5 times with 1.5 ml of 0.01 M Tr~s-HCl/0.85~
NaCl(pH 7.2) by applying a vacuum to the tip of the mini-. / 2~31054 _ 25 column. Vacuum was applied by inserting the tips of the mini-columns in the female luer connectors of a vacuum manifold (Lida, Inc., No. 2100-00) and applying -50 Kilo-Pascals of vacuum.
For PCR detection of the concentrated cells, the con A-cloth squares were removed from the mini-column using sterile forceps and placed in a 1.5 ml microfuge tube.
Fifty ~l of PCR buffer (devoid of primers and dNTPs) were pipetted onto the cloths followed by 50 ~l of 2% (w/v) TRITON~ X-100. The microfuge tube was then placed in a block heater and heated at 100C 10 min. Ten ~l of the resulting lysate were then subjected to PCR and the product was analyzed by agarose gel electrophoresis.
(ix) Immunoaffinity Concentration of Salmonella and Y.enterocolitica Cells on AntibodY-Cloth and PCR Detection Mini-columns were made by packing two bacteria-speci-fic antibody-cloth squares in a 1 ml pipet tip. The immunoaffinity concentration procedure was essentially the same as described in the previous example, with the following exception: after washing the captured cells, the cloths were removed from the mini-columns, placed in 100 ~l of PCR buffer containing 0.1% Triton X-100 and heated at 100C for 10 min. 50 ~l of lysate was then removed and added to a tube containing 50 ~l of 2 X concentrated PCR
mixture, then subjected to the PCR. Product was analyzed by agarose gel electrophoresis as before.

(x) Immunoaffinity Concentration of Chloramphenicol Conjuqate on Antibody-Coated Polyester Cloth-Packed Mini-Columns Immunoaffinity concentration of the chloramphenicol-fluorescein conjugate (CF) was attempted from either buffer solution or bovine milk (used as a model food sample) as follows: the mini-column was washed with several ml of 0.01 M Tris-buffered (pH 7.5)/0.85% NaCl (TBS) by gravity flow of the buffer through the column. Two ml of sample con-taining CF (buffer or milk spiked with CF) was then allowedto pass through the column by gravity flow at a rate of 1 ml/15 min, at room temperature. The column was then inserted in the port of a vacuum manifold (Lida, Inc., no.
2100-00) and washed by passage of 3 ml of TBS followed by 1 ml of 0.1 M borate buffer using -50 kiloPascals of vacuum. To recover the immunospecifically bound CF, the antibody-cloth packing was then removed from the mini-column and placed in a tube containing 50 ~l of 0.1 M
borate buffer, which was then heated at 100C for 5 min.
This heat-elution of the analyte works according to the following principle: antibodies on the cloth with bound analyte are exposed to the high temperature, which dena-tures the antibody, thus altering its conformation and causing dissociation of the antibody-analyte complex. The resulting liquid eluate was removed to a separate tube and then subjected to capillary electrophoresis (or any other suitable analytical chemistry technique, e.g.,~HPLC).

. 21310S`~
_ 27 (xi) Capillary Electrophoresis Analysis Eluate recovered from the antibody-cloth after immuno-affinity concentration was pipetted into a microvial, which was then placed in the P/ACE 2200 System and subjected to capillary electrophoresis. Prior to each sample injection, the capillary was treated with successive 2 min high pres-sure rinses of 0.1 M NaOH and 0.4 M sodium borate (pH
7.5)(run buffer). Samples were introduced into the capil-lary by pressure injection for 20 sec, then separated in the run buffer-filled capillary using an applied voltage of 20kV (with the anode located at the inlet end of the capil-lary). The temperature of the capillary was maintained at 25C throughout the runs.
In the accompanying drawings, Fig. 1 is a drawing of photographs of PCR assays showing the effects of enrichment broths on the PCR
detection of L. monocYloqenes;
Fig. 2 is a drawing of a photograph of PCR assays showing the effect of other Listeria sp in the affinity concentration of L. monocytoqenes cell;
Fig. 3 is a bar graph showing the comparative sen-sitivity of PCR detection of L. monocytogenes after the affinity concentration in con A-cloth;
Fig. 4 is a drawing of a photograph of PCR assays showing the PCR assay of S. tYphimurium cells using concentration on polyester cloth prior to performing the PCR;

. ~`~0~;4 Fig. 5 is a drawing of a photograph of PCR assay of Y.
enterocolitica cells using affinity concentration on anti-body-coated polyester cloth prior to performing the PCR;
Fig. 6 is a schematic flow chart showing the affinity concentration of cells on polyester cloth;
Fig. 7 is an electropherogram of a fluorescein-chlor-amphenicol test solution subjected to capillary electro-phoresis without prior concentration; and Fig. 8 is an electropherogram of a fluorescein-chlor-amphenicol test solution which was affinity-concentrated on an anti-chloramphenicol antibody-coated cloth-packed mini-column.
In carrying out the assay shown in Fig. 1, L.
monocytogenes cells were suspended at 5 X 107 cells/ml in various enrichment broths. A: samples of the suspensions were directly subjected to the PCR as described in Methods.
B: 1 ml samples of the suspensions were passed through Con A-cloth mini-columns and then subjected to the PCR as des-cribed in Methods. Lane m: 123 bp ladder DNA molecular weight marker; Lane 1: TSB; Lane 2: BHI; Lane 3: LEB; Lane 4: FB; Lane 5: PCR buffer.
In carrying out the assay shown in Fig. 2 L. monocy-togenes cells were suspended at 105 cells/ml in TSB contain-ing various concentrations of L. innocua cells. One ml samples were then passed through con A-cloth mini-columns and subjected to the PCR as described in Methods. Lane m:
123 bp ladder DNA molecular weight marker; Lane 1: 5 X 1o8 L. innocua cells/ml; Lane 2: 5 X 107 L. innocua cells/ml;

~31~S4 _ 29 Lane 3: 5 X 106 L. innocua cells/ml; Lane 4: 0 L. innocua cells/ml.
In carrying out comparisons shown in the bar graph of Fig. 3, L. monocytoqenes cells were diluted in TSB to give various cell concentrations, and 1-4 ml samples of each dilution were passed through con A-cloth mini-columns at the rate of 1 ml/30 min and subjected to the PCR. The dilutions were also directly assayed by the PCR without concentration. Each experiment was carried out in tripli-cate. Results are reported as mean detectability limit +standard deviation (n=3).
In carrying out the assay shown in Fig. 4 suspensions of S. tyPhimurium cells in BPW (-104 cells/ml, Lanes 1, 3 and 5; 103 cells/ml, Lanes 2, 4 and 6) were concentrated (1 ml) through either uncoated polyester cloth (Lanes 1 and 2) or anti-Salmonella antibody-coated polyester cloth (Lanes 3 and 4). The cloth columns were then washed with Tris-buffered Saline and the captured cells were lysed by briefly heating the cloths in PCR buffer, followed by PCR
amplification of the lysate. PCR product was analyzed by agarose gel electrophoresis. For comparison, 5 ~l of the unconcentrated cell suspensions were also subjected to the PCR (Lanes 5 and 6). Lane m is a 123 bp DNA ladder marker.
In carrying out the assay shown in Fig. 5, suspensions of Y. enterocolitica cells in TSB (104 cells/ml, Lanes 1 and 4; 103 cells/ml, Lanes 2 and 5; 102 cells/ml, Lanes 3 and 6) were concentrated (1 ml) through anti-Yersinia antibody-coated polyester cloth (Lanes 4, 5 and 6). The cloth _ 30 columns were then washed with Tris-buffered saline and the captured cells were lysed by briefly heating the cloths in PCR buffer, followed by PCR amplification of the lysate.
PCR product was analyzed by agarose gel electrophoresis.
For comparison, 5 ~1 of the unconcentrated cell suspensions were also subjected to the PCR (Lanes 1, 2 and 3). Lane m is a 123 bp DNA ladder marker.
Fig. 6 shows in schematic form how the affinity concentration of cells on polyester cloth may be carried out.
Fig. 7 shows the electropherogram obtained when a 2 pmol/ml solution of chloroamphenicol-fluorescein conjugate (CF) was subjected to capillary electrophoresis without prior concentration. Four major peaks, A, B, C and D are evident. Peaks A, B and C correspond to structurally dif-ferent forms of the chloramphenicol-fluorescein conjugate (the different forms probably resulting from the attachment of fluorescein at different sites on the chloramphenicol molecule during the conjugation reaction). A separate experiment where the CF preparation was spiked with free fluorescein and then subjected to capillary electrophoresis shows that peak D corresponds to unconjugated fluorescein, probably present as an artifact of the conjugation reac-tion.
Fig 8 shows the electropherogram obtained when a 0.37 pmol/ml solution of CF in TBS buffer was affinity-concen-trated on an anti-chloramphenicol antibody-coated cloth-packed mini-column, then heat-eluted from the cloth and 2~31054 _ 31 analyzed by capillary electrophoresis. It can be seen that the relative proportions of peaks A, B and C have changed after affinity-concentration (compare to Fig. 9), with peak C emerging from the mini-column as the largest entity.
This suggests that the anti-chloramphenicol antibody used in this procedure had a higher affinity for the chloram-phenicol-fluorescein form represented b-y peak C, which was enriched over the forms corresponding to peaks A and B. It is also of interest to note the complete absence of peak D
(representing unconjugated fluorescein devoid of a chlor-amphenicol moiety) after affinity-concentration, proving the specificity of the procedure for chloramphenicol.
EXAMPLE I - Cell CaPture and Washinq on Con A-Cloth The recovery of viable bacteria from foods and other samples for PCR detection often requires their pre-growth in an enrichment broth. Certain enrichment broths con-taining selective (FB) and non-selective (TSB, BHI and LEB) components will support the growth of L. monocytoqenes and can be used to cultivate this organism from a variety of samples. These enrichment broths constitute complex chemi-cal environments that may contain agents (e.g., salts, organic acids, etc.) that are inhibitory to the PCR. The experiments summarized below describe the application of con A-cloth (an inexpensive affinity adsorbent) in the capture and washing of _. monocYtoqenes cells suspended in various enrichment broths for their PCR detection.

2~31054 _ 32 Varying degrees of inhibition of the PCR occurred when the cells were suspended in FB, BHI and LEB and directly introduced into the PCR mixture (Fig. 1, lanes 2, 3 and 4).
FB produced the most severe inhibition, completely abro-gating the formation of PCR product` (730 bp amplicon)followed by BHI and then LEB. Suspending the cells in TSB
did not appear to affect the PCR (Fig. 1, lane 1), since the intensity and quality of the 730 bp amplicon band were similar to that produced when the cells were suspended in PCR buffer rather than broth (Fig. 1, lane 5). A dramatic improvement in the quantity and quality of the PCR product resulted when 1 ml samples of the same cell suspensions were concentrated and washed on mini-columns packed with con A-cloth (Fig. 1, lanes 1-5). The inhibitory effects of FB, BHI and LEB were eliminated after passage of the cells through con A-cloth (Fig. 1, lanes 2, 3 and 4). While it is not desired to be limited by theory this improvement appears to be due to a combination of the effects of con-centrating the cells from a larger volume of sample as well as the ability to wash the captured cells of the inhibitory enrichment broth components.
The dependence of cell binding on the presence of con A immobilized on the cloth was confirmed by passing 1 ml of L. monocYtogenes cell suspensions (5 x 105 cells/ml in TSB;
determined by viable counts) through mini-columns packed with uncoated polyester cloth instead of the con A-cloth, followed by PCR analysis as before. No det,ectable PCR
product was generated after concentration of the cells on _ 33 the uncoated cloth. The 730 bp amplicon was produced when the same cell suspension was concentrated on con A-cloth.
Thus, the immobilization of con A on the cloth surface was necessary to give a detectable PCR product at the cell con-centration tested.
EXAMPLE II - ComParative Detectability of the PCR Usinq Affinity Concentration of the Samples The preceding results indicate that con A-cloth was able to concentrate L. monocytogenes cells present in the broth samples applied to the mini-columns. The affinity concentration of bacterial cells present in large volumes would benefit the PCR detection system by enabling the sampling of a larger fraction of the total cells available in the enrichment culture, possibly resulting in an improvement in the effective detectability of the PCR test.
This possibility was tested by passing various volumes of L. monocYtoqenes TSB suspension containing different cell densities through mini-columns packed with con A-cloth, followed by PCR detection of the captured cells. For the purpose of comparison, the different cell suspensions were also directly subjected to the PCR without prior concen-tration. Table 1 (below) and Fig. 3 show that, when L.
monocYtoqenes was affinity concentrated on the con A-cloth from 1 ml of TSB suspension, the PCR could detect a minimum of about 6 x 103 cells/ml, compared to a minimum of about 2.5-5 x 104 cells/ml when the cells were sampled directly from the enrichment broth without concentration.

;~ 54 rABLE 1.
COMPARATIVE DETECTABILITY OF THE PCR AFI ER CONCENTRATION OF CE~LS
ON CON A-CLOTHn ConcG~lion (mlj Ce~l density (cdls/ml X 10'~ None 1 2 4 50' +,+,~ ~,~,++,+,+ ~,~,~
+ , +,+,+= ~,+,~ +,+,~
12.5 -. -. -' ~ + ++ + + + + ~
6.25 , + + ++ + + + + +
3. 12 -. -, - -. -. - + + -- + + +
-- I . 6 1 --,--, -- , , + +
0.80 --.--. ----.--. -- --.
O , . . . .

~ A broth culture of L. monocy~o~enes cells co..laining 5 X 10' cellsllnl (de~ermined by viable counts) was serially diiutc~l in TSB ~o givc variolls ccll (Icnsi1ies as in~licated. One ~o 4 ml s~mples of eacll cell suspl:slsioll werc passctl Illro~ Il con /~lolll mini-columlls at the ra~e of I
mll30 min, and 10 ~l of Iysale obtained from each column was tllen subjected lo the PCR as d~ c~ in Methods. For colnparison, 10 ~ll of Iysate obt uned ~rom the cell aust~r~ ns prior to ~u~ dlion OII con l~-clotll was also subjecled ~o ~he PCR. Tlle resul~s of 3 replicate ~;"t,~ s for each cell Sua~ iO~l testetl with an~l wi~hout prior concentration are p~ t~L
in tenns of the presence (+) or abscnce (-3 of lhc 730 bp ~mplicon upon analysis of the PCR
product by a~arose ~el cl~ uj~llol~;s.

A minimum of about 2 x 103 cells/ml could be detected by concentrating cells from 4 ml of TSB. Thus, the effec-tive detectability of the PCR (i.e., minimum cell density in the suspensions which produced detectable amplification) increased with increasing volumes of sample applied to the mini-columns.
EXAMPLE III - Specificity of the Combined AffinitY
Concentration-PCR Procedure In order to initially assess the specificity of the combined affinity concentration-PCR procedure, five differ-ent L. monocYtoqenes strains and several other Listeria sp.
(non-monocytogenes) were suspended in TSB at 5 ~ 108 cells/-ml (determined by viable counts), and 1 ml samples were ~ Z13~054 concentrated on con A-cloth and then subjected to the PCR
as before. In addition to the preceding results obtained with the cheese L. monocYtogenes isolate, the 730 bp ampli-con was detected with the environmental and other cheese isolate as well as the 2 ATCC strains, whereas none of the other Listeria sp. tested produced the amplicon.
It was thought that the presence of other bacteria in the enrichment broth might interfere with the capture of L.
monocYtoqenes cells on the con A-cloth, and hence, their subsequent assay by the PCR. Therefore, the effect on the PCR of concentrating L. monocytogenes cells on con A-cloth in the presence of L. innocua cells (a harmless common contaminant of food and environmental samples) was studied.
Figure 2 shows that no appreciable effect on the generation of PCR product occurred when L. innocua cells were present in a 50 to 500-fold excess of the L. monocytogenes cells (Fig. 2, lanes 2 and 3). While it is not desired to be limited by theory, this is probably due to the large sur-face area of the con A-cloth, which could fully accommodate both the competing cells and the target cells. However, a significant reduction in the amount of PCR product gen-erated occurred when the L. innocua cells were present in a 5000-fold excess (Fig. 2, lane 1). Similar results to those shown in Fig. 2 were obtained for the formation of PCR product when the same cell suspensions containing the different ratios of target to competing cells were sub-jected to PCR without prior affinity concentr,ation, sug-gesting that the availability of binding sites on the con A-cloth was not a limiting factor. While it is not desired to be limited by theory it would appear that the presence of non-target cells ~n very large excess of the target cells directly interferes with the PCR, independently of whether or not they are first concentrated on the con A-cloth.
The above examples show the effect when the canavalin ensiformis lectin concanavalin A (con A) was immobilized on macroporous polyester cloth to form an inexpensive high surface area adsorbent (con A-cloth) for the affinity con-centration of bacterial cells in aqueous suspensions.
It is believed that the cloth material used herein is advantageous for PCR analyses since, unlike microporous solid phases of the prior art (e.g., membranes, beads), such cloth permits the easy passage of the test sample by filtration and of wash buffer for the quick and effective washing of the cloth after capture of the cells to remove extraneous sample components which might otherwise inter-fere with the assay, without the use of vacuum apparatus or magnetic separators, etc. The cloth, owing to its macro-porosity, will not clog as readily as microporous solid phases upon passage of viscous, colloidal or particulate samples (e.g., enrichment broths from food samples). The cloth can be washed simply by placing it on an absorbent pad and rinsing with a small volume of wash buffer or by gravity flow of the wash buffer through the mini-column.
Macroporous cloth can accommodate a larger volume of liquid sample per area for more extensive and rapid reactions with 2~31054 the immobilized affinity ligand. Macroporous cloth pro-vides a much more accessible surface for analyte capture than microporous solid phases. Macroporous cloths of hydrophobic fibres provide a much larger surface area, accommodate a larger volume of sample more rapid reaction with analyte, and allow for easier passage of sample with-out clogging and also easier washing. Furthermore, such cloths are readily available and economical.
The above-described macroporous, hydrophobic, synthetic polymer cloth is selected from a broad group of textiles composed of hydrophobic, synthetic, woven or non-woven polymer fibres, e.g., polypropylene, polyester, nylon, and polyethylene. One specific and preferred example of such non-woven polyester cloths is that known by the trade-mark SONTARATM of Dupont. Typical properties of SONTARATM are well known to those skilled in the art.
When the con A-cloth was packed into a 1 ml pipette tip, the resulting mini-column could be used to concentrate large volumes of dilute L. monocytogenes cell suspensions, followed by the polymerase chain reaction (PCR) amplifi-cation of L. monocytoqenes-specific hly A se~uences from lysates of the captured cells. This improved the effective sensitivity of the PCR as compared to the assay of L. mono-cYtogenes in unconcentrated suspensions. It is believed that the simple and inexpensive affinity concentration method of this invention should be applicable to the PCR
detection of other pathogens in enrichment cultures.

213105~

._ The affinity concentration of L. monocYtoqenes cells on the above-described lectin-coated polyester cloth (i.e., con A-cloth) was demonstrated in the previous example.
Although effective, this method relied upon the capture of cells on the basis of the relatively weak interaction of the lectin with cell surface sugar residues. Since specific antibodies are known to display a much higher affinity (and, in the case of polyclonal antibodies, avidity) toward their corresponding antigens, their appli-cation as affinity ligands for the immunoaffinity concen-tration of bacterial cells on cloth-packed mini-columns might prove advantageous. Furthermore, the possibility of producing antibodies specific for virtually any antigen considerably broadens the number of different applications in which the affinity concentration system might be useful, for example, in the concentration of non-lectin agglu-tinable bacteria, viruses and other targets.
In the following example, the applicability of anti-body-coated macroporous, hydrophobic synthetic polymer cloth-packed mini-columns in the immunoaffinity concen-tration of Salmonella tYPhimurium cells was demonstrated.
The PCR amplification of invA gene sequences was studied as a model system.
EXAMPLE IV - Concentration of Salmonella Cells The concentration and PCR analysis of Salmonella on polyester cloth mini-columns was studied by passing suspen-sions of S. typhimurium cells at various densities through mini-columns packed with uncoated or anti-Salmonella anti-21310~4 _ 39 body-coated polyester cloth. Fig. 4 shows that concen-trating the Salmonella suspensions on the cloth mini-columns permitted the detection of the cells at both dilu-tions tested, with stronger signals obtained when the cloth was coated with anti-Salmonella antibodies (anti-serogroup B). Performing the PCR directly on the suspensions (no concentration) gave a faint amplicon band at ca. 104 cells/ml, but no visible band at ca. 103 cells/ml. Thus, prior concentration of the cell suspensions through either uncoated or anti-Salmonella antibody-coated polyester cloth improved the effective sensitivity of the PCR, permitting the detection of a lower cell concentration in the suspen-sions.
This example shows that Salmonella typhimurium cell suspensions were concentrated by passage through anti-serogroup B antibody-coated polyester cloth-packed mini-columns, followed by successful PCR detection of the cap-tured cells. A degree of concentration was also achieved when the cells were passed through uncoated polyester cloth-packed mini-columns, possibly due to hydrophobic interaction occurring between hydrophobic sites present on the bacterial cell surface and the uncoated cloth. The PCR
system used in this example targeted lnvA gene-specific sequences. Concentration of the cells on the antibody-cloth greatly improved the effective sensitivity of thePCR, as compared to performing the PCR on unconcentrated cell suspensions. This high affinity capture system effec-. Z13~054 tively demonstrated the advantages of immunoaffinity con-centration of the target cells for their PCR detection.
In the previous example, the immunoaffinity concen-tration and PCR detection of Salmonella cells was demon-strated. In principle, the availability of antibodies to antigens of other bacteria should enable the design of immunoaffinity systems for their detection. Now, the following example further demonstrates the breadth of possibilities for this novel technique, where an immuno-affinity concentration system for the PCR detection of virulent Yersinia enterocolitica strains was studied. As an antibody source for the immunospecific capture of the cells, a rabbit antiserum prepared against serotype 0:3 Y.
enterocolitica cells was used. The target for the PCR
amplification was the pYV plasmid-borne y~_ gene, a known virulence marker.
EXAMPLE V - Immunoaffinity Concentration of Y. enterocoli-t _ Cells on AntibodY-Cloth and PCR Detection The immunoaffinity concentration and PCR analysis of virulent Y. enterocolitica serotype 0:3 cells was studied.
Suspensions of different cell concentrations were passed through anti-serotype 0:3 antibody-coated polyester cloth mini-columns. The captured cells were lysed and then sub-jected to the PCR. Fig. 5 shows that the PCR gave positive results with suspensions containing 102-103 cells/ml after concentration on the antibody-cloth, whereas the assay required a minimum of 103 cells/ml to give a pos~tive result with the unconcentrated suspensions. Thus, as in the pre-` 21310S4 vious examples, the immunoaffinity concentration of the cells increased the effective sensitivity of the PCR
analysis.
When the cells were concentrated on mini-columns packed with uncoated polyester cloth, no PCR signal could be recovered at the concentrations tested indicating that an antibody-coating on the cloth was necessary for cell capture for this particular organism.
Polyester cloth coated with anti-Yersinia enterocoli-tica serotype 0:3 antibodies (rabbit antiserum) was usedfor the immunoaffinity concentration of cells suspended in enrichment broth. Concentration of 1 ml of cells permitted their detection in the range of 102-103 cells/ml by PCR, as compared to an effective detectability limit of 103 cells/ml for the unconcentrated suspensions. Concentration of the cells on uncoated polyester cloth failed to give a detect-able PCR signal at the cell concentrations tested. Thus, the immunoaffinity concentration procedure gave a small increase in the effective sensitivity of the PCR test, and it is surmised that the use of a better quality antibody (e.g., higher affinity, greater titer and purity) to coat the cloth might greatly improve the gain in sensitivity.
These Examples demonstrated the applicability of macroporous hydrophobic polyester cloth coated with an affinity ligand (e.g., lectin or antibody) in the affinity concentration of bacterial cells for their PCR detection.
This simple and inexpensive method not only increased the effective sensitivity of the PCR by allowing the sampling 213~054 -of cells from a larger fraction of the available sample suspension, but also permitted the washing of the captured cells in order to remove PCR inhibitors arising in some of the enrichment broths or sample matrices.
This invention may also be used in the form of anti-body-coated polyester cloth-packed mini-columns for the detection of lower molecular weight chemical residues in foods and other samples using standard analytical chemistry techniques.
The following Examples relate to the applicability of antibody-coated polyester cloth-packed mini-columns in the affinity-concentration and clean-up of chemical residues present in foods and other samples, where such residues need to be detected by analytical chemistry techniques, e.g., capillary electrophoresis and high-performance liquid chromatography (HPLC).
EXAMPLE VI - ImmunoaffinitY Concentration of Chlorampheni-col on Antibody-Coated PolYester Cloth-Packed Mini-Columns This Example demonstrates the ability of antibody-coated polyester cloth effectively to concentrate low mole-cular weight organic chemical analytes in solution, for subsequent analyses by analytical chemistry techniques.
This Example provided the concentration of a chlorampheni-col-fluorescein conjugate (CF) on anti-chloramphenicol antibody-coated polyester cloth-packed mini-columns, followed by analysis of the concentrated conjugate by -capillary electrophoresis. Fig. 7 shows the electrophero-gram obtained when a 2 pmol/ml solution of CF was subjected to capillary electrophoresis without prior concentration.
Four major peaks, A, B, C and D are evident. Peaks A, B
and C correspond to structurally different forms of the chloramphenicol-fluorescein conjugate (the different forms probably resulting from the attachment of fluorescein at different sites on the chloramphenicol molecule during the conjugation reaction). A separate Example where the CF
preparation was spiked with free fluorescein and then sub-jected to capillary electrophoresis shows that peak D cor-responds to unconjugated fluorescein, probably present as an artifact of the conjugation reaction.
Fig. 8 shows the electropherogram obtained when a 0.37 pmol/ml solution of CF in TBS buffer was affinity-concen-trated on an anti-chloramphenicol antibody-coated cloth-packed mini-column, then heat-eluted from the cloth and analyzed by capillary electrophoresis. It can be seen that the relative proportions of peaks A, B and C have changed after affinity-concentration (compare to Fig. 7), with peak C emerging from the mini-column as the largest entity. It is believed that this suggests that the anti-chloramphe-nicol antibody used in this procedure had a higher affinity for the chloramphenicol-fluorescein form represented by peak C, which was enriched over the forms corresponding to peaks A and B. It is also of interest to note the complete absence of peak D (representing unconjugated rfluorescein devoid of a chloramphenicol moiety) after affinity-concen-44tration, proving the specificity of the procedure for chloramphenicol.
As noted in the above Example VI, the ability of anti-body-coated polyester cloth effectively to concentrate low molecular weight organic chemical analytes in solution, for subsequent analyses by analytical chemistry techniques, was studied using the concentration of a chloramphenicol-fluor-escein conjugate (CF) on anti-chloramphenicol antibody-coated cloth-packed mini-columns followed by analysis of the concentrated CF by capillary electrophoresis.
EXAMPLE VII - ImProvement of the Sensitivity of Detection bY the AffinitY-concentration Procedure In a second set of experiments, the improvement in the sensitivity of detection of the CF by capillary electro-phoresis after affinity-concentration was studied. It was surmised that since this system uses a flow-through of a liquid sample through the antibody-coated cloth-packed mini-columns, it should be possible to observe an increase in the sensitivity of detection of the analyte after passage of a larger volume of sample, as compared to per-forming analysis of the sample directly without prior con-centration. Two ml samples of TBS buffer containing dif-ferent concentrations of CF were passed through the anti-chloramphenicol antibody-coated cloth-packed mini-columns, heat-eluted as before, and then analyzed by capillary elec-trophoresis.

Z~31054 _, The results are shown below in Table 3.
TAB~E 3 Sensitivity of ~et~tion after (~on~ oll on antibody~loth-packed mini~ol~-mn~

' Pça~
CF concr,.~ ;ol1 With conc~ntration Withoutcono~n~r~ti-n (pmol/ml) A B C D A B C D

1.86 7 35 100 0 2 18 14 4 0.37 1 15 69 0 0 5 4 0.18 0 7 30 0 0 3 2 0 ' Migration times for the different peaks: A, 4.Z min; B, 4.5 min; C, S min; D, 5.8 min (Iefer to Fig. A and B).

Table 3 shows the peak areas obtained in an electro-pherogram when the different concentrations were analyzed by capillary electrophoresis with and without prior affi-nity-concentration. For each concentration of CF tested, significantly higher values were obtained for the areas of peaks A, B and C when the samples were concentrated on the antibody-cloth-packed mini-columns prior to analyses, as compared to the peak area values obtained for the same samples without prior concentration. A CF concentration as low as 0.18 pmol/ml gave significant areas for peaks B and C after concentration, whereas a concentration of about 1.86 pmol/ml was required to give roughly the,same values when CF was analyzed without prior concentration. Thus, - ~ Z131054 the gain in sensitivity was approximately ten-fold when the analyte was concentrated on the antibody-cloth-packed mini-columns. It is believed that this is because concentrating CF on the antibody-cloth permitted the recovery and subse-quent analysis of the CF molecules present in a much largervolume of sample (e.g., 2 ml) than is otherwise introduced into the capillary electrophoresis instrument when the unconcentrated sample is injected directly (e.g., a few nanoliters). It is believed that even higher gains in sensitivity could be achieved by passing greater volumes (e.g., 10-100 ml) of sample through the columns.
EXAMPLE VIII - Affinity-Concentration from Milk The preceding experiments show that antibody-coated cloth-packed mini-columns can be used for the affinity con-centration and subsequent analyses of an organic chemicalanalyte suspended in buffer. However, the applicability of this procedure to the detection of analyte in an actual food sample matrix (e.g., milk) needs to be demonstrated.
Therefore, the affinity-concentration and subsequent capillary electrophoresis analysis of CF suspended in milk was studied as a model system. CF was spiked into bovine milk at a concentration of 1.86 pmol/ml, and a 2 ml sample was then subjected to the affinity concentration and capillary electrophoresis analysis as before.

Z13~054 The results are shown below in Table 4 Affinity col~c~ lion of chlG.;~ ff,lo~s~eill conjugate in n~iLtc.

Peak area sample A B C D

Milk 2 20 127 0 TBS buffer 7 35 lOO 0 ' Migration times for the diffe~ent peal~s are as gi:~en in Table 3 Table 4 shows the ~ea~ areas obtained in an electro-pherogram when the CF was concentrated from milk and analyzed by capillary electrophoresis. For the purpose of comparison, the peak areas obtained in a similar experiment in which the CF was suspended in TBS buffer instead of milk are also shown in Table 3. It can be seen from these data that the efficiency of the anti-chloramphenicol antibody-coated cloth-packed mini-columns in recovering the CF from milk is about equivalent to that for recovery of the same in TBS buffer. When CF in milk was directly subjected to the capillary electrophoresis analysis without prior con-centration, no distinct CF peaks were seen on an electro-pherogram, suggesting that the presence of extraneous milk sample components interfered with the resolution of the CF.
Thus, the affinity-concentration procedure of the present invention proved extremely valuable in the analysis of an analyte (e.g., CF) in a complex food matrix (e'g., milk).

.. 2131054 As examples of the applicability of the device of this invention in the PCR detection of bacteria, systems for the affinity concentration of several important food pathogens including Listeria monocYtogenes, Salmonella and Yersinia enterocolitica have been described in examples hereinabove.
Using the process of this invention, it was shown to be possible to increase the effective detectability of the PCR
(i.e., ~;nimum cell concentration detected) after concen-trating one to several ml of the cells. Furthermore, as will be described hereinafter, cells captured on the cloth could be washed with a buffer to remove PCR inhibitors present in some of the enrichment broths examined. This allowed the detection of the cells in chemical environments prohibiting the PCR detection of unwashed cells, thus enhancing the reliability of the PCR. The entire affinity concentration and clean-up process for the analysis of bacteria by PCR is shown schematically in Fig. 6. Provided that a suitable affinity ligand is available, the simple and inexpensive device of the present invention should be broadly applicable to the affinity concentration and PCR
analysis of a wide variety of bacteria, viruses and other targets of clinical and economic interest.
A simple inexpensive affinity concentration device has thus been developed for use in the polymerase chain reac-tion of target cells in dilute suspension. The deviceconsists of a 1 ml disposable pipet tip packed with small segments of a macroporous hydrophobic polyester cloth coated with suitable target cell specific capturing agent.

Z~31054 _ 49 Clogging is minimized and several milliliters of sample can be concentrated. Through this method the sensitivity of the diagnostic assay can be increased. Cells can be washed with buffer to remove PCR inhibitors, which increased the possible sample types to include those in chemical environ-ments which otherwise would have precluded PCR analysis.
The cloth and preparation with suitable ligand is inexpensive. Typically, the lectin concanavalin A or suit-able antibody are used to coat the cloth. The cloth is macroporous and clogging of the mini-column is minimized.
This method has application to a wide variety of target bacteria, viruses and other targets that may be of interest in biomedical, food or environmental laboratories. Because samples are concentrated, the assay sensitivity is lS increased.
The separation of the target cells from enrichment broth components prior to performing the PCR is believed to be the basis of the present invention as a pre-requisite to achieving reliable and sensitive results.
The applicability of macroporous polyester cloth coated with the lectin con A in the affinity concentration of L. monocytogenes cells for their PCR detection has been provided by the present invention. This simple and inex-pensive method not only increased the effective sensitivity of the PCR by allowing the sampling of cells from a larger fraction of the available sample suspension, but also permitted the washing of the captured cells in order to remove PCR inhibitors arising in some of the enrichment 2~31054 -broths. This method is believed to be useful in washing away possible inhibitors originating from food and environ-mental sample matrices. The method of this invention may be suitable for the concentration of cells in enrichment cultures of food and environmental samples, where the effect of possible interfering sugars originating from the sample matrix is minimized by dilution of the sample in enrichment broth.
When the polyester cloth was coated with specific antibodies, it was possible to achieve the immunoaffinity concentration of Salmonella and Yersinia enterocolitica cells, improving the effective sensitivities of the PCR
tests. Since these systems relied upon the interaction of the cells with immobilized antibodies, the presence of interfering sugars in the test sample should not be a problem, as might be the case where a lectin is used as the affinity ligand. Furthermore, since one can readily raise antibodies against any antigen, it is believed to be possi-ble to design immunoaffinity systems for the concentration of virtually any target cell, virus, etc.
The use of cloth-packed mini-columns in the affinity concentration of target cells is believed greatly to improve the diagnostic applications of the PCR technology in analytical, clinical and research laboratories.
According to the above examples, the usefulness of this device in routine analyses using analytical chemistry techniques, has been provided by examples of ~its applic-ability in the affinity-concentration and clean-up of a ;2~31054 _ 51 chloramphenicol-fluorescein conjugate tused as a model low molecular weight analyte) for analysis by capillary elec-trophoresis. The principle of this model system is as follows: a large volume (e.g., several ml) of a dilute solution of chloramphenicol-fluorescein conjugate is passed through an anti-chloramphenicol antibody-coated polyester cloth-packed mini-column, on which tXe chloramphenicol-fluorescein conjugate is retained by virtue of the immuno-specific interaction of the antibody on the cloth with the chloramphenicol moiety on the conjugate. The column is then washed with a buffer to remove unbound extraneous sample components, and the chloramphenicol-fluorescein conjugate retained on the column is then eluted by heat (heat dissociates the antibody-antigen complex). The elu-ate is then injected into the capillary electrophoresisinstrument, and the analyte is detected by an on-line fluorescence detector (built into the instrument) which responds to the fluorescein tag linked to the chloram-phenicol.
Because of the macroporosity of the polyester cloth and its excellent filtration characteristics, clogging by sample solids and particulate matter is virtually elimi-nated, obviating or minimizing the need for sample prepara-tion prior to passage through the column. Thus, this device is advantageous over previous systems in that it is simple to prepare (the antibody is immobilized on the poly-ester cloth surface by hydrophobic interaction and, unlike the prior art which utilizes the microporous beads, etc., . 213~054 does not require chemical activation of the support to achieve immobilization), the device uses inexpensive materials (e.g., polyester cloth and disposable pipette tips), the cloth support is macroporous (minimizing clogging by the sample), and the device is easy to handle and does not require any special equipment.
These examples demonstrate that a macroporous hydro-phobic cloth (e.g., polyester cloth) with an adsorbed specific antibody can be packed into an inexpensive column (e.g., disposable pipette tip) to create an effective affinity-concentration device for the recovery and analyses of specific organic chemical analytes present in complex samples (e.g., foods), such analytes then being recovered by heat-elution from the column so that they may subse-quently be analyzed by powerful analytical chemistry tech-niques (e.g., capillary electrophoresis, HPLC, etc.). It is well known that analytes present in chemically complex matrices, e.g., foods cannot be directly analyzed by most analytical chemistry techniques, especially the chromato-graphic methods, due to interference caused by the presenceof extraneous sample components (e.g., salts, ions, pro-teins, etc.). The device of the present invention is believed to be of great use in the separation of the analyte of interest from extraneous sample components, since passing the sample through the antibody-cloth-packed mini-columns causes the analyte to be selectively retained on the column by virtue of the specific antibody-analyte interaction on the cloth matrix) while the other sample components pass through and are washed off. As an example here, the affinity-concentration and subsequent capillary electrophoresis analysis of a chloramphenicol-fluorescein conjugate in milk has been demonstrated. Provided that antibodies of suitable specificity are available for coating the cloth, virtually any chemical analyte antibio-tics, pesticides, and organic chemical pollutants can be affinity-concentrated for subsequent analyses using this device.
As disclosed in detail above, in one embodiment of this invention, antibody- or lectin-coated polyester cloth is used for the affinity concentration of bacterial cells for their subsequent analyses by the polymerase chain reac-tion. This not only improves the effective sensitivity of lS the polymerase chain reaction technique, but also permits the separation of target cells for inhibitory substances present in the sample matrix which might otherwise inter-fere with the analysis.
In another embodiment, antibody-coated polyester cloth is used for the affinity concentration and clean-up of organic chemical analytes (e.g., antibiotics) present in foods and other samples, where such analytes need to be detected by standard analytical chemistry techniques. The macroporous hydrophobic synthetic polymer woven or non-woven cloth used in the present invention is preferablycharacterized by having a thickness of more than about 200 ~m and having spaces between the fibres exceeding 20 ~m in diameter and a Frazier Air Permeability in CFM/ft2 at 0.5"

_ 54 HLO of about 215 for a cloth thic~ness of about 40 mils, the cloth thereby having such porosity that it can accom-modate a large volume of liquid per surface area thereof, that it has a large surface area for binding to the speci-fic capturing agent (affinity ligand) and that it has a minimum flow resistance.

Claims (22)

1. An affinity concentration device for use in the analysis of target cells or organic chemical analytes in dilute suspension which comprises: a macroporous, hydro-phobic synthetic polymer cloth coated with a suitable, target-cell-specific capturing agent, or a suitable analyte-specific capturing agent, or any other affinity ligand.

2. The affinity concentration device of claim 1 wherein said target-cell-specific capturing agent, or said suitable analyte-specific capturing agent is an antibody.

3. The affinity concentration device of claim 1 wherein said target-cell-specific capturing agent is a lectin.

4. The affinity concentration device of claim 3 wherein said lectin is con A or antibodies.

5. The affinity concentration device of claim 1 coated with a lectin for use in the affinity concentration of cells prior to polymerase chain reaction detection.

6. The affinity concentration device of claim 5 wherein said lectin is con-A or antibodies.

7. The affinity concentration device of claim 1 for use in the affinity concentration and clean-up of organic chemical analytes prior to analyses by analytical chemistry techniques.

8. The affinity concentration of claim 6 wherein said lectin is con-A or antibodies.

9. A mini-column device for the affinity concen-tration of bacterial cells and their subsequent detection by the polymerase chain reaction comprising: a pipette tip or other suitable column loaded with the macroporous hydro-phobic synthetic polymer cloth as claimed in claim 1.

10. A mini-column device for the affinity concen-tration of antibiotics (e.g., chloroamphenicol) and the subsequent detection by analytical chemistry techniques (e.g., capillary electrophoresis), comprising: a pipette loaded with a macroporous hydrophobic synthetic polymer cloth as claimed in claim 1.

11. A method for carrying out a polymerase chain reaction of target cells or organic chemical analyses of samples in dilute solution which includes the preliminary step of separation of the target cells or organic chemical analytes from sample solution using the affinity concentra-tion device of claim 1.

12. An improvement in the application of analytical techniques for the detection of bacterial cells or chemical residues (antibiotics) comprising the following three main steps: (1) capturing bacteria or chemical residue analytes on the affinity ligand-coated cloth as claimed in claim 1 by filtration; (2) desorbing said captured analytes from said cloth; and (3) analyzing said desorbed analytes by suitable means.

13. The method of claim 12 wherein said desorbing step is carried out by lysis of bacterial cells on the cloth.

14. The method of claim 12 wherein said desorbing step is carried out by exposure to chemical or physical agents.

15. The method of claim 14 wherein said chemical or physical agent is an acid or heat.

16. The method of claim 12 wherein said analyzing is carried out by PCR for bacterial cells.

17. The method of claim 12 wherein said analyzing is carried out by high performance ligand chromatography for chemical analytes.

18. The method of claim 17 wherein said chemical analytes are antibiotics.

19. The method of claim 12 wherein said analyzing is carried out by capillary electrophoresis.

20. The method of claim 19 wherein said chemical analytes are antibiotics.

21. A method for affinity concentration for use in the analysis of target cells and organic chemical analytes in dilute suspensions by microbial assay and analytical chemistry techniques, comprising: passing a sample sus-pension (e.g., enrichment broth or body fluids) containing target cells (e.g., bacteria, virus particles) through a mini-column packed with small segments of a macroporous hydrophobic polyester cloth as claimed in claim 1, thereby capturing said target cells on said cloth surface; lysing said captured cells on the cloth by briefly heating in a detergent solution; and subjecting said lysate to PCR
analysis.

22. A method for affinity concentration for use in the analysis of target cells and organic chemical analytes in dilute suspensions by microbial assay and analytical chemistry techniques, comprising: passing a sample sus-pension (e.g., milk or other liquid food sample) through a mini-column packed with small segments of an antibody-coated macroporous hydrophobic polyester cloth as claimed in claim 1, thereby to capture said analytes on said cloth surface; eluting said captured analyte by exposure to heat or a mild acid; and subjecting said analyte to analysis using standard analytical chemistry techniques, (e.g., capillary electrophoresis).