WO1990002202A1 - Reduction of peroxidatic and catalatic interference with assays of peroxidatic activity - Google Patents

Abstract

Described are mild reagents and gentle methods for inactivating background peroxidatic activity in a test sample before analysis by a peroxidase-linked specific binding assay. Reagents are provided which inactivate plant peroxidases, hemoglobin, methemoglobin, metmyoglobin, leucocyte peroxidases, hematin, and iron salts, individually or in combination. Also described are improved methods for detecting the presence of blood or the occurrence of hemolysis in a test sample, which use specific inactivating reagents or unique assay reaction kinetics to distinguish hemoglobin and methemoglobin from other peroxidatic catalysts potentially present in a test sample. The invention provides simple methods for blocking catalase interference with either the background-reduction reactions described above or the assay of peroxidatic activity, and for stopping signal generation in solid-phase assays of the peroxidatic activity of hemoglobin, methemoglobin, or plant peroxidases. The invention includes a method for using the specific and permanent inactivation of plant peroxidases to permit serial probing of a test sample for different analytes in a peroxidase-linked specific binding assay.

Description

Title

REDUCTION OF PEROXIDATIC AND CATALATIC INTERFERENCE WITH ASSAYS OF PEROXIDATIC ACTIVITY

Background of the Invention Field of the Invention

This invention describes novel reagents and methods in the fields of forensic chemistry, analytical chemistry, clinical chemistry, cytochemistry, and histochemistry. More specifically, the invention relates to the reduction of interfering peroxidatic and catalatic activities in test samples being analyzed by peroxidatic assay, and to control of the peroxidatic catalysts generating the analyte-specific signal.

Description of Background and Related Art Peroxidatic oxidation usually occurs according to one or the other of the following reaction schemes:

AH₂ + ROOH -— A + ROH + H_jO;

H₂0 + AH₂ + ROOH — \> AH₂ ⁺² + ROH + 20H~;

in which AH₂ is an electron donor and ROOH is a hydroperoxide. Catalysts of peroxidatic oxidation are said to have "peroxidatic activity". Hydroperoxide dismutation occurs according to the following reaction scheme: 2 ROOH — 0₂ + 2 ROH. Because this reaction is catalyzed by the enzyme, catalase, catalysts of hydroperoxide dismutation are said to have "catalatic activity". When these reactions are catalyzed, the reactants are called "substrates" for the catalyst. The catalysts of both reactions fall into five main classes: certain transition metal ions and their complex ions, hematin compounds, hemoproteins, peroxidase enzymes, and catalase enzymes. Catalatic reactions are really a subset of peroxidatic reactions, in which a hydroperoxide serves as both electron donor and oxidant. A given catalyst in one of the above- mentioned classes is likely to accelerate both catalatic reactions and non-catalatic peroxidatic reactions, being more effective with one reaction than with the other and being more effective with some substrates than with others. In practical terms, the most active and most commonly used hydroperoxide for either kind of reaction is hydrogen peroxide. Urea hydrogen peroxide, a crystalline but highly water soluble form of hydrogen peroxide, is also commonly encountered in clinical chemistry. Once dissolved in water, it is functionally equivalent to hydrogen peroxide. When it is desired to assay for peroxidatic activity, addition of a chromogenic or fluorogenic electron donor and a hydroperoxide to a sample results in the generation of an absorbance or fluorescence signal, often visibly colored to the naked eye. However, simultaneous presence of catalatic activity can interfere with assay of peroxidatic activity because the hydroperoxide needed for detection of the latter may be consumed by the former.

The peroxidase from horseradish (HRP) is widely used in analytical, biological, and clinical chemistry as a signal generator in enzyme-linked assays for three reasons. It accepts a wide range of chromogenic or fluorogenic electron donors; it shows unusually high catalytic activity (leading to high analytical sensitivity); and it is an especially cheap and durable protein, easily coupled to other molecules. In a typical analytical application, the enzyme has been attached to a molecule (called a binding moiety) which binds specifically to the analyte to make a "probe" for specific detection of the analyte; and a method has been devised to separate complexes between the analyte and the probe from uncomplexed probe. When a chromogenic electron donor and a hydroperoxide are added to the purified analyte-probe complexes, the rate of color formation or the total amount of color generated in a set interval reports on the quantity of analyte present. In many applications only the presence or absence of color is noted as a qualitative indication of analyte; if the dye is insoluble, the location of the immobilized dye on a solid phase may also convey valuable information identifying the analyte. Many variations of this basic "specific binding assay" design exist, depending on how the enzyme is linked to the binding moiety, how the analyte-probe complex is separated from excess probe, what chromogenic or fluorogenic electron donor and hydroperoxide are used, and how the analytical signal is recorded. For example, probing often occurs in two steps. A primary binding moiety is bound to the analyte, and is detected by a secondary probe, consisting of a conjugate between enzyme and a second binding moiety which binds specifically to the primary binding moiety. The binding moiety most commonly used is an antibody or antibody fragment for which the analyte is an antigen; such assays are known by several names, including enzyme immunoassays (EIA's), enzyme-linked immunosorbent assays (ELISA's), enzyme-linked im unoblots, and im unoenzyme histochemical or cytochemical staining methods. However, antibody analytes are probed with the help of antigen binding moieties; and nucleic acid analytes may be detected or quantitated by enzyme-linked nucleic acid hybridization, or DNA-probe, methods, in which the binding moiety is a nucleic acid containing a base sequence complementary to the analyte base sequence. There is a second kind of assay which depends on peroxidatic signal generation; here the peroxidatic catalyst is the analyte. For example, in clinical chemistry the presence in stool or urine of hemoglobin, a peroxidatic catalyst, is a valuable sign of diseases which result in internal bleeding, such as cancer or urinary tract infections. Many methods exist which use the color change of a mixture of an hydroperoxide and a chromogenic electron donor to indicate the presence of "occult blood" in stool or urine specimens or of hemolysis-derived hemoglobin in plasma or serum from stored blood. (See, for example, U.S. Patents 4,077,772 and 4,447,542). Peroxidatic assays for hemoglobin also help forensic scientists to determine whether stains contain blood. Often the hemoglobin in clinical and forensic test samples will have been oxidized to a form known as

"methemoglobin". Although it lacks normal oxygen-binding function, methemoglobin is practically indistinguishable from hemoglobin in peroxidatic catalysis. Neuroscientists use HRP injected into neurons to trace neuronal extensions histochemically. Here too, the catalyst is the analyte.

In many assays which depend on peroxidatic signal generation, the analyte is not completely separated from the other substances in the sample matrix before the color- generating reaction is run. The assay may yield false negative results if catalatic activity in the sample destroys the hydroperoxide before the hydroperoxide can oxidize the chromogenic electron donor. A false positive result may be seen if significant endogenous peroxidatic activity remains in the sample. The potential for interference from peroxidatic or catalatic background reactions prevents the use of peroxidatic signal generation in some analytical applications which would benefit from the above-mentioned advantages of HRP as a detection enzyme, and reduces the reliability of some assays in which the analyte is a peroxidatic catalyst.

In test samples containing materials of mineral origin, most likely to be tested for analytes of environmental or forensic interest, the major source of interfering peroxidatic activity probably will be transition metal ions, especially Fe(III). In test samples from plants, the major sources of background activity are plant peroxidases similar to HRP in structure and function, although nitrogen-fixing root nodules may also contain a hemoprotein called "leghemoglobin". In test samples from animals, including human clinical samples, two sources of background peroxidatic activity dominate all others: oxygen transport and storage hemoproteins such as hemoglobin and myoglobin, and peroxidase enzymes (also hemoproteins), which are especially enriched in white blood cells and related cells of the reticuloendothelial system. The neutrophil sub- population of white cells is the major source of leucocyte peroxidatic activity, in the form of an enzyme known as myeloperoxidase (MPO) . Eosinophils are a separate class of white blood cells, much less prevalent than neutrophils in most blood samples, which contain a peroxidase similar to but distinct from MPO. Monocytes and macrophages also contain peroxidases. Red and white blood cells as well as many other tissues possess abundant catalase. Many analytical samples, especially in clinical chemistry, contain intact or lysed red or white blood cells and therefore contain large amounts of catalase and hemoglobin or leucocyte peroxidases. For example, pus, commonly present at sites of infection, is composed primarily of live and dead neutrophils. The presence and function of MPO in leucocytes and in pus are the subject of an extensive literature, reviewed by Clark ((1983) Advan. Inflamm. Res., 5_: 107-146) and Klebanoff ((1975) Seminars in Hematology, 12:117-142). Although MPO is a haloperoxidase, for which halide ions are the normal electron donor, it also effectively oxidizes chromogenic electron donors (Andrews and Krinsky (1982) Anal. Biochem. , 127:346-350). Diagnosis of respiratory or genitourinary tract infections is increasingly based on EIA analysis of swabs of the putatively infected regions, which often contain pus or blood. As the infectious organisms may be present in vanishingly small amounts, it is desirable to use the most sensitive available signal-generating system in a pathogen-specific EIA. HRP, in conjunction with certain chromogenic electron donors such as 3,3*,5,5'- tetramethylbenzidine (TMB), appears to represent the most sensitive commercially available signal generator. However, it cannot be used unless all background peroxidatic activity has been removed or inactivated. Peroxidatic assays for occult blood (reviewed by

Irons and Kirsner (1965) American Journal of Medical Sciences, 249:247-260; see also Ostrow et al. (1973) American Journal of Digestive Diseases, Ijϊ:930-940; Glober and Peskoe (1974) American Journal of Digestive Diseases, 1 :399-403) may yield false positive results as a consequence of diet or metabolic irregularities which leave residues of peroxidatic catalysts other than hemoglobin such as undigested myoglobin (from ingested meat), undigested plant peroxidases, or hematin compounds (from digested myoglobin or plant peroxidases or from hemolytic disease states) in the sample. False negative results, much harder to evaluate, may result from a common approach to reducing false positives, namely, use of a relatively insensitive chromogen. Catalase released along with hemoglobin in hemolysis or in colonic or urinary bleeding also could reduce the sensitivity of tests for hemoglobin peroxidatic activity. Such assays would benefit from procedures which distinguish hemoglobin from other common peroxidatic catalysts and which block catalatic activity, particularly if such measures permitted the use of the most sensitive chromogens. It is generally recognized that benzidine and its modified forms, such as o-tolidine and TMB, are the most sensitive chromogens for detection of occult blood (Irons and Kirsner, supra) . Cytochemists and histochemists are responsible for most prior progress in blocking peroxidatic activity, reviewed by several authors (Farr and Nakane (1981) J. Immun. Meth., 4_7:129-144; Taylor (1978) Arch. Pathol. Lab. Med., 10_,2:113-121; Bourne (1983) Handbook of

Immunoperoxidase Staining Methods, DAKO Corporation, Santa Barbara, p. 13; Tijssen (1985) Practice and Theory of Enzyme Immunoassays, Elsevier, Amsterdam, pp. 484-485). Six major treatments have been developed: (1) incubation in methanol containing 0.2 M ΞC1;

(2) incubation in methanol containing 0.03 M sodium nitroprusside and 0.2 M acetic acid;

(3) incubation serially (10 minutes each) in 0.01 M periodic acid and 0.003 M sodium borohydride; (4) incubation in methanol containing 0.15 M

H 0₂ for 30 minutes;

(5) incubation in 0.8 M H₂0₂ for 10 minutes; and

(6) addition of 5 mM NaN₃ to the peroxidase substrate solution. The first five remedies are performed before incubation of test samples (generally smears or tissue sections) with immunological reagents; the last approach allows visualization of hemoglobin or myoglobin but would block MPO or HRP, which are reversibly inhibited by azide. The first five methods are suboptimal for cytochemical or histochemical analyses and very poorly suited for EIA's, because they are highly denaturing for protein antigens or are chemically destructive to carboTiydrate antigens. As Bourne, supra notes, method (5) can physically disrupt histochemical and cytochemical specimens; endogenous catalatic activity dismutes H₂0₂ to H₂0 and 0₂, causing strong effervescence which distorts tissue and cellular structures. At least some of these procedures take longer than is desirable for a new generation of EIA-based clinical diagnostic tests, in which total assay time is less than 30 minutes (often less than 10 minutes) . The following five bodies of literature relate to the ablation of endogenous peroxidatic and catalatic activity in test samples.

(1) Hemoproteins and hematin compounds tend to self-inactivate while catalyzing peroxidatic or catalatic reactions because reactive intermediates irreversibly inhibit the active site, usually by covalently modifying the tetrapyrrole ring of the hematin prosthetic group. Substrates which are especially effective in causing catalytic self-inactivation are known as "suicide substrates". The uniqueness of the chemical environment surrounding the hematin group of each hemoprotein causes different hemoproteins to have different optimal suicide substrates.

The most detailed study of suicide inactivation of hemoproteins has been done on the cytochromes P-450

(Ortiz de Montellano and Correia (1983) Am. Rev. Pharmacol. Toxicol. 2_3:481-503), a group of enzymes which bear little resemblance to the hemoprotein peroxidatic catalysts of concern .here beyond the use of a hematin prosthetic group and the ability to experience one-electron and two-electron transfers. Nevertheless, the cytochrome P-450 system provides the most detailed suicide models for what might be expected with peroxidatic and catalatic catalysts (including free hematin compounds). Another model is provided by the "coupled oxidation" of free and protein- bound hematin by 0₂ and ascorbate, a reaction shown to cleave tetrapyrrole rings at the methine carbons (O'Carra, pp. 123-151 in Smith (ed.) (1975) Porphyrins and Metallo- porphyrins, Elsevier, Amsterdam). Although it has been known for a long time that H₂0₂ inactivates MPO (Agner (1963) Acta Chem. Scand., 17:332-338), detailed characterization of this reaction as a catalytic suicide is fairly recent (Naskalski (1977) Biochim. Biophys. Acta, 485:291-300) . HRP undergoes time-dependent inactivation during catalysis of oxidation of a variety of chromogens (Porstmann et al. (1981) J. Clin. Chem. Clin. Biochem., 19_:435-439; Gallati and Pracht (1985) J. Clin. Chem. Clin. Biochem., 22k453-460). This reaction has been described as "H₂0₂-inactivation", but experiments by the present inventor indicate that it is truly a case of catalytic suicide, strongly dependent on the identity of the chromogen as well as the H₂0₂ concentration. In addition, HRP is especially vulnerable to self-inactivation in the presence of either of two unusual substrate systems, H₂0₂ plus cyclopropanone hydrate (Wiseman et al. (1982) J. Biol. Chem., 297:6328-6332) and hydroxymethyl hydroperoxide (Marklund (1971) Eur. J. Biochem., £1:348-354). Catalytic suicide by the "pseudoperoxidases", myoglobin and hemoglobin, has not been studied in detail beyond the "coupled oxidation" reaction described above, but the precedents of MPO and HRP suffice to suggest that hydroperoxides might be suicidal for myoglobin and hemoglobin. Indeed, there is one brief report that H₂0₂ alone inactivates hemoglobin with low efficiency (Keilin and Hartree (1935) Proc. Roy. Soc. Lond., 117B, 1-15).

(2) The catalytic effectiveness of catalase with alkyl hydroperoxides is much lower than with hydrogen peroxide and declines as alkyl group bulk increases (Schonbaum and Chance (1976) The Enzymes, 3rd. Ed.

13:363-408). In the event that alkyl hydroperoxides were effective peroxidatic suicide substrates, this phenomenon suggests that they might not require protection by a separate specific catalase inhibitor. Primary and secondary alkyl hydroperoxides are too unstable to be available commercially, so one must look to the tertiary alkyl hydroperoxides as potential practical peroxidatic suicide substrates which would be indifferent to the presence of catalase. Cumene hydroperoxide has been popular in occult blood tests, presumably because of its stability and low water solubility (see, for example, U.S. Patent 4,447,542). However, this application of tertiary alkyl hydroperoxides in indicators of peroxidatic activity differs from the use disclosed in the present invention, as a catalase-insensitive hemoglobin suicide substrate applied prior to addition of a chromogen intended to detect other peroxidatic catalysts.

(3) Hydroxylamine is a potent catalase inhibitor. Several compounds are generally recognized as inhibitors of peroxidase and catalase enzymes, including azide and 3-amino-l,2,4-triazole (Evans and Rechcigl (1967) Biochem. Biophys. Acta, 148:243-250; Schonbaum and Chance, supra) . Hydroxylamine inhibits not only catalase (Blaschko (1935) J. Physiol. (London) 4:52P-53P; Keilin and Hartree (1945) Biochem. J., 39_:148-157) , but also HRP (Keilin and Mann (1937) Proc. Roy. Soc. Lond. B, 122:119-133) and MPO (Lemberg and Legge (1949) Hematin Compounds and Bile Pigments, Interscience, N.Y., p. 432). This reported nonspecificity creates a concern that hydroxylamine might not block catalase interference with peroxidatic assay or the catalytic suicide of peroxidatic catalysts in a practical sense because it also would stop the reactions which it is intended to protect.

(4) Transition metal ions have been heavily studied as catalysts of hydroperoxide dismutation and have been implicated in non-catalatic peroxidatic catalysis as well (reviewed by Brown, et al. (1970) Progress in Inorganic Chemistry, 13:159-204; see also Walling (1975) Accounts of Chemical Research, 8_^:125-131; Sigel, et al. (1979) Inorganic Chemistry, 18:1354-1358). The relevant transition metals are those which easily undergo 1-electron and 2-electron changes in oxidation numbers among their stable ionic forms, and which are relatively common in test samples: Cr, Mn, Fe, Co, Ni, and Cu. Chelators influence 5 the rates of peroxidatic and catalatic catalysis by transition metal ions, sometimes accelerating and sometimes inhibiting the reaction in a manner which depends on the reaction being catalyzed, the specific transition metal ion added, and the specific chelator used. For example, EDTA

10 has been used to stablize H₂0₂ at high pH (Koubek, et al. (1963) J. Amer. Chem. Soc, 9:2263-2268), yet Fe(III)*EDTA complexes are clearly catalatic and peroxidatic catalysts (Walling, supra). Several chelators have been observed to stimulate or inhibit non-catalyzed lipid peroxidation in a

15 concentration-dependent manner, with only one of them, desferrioxamine B (deferoxamine, DFA), being inhibitory at all concentrations tested (Gutteridge, et al. (1979) Biochemistry Journal, 184:469-472). The ability of four chelators, DFA, bathophenanthroline ^'sulfonate (BPS),

20 diethylenetriaminepentaacetic acid (DTPA), and ethylenediaminetetraacetic acid (EDTA), to retard iron- catalyzed auto-oxidation of several catecholamines depended on the catecholamine and the chelator. In this case, EDTA and BPS were capable of accelerating the reaction; only DFA

25 and DTPA were consistently inhibitory, usually to similar degrees. (Heikkila and Cabbat (1981) Biochemical Pharmacology 30:2945-2947)

The ability of transition metal ions to catalyze the reaction of hydroperoxides with chromogenic electron

30 donors such as substituted and unsubstituted benzidine has long been understood, and efforts to inhibit this reaction by adding chelators to peroxidatic activity indicator solutions have been reported with variable success (van Duijn (1955) Receuil des Travaux Chimi ues des Pays-Bas,

35 74_:771-778; Mesulam (1978) J. Histochem. Cytoche ., 26:106- 117; Porstmann et al, supra; European Patent Publication No. 123,902; PCT Publication No. WO86/04610). Although the focus of these efforts has been on improving the shelf life of peroxidatic activity indicator solutions, the fundamental issue is the same as one addressed by the present invention: how to suppress transition-metal-ion catalysis of hydroperoxide oxidation of chromogenic electron donors. However, the two situations have significant practical differences. The trace metal contamination which limits the storage stability of indicator solutions on the time scale of hours to months is orders of magnitude lower than that needed to create false- positive interference in peroxidatic assays occurring on the time scale of minutes. In addition, transition metal ligand exchange reactions often occur on the time scale of minutes. Therefore chelators which storage-protect indicator solutions may not suffice-to inhibit metal ions introduced by the test sample, being stoichiometrically insufficient or too slow in reacting to replace the metal ligands in the test sample before the assay is under way. Finally, the potential variety of metal ligands in test samples is much greater than in carefully formulated indicator solutions. The quantity and identity of transition-metal-ion salts and complexes presented by test samples cannot be predicted or controlled with any certainty.

(5) The potential for catalase interference with peroxidatic assays has been recognized and addressed by inclusion of specific catalase inhibitors in peroxidatic activity indicator solutions as well as by the use in such indicator solutions of chromogens or hydroperoxides which inhibit catalase or which are not readily destroyed by catalase (Saunders, et al. (1984) Peroxidase, Butterworths, London, p. 163). Geoghegan, et al. ((1983) J. Immunol. Methods, 50_:61-68) pointed out that the selectivity of sodium azide in inhibiting catalase but not peroxidase depends on indicator .solution pH. It is usefully selective near neutrality, but not at the acid pH values where HRP is most active with preferred chromogenic electron donors.

So far the discussion has concerned peroxidatic and catalatic interferences introduced by the test sample being analyzed in a peroxidatic assay. The assay procedure itself may generate conditions reducing its sensitivity or utility. In at least three situations, failure to stop the peroxidatic reaction completely after a convenient interval will interfere with the assay.

(a) If one must compare different test samples quantitatively or run many tests simultaneously, it is most convenient to record the peroxidatic signal at a single time point rather than continuously.

(b) It may be necessary to keep a durable visual record of the peroxidatic signal which will not change during subsequent storage, especially if the chromophoric reaction product is immobilized on a solid phase and is evaluated by visual inspection.

(c) Many peroxidatic assay formats compare the signal obtained with the test sample to that of a "negative control" reaction, in which the test sample has been omitted or inactivated in some way; the difference between the two signals indicates the presence or amount of analyte. In an enzyme-linked analyte-specific binding assay, the negative control will have been treated with probe(s) identically to the test sample. The wash procedure removing probe molecules which have not bound to analyte molecules rarely is completely effective, because macromolecular probes tend to bind nonspecifically to the surface of the container or solid phase in or on which the test reactions occur. Therefore the negative control will generate peroxidatic signal, simply doing so more slowly than a test sample which contains a significant amount of analyte. Failure to stop the peroxidatic reaction before the negative control signal rises results in a background which obscures the difference between test sample and negative control signals. Such a background increases the assay detection limit and the frequency of false negative reports.

Timed peroxidatic assays have been stopped by acidification (Porstmann et al., supra; Gallati and Pracht, supra; Geoghegan, et al., supra) , addition of sulfite to destroy unreacted hydrogen peroxide (Porstmann, et al., supra), addition of reversible peroxidase inhibitors such as sodium cyanide (Engvall, et al. (1971) Biochem. Biophys. Acta, 251:427-434), and filtration or decantation, usually followed by a solvent wash, to remove physically the unreacted substrates (Valkirs and Barton (1985) Clinical Chemistry, 3L:1427-1431; Hawkes et al. (1982) Analytical Biochemistry, 119:142-147) . These stopping processes are impractical, unsafe, and/or impossible for various peroxidatic assay formats. Although acidification works well for ELISA's where the peroxidatic signal is evaluated for a chromophore which is soluble in the indicator solution, it dissolves some of the insoluble chromophoric reaction products preferred in formats where the signal is deposited on a solid phase. Relatively unstable or unsafe stopping agents like sodium cyanide and acidified sulfite are hard to formulate for commercial test kits. Filtration may be inefficient in rapid filter-binding immunoassays such as that described by Valkirs and Barton (supra) ; back- diffusion of substrates from the absorbent pad beneath the filter can interfere with efforts to remove substrates from the peroxidatic catalyst permanently. Additional washing steps after the peroxidatic signal-generating reaction may be inconveniently time-consuming or require inconvenient volumes of wash solvents.

A second assay format where the assay reagents, rather than the test sample, introduce interference is presented by enzyme-linked specific binding assays which probe for multiple analytes. Here the analyte-specific signal, rather than background signal from nonspecific probe binding, interferes with the subsequent detection of other analytes. A single test sample is incubated, usually in succession, with several probes specific for different analytes, performing all of the necessary washing and signal-generating steps after each incubation. For such "multi-probing" to work, two technical conditions must be satisfied: the enzyme activities associated with the different analyte-specific binding moieties must not interfere, and the signals generated_, by the different activities must not interfere. For immunological probes, the first condition is customarily addressed by disruption of the analyte-probe interaction, after the signal generated by specific binding of a given probe has been recorded, by any of several denaturants, including chaotropes, strong detergent, or a low-pH buffer (reviewed by Gershoni and Palade (1983) Anal. Biochem., 131:1-15; see also Nakane (1968) J. Histochem. Cytochem., 16:557-560; Geysen et al. (1984) Electrophoresis, 5_:129-131). Several of these authors warned of the need to show that the denaturing conditions do not destroy the analyte structure being probed. An alternative multi-probing approach is to use different enzymes with the different analyte-specific probes (reviewed by Mason et al. (1973) pp. 113-128 in Polak and Van Noorden (eds) Immunocytochemistry, Wright PSG, Bristol), a method which necessarily limits how many analytes can be probed from a single test sample to just two or three. The second technical condition for multi-probing has been met by using chromogenic electron donors which yield differently colored oxidation products (Nakane, supra; Geysen, et al. supra) , an approach which limits the number of probings because only 2 or 3 suitable chromogens exist, and which has been limited in sensitivity by the fact that most of the suitable chromogens are far from the most active ones available for peroxidase. Copending U.S. Patent Application Serial No. 896,577 filed August 20, 1986, discloses an alternative tactic for meeting the second condition in those specific binding assays where the colored oxidation product is immobilized on a solid support. If the chromophore is trapped as a salt or ionic complex, it can be washed away by electrolyte solutions containing an effective concentration of a non- precipitating salt. This approach permits use of the various unsubstituted and substituted benzidine chromogens including TMB, which has the advantages of very high sensitivity and safety. The present invention meets the first condition by suicide inactivation of the probe enzyme after signal generation is complete, an approach which is rapid, chemically mild, irreversible, and unlikely to disrupt the structures of analytes targeted for subsequent probing or to wash analytes from the test device. The present invention discloses that a single chemical phenomenon, the relatively specific and chemically mild inactivation of various peroxidatic and catalatic catalysts by several reagents, most of which act via catalytic suicide of the different catalysts,can be used in a variety of ways to improve a wide range of peroxidatic assays. This inactivation can be used (a) to eliminate background peroxidatic activities introduced by the test sample, (b) to sharpen the discrimination of hemoglobin from other endogenous peroxidatic catalysts in test samples being analyzed for occult blood, (c) to prevent catalase in the test sample from interfering with assay of peroxidatic activity or with specific inactivation of background peroxidatic activity, (d) to stop permanently the signal- generation step of peroxidatic assays, and (e) to improve the use of HRP for signal generation in multi-probed analyte-specific binding assays.

Summary of the Invention

In a first aspect, the invention relates to methods for minimizing the analytical background in peroxidase-linked specific binding assays by inactivating endogenous peroxidatic catalysts in the test sample. The test sample is incubated with an effective amount of an inactivation reagent for an effective interval (minimally on the order of a minute) at an effective temperature, which normally needs not exceed approximately 60C. The inactivation reagent may comprise (a) an organic hydroperoxide (preferably a tertiary alkyl hydroperoxide), or (b) a combination of a non-peroxide catalase inhibitor with hydrogen peroxide or urea hydrogen peroxide, or (c) a combination of 4-chloronaphthol with hydrogen peroxide or urea hydrogen peroxide, or, for greatest breadth of effectiveness, (d) a combination of an organic hydroperoxide or 4-chloronaphthol with a non-peroxide catalase inhibitor and hydrogen peroxide or urea hydrogen peroxide. Optional additives include chloride or bromide Ion, and/or certain chelating agents.

This aspect of the invention also relates to the following compositions useful for carrying out the methods just described: (a) a combination of an organic hydroperoxide

(preferably a tertiary alkyl hydroperoxide) with a test sample; (b) a combination of hydrogen peroxide or urea hydrogen peroxide with a non-peroxide catalase inhibitor;

(c) a combination of hydrogen peroxide or urea hydrogen peroxide with 4-chloronaphthol and a test sample; (d) a combination of an organic hydroperoxide or

4-chloronaphthol, hydrogen peroxide or urea hydrogen peroxide, and a non-peroxide catalase inhibitor; and

(e) combinations of any of the above with chloride or bromide ion and/or certain chelating agents. This aspect of the invention also includes test kits for peroxidase-linked specific binding assays which use the methods or compositions of the invention to eliminate interfering peroxidatic activity in the test sample. In a second aspect, the invention uses the inactivation chemistry of the first aspect to create improved methods for analyzing for the presence of blood or hemolysis in a test sample. Such analyses report on the endogenous peroxidatic activity of hemoglobin or methemoglobin in the test sample; they are vulnerable to false negative results if the test sample is rich in catalase and to false positive results from high concentrations of myoglobin or metmyoglobin, leucocyte peroxidases, plant peroxidases, iron salts, and/or hematin compounds. One method for improving the discrimination of hemoglobin or methemoglobin is (a) to form two separate portions from the test sample, (b) to treat one portion with an effective concentration of an organic hydroperoxide (preferably a tertiary alkyl hydroperoxide) for an effective time interval (minimally about one minute) at an effective temperature (normally not exceeding 60C), leaving the other portion untreated, and (c) to assay both samples for peroxidatic activity. Because organic hydroperoxides inactivate hemoglobin and methemoglobin selectively, loss of peroxidatic activity in the treated portion relative to the untreated portion indicates the presence of hemoglobin or methemoglobin in the test sample. Equally importantly, failure of organic hydroperoxide treatment to reduce peroxidatic activity clearly indicates the absence of hemoglobin and methemoglobin in the test sample.

Another method for improving the discrimination of hemoglobin or methemoglobin is (a) to treat the test sample with a combination of 4-chloronaphthol and hydrogen peroxide or urea hydrogen peroxide, under concentration, time, and temperature conditions which preserve significant hemoglobin and methemoglobin activity but largely inactivate other endogenous peroxidatic catalysts, and (b) to assay the treated sample for peroxidatic activity. Inclusion of a non-peroxide catalase inhibitor and/or chloride ion or bromide ion and/or certain chelating agents in the inactivation incubation may improve the blocking of non-hemoglobin peroxidatic activity.

This second aspect of the invention includes kits for measuring the presence of hemoglobin or methemoglobin as a test for blood or hemolysis, which kits employ at least one of the methods just described for improving specificity for hemoglobin and methemoglobin.

A third aspect of the invention relates to a method of detecting the presence of hemoglobin or methemoglobin in a test sample by analyzing the shape of the kinetic trace obtained when a portion of the test sample is used to catalyze the oxidation of TMB by hydrogen peroxide. The composition of the peroxidatic activity indicator solution is adjusted so that hemoglobin and methemoglobin experience an exponential loss of at least 80% of their initial activity, which loss occurs with a half-time at least three times that observed for metmyoglobin and no more than a third of that seen for HRP or hematin. Preferably this "rate-relaxation half-time" for hemoglobin or methemoglobin, will lie between approximately 30 and approximately 60 seconds. Such behavior is obtained by control of the assay pH between approximately 3 and 6 pH units, control of the identity and concentration of organic cosolvent in the assay solution, and control of the identity and concentration of detergent in the indicator solution. Once an effective assay solution composition has been chosen, the shape of the spectrophotometrically monitored kinetic trace obtained after a portion of test sample has been added to the indicator solution is analyzed for the presence of a transient component of activity loss with the expected rate-relaxation half-time for hemoglobin and methemoglobin. If the transient is not visually evident from a graph of absorbance or rate versus time,, numerical analysis of one of these graphs may be performed in a digital computer, finding the nonlinear least-squares best fit to a function containing at least one exponential decay and one time-invariant term in reaction rate. The presence of hemoglobin or methemoglobin can be inferred if the best- fit value of the amplitude of a transient with a half-time within the expected range for methemoglobin and hemoglobin is significantly larger than the standard deviation for that amplitude, also determined by the least-squares procedure.

A fourth aspect of the invention relates to compositions for peroxidatic activity indicator solutions which reduce the vulnerability of peroxidatic assays to interference by catalase, comprising simply a combination of hydroxylamine or an O-alkyl hydroxylamine with a chromogenic or fluorogenic electron donor and a hydroperoxide, preferably hydrogen peroxide or urea hydrogen peroxide. The hydroxylamine inhibits catalase, which might consume the hydrogen peroxide, without interfering with catalysis of the peroxidatic reaction of the other two components. Optional addition of certain chelating agents to the indicator solution may help to reduce peroxidatic interference by transition metal ions in the test sample.

The specification of "certain chelating agents" in particular aspects of the invention recognizes that not all chelators are equally effective in blocking the peroxidatic activity of transition metal ions, and that chromogenic electron donors have their own specificities with respect to what chelators effectively protect them.

In a fifth aspect, the invention relates to methods for stopping the peroxidatic color-forming reaction in peroxidatic assays, comprising addition to a solid phase, to which is attached the peroxidatic catalyst and/or the chromophoric oxidation product, of an effective amount of a substance which inactivates the catalyst permanently. The color-forming indicator solution has been separated from the solid phase before addition of the inactivator, to minimize competition between the indicator and the inactivator for the catalyst. If the catalyst is hemoglobin or methemoglobin, the inactivator is either an organic hydroperoxide, preferably a tertiary alkyl hydroperoxide, or a combination of hydrogen peroxide or urea hydrogen peroxide with a non-peroxide catalase inhibitor. If the catalyst is a plant peroxidase, the inactivator is 4-chloronaphthol in combination with an effective amount of hydrogen peroxide or urea hydrogen peroxide. In a sixth aspect, the invention relates to a method for analyzing a single test sample for more than one analyte in an analyte-specific binding assay, wherein (a) a plant peroxidase, preferably HRP, is linked directly or indirectly to binding moieties specific for more than one analyte to make analyte-specific probes, (b) a test sample potentially containing a collection of analytes is incubated with a solid phase in such a way that the solid phase captures the analytes (or else the solid phase naturally contains the analytes, as in histochemistry and cytochemistry), (c) the solid phase is incubated serially with the separate analyte-specific probes, (d) after each such incubation, the solid phase is washed to remove excess probe and treated with a peroxidatic activity indicator solution, (e) after such treatment, the peroxidatic signal associated with that probe and therefore with its analyte is recorded, (f) after signal recording, the solid phase is incubated with 4-chloronaphthol and hydrogen peroxide or urea hydrogen peroxide to destroy the peroxidatic activity responsible for that signal, and (g) after such inactivation, the solid phase is washed to remove the inactivating reagent before it is incubated with the next probe. Accompanying this method must be a method for preventing the peroxidatic signal generated by one probe from interfering with that from subsequent probes.

Preferred among options for the accompanying method is the use in the indicator solution of a chromogenic electron donor which forms a soluble colored product or an insoluble colored product which can be dissolved under chemically mild conditions in an aqueous solvent. Then there is no practical limit to the number of analytes which can be probed from a single test sample.

Brief Description of the Drawings

Figure 1 depicts the effects of catalase, with and without hydroxylamine, on peroxidatic assay of hemoglobin.

Figure 2 depicts the absorbance kinetic traces for catalysis by six peroxidatic catalysts of the oxidation of TMB by H₂0₂. Figure 3 depicts rate kinetic traces for catalysis by six peroxidatic catalysts of the oxidation of TMB by H₂0₂.

Figure 4 depicts tertiary-butyl hydroperoxide inactivation of the peroxidatic activity in whole blood as a function of time, temperature, and tertiary-butyl hydroperoxide concentration.

Figure 5 depicts cumene hydroperoxide inactivation of the peroxidatic activity in whole blood as a fuction of time, temperature, and cumene hydroperoxide concentration.

Description of the Preferred Embodiments

Advantages of the Invention

In a first aspect, the present invention is an improvement in several specific ways over what was previously known for inhibiting background peroxidatic or catalatic activity.

(1) The invention provides chemically mild reagents (dilute organic hydroperoxides) for inactivating hemoglobin and methemoglobin in test samples, reagents which are relatively ineffective in inhibiting peroxidatic activity of HRP, hematin compounds, and MPO and which resist catalatic destruction.

(2) The invention provides a chemically mild reagent (dilute hydrogen peroxide in combination with chloride or bromide ion) for inactivating MPO in test samples, a reagent which does not completely inactivate hemoglobin peroxidatic activity.

(3) The invention provides a chemically mild reagent (a combination of dilute 4-chloronaphthol with dilute hydrogen peroxide or urea hydrogen peroxide) for completely inactivating plant peroxidases such as HRP, a reagent which does not completely inactivate hemoglobin peroxidatic activity.

(4) The invention provides chemically mild reagents (dilute hydroxylamine and O-alkyl hydroxylamines) for inhibiting catalase in test samples, reagents which do not interfere with HRP or hemoglobin catalysis or the catalytic suicide of HRP, hemoglobin, MPO, and hematin compounds.

(5) The suppression of catalatic activity accomplished by the invention assures that effervescent 0₂ evolution will not disrupt structures in the test sample or disperse the test sample in a way which might create a biohazard.

(6) As the reagents for inactivating peroxidatic activity are hydroperoxides, the suppression of catalatic activity assures that all of the added peroxide will be available for peroxidatic background reduction. None will be wasted on dismutation.

(7) The reagents used for the invention are commercially available, relatively safe, relatively stable, and mutually compatible, so that economical and practical reagent mixtures can be designed to inactivate or inhibit all likely background peroxidatic and catalatic activity in test samples. (8) The reagents herein are effective over a wide pH range bracketing neutrality, at relatively mild temperatures bracketing room temperature, in the absence of non-aqueous solvents or other protein denaturants. Therefore, there is little concern that they will disrupt the analyte structures, such as antigenic determinants, which are targeted by analyte-specific binding moieties. (9) The reagents herein can be effective on the time scale of several minutes, so that they are compatible with rapid clinical diagnostic tests.

In second and third aspects, the invention offers for the first time the opportunity to show whether the peroxidatic activity in an occult blood test contains a significant contribution from hemoglobin or methemoglobin. The ability to filter out contributions from hematin, plant peroxidases, myeloperoxidase, complexed iron, and metmyoglobin should directly reduce the incidence of false positive results. When these contributions are greatly reduced, peroxidatic assay sensitivity can be increased to lessen the possibility of false negative results when low amounts of hemoglobin are present. Furthermore, the current common and inconvenient practice of dietary restriction (primarily with respect to red meat) for several days prior to testing stool samples for occult blood may be dropped if specificity for the peroxidatic activity of hemoglobin and methemoglobin is increased. In a fourth aspect, the invention protects peroxidatic activity indicator solutions from catalase interference with a catalase inhibitor which does not substantially reduce assay sensitivity to any of the commonly assayed peroxidatic catalysts. In a fifth aspect, the invention provides for the first time specific and chemically mild reagents for stopping timed peroxidatic assays of hemoglobin or methemoglobin and plant peroxidases, reagents which are relatively safe and storage stable and which also are compatible with formats in which the catalyst and/or the chrortlophore created during the peroxidatic reaction are immobilized on a solid support.

In a sixth aspect, the invention establishes the capability of probing test samples in specific binding assays any number of times for different analytes without concern that the signal reporting the presence of one analyte will interfere with that reporting on another analyte by giving false positive results, or concern that the method of preventing such interference will itself interfere by giving false negative results for other analytes.

Definitions:

As used herein, "peroxidatic activity" refers to the ability of certain substances to accelerate the reaction of hydroperoxides with electron donors, especially colorless electron donors which become fluorescent or visibly colored after oxidation by an hydroperoxide.

"Peroxidatic catalyst" refers to a substance possessing peroxidatic activity. Such activity accelerates the oxidation of an electron donor by an oxidizing agent such as a hydroperoxide. It most commonly is detected by using a colorless and non-fluorescent electron donor, the oxidation product of which is fluorescent or visibly colored. Included among these substances are various oxidation states of certain transition metals, such as Ni, Mn, Fe, Co, Cr, Mo, and Cu, and hematin-containing molecules, such as hematin compounds and hemoproteins. The peroxidatic activity of oxidized transition metals depends strongly on what ligands occupy their inner coordination spheres.

As used herein, "hematin compound" refers to a porphyrin ring together with a bound iron atom, regardless of the porphyrin side chains or the oxidation state of the iron. Hemoproteins are exemplified by hemoglobin, myoglobin, the leucocyte peroxidases, and the plant peroxidases. Native hemoglobin and myoglobin contain Fe(II). Once they are released from the reducing environment in the cell, they slowly are oxidized to Fe(III) forms known as "methemoglobin" and "metmyoglobin", respectively.

"Endogenous peroxidatic activity" refers to the action of such peroxidatic catalysts as are present in a test sample and are not added by the analyst.

"Peroxidatic assay" refers to any analytical procedure which relies on peroxidatic activity to create the signal which is detected or measured to infer the presence of analyte and/or the amount of analyte present. The two major classes of peroxidatic assay are (a) those in which a peroxidatic catalyst is the analyte, and (b) peroxidase-linked specific binding assays.

"Chromogenic electron donor", or "chromogen", refers to a compound which undergoes an easily observed change in color upon oxidation by an oxidizing agent such as an hydroperoxide. Chromogenic electron donors are exemplified by four classes of substances: (1) the benzidine compounds, including -benzidine, 3,3'- dimethylbenzidine (o-tolidine) , 3_*3'-dimethoxybenzidine (o-DAD), 3,3'-diaminobenzidine (DAB), 3,3' ,5,5'- tetramethylbenzidine (TMB), and 2,7-diaminofluorene (DAF), (2) the phenylene diamines, including o-phenylenediamine (o-PD), (3) 2,2'-azino-di{3-ethyl—benzthiazoline sulfonate (ABTS), and (4) aminoethyl carbazole (AEC) .

"Fluorogenic electron donor", or "fluorogen", refers to a compound which undergoes an easily observed change in fluorescence upon oxidation by an oxidizing agent such as an hydroperoxide. Usually this compound will show negligible visible florescence until it is oxidized.

"Hydroperoxide" refers to a compound of the general formula, ROOH, wherein the R group is an aryl, alkyl, or acyl group or a hydrogen atom. If R is a hydrogen atom, the hydroperoxide is known as hydrogen peroxide. If the R group is an aryl, alkyl, or acyl group, the compound is an "organic hydroperoxide". If R is the formula, R-_j^R C where R-^, R₂ and R3 is any combination of aryl and alkyl groups other than hydrogen, the hydroperoxide is a tertiary alkyl hydroperoxide. If R^ is a phenyl group and R₂ and R3 are methyl groups, the hydroperoxide is cumene hydroperoxide. If _lf R₂ and R₃ are methyl groups, the hydroperoxide is tertiary butyl hydroperoxide. "Peroxidatic activity indicator solution" comprises a hydroperoxide in combination with an electron donor in liquid solution, which undergoes a measurable chemical or physical change when contacted wiht a peroxidatic catalyst. Preferred electron donors are chromogens or fluorogens. Preferred solvent is water, often in combination with buffer, a chelator, and/or an organic cosolvent.

"Catalytic suicide" refers to the self- inactivation, usually permanent, of certain catalysts during their action. Hematin-based peroxidatic and catalatic catalysts are particularly susceptible to such self-inactivation. The compounds transformed by catalysts are often called "substrates". Substrates often differ in their ability to promote catalytic suicide. Those which are especially effective are called "suicide substrates". "Inactivation" of catalysts, e.g., by suicide substrates, refers to chemical transformations resulting^' in loss of activity which is permanent or difficult to reverse.

An "effective amount" of a suicide substrate is a quantity which reduces the peroxidatic activity of the catalysts for which it is targeted by at least about 80% under specified conditions of incubation time, incubation temperature, solvent composition, and (where appropriate) co-substrate concentration. For example, 4-chloronaphthol targets plant peroxidases, such as HRP, and requires a hydroperoxide, preferably hydrogen peroxide, as a co- substrate. Tertiary alkyl hydroperoxides target hemoglobin and methemoglobin, and require no co-substrate. An "effective temperature" for suicide inactivation of a peroxidatic activity is one where at least about 80% of the original activity is lost upon incubation of the test sample with an inactivating reagent under specified conditions of incubation time, reagent concentration, and solvent composition.

"Inhibition" of catalysts, by substances referred to as "inhibitors", refers to the temporary inactivation of catalysts in the presence of such substances, such that removal of an inhibitor substantially restores catalytic activity.

"Non-peroxide catalase inhibitors" refers to inhibitors of the enzyme, catalase, other than hydroperoxides; examples include cyanide ion, azide ion, 3-amino-l,2,4-triazole, hydroxylamine, and the O-alkyl hydroxylamines. An O-alkyl hydroxylamine has the formula, H₂N-0-R, where R is methyl, ethyl, propyl, butyl, or some other aliphatic organic moiety.

An "effective amount" of a non-peroxide catalase inhibitor is one which blocks the dismutation of hydrogen peroxide by the enzyme, catalase, by at least approximately 80% under the given conditions of temperature and solvent composition.

A "chelating agent", or "chelator", is a compound which binds tightly to metal ions possessing at least two positive charges, by virtue of the fact that it carries at least two negative or electron-rich reactive groups which are directly bonded to the chelated metal ion. An "effective chelating agent" is one which, when added to an aqueous solution of an Fe(III) salt at a concentration no lower than that of Fe(III), lowers the peroxidatic activity of Fe(III) toward a given chromogen and H₂0₂ to a value no greater than approximately 1% of that of unchelated Fe(III). Different chromogens possess different sets of effective chelators. For chromogens which are benzidine compounds, examples include ethylenediaminetetraacetic acid (EDTA), trans-l,2-diaminocyclohexane-N,N,N' ,N'-tetraacetic acid (CDTA), ethylenediamine-N,N'-diacetic acid-N,N*-di- alpha-propionic acid (EDADP-alpha), ethylenediamine-N,N'- diacetic acid-N,N*-di-beta-propionic acid (EDADP-beta), ethylenediamine tetra-beta-propionic acid (EDTP-beta), ethylenediamine tetra-alpha-propionic acid (EDTP-alpha), N,N'-bis(2-hydroxybenzyl)ethylene-diamine diacetic acid (HBED), ethylenediamine-di(o-hydroxyphenylacetic acid) (EDDHA), desferrioxamine B (deferoxa ine, DFA), and pyrophosphate (PP^).

By "analyte-specific binding assay", or "specific binding assay", is meant an analytical procedure for detecting and/or quantitating a particular substance which depends on the binding to that substance of a second substance, known as a "binding moiety", which is not expected to bind to any other substance in the test sample or the test apparatus. The binding moiety is directly or indirectly linked to a signal-generating moiety, which is an enzyme in an "enzyme-linked analyte-specific binding assay". For the purpose of the present invention, the enzyme must be a peroxidase, preferably from a plant, most preferably from horseradish root. If the analyte is an antigen and the binding moiety is an antibody, or vice- versa, the enzyme-linked analyte-specific binding assay is known as an "enzyme immunoassay" (EIA) or an "enzyme-linked immunosorbent assay" (ELISA). If the analyte is an antigen or antibody attached to a cell or a tissue structure in a cytochemical smear or histochemical section and the binding moiety is an antibody or antigen specific for the analyte. the peroxidase linked analyte-specific binding assay is known as an "immunoperoxidase cytochemical or histochemical staining procedure". If the analyte is an antigen or antibody which has been captured on or in a manufactured solid support, the enzyme-generated signal is expected to stick to the solid support at the point of generation, and the binding moiety is an antibody or antigen specific for the analyte, the analyte-specific binding assay is known generally as an "immunoblot", or specifically as a "filter- trapped immunoassay", "immunodot blot", or "Western blot", depending on how capture occurs. If the analyte is a specific sequence of DNA or RNA bases and the binding moiety is a nucleic acid containing a base sequence complementary to at least part of the analyte sequence, the analyte-specific binding assay is known as a "nucleic-acid probe" procedure. If the nucleic acid analyte has been captured on a manufactured solid support and the enzyme- generated signal is expected to stick to the support at the point of generation, the nucleic acid probe procedure is known as a nucleic-acid-probe dot blot or as a Southern or Northern blot, depending on how capture occurs and on whether the analyte is DNA or RNA. If the nucleic acid analyte is part of a cytochemical smear or a histochemical section, the nucleic-acid-probe procedure is known as an Ln situ nucleic acid hybridization assay.

"Probe" in an analyte-specific binding assay refers to a binding moiety specific for the analyte or for another binding moiety specific for the analyte when either binding moiety is used to link the analyte to a signal- generating moiety, such as an enzyme. At least one probe in a specific binding assay is coupled to the signal- generating moiety by a covalent bond or a very strong noncovalent bond, such as that between biotin and avidin or between an antibody and its antigen. "Primary probe" binds specifically to the analyte. "Secondary probe" binds specifically to primary probe.

"Analytical background" in a peroxidatic assay refers to any signal resulting from peroxidatic activity which cannot be linked, directly or indirectly, to the presence of analyte. For example, in a peroxidase-linked specific binding assay, analytical background can be contributed by any endogenous peroxidatic catalyst in the test sample which is not removed or inactivated before the signal forming step in the assay. Nonspecific binding of probe(s) to the test device in a specific binding assay also can result in analytical background.

"Test kit", refers to any combination of equipment, reagents, and/or instructions for the use of equipment and/or reagents to assay for the presence or amount of an analyte. Examples include test kits for performing peroxidase-linked analyte-specific binding assays such as EIA's, immunoperoxidase histochemical or cytochemical staining, and nucleic acid probes analyses, and for assaying for the presence of blood or hemoglobin in forensic or clinical test samples.

"Test sample" refers to any material which contains an analyte of interest and is suspected to contain interfering endogenous peroxidatic catalysts. Examples include animal-derived material such as stool, urine, biopsied or autopsied tissue, pus, sputum, semen, gastric fluid, synovial fluid, cerebrospinal fluid, whole blood, blood serum or plasma, swabs taken from the mouth, eye, upper respiratory tract, lower genitourinary tract, or lower gastrointestinal tract, and cytochemical smears or histochemical sections. The materials can also be derived from plants; and in addition, the test sample can be drawn for environmental analysis as required for the determination of pollutants in soil, water, or air. Test samples for forensic analysis may consist of solid objects (e.g., weapons or clothing) which might contain blood or semen. Also included in the definition are liquid extracts of the preceding examples, wherein the extracting liquid preferably is an aqueous solvent, and solid phases resulting from the contact of samples such as those listed above with a solid in such a way that the analyte of interest would be attached to the solid.

"Kinetic trace" in the peroxidatic assay of reaction of a chromogenic or fluorogenic electron donor refers to a graph of absorbance or fluorescence versus time after mixing of a test sample containing one or more peroxidatic catalysts with the electron donor and a hydroperoxide. "Rate kinetic trace" refers to a graph versus time of the rate of change of absorbance or fluorescence.

"Catalytic activity" of a peroxidatic catalyst is measured by the magnitude of the slope of the tangent to the kinetic trace observed when that catalyst is mixed with a chromogenic or fluorogenic electron donor and a hydroperoxide.

When the catalytic activity of a peroxidatic catalyst undergoes an exponential decay over time after mixing with an electron donor and a hydroperoxide, "rate- relaxation half-time" refers to the interval required for the catalytic activity to decline 50% of the way toward its value at long reaction times, which value may be zero or positive. For mixtures of peroxidatic catalysts, the kinetic traces for which may be fitted by the function described below under "curve-fitting", the fitting parameters, D, F, and H, are "rate-relaxation rate constants", which can be transformed into rate-relaxation half-times for the respective catalysts by division into the constant, 0.69 (the natural logarithm of 2). "Curve-fitting" refers to the process of finding the values of the parameters in a mathematical function which produce the least deviation of a graph of the function from another graph obtained by some other means, preferably experimental. The preferred criterion for deviation minimization is that of "least squares", which minimizes the sum of the squares of the differences between the experimental dependent variable values and the dependent variable values calculated with the fitting function from the experimental independent variable values. In the present context, the experimental graph is a kinetic trace for a peroxidatic assay; and the fitting function is of the form:

y=A+Bt+C(l-e^"Dt)+E(l-e^"Ft)+G(l-e~^Ht),

where t is time, y is absorbance or fluorescence, and A, B, C, D, E, F, G, and H are parameters to be adjusted to minimize the deviation of a graph of the function from the kinetic trace. Preferably, the eight parameters will be mathematically constrained to reduce the number of parameters varied independently during the curve-fitting process. For example, the parameters D, F, and H can be fixed at the experimental values obtained for the rate- relaxation rate constants of pure metmyoglobin, hemoglobin or methemoglobin, and HRP or hematin, respectively, where rate-relaxation rate constants are related to rate- relaxation half-times by the expression: rate constant = 0.69/half-time.

Modes for Carrying Out the Invention

In a first aspect, the invention is realized by treating a test sample with an aqueous solution of one or more suicide substrates for endogenous peroxidatic catalysts prior to analysis with an analyte-specific binding assay which employs a peroxidatic catalyst as a signal generator. The treatment temperature lies preferably between approximately 20 and 60C; the treatment time lies preferably between approximately one and 20 minutes.

The suicide substrates which are selectively effective against hemoglobin and methemoglobin are organic hydroperoxides, preferably a tertiary alkyl hydroperoxide such as cumene hydroperoxide or tertiary butyl hydroperoxide. The organic hydroperoxide is normally effective at a concentration between approximately 10^"^ M and 1 M in an aqueous solution which otherwise has few compositional restrictions. The organic hydroperoxides may be used alone with test samples which might contain hemoglobin and/or methemoglobin and are unlikely to contain effective amounts of other peroxidatic catalysts.

In addition, organic hydroperoxides show significant effectiveness against metmyoglobin and modest effectiveness against hematin compounds, but require harsher conditions (higher temperature, higher concentration, longer incubation times) to achieve complete inactivation of those peroxidatic catalysts than are needed to inactivate hemoglobin and methemoglobin.

The suicide substrates which are most effective for leucocyte peroxidases are hydrogen peroxide and urea hydrogen peroxide. These substrates are effective at concentrations between approximately 10~³ M and 1 M. Suicide inactivation of leucocyte peroxidases by hydrogen peroxide or urea hydrogen peroxide is promoted by chloride and bromide ions. While test samples may have a high enough concentration of one of these ions to obviate the need for their addition, it is preferred that chloride ion or bromide ion also be added to a concentration between approximately 10^""^ M and 1 M. The sodium or potassium salts of chloride and bromide are the most convenient sources of these anions. During treatment of a test sample with hydrogen peroxide or urea hydrogen peroxide and possibly chloride ion or bromide ion, a catalase inhibitor must also be included, to prevent catalase, frequently present in biological test samples, from destroying hydrogen peroxide before the peroxide can completely inactivate leucocyte peroxidases. Hydroxylamine or an O-alkyl hydroxylamine is preferred for catalase inhibition because these compounds do not inhibit peroxidatic catalysts significantly under certain conditions where they effectively inhibit catalase (for example, at concentrations below about 10~³ M at a pH value of about 5), and therefore do not interfere with the catalytic suicide inactivation by hydrogen peroxide or urea hydrogen peroxide and chloride ion or bromide ion. These hydroxylamine compounds may be supplied either as the free bases or as strong-acid, e.g., HC1, salts. When protected from catalase, hydrogen peroxide and urea hydrogen peroxide are effective suicide substrates for hemoglobin or methemoglobin; they also show significant effectiveness against metmyoglobin, and modest effectiveness against plant peroxidases.

The suicide substrate which is most effective against and most selective for plant peroxidases is the chromogenic electron donor, 4-chloronaphthol, in combination with hydrogen peroxide or urea hydrogen peroxide. The 4-chloronaphthol is effective at concentrations above approximately lO^-**** M in combination with hydrogen peroxide at concentrations above approximately 2 x 10^"-^*** M, in an aqueous solution which otherwise has few compositional restrictions. Preferably, hydroxylamine or an O-alkyl hydroxylamine also is present at a concentration above approximately 10~⁴ M, in order to inhibit catalase which, if present in the test sample, might reduce inactivation effectiveness by consuming the hydrogen peroxide. This inactivation reagent has significant effectiveness against hemoglobin and methemoglobin, which as animal proteins, are unlikely to contaminate a test sample containing endogenous plant peroxidases; and modest effectiveness against hematin compounds and leucocyte peroxidases. It is likely that these less effective inactivation reactions are caused by the hydrogen peroxide component of the inactivation reagent without any contribution from the 4-chloronaphthol, which imparts the selectivity for plant peroxidases.

Suicide inactivation of peroxidatic catalysts with 4-chlorophenol and hydrogen peroxide generates a blue dye which is water-insoluble. However, the catalytic activity of plant peroxidases such as HRP with

4-chloronaphthol is several orders of magnitude lower than their catalytic activity with the most sensitive chromogenic electron donors, such as TMB. Consequently, the amount of blue color generated during inactivation of endogenous plant peroxidases in test samples is likely to be negligible provided that the peroxidase-linked specific binding assay which subsequently is performed on the test sample uses TMB as a chromogen. Assay conditions can be adjusted to assure that any blue color from 4-chloronaphthol is negligible, for example, by limiting the amount of test sample or of analyte-specific probe used in the test. HRP suicide inactivation by 4-chloronaphthol and hydrogen peroxide is improved at temperatures mildly elevated above room temperature, and below approximately 60C, producing greater inactivation with less color development. In addition, background blue signal generated during the inactivation step can be resolved from the analyte-specific signal by any of three approaches:

(a) use a peroxidatic detection chromogen which produces a color other than blue, such as diaminobenzidine (brown), aminoethyl carbazole (red), ABTS (green), or o-phenylenediamine (yellow);

(b) use TMB as a peroxidatic chromogen under conditions where the blue reaction product is water-soluble and therefore easily separable from the insoluble blue product of 4-chloronaphthol oxidation;

(c) use TMB as a peroxidatic chromogen and quench the color-forming reaction with a reagent, such as a mineral acid, which transforms the blue TMB oxidation product into a soluble yellow compound.

Alternatives (a) and (c) are well described in the research literature on peroxidatic assays and are generally known to those skilled in the art of peroxidatic assay. Alternative (a) is not preferred because these chromogens are less sensitive than TMB and, unlike TMB, are cancer-suspect agents. Alternatives (b) and (c) are the normal mode of using TMB as a chromogenic electron donor in peroxidatic assays, and are well known to those skilled in the art. Co-pending U.S. Patent Application 896,677, the disclosure of which is incorporated herein by reference, describes methods for rendering the blue TMB oxidation product insoluble, methods which might not be applicable in an assay where 4-chloronaphthol plus hydrogen peroxide are used to inactivate endogenous plant peroxidases, if the latter were so concentrated as to produce an easily visible blue deposit.

While each of these treatments alone is most effective for inactivating a different peroxidatic activity, the user most often will not know in advance what endogenous peroxidatic catalysts are present in a test sample. Therefore, incubating the test sample with a combination of an organic hydroperoxide (preferably a tertiary alkyl hydroperoxide) with hydrogen peroxide (again with a non-peroxide catalase inhibitor and preferably with chloride or bromide ion) is preferred because such treatment inactivates four of the major endogenous peroxidatic catalysts: hemoglobin, methemoglobin, metmyoglobin, and leucocyte peroxidases. It substantially inactivates hematin compounds as well. 4-chloronaphthol optionally may be included, in order to assure effectiveness of this combination reagent against plant peroxidases. The respective effective concentration ranges are those specified above. The exact conditions used for inactivating endogenous peroxidatic activity are optimized for each separate application, depending on (a) the maximum amounts of hemoglobin, methemoglobin, myoglobin, hematin compounds, plant peroxidases, and/or leucocyte peroxidases which might be found in the test sample, (b) the length of time for which it is convenient to run the background inactivation reaction, (c) what stabilizing agents, such as chelators, are necessary to preserve the inactivation reagents beween manufacture and use, (d) what other components, such as detergents, salts, chaotropic agents, organic cosolvents, acids, and bases are needed to extract the analyte from the test sample (should the particular analytical method require extraction), (e) how well the analyte tolerates the oxidizing conditions presented by hydroperoxides, (f) how much risk the user may reasonably assume in handling the reagents, which are individually somewhat toxic at high concentrations, and (g) whether or not it is convenient to subject the test sample to mild heating (between ambient temperature and approximately 60C) during treatment. Generally speaking, the speed and completeness of the inactivation reactions are increased as organic hydroperoxide, 4-chloronaphthol, hydrogen peroxide, and/or non-peroxide catalase inhibitor concentrations are increased and as the temperature is increased. If test samples are expected to contain high levels of endogenous peroxidatic activity, high reagent concentrations and/or higher temperatures may be needed in order to be confident that any surviving activity will be too low to interfere with the peroxidatic signal indicating the presence of analyte. In addition, if it is desired to complete the inactivation reaction in a few minutes, high reagent concentrations and/or treatment temperature may be chosen. On the other hand, lower reagent concentrations and/or treatment temperatures reduce the chance (already low) of damaging the analyte or exposing the user to hazardous conditions, and may be preferred if longer treatment times (over 10 minutes) do not inconvenience the user. For example, the invention can be realized by storing the test sample with inactivation reagent, between time of collection and time of analysis, even though the interval may amount to hours or days.

This aspect of the invention is most likely to be realized as one step in a multi-step specific binding assay for a particular analyte. In such assays, the analyte is detected by binding to it a specific binding moiety which distinguishes the analyte from other substances which might be present in the test sample. A signal-generating moiety is covalently coupled to the analyte-binding moiety or, alternatively, is covalently coupled to a second binding moiety which distinguishes the analyte-binding moiety from other substances in the test sample by binding specifically to it. Plant peroxidases, especially from horseradish, are preferred as signal generators because of their high sensitivity, safety, durability, ease of chemical coupling, and relatively low cost. Examples of enzyme-linked specific binding assays include EIA's, ELISA's, immunodot blots, immunoenzyme histochemical and cytochemical procedures, nucleic acid probe dot blots, and i situ nucleic acid hybridization procedures, methods for which are well documented in the research literature. These assays have in common the features that (1) the analyte is contained in or trapped by the analyst on a solid phase which might also contain or capture endogenous peroxidatic catalysts, and (2) the final chemical step in the procedure is incubation of the solid phase with chromogenic or fluorogenic substrate for the signal-generating enzyme. Because the analyte is immobilized, the background- inactivating reagents of this invention can be removed from the analyte by filtration or decantation after their job is done, and the analyte-containing solid phase can be washed with an aqueous solvent to assure complete removal. That way the inactivating reagents cannot interfere with the peroxidatic detection chemistry of the peroxidase-linked specific binding assay. The invention is indifferent to the analyte- binding moiety pair utilized by the peroxidase-linked specific binding assay. The analyte could be an antigen; and the binding moiety could be an antibody or antibody fragment which binds specifically to it, or vice versa. If the analyte is a specific sequence of DNA or RNA, the binding moiety will be a nucleic acid molecule, preferably DNA, containing a complementary single-stranded nucleic acid sequence. However, other analyte-binding-moiety pairs suitable for peroxidatic signal generation and therefore for the background reduction provided by the invention include the following: carbohydrate-lectin, immunoglobulin-Staphyloccal Protein A, immunoglobulin- complement, enzyme-inhibitor, enzyme-substrate, nucleic acid-binding protein, hormone-receptor, hormone-transport protein, cytokine-receptor, neurotransmitter-receptor, and drug-receptor. Antigen-antibody binding is not the only case where the polarity of the analyte-binding moiety relationship is unimportant. For example, peroxidase- tagged carbohydrate could be used to detect particular lectins; or peroxidase-tagged lectin could be used to locate particular carbohydrates.

The test sample can be almost any material which might contain the targeted analyte. Most commonly it will be biological material of animal origin, such as blood or a blood fraction, cerebrospinal fluid, synovial fluid, sputum, saliva, urine, pus, a biopsy or autopsy specimen, a nasopharyngeal, ocular, or genitourinary swab, a cytochemical smear, or a histochemical section, each of which is collected or prepared by methods well established in the arts of medicine, clinical chemistry, histochemistry, or cytochemistry. However, it might also represent tissue or an extract from any part of a plant. In addition, test samples in environmental or forensic science might contain no or negligible amounts of biological material. Finally, the test sample which is treated to reduce peroxidatic background may be a liquid (preferably aqueous) extract of any of the examples listed above; such extraction often helps to reduce interference. The reagents and methods of the invention permit great flexibility with respect to where in the total analytical procedure the inactivation of background peroxidatic activity occurs. With few exceptions, they may be employed at any time prior to the addition of the signal-forming peroxidatic substrates, although preferably inactivation of peroxidatic background will be done before the test sample is incubated with the analyte-specific binding moiety. If 4-chloronaphthol and hydrogen peroxide are used to inactivate endogenous plant peroxidases in the test sample, they must be applied and then completely removed before adding the peroxidase-containing detection reagent.

The invention also includes reagent solutions for inactivating endogenous peroxidatic activity. A preferred mode is to formulate the organic hydroperoxide, 4-chloronaphthol, hydrogen peroxide, non-peroxide catalase inhibitor, and/or any. necessary stabilizers or analyte extractants as a stock solution which is approximately 2 to 20 times more concentrated with respect to each component than is desired in the inactivation reaction. Not only is such a stock solution more convenient to store, but it can be applied with minimal dilution of the test sample. This mode is most likely to be used for applications, such as ELISA, where the test sample is in liquid form. For applications such as immunohistochemistry, immunocytochemistry, and jLn situ nucleic acid hybridization, where the test sample is solid, the reagents are just as conveniently formulated at use concentration because dilution of the test sample is not an issue. Although there are few restrictions on the pH at which the inactivation reagents are used, formulation pH preferably is mildly acidic, below 7 (preferably below 5) and above 2, because the rate of spontaneous dismutation of hydroperoxides varies directly with pH. Therefore lower formulation pH improves storage stability.

In some cases (e.g., hydrogen peroxide in combination with a non-peroxide catalase inhibitor; hydrogen peroxide in combination with 4-chloronaphthol), the chemical components of the inactivation reagents must be present at the same time during use. However, they need not be formulated together if storage stability or convenience is promoted by preparing them in separate solutions which are mixed at the time that the inactivation reaction must proceed. There are very few practical restrictions on the modes of preparing the inactivation reagent solutions, as the essential ingredients are all reasonably water soluble and are chemically compatible with one another.

In a second aspect, the methods and reagents of the invention reduce the interference in forensic and clinical tests for blood or hemolysis. Currently these tests are carried out by incubating the test sample with an indicator solution or solid phase containing an hydroperoxide and a chromogenic electron donor and observing whether a color is formed; such a signal indicates the presence of a peroxidatic catalyst (inferred to be hemoglobin, methemoglobin, or a hematin compound) which has promoted oxidation of the chromogen. A wide variety of such tests is available for use in the forensic or clinical laboratory and even by scientifically untrained individuals at home, using a variety of hydroperoxides and chromogens in a variety of formats. The most common peroxidatic assay for blood employs stool as a test sample and is intended to detect intestinal bleeding characteristic of colon cancer. In this case interference by myoglobin or metmyoglobin and perhaps by hematin compounds is reduced by dietary avoidance of red meat for several days prior to sample collection. Although hemoglobin can be digested to release hematin compounds, it is preferable that an occult blood test be insensitive to hematin compounds because they are most likely to arise from digestion of dietary hemoglobin or myoglobin in the upper digestive tract, not from colonic bleeding.

One embodiment of the invention reduces potential catalatic interference encountered with such assays by including in the indicator solution a non-peroxide catalase inhibitor which does not block the peroxidatic activity of hemoglobin or methemoglobin. Preferably the catalase inhibitor is hydroxylamine or an O-alkyl hydroxylamine at a concentration between approximately 10 ~⁴ M and 10^"2 M. Although any chromogenic electron donor is appropriate on which hemoglobin and methemoglobin are active, 3,3*,5,5'- tetramethylbenzidine (TMB) is greatly preferred because of its high sensitivity and low toxicity. Depending on the assay format, the TMB may be dissolved, preferably at a concentration between 10 ~⁵ M and lO^"*** M in an aqueous buffer which may or may not contain an organic cosolvent to increase TMB solubility, or may be immobilized on a solid support. TMB solubility depends strongly on pH and cosolvent concentration, increasing as the pH is reduced below 5 and/or the cosolvent concentration is raised. Preferred among organic cosolvents is N-methyl pyrrolidone- (ϋ.S. Patent No. 4,596,770, the disclosure of which is incorporated herein by reference) Incorporation of a non- peroxide catalase inhibitor is useful primarily for those tests in which the hydroperoxide is hydrogen peroxide, because organic hydroperoxides are not rapidly dismuted by catalase. Therefore the presence of catalase in a test sample cannot greatly reduce the sensitivity of a test relying on an organic hydroperoxide. However, hydrogen peroxide is preferred in indicator solutions for hemoglobin and hematin compounds because it gives greater sensitivity than do the organic hydroperoxides; and a consequence of this embodiment is that tests for blood which use hydrogen peroxide for color generation become even more preferred over those using an organic hydroperoxide because their vulnerability to catalase interference is reduced.

A further embodiment of this second aspect is the application of the methods and reagents of the invention in an assay which discriminates between hemoglobin or methemoglobin and other peroxidatic catalysts which might be present in the test sample. A fraction of the test sample is incubated with an organic hydroperoxide, preferably a tertiary alkyl hydroperoxide such as cumene hydroperoxide or tertiary butyl hydroperoxide, before treatment with an indicator solution containing a chromogenic electron donor and a hydroperoxide. Both treated and untreated portions of the test sample are exposed to chromogen and hydroperoxide, using any of many sets of conditions established in the art for assaying peroxidatic activity. If the untreated portion gives a significant color reaction and the treated one does not, the user infers that the only peroxidatic catalyst in the test sample was hemoglobin or methemoglobin. If both portions give a similar amount of color, the peroxidatic catalyst is something other than hemoglobin or methemoglobin. If neither portion gives significant color, no peroxidatic catalyst is present above the detection limit of the test. If treatment reduces the signal intensity significantly but not completely, then the sample might contain hemoglobin or methemoglobin in combination with another peroxidatic catalyst, or the peroxidatic catalyst might consist largely of metmyoglobin or a hematin compound. Interpretation of this fourth possible outcome depends on the specific context of the assay, which determines whether it is reasonable to expect metmyoglobin or hematin compounds to be present. When the test sample is stool col'lected after several days of dietary exclusion of red meat, metmyoglobin should not be present; and hematin compounds are unlikely to be present.

Many modes are possible for carrying out the assay. The only significant restriction is that the organic hydroperoxide, preferably cumene hydroperoxide or tertiary butyl hydroperoxide, be present at a concentration between approximately 10^"^ M and 1 M. As cumene hydroperoxide has an aqueous solubility near 20 mM in the absence of detergent and organic cosolvents,- higher concentrations will require formulation in the presence of enough detergent or cosolvent to reach the targeted concentration. l-Methyl-2-pyrrolidinone and epsilon- caprolactam are preferred as organic cosolvents for this purpose. A wide range of treatment solvent compositions, including pH, is possible. A preferred mode is to formulate the organic hydroperoxide at a concentration which is two to fifty times the use concentration, so that the test sample, which often consists of stool, urine, or blood serum or plasma, is not significantly diluted by the reagent. Mildly acidic (pH 2-6) formulation pH values are preferred in order to improve reagent storage stability. Mild heating (up to approximately 60C) is preferred during use because it accelerates the inactivation reaction and promotes complete inactivation of any hemoglobin present. Longer treatment times are also preferred, but incubation longer than 20 minutes is unlikely to improve discrimination.

In another embodiment of the second aspect, the methods and reagents of the invention provide a second way to improve discrimination among peroxidatic catalysts which might be present in a test sample being assayed for blood or hemolysis. The total sample is incubated with a mixture of 4-τchloronaphthol, hydrogen peroxide or urea hydrogen peroxide, and optionally a non-peroxide catalase inhibitor (preferably hydroxylamine or an O-alkyl hydroxylamine). As before, peroxidatic activity is then monitored by adding to the treated test sample a hydroperoxide, preferably hydrogen peroxide, and a chromogenic electron donor, preferably TMB. If, in the inactivation reaction the 4-chloronaphthol concentration lies between approximately 10~⁵ and approximately 10~³ M, the hydrogen peroxide concentration lies between approximately 2 x 10~⁵ M and approximately 2 x 10^"3 M, and the non-peroxide catalase inhibitor (if present) has a concentration between 10~⁴ M and 10~² M, plant peroxidases in the test sample should be completely inactivated while a significant fraction (greater than 15%) of the hemoglobin and methemoglobin activity should survive treatment. What hemoglobin and methemoglobin inactivation does occur is caused by the hydrogen peroxide more than the 4-chloronaphthol. There is no need in the inactivation reaction for a hydrogen peroxide concentration exceeding the 4-chloronaphthol concentration by more than a factor of about two. Therefore hemoglobin and methemoglobin activity can be maximally preserved by maintaining this ratio and using only about as much 4-chloronaphthol as is observed to be 5 necessary for worst-case test samples. Provided that sufficiently sensitive peroxidatic assay conditions are used (for example, a TMB concentration between 2 x 10^""4 M and 8 x 10^""4 M, a hydrogen peroxide concentration between 10~³ M and 10~² M, a hydroxylamine concentration between

10 10^"4 M and 10~² M, and a pH between 3.5 and 4.5), such an assay is sensitive enough for the needs of many applications despite the fact that as much as 85% of the hemoglobin and methemoglobin activity is lost in the process of completely inactivating plant peroxidases. This

15 embodiment is ineffective in eliminating background peroxidatic activity caused by metmyoglobin and only partly effective in inactivating hematin compounds and leucocyte peroxidases, and so is not preferred for applications where there is a possibility that peroxidatic activity could -20 arise from these sources.

In a third aspect, the invention detects the presence of hemoglobin or methemoglobin in a test sample simply by analyzing the shape of the kinetic trace obtained when a portion of the test sample is assayed for

25 peroxidatic activity by adding it to an indicator solution containing TMB and hydrogen peroxide or urea hydrogen peroxide. The background-inactivation pre-treatments of the second aspect are unnecessary and generally undesirable in the third aspect. However, here the colored signal

30 indicating the presence of a peroxidatic catalyst must be monitored in a spectrophotometer or colorimeter which records absorbance as a function of time. Furthermore, the test sample must be diluted or concentrated as necessary to obtain a spectral change on the chosen time scale

35 (preferably 3 to 10 minutes) which is large enough to measure accurately but not too large to remain within the range of the instrument; in practice, a total absorbance change between approximately 0.2 and 2 is normally needed. For the first and second aspects of the invention, visual detection of the peroxidatic signal often is adequate and even preferred (for reasons of simplicity), and the effective time scale for generating the peroxidatic signal could range from a few seconds to a few hours.

The third aspect of the invention derives from the observation that when peroxidatic assay with TMB and hydrogen peroxide is performed in certain solvents, hemoglobin and methemoglobin uniquely generate a kinetic trace in which approximately 90% of the catalytic activity is lost in an exponential decay over several minutes. Plant peroxidases, leucocyte peroxidases, and hematin compounds give kinetic traces in which catalytic activity decays much more slowly, if at all; and metmyoglobin, while showing a loss of catalytic activity analogous to that of hemoglobin, often does so on a time scale of less than a minute. Therefore, any method of testing the peroxidatic assay kinetic trace for the presence of an exponential loss of catalytic activity on the time scale of a few minutes can realize this third aspect of the invention. If the peroxidatic catalyst is essentially pure, visual inspection of the absorbance kinetic trace may suffice to detect the presence or absence of significant curvature on the time scale of a few minutes. However, this visual discrimination is greatly improved by graphing catalytic activity, the rate of change of absorbance, versus time. This process is facilitated by collection of the kinetic trace^' in a microprocessor-controlled spectrophotometer with a pre-programmed operation for taking the derivative of absorbance with respect to time (example: the HP8450A spectrophotometer of the Hewlett Packard Company), or in a computer-interfaced spectrophotometer in which the computer is programmed to perform this transformation.

However, because many test samples might contain a mixture of hemoglobin or methemoglobin with other peroxidatic catalysts, this third aspect of the invention is best realized by curve-fitting the absorbance kinetic trace to a mathematical function which is a sum of the integrated rate expressions for the individual peroxidatic catalysts which might be present in the test sample. The following function should suffice for this purpose:

y = A+Bt+C(l-e~^Dt)+E(l-e^"Ft)+G(l-e^"Ht).

In it, t represents time after mixing the appropriately diluted or concentrated test sample with the TMB and hydrogen peroxide; y represents the absorbance;^' and A, B, C, D, E, F, G, and H represent the fitting parameters which must be adjusted in value to optimize the fit of the kinetic trace to the function. The least-squares criterion for best fit suffices for this procedure; and the simplest procedure for finding the best fit is by digital computation using a nonlinear least-squares algorithm, now commonly included in commercial statistical and scientific subroutine packages (for example, in RS-1, commercial tradename of Bolt Beranek and Newman, Inc., in Statgraphics, Version 1.2, commercial tradename of STSC, Inc., and in SYSTAT, Version 3.0, commercial tradename of SYSTAT, Inc.) and well known to those skilled in the art of numerical analysis (Bevington (1969) Data Reduction and Error Analysis for the Physical Sciences, McGraw-Hill, New York, pp. 232-246, the disclosure of which is incorporated herein by reference).

The fitting parameters, A-H, have simple interpretations. A is the absorbance at zero time in the kinetic trace. B is the sum of the remaining catalytic activities of all peroxidatic catalysts in the test sample after they have undergone their individual exponential decays in catalytic activity. D, F, and H are the rate- relaxation rate constants of three kinetic classes of peroxidatic catalyst: metmyoglobin, which decays in less than a minute, hemoglobin and methemoglobin, which decay on the time scale of a few minutes, and plant peroxidases (such as HRP) and hematin compounds, which decay on the time scale of one or a few hours. Although there may be a significant difference in rate-relaxation rate constant between catalysts of a kinetic class (hemoglobin versus methemoglobin; plant peroxidases versus hematin compounds), these differences are too small to reduce the quality of fit significantly if an average rate constant value is applied to each kinetic class. The parameters, C, E, and G are the amplitudes of the transient components of the kinetic trace contributed by the three kinetic classes_,of peroxidatic catalyst. Nonlinear least-squares fitting algorithms include operations for estimating the standard deviations of the fitting parameters. The goal of the curve-fitting procedure is to obtain a best-fit estimate of the transient amplitude corresponding to the rate relaxation rate constant of hemoglobin and methemoglobin. If this value is significantly greater than its standard deviation (where significance is evaluated by methods well known to those skilled in the art of statistical analysis), then the analyst can conclude that the test sample contained hemoglobin or methemoglobin.

The least-squares curve fitting procedure can be simplified in several ways which may improve confidence in the determination of the presence of hemoglobin or methemoglobin. One way is to fix the values of as many parameters as possible. Ideally, only the parameters, B, C, E, and G need be optimized. A can be fixed at the initial absorbance value of the kinetic trace unless the trace is noisy. D, F, and H can be fixed at rate- relaxation rate-constant values measured for the pure catalysts in calibration reactions. The other mode of simplification is to restrict the kinetic trace to a time window which excludes as much of the curvature due to the interfering peroxidatic catalysts as possible. If the first 30-60 seconds after mixing are eliminated, metmyoglobin rate decay is essentially complete, so that the transient amplitude parameter assigned to metmyoglobin can be fixed at a zero value, completely eliminating one of the three exponential decay terms from the fitting function. In addition, limiting the total kinetic trace to approximately five minutes of duration after mixing reduces the contribution of the exponential decay term for plant peroxidases and hematin compounds. These steps maximize the sensitivity of the curve fitting operation to the presence of hemoglobin or methemoglobin.

The power of this third aspect of the invention for discriminating the presence of hemoglobin or methemoglobin in the test sample is increased by adjusting indicator solution solvent composition to maximize the separation of the rate-relaxation half-times of hemoglobin and methemoglobin from that of metmyoglobin (on a shorter time scale) and those of HRP and hematin compounds (on a longer time scale). Metmyoglobin catalytic activity should decay at least three times faster than those of hemoglobin and methemoglobin, which should decay at least three times faster than those of plant peroxidase and hematin compounds. Rate-relaxation half-time is controlled by at least five assay solvent composition variables: pH, identity and concentration of organic cosolvent, and identity and concentration of detergent. In the absence of cosolvent and detergent, a peroxidatic assay of pH of 4.0 gives good resolution of the three kinetic classes of peroxidatic catalyst: metmyoglobin has a rate-relaxation half-time of less than ten seconds, such that its rate relaxation is essentially complete in less than one minute; hemoglobin and methemoglobin have rate-relaxation half- times of 30-40 seconds, such that their rate relaxations are complete in 3-4 minutes; HRP and hematin have rate relaxation half-times exceeding five minutes. As the assay pH is increased from 4 to 5, the rate-relaxation half-times of metmyoglobin, hemoglobin, and methemoglobin increase until they approximate those of hematin and HRP, destroying any ability to resolve the various peroxidatic catalysts on the basis of curvature of the kinetic trace. At pH 5, the ability to distinguish the kinetic classes of peroxidatic catalyst is restored by adding to the indicator solution certain concentrations of certain organic cosolvents and/or by adding a certain concentration of practically any detergent. Because the rate-relaxation half-times vary continuously with concentration of effective cosolvent or detergent, each additive has a narrow concentration range over which the rate-relaxation half-time of hemoglobin lies in the range of 35-45 seconds. For example, a 10% concentration of the organic cosolvent, epsilon- caprolactam, an 0.0013% concentration of the detergent, sodium dodecyl sulfate, and an 0.19% concentration of the detergent, Tween 20, all give a hemoglobin rate-relaxation half-time of 35-45 seconds at pH 5.0. The methemoglobin rate-relaxation half-time is generally 10-20 seconds shorter than that of hemoglobin under any of these conditions. These additives control the rate-relaxation half-time of metmyoglobin as well. Although this effect is analogous to that on hemoglobin and methemoglobin, quantitative difference among detergents render some of them especially effective in resolving metmyoglobin kinetically from hemoglobin and methemoglobin. Especially preferred at pH 5 are sodium dodecyl sulfate at 0.0013%, Zwittergent 3-12 (registered trademark of Calbiochem Behring) at 0.068%, Neodol 25-3S (registered trademark of Shell Chemical Company) at 0.00068%, cetyl trimethylammonium bromide at 0.0023%, and octyl-beta-D- thioglucopyranoside at 0.19%. However, practically every detergent should have an effective concentration for generating a hemoglobin rate-relaxation half-time of 35-45 seconds, and some detergents not yet tested may resolve the three kinetic classes of peroxidatic catalyst better than do those listed here. Hydrogen ion, cosolvent, and detergent independently control the rate relaxation of metmyoglobin, hemoglobin, and methemoglobin, and the concentration of each can be independently varied within the scope of this third aspect of the invention.

In practice, the third aspect of the invention is realized by performing the following sequence of operations.

(a) A test sample is obtained as a liquid essentially free of particulate matter. If the original test sample, such as a stool specimen, contains solid matter, it must be extracted with an aqueous solvent by thorough mixing followed by gravitational settling, centrifugation, or filtration. In this case, preferably the volume ratio of liquid to solid will not exceed approximately 2:1, to minimize dilution of extracted peroxidatic catalysts. The extraction solvent may consist of the preferred buffer for performing the peroxidatic assay, and may contain detergent, organic cosolvent, or other additives to promote red cell lysis. Maintenance of a low ionic strength (below approximately 0.02) also favors red cell lysis.

(b) A portion of the clear liquid test sample is added to a mixture of TMB, hydrogen peroxide, buffer, possibly organic cosolvent, and possibly detergent which comprises an indicator solution of an effective composition for resolving the rate-relaxation half-times of hemoglobin and methemoglobin from those of metmyoglobin, plant peroxidases, and hematin compounds. This addition is performed in a spectrophotometer cuvette, and the peroxidatic reaction is followed in a spectrophotometer at an effective wavelength between approximately 350 nm and 700 nm for an interval of approximately four to approximately ten minutes, possibly excluding the first 30 to 60 seconds after mixing in order to eliminate any rate relaxation contribution from metmyoglobin. The most sensitive wavelengths for following TMB oxidation are 370 nm and 652 nm, but almost any wavelength in the visible spectral region can be chosen to give a total absorbance increase over the assay interval of between approximately 0.2 and 2.

(c) If the test sample has such a high concentration of peroxidatic catalysts that the absorbance change exceeds approximately 2, even at wavelengths away from the absorbance maxima of 370 and 652 nm, it should be diluted in an aqueous solvent and reassayed.

Alternatively, the assay can be repeated in a cuvette with a shorter light path than the 1 cm value most commonly used. If the test sample has too low a concentration of peroxidatic catalysts to give an absorbance change exceeding approximately 0.2, even at the absorbance maximum of 370 nm, it should be concentrated by vacuum, pressure, or centrifugal ultrafiltration, using an ultrafiltration membrane with a molecular weight cutoff not exceeding approximately 60 kilodaltons. Then it should be re- assayed. Alternatively, it may be possible to increase the volume ratio of test sample to indicator solution reagents in the assay cuvette, or to use a cuvette with a light path longer than 1 cm, during reassay.

(d) If no practical concentration of the test sample gives a kinetic trace with an amplitude exceeding a few hundredths of an absorbance unit at 370 nm or 652 nm, it is concluded that the test sample contains insignificant peroxidatic catalyst of any kind.

(e) If a kinetic trace has been obtained with a total absorbance change large enough for accurate numerical analysis, it is examined visually for signs of curvature over the time scale of one to five minutes after mixing. If it is essentially linear during this interval, further numerical analysis probably is unjustified; and the absence of hemoglobin and methemoglobin in the test sample can be concluded. If it shows significant curvature, the kinetic trace should be used as input for nonlinear least-squares curve fitting in a digital computer, as described above. Preferably, the time window fitted will be restricted and some of the fitting parameters will be fixed, as described above.

(f) The output of the curve-fitting operation should contain the best-fit value and standard deviation of the amplitude parameter for the exponential decay component with the rate-relaxation rate constant of hemoglobin or methemoglobin (or an average of the two rate-relaxation rate constants). The best-fit amplitude is compared to its standard deviation according to customary statistical procedures to determine whether it is significantly greater than zero. If it is, a conclusion is made that hemoglobin or methemoglobin was present in the test sample. If it is not, whatever peroxidatic activity is present in the test sample must be contributed by some other catalyst.

Because of the effort and instrumentation required for this third aspect of the invention, it is most economically applied only to those test samples which have tested positive by a more qualitative occult blood test, such as that described under the second aspect of the invention. A fourth aspect of the invention encompasses indicator solutions for peroxidatic activity which contain hydroxylamine or an O-alkyl hydroxylamine, a chromogenic or fluorogenic electron donor, preferably TMB, and a hydroperoxide, preferably hydrogen peroxide; such solutions are protected from catalase in the test sample which might dismute the hydroperoxide, lowering the assay sensitivity. The hydroxylamine or O-alkyl hydroxylamine concentration preferably will be between about 10^""4 M and 10~² M. The chromogen or fluorogen concentration will be above about 10^"6 M, determined by its solubility or the kinetic properties of the targeted peroxidatic catalyst. The hydroperoxide concentration will lie above about 10~⁴ M, also optimized to match the kinetic properties of the targeted catalyst. The other aspects of indicator solution composition, such as ionic strength, buffer ion, or the inclusion of organic cosolvent, also may be optimized for the targeted peroxidatic catalyst without greatly reducing the ability of the hydroxylamine to block catalatic activity.

The most common mode of using such a solution, well known to practitioners of the art of analytical chemistry, is to add a small amount of a test sample thought to contain a peroxidatic catalyst to such an indicator solution and immediately to observe the change in visible color or fluorescence. Often the test sample is diluted in an aqueous solvent, such as a buffer with a pH in the range of 5 to 8, before assay in order to create an accurately measured signal change on a convenient time scale (minimally approximately one minute). This change can be quantitated in instruments such as colorimeters, spectrophotometers, and fluorometers, or can be recorded qualitatively by visual inspection. In a variant of this mode of use, the peroxidatic reaction is quenched after a set time interval, usually by adding to the reaction mixture an inhibitor of peroxidatic activity, such as strong acid, cyanide, or azide. In another variant, the colored or fluorescent oxidation product is insoluble or adsorbs to a convenient solid phase, such that it can be separated from the rest of the reaction mixture by centrifugation, decantation, or filtration. Then the color-forming reaction can be stopped by performing one of these separation operations, usually followed by a solvent wash to remove any residual reaction mixture. None of these various modes of using this modified indicator solution affects its novel and useful property of being indifferent to the presence of catalase in the test sample being assayed.

The preceding four aspects of this invention can sometimes be rendered more effective by addition of certain chelators to the background inactivation reagents and/or peroxidatic activity indicator. These chelators serve to reduce the background peroxidatic activity of certain transition metal ions, especially Fe(III), which are active in oxidation-reduction reactions and which may be present in the test sample as insoluble hydroxides or salts or as complex ions which are more labile than the chelated Fe(III) in hematin compounds. These chelators work by displacing the ligands which complex the transition metal ions in the test sample, forming chelate complexes of especially low peroxidatic activity. As different chelate complexes of a transition metal ion may have-peroxidatic activities ranging over more than three orders of magnitude with a specificity which depends on the choice of chromogenic electron donor for the indicator solution, it is important to choose chelators which effectively reduce peroxidatic activity for the chromogen used. Especially preferred for suppressing the peroxidatic activity of Fe(III), which is the most common transition metal ion in most test samples and chemical reagents, when the chromogen is a benzidine compound such as TMB, are DFA, EDDHA, EDADP- alpha, CDTA, EDTA, EDADP-beta, HBED, EDTP-alpha, EDTP-beta, and PP^- These chelators, which may be added as free acids or alkali-metal salts, as convenient, are effective at concentrations above approximately 10^"*' M and rarely need be added at concentrations exceeding 10~² M. A concentration of 10^"3 M is preferred for most applications, depending partly on the concentration of catalytically active transition metal ion in a test sample. Chelator concentrations higher than needed to sequester transition metal ions in the test sample can be deleterious because many commercially supplied chelators are significantly contaminated with Fe(III) to begin with. When simple procedures exist for removing metal ion contaminants from a chelator, it is preferable to repurify the chelator before use. For example, transition metal ion can be .removed from EDDHA by the procedure of Rogers [(1973) Infection and Immunity 7, 445-456], the disclosure of which is incorporated herein by reference.

Many of the chelators listed above are available from suppliers of chemical reagents. DFA is supplied by CIBA-GEIGY under the commercial trademark of Desferal. The synthesis of HBED was described by L'Eplattenier et al. ((1967) J. Amer. Chem. Soc, J39:837-843), the disclosure of which is incorporated herein by reference. EDTP-alpha and EDTP-beta are synthesized by slight variations of the standard procedure for making EDTA. For example, alpha- chloropropionic acid or beta-chloropropionic acid is heated in dilute NaOH with ethylene diamine at a molar ratio of acid to diamine in modest excess of 4:1. The cooled reaction mixture is acidified with mineral acid such as HCl to precipitate the chelator as the tetra-acid. If all synthetic steps are performed with care to eliminate transition-metal contamination (e.g., by using the purest possible chemical reagents and water in carefully cleaned plastic containers without exposure to metal surfaces or magnetic stir bars which might carry iron filings), the resulting chelators will have maximum effectiveness in blocking transition-metal-ion interference in peroxidatic assays.

Preferably test samples are incubated with an effective concentration of an effective chelator for the longest practical interval (minimally approximately one minute) before exposure to the peroxidatic activity indicator solution, because some catalytically active transition metal ion complexes may have such high stability constants and low dissociation rate constants that an interval of at least a few minutes is required for complete trapping of the transition metal ions in catalytically inactive chelate complexes. If first exposure of the test sample to chelator occurs during the peroxidatic assay itself, the assay may have proceeded significantly before the chelator has had a chance to reduce the transition- metal-ion-generated background activity. The fifth aspect of this invention entails the stopping of the color-development step in peroxidatic assays by addition of an effective amount of a suicide • substrate specific for the peroxidatic catalyst. The stopping reaction should be complete between about 1 and about 10 minutes, being accelerated by higher temperatures (in a practical range between about 20C and about 60C) and higher suicide substrate concentrations. These reaction conditions and less important details, such as reaction pH (between about 4 and about 8, preferably between about 4 and about 5) and the other components of the stopping reagent (buffers, detergents, cosolvents, and other chemical components) should be optimized for a given application. The stopping reaction normally occurs at the end of a multi-step analytical procedure, such as an analyte- specific binding assay or a test for occult blood, and is largely insensitive to details of the procedure. Preferably the stopping reaction is performed after a peroxidatic activity indicator solution has been incubated for a timed interval with the peroxidatic catalyst, which normally is immobilized on or contained in a solid phase. The stopping reaction is performed after a peroxidatic activity indicator solution has been incubated with the test sample or in the test device for enough time to detect or quantitate any peroxidatic catalyst which might be present. It is useful in formats where the catalyst and/or the chromophore produced in the indication reaction is immobilized on a solid phase, and where it is desired to prevent continued reaction by indicator solution which is hard to remove completely from the immobilized catalyst or by catalyst which is hard to remove completely from the immobilized chromophore. Examples of such assays are filter-binding enzyme immunoassays (Valkirs and Barton, supra), dipstick assays for occult blood (U.S. Patent 4,447,542), or any peroxidatic assay in which the chromophore indicating the presence of peroxidatic activity is trapped in a porous matrix. For the stopping reaction to be effective, the indicator solution must be removed (e.g., by decantation or filtration) from the solid phase carrying the catalyst and/or chromophore before the solid phase is contacted with the stopping solution. Otherwise remaining chromogen in the indicator solution will block access of the suicide substrate to the catalytic active site._^ Because the stopping solution is not substantially diluted by addition to the test sample or test device, it is conveniently formulated at use concentrations of the suicide substrates. If the peroxidatic catalyst is a plant peroxidase such as HRP, the stopping reagent is 4-chloronaphthol at a use concentration between about 10^"^ and about 10~³ M,' in combination with hydrogen peroxide or urea hydrogen peroxide at a use concentration between about 2 x lO^"**' M and about 10~² M. Stopping reagent formulation may be affected by the fact that 4-chloronaphthol solubility in aqueous solutions at a pH below about 8 is about 1 mM in the absence of detergents and organic cosolvents. If formulation at a concentration above 1 mM is desired, l-methyl-2-pyrrolidone and epsilon-caprolactam are preferred as organic cosolvents which may be mixed with water at any concentration needed to meet a 4-chloronaphthol solubility requirement. As noted above, inactivation of HRP with 4-chloronaphthol and H₂0₂ can result in the formation of a blue deposit. If this material interferes with the assay, it can be minimized by performing the stopping reaction at a mildly elevated temperature, which normally need not exceed about 60C. If the peroxidatic catalyst is hemoglobin or methemoglobin, the stopping reagent is either a tertiary alkyl hydroperoxide, preferably t-butyl hydroperoxide or cumene hydroperoxide, or a combination of hydrogen peroxide or urea hydrogen peroxide with a non-peroxide catalase inhibitor, preferably hydroxylamine or an O-alkyl hydroxylamine. The use concentration of the tertiary alkyl hydroperoxide is between about 10~³ M and about 10^"1 M, and of the hydroxylamine or O-alkyl hydroxylamine is between about 10~⁴ M and 10~² M. Formulation of cumene hydroperoxide is limited by the fact that its aqueous solubility at pH values below about 8 is near 20 mM. Organic cosolvents, preferably l-methyl-2-pyrrolidinone or epsilon-caprolactam, are useful for obtaining higher concentrations. The sixth aspect of the invention employs the properties of 4-chloronaphthol and hydrogen peroxide as plant peroxidase suicide substrates to create a multi- probing format for peroxidase-linked analyte-specific binding assays. In general terms, a solid phase to which analytes from a test sample may have been attached is incubated with a series of separate peroxidase-linked probes specific for different analytes; and the peroxidatic signal associated with the binding of each probe to the test sample is recorded. After each analyte-associated signal is recorded and before incubation of the solid phase with the next probe, the peroxidase activity responsible for that signal is destroyed by the suicide reaction, so that peroxidase attached to the solid phase through the binding to one analyte by its specific probe cannot generate a signal which will confuse the outcome of later testing for other analytes.

Many of the steps in a multi-probing assay format are identical to those well established in the art of specific binding assays probed for a single analyte, including such assays as ELISA's, rapid filter-binding enzyme immunoassays, peroxidase-linked immunoblots and nucleic acid probe assays, and peroxidase-linked histochemical and cytochemical procedures. Such steps include immobilization of putative analytes from a test sample on a solid phase, fixation of histochemical or cytochemical test samples, preparation of analyte-specific probes, incubation of the solid phase containing the putative analytes with the probes, washing of the solid phase to remove probe not specifically associated with analyte, incubation of the test probed solid phase with a peroxidatic activity indicator solution, and recording of the peroxidatic signal.

The novel features in a multi-probed specific binding assay are the following: (a) the solid phase is incubated separately with more than one probe, each probe being specific for a different analyte; (b) after the peroxidatic signal generated by a single analyte-specific probe has been recorded, the solid phase to which that probe has been bound is incubated with effective concentrations of 4-chloronaphthol and hydrogen peroxide or urea hydrogen peroxide, for an interval of about one minute to about 10 minutes, at a temperature of between about 20C and about 60C; (c) after this suicide inactivation of the peroxidase signal generator, the solid phase is washed with an aqueous solvent to remove the inactivation reagents before exposure to the next probe; (d) apropriate measures are taken, as discussed below, to prevent interference of the peroxidatic signal from one probe with that from any other probe.

Preferred use concentrations are between about 10~⁵ M and 10~³ M for 4-chloronaphthol and between about 2 x 10~ M and 10~² M for hydrogen peroxide or urea hydrogen peroxide. Inactivation of peroxidatic activity is likely to be faster or more complete, the higher the temperature and the higher the suicide reagent concentrations. There is little advantage to incubating longer than 10 minutes, and the suicide reaction is largely completed within three minutes. There are few practical constraints with respect to other elements of the suicide substrate reagent, such as pH, ionic strength, or the choice of buffer or other components such as detergents and organic cosolvents. The peroxidatic activity indicator solution must be separated from the solid phase (e.g., by decantation or filtration) before adding the suicide reagent to prevent the chromogen from _^interfering with the inactivation reaction. Accordingly, the suicide reagent is most conveniently prepared at use concentrations and added to the solid phase after removal of the indicator solution. For this sixth aspect of the invention, it is preferred that the analyte-specific probe be coupled directly to peroxidase through a covalent bond. Use of two-step probing, such as serial incubation of the solid phase with an analyte-specific primary probe and an enzyme- tagged secondary probe specific for the primary probe, raises the possibility that later exposure of the solid phase to the secondary probe while probing for another analyte will result in linkage of peroxidase to unblocked sites on the primary probe introduced in an earlier probing. It is not always possible to assure that while two-step probing for one analyte, secondary probe will saturate all of the sites on primary probe to which it might bind. Such cross-talk between efforts to probe for different analytes can result in a false-positive signal for any analyte except the first one probed.

Use of 4-chloronaphthol as a peroxidase suicide substrate presents another potential source of false- positive results or high background signals. Peroxidatic oxidation of 4-chloronaphthol creates a water-insoluble blue dye which normally is deposited wherever peroxidatic activity is located on the solid phase. If an assay format is used in which the chromophoric product of peroxidatic catalyst action on the indicator solution is soluble in that indicator solution, the soluble peroxidatic signal can be recorded without interference from insoluble dye. If the chromophore is deposited on the solid phase, several tactics can minimize such interference. One is to perform the inactivation reaction at a mildly elevated temperature, between about 40C and 60C. Another is to use the lowest 4-chloronaphthol concentration which gives acceptably complete suicide. A third is to use a chromogen such as diaminobenzidine or aminoethylcarbazole, which generates a non-blue insoluble product distinguishable from the 4-chloronapthol product. However, these chromogens suffer practically from being potential carcinogens and from being less than maximally sensitive. Preferably, TMB, which is a safe and much more sensitive chromogen, will be used, its oxidation product being immobilized on the solid phase as an insoluble salt or complex ion as disclosed in U.S.

Patent Application No. 896,677, the disclosure of which is incorporated herein by reference. Although TMB oxidation - generates a blue chromophore, optimized peroxidatic assays using TMB generate blue color so much more efficiently than does 4-chloronaphthol oxidation that background from the suicide reaction is much fainter than the signal from analyte-associated TMB oxidation. Specific binding assays using TMB as chromogen can be adjusted to give a conveniently intense analyte-associated signal in any of several ways, including reduction of the size of the test sample used and reduction in the concentration or activity of probes used. Under such conditions, any additional.blue color from 4-chloronaphthol oxidation is likely to be invisible or barely visible. Furthermore, the analyte- specific signal of oxidized TMB can be distinguished from oxidized 4-chloronaphthol because it and only it can be washed from the solid phase by salt solutions, as described below.

Multi-probing generates one other practical concern not considered when a specific binding assay is probed for a single analyte. The peroxidatic signal created after probing for one analyte must not interfere with later efforts to detect other analytes. If the chromophoric product of the peroxidatic reaction is soluble in the indicator solution, it can be removed by decantation or filtration, preferably followed by at least one wash with an aqueous solvent, before continuing to the next probe. If the chromophore is insoluble, two tactics exist to resolve the signals associated with different analytes. Much less preferred is the approach of Geysen (supra) or Nakane (supra) , the disclosures of which are incorporated herein by reference, in which chromogens yielding different colored chromophores are used after probing for different analytes. Not only are these chromogens suboptimally sensitive and potentially toxic, but only a few non-blue colors may be used before interference among different probings arises. Furthermore, the colored products from oxidation of 4-chloronaphthol, diaminobenzidine, and a inoethylcarbazole are so insoluble than they can be cleared from the solid phase only by use of chemically harsh conditions. Much preferred is the use of TMB as a chromogen as disclosed in U.S. Patent Application Serial No. 896,677, where the chromophore is immobilized in ionic form, readily released from the solid phase by brief incubation with a chemically mild solution of a salt which does not precipitate the chromophore, such as 0.5 M sodium acetate. Washing the solid phase to remove the colored product of TMB oxidation can be done with the same solution used to wash away the suicide substrate reagent before incubation of the solid phase with the next probe. This way of avoiding interference among peroxidatic signals from different probes requires that any permanent record of the result from probing any analyte (except the last one probed) be separate from the test device (e.g., a photograph, photocopy, or notebook annotation). However, it allows a practically unlimited number of analytes from a single test sample to be probed in a single test device.

Test kits which utilize the methods or reagents as defined above are an important embodiment of all aspects of the invention. The kit may consist of the reagents or methods specified herein alone or more preferably as part of a larger set of reagents required in the peroxidatic assay for a particular analyte. For example, the kit may comprise any collection of apparatus, reagents and/or instructions designed to facilitate assay of a particular analyte of biological, clinical, or forensic interest.

The process of the present invention is further described by the following examples. The examples are provided for purposes of illustration only and are not intended to limit the invention in any manner.

EXAMPLE I

Effects of Hydroxylamine and of Catalase on Assay of Peroxidatic Catalysts Six peroxidatic activity indicator solutions were made. All contained 10^"2 M Na citrate, 10^"3 M NaCDTA, 0.4 mM TMB, and 3.0 mM H₂0₂. Three were adjusted to pH 4.00 with solid Na₂C0₃*H₂0 and concentrated HCl and also contained either no hydroxylamine (0M) or 10~³ M or 10~² M NH₂OH, supplied as the HCl salt. Three were adjusted to pH 5.00 and contained 10% l-methyl-2-pyrrolidinone, also containing either none or 10~³ M, or 10~² M NH₂OH. [The organic cosolvent is needed to maintain TMB solubility.]

Six peroxidatic catalyst preparations were tested for sensitivity to NH OH in the assay solution. Purified horseradish peroxidase (HRP) (Sigma Chemical Company, Type VI) was diluted to a concentration of 22 ng/ml (based on spectrophotometric analysis of more concentrated solutions, assuming an A₄₀₂ of 2.38 cm²mg^_1) in 10~³ M NaP0₄, 10"¹ M NaCl, pH 6.0. Purified horse metmyoglobin (Sigma Chemical Company Type I) was diluted to a concentration of 0.85 mg/ml or 0.17 mg/ l (assuming an A_4Q5 of 8.5 cm²mg^_1) in 10~² M Na citrate, pH 5.0. Purified human methemoglobin (Calbiochem) was diluted to a concentration of 0.32 mg/ml or 0.16 mg/ml (assuming an A_4Q5 of 10.3 cm²mg^_1) in 10^""2 M Na citrate, pH 5.0; the lyophilized protein did not dissolve completely, so that suspended material had to be removed by centrifugation before spectral assay and use. Hematin (Sigma Chemical Company) was dissolved in water by adding almost two equivalents of NaOH to create an aqueous solution which was approximately 10~² M (by weight). It was diluted to 5 x 10^"4 M in 10^"2 M Na citrate, pH 5.0. ΞDTA-stabilized whole human blood, stored at 4C, was diluted 1/300 or 1/900 into 10^"2 M Na citrate, pH 5.0 to serve as a source of unpurified reduced hemoglobin. [An experiment in Example 3 shows that leucocyte peroxidases contributed negligibly to its peroxidatic activity.] HL-60, a promyelocytic human leukemia cell line [ATCC No. CCL 240 was grown in RPMI 1640 medium (Moore and Woods (1977), Tissue Culture Association Manual 3, 503-509, incorporated herein by reference), supplemented with 2 mM L-glutamine, 10% fetal bovine serum, 10~² units/ml penicillin G_f 10~² ug/ml streptomycin, and 0.25 ug/ml a photericin B, to densities between 1 and 4 x 10*^> cells ml^-1 (determined by hemocytometer), harvested by centrifugation at 10³ rpm for 10 minutes, and resuspended at a density of 5 x 10⁷ cells/ml in either of two transport media designed to preserve Chlamydia organisms, one based on sucrose and phosphate, (2SP) and one based on sucrose, phosphate and glutamate (SPG) [Lennette et al. (1985), Manual of Clinical Microbiology, 4th Edition, American Society of Microbiologists, p. 1061 incorporated herein by reference]. Resuspended stored HL-60 cells were diluted 1/14 into deionized H₂0 to serve as a source of unpurified myeloperoxidase.

Peroxidatic assay was performed by diluting one of the six peroxidatic catalyst preparations 1/51 into one of the six indicator solutions at ambient temperature (24- 26C) and monitoring the increase in A_g52 (1 cm light path) for five minutes (ΔA₆₅₂/5 min.) in an Hewlett-Packard spectrophotometer, HP8450A. Although initial and final rate values were recorded for each kinetic trace, the traces were sufficiently curved (due to catalytic suicide in the assay cuvette) and the curvature was sufficiently dependent on catalyst identity and assay pH that ΔA_g52/5 min. was used as a general measure of peroxidatic activity. Table 1 summarizes the activities of the six peroxidatic catalyst preparations at the two pH values and three hydroxylamine concentrations. At pH 4, hydroxylamine up to 10~² M had no practical effect on the sensitivity of this assay substrate formulation to HRP, metmyoglobin, hematin, methemoglobin, or reduced hemoglobin. Usually there was a slight rise in activity between 0 and 10~³ M NH₂OH and a small drop in activity between 10~³ M and 10~² M NH₂OH. In contrast, the sensitivity to impure myeloperoxidase from HL-60 cells was actually increased several fold by hydroxylamine at pH 4. At pH 5, impure myeloperoxidase showed a two-fold activation between 0 and 10^"3 M NH OH and a 40% loss of activity between 10^""3 and 10~² M NH₂OH. The other five preparations showed modest (2-20%) drops in activity between 0 and 10^"3 M NH₂OH and much more significant inactivation (17-72%) between 10^{" 3} and 10^*"2 M NH₂OH. Hematin and metmyoglobin were the most sensitive to NH₂OH. The hemoglobin in whole blood was about half as sensitive as methemoglobin. Examination of the kinetic traces showed that for all six peroxidatic catalyst preparations, the bulk of the inhibitory effect of NΞ₂OH was due to increased catalytic suicide during the assay (curvature of the kinetic trace), not reduced initial catalytic activity. At both pH values, myeloperoxidase shov/sd negligible catalytic suicide in the absence of NH₂OH and significant suicide during assay at either NH₂OH conce^'ntration despite a large (6-9 fold) NH₂OH-induced increase in initial velocity.

These data show that despite reports that hydroxylamine inhibits HRP and myeloperoxidase [Keilin and Mann, supra; Lemberg and Legge, supra, disclosures incorporated by reference], one may, in fact, include hydroxylamine in indicator solutions for these peroxidatic catalysts without losing significant analytical sensitivity. At pH 4, NH₂OH concentrations as high as 10~² 5 M may be used, whereas at pH 5 optimal sensitivity is limited to the NH₂OΞ concentration of only about 10~³ M for hematin, metmyoglobin, and methemoglobin. HRP and reduced hemoglobin can be safely assayed at NΞ₂OΞ concentrations as high as 10~² M at pH 5. Hydroxylamine actually increases 10 sensitivity to myeloperoxidase several fold; this enhancement is especially dramatic for short assay intervals (1/4 minute-two minutes), before hydroxylamine- induced suicide becomes significant.

As a bonus, these data suggest two other 15 improvements which NH OH might make in peroxidatic assays. At pH 5, 10^""2 M NH₂OH improves the discrimination in favor of HRP or of reduced hemoglobin, as opposed to metmyoglobin or hematin in assays where the last two catalysts might provide interference. In addition, the 20 tendency of hydroxylamine to increase catalytic suicide during assay of all of these peroxidatic catalysts suggests that it might also enhance pre-assay catalytic suicide where inactivation of peroxidatic background is desired in peroxidase-linked specific binding assays.

25 The major improvement introduced by the inclusion of hydroxylamine in peroxidatic assay solutions is the elimination of potential interference from catalase in the test sample, which might otherwise consume hydrogen peroxide needed for the peroxidatic color-forming

30 reaction. The following experiment showing the effect of catalase with and without NH₂OH on peroxidatic assay of hemoglobin (Figure 1) illustrates such an improvement. A 1/300 dilution of whole human blood into 10^""2 M citrate, pH 5.0, was assayed by 1/51 dilution into 10^"2 M Na citrate,

35 10^"3 M NaCDTA, 10% l-methyl-2-pyrrolidinone, 0.4 mM TMB 3.0 mM H₂0₂, pH 5.0 containing no NH₂OH or 10^""3 M NH₂OH. In addition, 15 seconds before adding the hemoglobin preparation, a 1 mg/ml solution in 10~² M NaP0₄, 10~² M Na citrate, pH 5.0 of purified bovine liver catalase (Sigma Chemical Company) was diluted 1/101 into the assay solutions. Control assays omitted the catalase. The results are shown in Figure 1 where hemoglobin activity is- indicated; alone (trace B), in the presence of catalase (trace D), in the presence of 10~³ M NH₂OH (trace C) and in the presence of 10^""3 M NH₂OH plus catalase (trace A).

Figure 1 illustrates the ability of catalase to reduce the sensitivity of the indicator solution to hemoglobin and the ability of 10^""3 M NH₂OH to reverse this effect. For a five minute assay, this quantity of catalase reduced assay sensitivity by 70%. For longer assay intervals, the catalase effect would become more dramatic and the NH₂OH rescue would become even more beneficial, because of the difference in curvature of the kinetic traces. Lower concentrations of catalase were shown to produce less interference under these assay conditions, whereas higher concentrations produced more interference. Catalase interference became more significant as the assay pH was raised. Similar catalase interference and hydroxylamine reversal were demonstrated in peroxidatic assays of HRP, as ell as of hemoglobin.

Table 1

Effect of Hydroxylamine in the

Indicator Solution on Assay of

Peroxidatic Catalysts

.625 .614 .510

.796 .643 .191

Whole Blood (1/15300) (1/45400)

HL-60 (7 x 10⁴ cells/ml)

*Final concentrations in cuvette are indicated in parentheses: molarity of purified catalysts, dilution of whole blood, and HL-60 cell density. Methemoglobin molarity is in terms of subunits (MW 16,100).

EXAMPLE II

Inactivation of Peroxidatic Catalysts During Assay With TMB and Ξ₂0₂

A single peroxidatic activity indicator solution was made, containing 10^""2 M Na citrate, 10~³ M Na CDTA, 0.4 mM TMB, 3.0 mM H₂0₂, adjusted to pH 4.00 with concentrated HCl, and solid Na₂C0₃.H₂0. Six peroxidatic catalyst stock solutions were made as in Example I, except that the HL-60 cells were diluted 1/14 into 10^""2 M Na citrate, pH 5.0 instead of into H₂0. For each catalyst, 10 ul of stock solution was diluted into 500 ul of indicator solution, after which a kinetic trace at 652 nm, 1 cm light path, was recorded at 24-26C for five minutes. Figure 2 shows examples of these kinetic traces. The kinetic traces for hematin (trace I; vertical scale is 1.0 Abs. Units Full Scale (AUFS)), HRP (trace H; 1.5 AUFS), and HL-60 cells (unpurified myeloperoxidase) (trace J; 0.5 AUFS) showed very little curvature. The metmyoglobin trace (trace E; 0.5 AUFS) showed an approximately 90% drop in rate over the first 30 to 40 seconds (half-time of approximately eight seconds) and was essentially linear for the remaining 4 to 4.5 minutes. The methemoglobin trace (trace G; 1.0 AUFS) showed an approximately 90% fall in rate over three to five minutes, with a half-time of approximately 30 seconds. The whole blood kinetic trace (trace F; 1.5 AUFS) showed an approximately 90% fall in rate over three to five minutes with a half-time of approximately 40 seconds. This half- time of activity loss is referred to as the "rate- relaxation half-time". These kinetic traces were recorded with an

Hewlett-Packard spectrophotometer, HP84450A, which permits a one-step transformation of absorbance kinetic traces into graphs of rate (V) versus time. Figure 3 shows such graphs for the six catalysts. The traces for whole blood (F), methemoglobin (G), metmyoglobin (E), and HL-60 cells (J) are displayed at a vertical scale of 0.015 absorbance units/second full scale. The traces for HRP (H) and hematin (I) are graphed at threefold higher sensitivity, 0.005 absorbance units/second full scale. Even better than the absorbance traces, the rate traces demonstrate the potential for discriminating among the various peroxidatic catalysts which might be present in a test sample on the basis of the rate of inactivation of the catalyst during the peroxidatic assay reaction itself. In practice, reduced hemoglobin or methemoglobin are the two catalysts which one is most likely to want to detect in the presence of a background of other endogenous activities in a test sample. They also have the most distinctive kinetics of self-inactivation during assay with TMB and H₂0₂. The 30- 40 second rate relaxation half-times stand well apart from the approximately eight-second half-time for metmyoglobin and the negligible or relatively minor catalytic suicide of myeloperoxidase, hematin, and HRP. The unique biphasic kinetics of metmyoglobin, methemoglobin, and hemoglobin catalysis of TMB oxidation by H₂0₂ can be controlled by the following conditions in the assay solution: pH, identity and concentration of added detergent, and identity and concentration of added organic cosolvent. The hemoglobin kinetic trace in Figure 2 shows the characteristic biphasicity at pH 4 with a half-time near 40 seconds. The pH 5 hemoglobin kinetic traces in Figure 1 taken in the absence of catalase or with NH₂OH inhibition of catalase show much less curvature on the time scale of several minutes; the rate-relaxation half-time is three to four minutes. The metmyoglobin and methemoglobin oxidation kinetics show an analogous pH dependence; shift from pH 4 to pH 5 increases the half-time for activity loss from 8-30 seconds to over three minutes. For those pH 5 experiments, 10% N-methyl pyrrolidinone was used as a cosolvent in the indicator solution to improve TMB solubility. If 10% epsilon-caprolactam was used at pH 5 instead of N-methyl pyrrolidinone, the hemoglobin kinetic trace more closely approximated that seen at pH 4 without cosolvent.

The effect of detergent on the shape of the kinetic trace at pH 5 was studied in experiments identical in design to those performed in Example 1 in the absence of NH₂OH. Stock solutions of detergent in distilled water were diluted from 50 to 250 fold into the indicator solution immediately before diluting a stock solution of peroxidatic catalyst 50 fold into the same solution. Every detergent examined had the same effect at pH 5 as epsilon- caprolactam. Increasing the detergent concentration (in the presence of 10% N-methylpyrrolidinone cosolvent) caused a systematic drop in the rate-relaxation half-time for hemoglobin without affecting the catalytic rate at the end of the exponential loss of activity or the initial catalytic rate (extrapolated through the time of the mixing). At sufficiently high detergent concentration, the exponential loss of catalytic activity was completed within the time of mixing, so that only the lower-bound catalytic rate was seen in the spectrophotometer. Table 2 compares the detergents with respect to the concentration in the indicator solution required to reduce the hemoglobin rate- relaxation half-time from three to four minutes to 35-45 seconds. Although the effective concentrations range over two orders of magnitude, the detergents have identical effects at the indicated concentrations. Because all detergents tested so far have shown this effect, one skilled in the art could easily determine, using the methods of the instant invention, effective concentrations for any other detergent.

All of these detergents were tested for effect on the metmyoglobin-catalyzed kinetic trace for H₂0₂ oxidation of TMB at pH 5.0. Although their effects paralleled those on the hemoglobin trace in a semi-quantitative manner, there were significant quantitative differences which would affect the ability to resolve hemoglobin or methemoglobin from metmyoglobin at pH 5. [It is desirable to achieve a metmyoglobin rate-relaxation half-time no more than 1/3 as large as that for either hemoglobin species, and preferably to accelerate the metmyoglobin rate relaxation so much that it is complete within the mixing time of several seconds.] The detergents in Table 2 are ordered from most favorable to least favorable with respect to resolution of hemoglobin from metmyoglobin, with detergents below octyl- beta-D-thioglucopyranoside having a metmyoglobin rate relaxation which overlaps that of hemoglobin to an inconvenient degree. The five best detergents in Table 2 also were compared for effect on the peroxidatic assay of hematin, HRP, and methemoglobin. The effects on methemoglobin exactly paralleled those on hemoglobin, giving a rate-relaxation half-time 10-20 seconds smaller than that for hemoglobin. The effects on the hematin and HRP kinetic traces were slight to nonexistent; catalytic suicide with TMB and H₂0₂ was completely unaffected, so that kinetic trace shape can be used to discriminate hemoglobin and methemoglobin from hematin or plant peroxidases as well in detergent at pH 5 as in the absence of detergent at pH 4.

Kinetic traces for mixtures of hemoglobin and HRP containing approximately equal amounts of peroxidatic activity from the two catalysts could be visually discriminated from solutions lacking hemoglobin, especially when traces of rate versus time, like those in Figure 3, were compared. Mixtures containing smaller proportions of hemoglobin activity would require numerical analysis, for example by nonlinear least-squares curve fitting in a digital computer, to resolve the hemoglobin rate relaxation from the slower catalytic suicide of HRP. Because the hemoglobin rate-relaxation half-time can be varied continuously over one to two orders of magnitude by control of pH, detergent identity and concentration, and organic cosolvent identity and concentration, it should be possible to find an indicator solution composition vhich optimally resolves the catalytic suicide kinetics of hemoglobin and " methemoglobin from those of the other peroxidatic catalysts potentially present in samples being tested for the presence of occult blood or hemolysis.

Table 2

Effect of Detergents on the Kinetics of Hemoglobin- Catalyzed Oxidation of TMB by H₂0₂ at pH 5

Detergent Concentration Giving a Rate-Relaxation Half-Ti e

Detergent Half-Ti e of 35-45 Seconds Mass Percent Molarit

Sodium dodecyl sulfate

Zwittergent 3-12* Neodol 25-3S⁺

Cetyl trimethyl ammonium bromide

Octyl-beta-D-thioglucopyranoside

Zwittergent 3-16*

Cetyl pyridinium chloride Tween 20

Sodium taurodeoxycholate

Triton X-100 #

* Registered trademark of Cal biochem Behring

+ Registered trademark of Shell Chemical Company

# Registered trademark of Rohm and Haas

EXAMPLE III

Effect of Incubation Time, Hydroperoxide Concentration, and Temperature on Inactivation of Hemoglobin ^* by Tertiary Alkyl Hydroperoxides EDTA-stabilized human blood was gently agitated and then diluted 1/150 into a lysis solvent containing 10^"2 M Na citrate 10~³ M Na EDTA, 1% Tween 20, pH 5.0 plus one of the following conditions: nothing (control), 10 mM tertiary butyl hydroxide (t-BuOOH), 40 mM t-BuOOH, 5 mM cumene hydroperoxide (cumOOH), 20 mM cumOOH. Immediately after mixing, the dilution was incubated at 26C, 37C, or 49C, the latter two temperatures being set by a water bath. At approximately six minute intervals, starting three or four minutes after dilution, aliquots of the incubated dilution were diluted 1/51 into a peroxidatic activity indicator solution of the following composition: 10^"2 M Na citrate, 10^"3 M Na EDTA, 10% epsilon-caprolactam, 0.6 mM TMB, 3 mM H₂0₂, pH 5.0. The peroxidatic reaction, at 26C, was monitored spectrophotometrically (1 cm light path) at 652 nm for five minutes.

Hydroperoxide-containing lysates turned brown, then pale yellow during the 30 minute incubation period, at a rate which increased with temperature or hydroperoxide concentration. Control lysates remained bright red, regardless of incubation temperature. As long as detergent was included in the lysate, no precipitate formed during the incubation. Control and some four-minute assay kinetic traces were sharply biphasic, with an approximately ten¬ fold drop in rate over the first two minutes. At high degrees of inactivation, the assay kinetic traces were only slightly curved, with rates below that of the slow phase of the control assays. Figures 4 and 5 graph the inactivation time dependence for each of the incubation conditions of hydroperoxide concentration and temperature. Figure 4 illustrates t-BuOOH inactivation of the peroxidatic activity in whole blood as a function of time, temperature, and t-BuOOH concentration. The independent variable in Figure 4 is incubation time at specified condition before assay of peroxidatic activity. The dependent variable is total change in A_g52 over five minutes. The label of each kinetic trace shown in Figure 4 is assigned the following t-BuOOH concentration and incubation temperature.

Figure 5 shows cumene hydroperoxide inactivation of the peroxidatic activity in whole blood as a function of time, temperature, and cumOOH concentration. The independent and dependent variables are the same as in Figure 4. The label of each kinetic trace shown in Figure 5 is assigned the following cumOOH concentration and incubation temperature.

Tem erature

^kN/D:

The results graphically illustrated in the two figures lead to the following conclusions. (1) In the presence, but not in the absence, of hydroperoxide, the peroxidatic activity of hemoglobin in the lysates declined to almost zero in an approximately exponential manner on the time scale of 3-30 minutes. (2) The rate of activation increased directly with (and approximately proportionately to) hydroperoxide concentration.

(3) The rate of activation increased directly with temperature. (4) Cumene hydroperoxide was approximately twice as effective as t-butyl hydroperoxide on a molar basis.

Because whole blood includes peroxidase- containing leucocytes as well as hemoglobin-containing red cells, there was some concern about the identity of the peroxidatic catalyst(s) inactivated in Figures 4 and 5.

This question was answered by fractionating EDTA-stabilized whole human blood to separate red cells from white cells and then assaying each fraction for peroxidatic activity. Fourteen ml of whole blood was mixed at room temperature with 10 ml of a 2% solution of dextran in phosphate- buffered saline and allowed to stand 45 minutes at room temperature. The red cells agglutinated to form a pellet beneath a yellow supernatant containing the white cells and plasma. After removal of the supernatant, the red cells were resuspended once in 15 ml of phosphate-buffered saline and once in 7 ml of the same buffer, centrifuging at 1200 rpm for 10 minutes at room temperature after each resuspension, to wash out the dextran. Finally the red cells were resuspended in phosphate-buffered saline to a volume of 14 ml. This volumn was expected to contain the same hemoglobin concentration as the original blood sample. The yellow supernatant containing the white cells and plasma was centrifuged at 1200 rpm for 10 minutes at room temperature to pellet the white blood cells together with a similar small volume of red cells; the supernatant, containing plasma proteins, was discarded. The pellet was resuspended in 1 ml deionized Ξ₂0 to lyse the remaining red cells, then diluted with 15 ml of phosphate-buffered saline to block white cell lysis. A short centrifugation separated the white cells from the red hemoglobin released by red cell osmotic lysis. The pellet was resuspended in 0.3 ml of phosphate-buffered saline.

The two cell fractions were assayed for peroxidatic activity by the same method used for Figures 4 and 5, above. The red cells and a separate sample of unfractionated whole blood were diluted 1/150 into citrate- EDTA (no Tween 20) and then 1/51 into the indicator solution. The white cells were diluted 1/30 or 1/150 into citrate-EDTA or citrate-EDTA-1% Tween 20 and then 1/51 into the indicator solution. No organic hydroperoxide was present, so that the assay kinetic traces corresponded to the 26C control reactions of Figures 4 and 5. The red cell fraction and the unfractionated whole blood gave identical kinetic traces revealing the normal biphasicity. The white cell fraction gave 1-3% of the activity of red cells or whole blood, without the characteristic biphasicity. These data show that hemoglobin from red cells is responsible for essentially all of the peroxidatic activity of whole blood and therefore for the sensitivity which that activity shows toward organic hydroperoxides. Commercially supplied methemoglobin (Calbiochem) also gave a kinetic trace similar to that seen for whole blood or red blood cells when adjusted to approximately the same concentration, as judged by absorbance at 404 nm. However, its initial rate of TMB oxidation was only 2-3 times the rate seen after two minutes of reaction, whereas whole blood or red blood cells gave an initial/final rate ratio of 5 to 10.

Two kinds of control experiments helped to show that Figures 4 and 5 report on the kinetic control of hemoglobin inactivation by tertiary alkyl hydroperoxide in the incubation. To assure that the alkyl hydroperoxides were not simply inhibiting hemoglobin catalytic activity in the cuvette, 80 mM t-butyl hydroperoxide or 20 mM cumene hydroperoxide was diluted 1/50 into peroxidatic activity indicator solution immediately before addition at 1/51 dilution of the control whole-blood lysate, containing no hydroperoxide. In each case, the kinetic trace was superposable on that of a control lysate assay in the absence of added hydroperoxide. Therefore the action of the hydroperoxides on hemoglobin activity must have occurred before addition to the assay cuvette. This conclusion is also consistent with the facts that inactivation was strongly dependent on incubation temperature and time and that hemoglobin was discolored during incubation on the same time scale as the loss in activity.

To eliminate the possibility that the degree of inactivation of hemoglobin was stoichiometrically controlled by the amount of tertiary alkyl hydroperoxide (i.e., that lower molar ratios of hydroperoxide to hemoglobin would have given lesser degrees of inactivation and higher molar ratios would have given greater inactivation), inactivation incubations were performed at 26C in which EDTA-stabilized whole human blood was diluted either 1/30 or 1/300 into lysis buffers containing 0, 2.5, 5, 10, or 20 mM cumene hydroperoxide. Immediately before assay, aliquots of the 1/30 dilutions were diluted 1/10 into peroxide-free lysis buffer. Then samples experiencing these two dilution protocols were diluted 1/51 into peroxidatic activity indicator solution.

This experimental design assured that hemoglobin experiencing the two different dilution protocols was assayed at the same hemoglobin concentration and was inactivated by the same hydroperoxide concentration but at ten-fold different molar ratios of hydroperoxide to hemoglobin. The inactivation kinetics were almost independent of dilution protocol for the 10 mM cumOOH and essentially independent for the 20 mM cumOOH; in the latter case, the kinetics matched the 26C, 20 mM cumOOH data of Figure 5. For 5 mM cumOOH, the 1/300 dilution gave the same kinetics as the 26C, 5 mM cumOOH peroxide data of Figure 5 (which used a 1/150 dilution), whereas the 1/30 dilution clearly failed to approach complete inactivation because there was insufficient hydroperoxide to kill all of the hemoglobin. If the 1/30 dilution of whole blood into 10 mM cumOOH is taken as an approximate stoichiometric endpoint, it may be calculated that the hemoglobin in 1 ml of whole blood requires approximately 0.3 mmoles of cumOOH for complete inactivation at 26C. A similar stoichiometric endpoint was obtained for t-BuOOH.

As the incubations of Figures 4 and 5 contained 0.75-3 moles of cumOOH or 1.5-6 mmoles of t-butyl hydroperoxide per ml of whole blood, clearly they could not have been stoichiometrically limited in hydroperoxide. Therefore, similar inactivation kinetics can be expected in future reactions under these concentration and temperature conditions, as long as at least 0.3 mmoles of hydroperoxide are provided per ml of whole blood.

EXAMPLE IV Effect of Non-peroxide Catalase Inhibitors on Suicide Inactivation of Hemoglobin and Myeloperoxidase in Catalase-Containing Samples

The test samples consisted of EDTA-stabilized whole human blood and HL-60 cells at a density of approximately 2.5 x 10⁷ cells/ml in SPG transport medium (described in Example I). The cell lysis solutions for blood consisted of (a) 10~² M Na citrate, pH 5.0 (control); (b) 10^"3 M NaN₃, 10^""2 M Na citrate, pH 5.0; (c) 10^""2 M H₂0₂, IO^"2 M Na citrate, IO^"3 M Na EDTA, pH 5.0; (d) IO"² M H₂0₂, IO^"3 M NH₂0H^*HC1, IO^"2 M Na citrate, IO^"3 M Na EDTA, pH 5.0; and (e) IO^"2 M H₂0₂, IO^"3 M NH₂0H^**HC1, IO^"2 M "Na citrate, 10~³ M Na EDTA, pH 5.0. In most inactivation reactions, the whole blood was diluted 1/40 into a lysis solution at 25C to start the experiment. At 4 and 11 minutes after initial dilution, aliquots of the lysate were diluted 1/10 into 10~² M Na citrate, pH 5.0, and assayed ^" immediately at 25C for peroxidatic activity by 1/51 dilution into the following indicator solution: 0.4 mM TMB, 3.0 mM H₂0₂, 1 mM Na EDTA, 10 mM Na citrate, pH 4.0. Absorbance at 652 nm was recorded for five minutes with a 1 cm light path in a Hewlett Packard spectrophotometer, HP8450A. To test for the possibility that H₂0₂ stoichiometrically limited suicide inactivation in this double-dilution assay protocol, whole-blood inactivation incubations also were performed with single 1/400 dilutions into lysis solutions (c) and (d).

The cell lysis solutions for HL-60 cells consisted of (a) IO^"2 M Na citrate, 10^"1 M KBr, pH 5.0 (control); (b) 10~² M H₂0₂, 10^"2 M Na citrate, 10~³ M Na EDTA, 10^"1 M KBr, pH 5.0; (c) IO^"2 M H₂0₂, IO^"3 M NH₂OH'HCl, IO^"2 M Na citrate, IO^"3 M Na EDTA, 10^"1 M KBr, pH 5.0; and (d) IO^"2 M H₂0₂, 10~³ M NaN₃, 10~² M Na citrate, IO^"3 M Na EDTA, 10^"1 M KBr, pH 5.0. To start an experiment, HL-60 cells were diluted 1/10 into a lysis solution at 25C. Four and 11 minutes later, aliquots of lysate were diluted 1/10 into IO^"2 M Na citrate, pH 5.0 and assayed immediately for peroxidatic activity in the same manner as for whole blood except that the HL-60 indicator solution also contained 0.02 M KBr and 10~³ M NH₂OH*HCl.

Table 3 summarizes the suicide inactivation results for both HL-60 cells, where the peroxidatic catalyst was myeloperoxidase, and for whole blood, where the peroxidatic catalyst was reduced hemoglobin; the values in the table are percentages of control activity remaining after 11 minutes incubation in lysis solution, where activity was calculated as total absorbance change over five minutes, minus the same measurement made in assay solution to which no catalyst had been added. [This correction was needed because the cuvette walls contained sufficient catalytically active transition metal ion to give a ΔAgg₂/5 minutes of 0.015.] Table 3 also includes two corrections for the values of surviving peroxidatic activity of hemoglobin inactivated in the presence of NH OH and NaN₃. In the first case, H₂0₂ was shown to be stoichiometrically limiting catalytic suicide in the 1/40 x 1/10 dilution series, because a single 1/400 dilution (containing the same H₂0₂ concentration but ten times the molar ratio of H₂0₂ to hemoglobin) gave more complete inactivation. [In the absence of catalase inhibitor, something else, presumably catalase activity, limited catalytic suicide, because the surviving peroxidatic activity was slightly higher for the 1/400 dilution than for the 1/40 x 1/10 dilution.] In the case of NaN₃ and hemoglobin, comparison of lysis solutions (a) and (b) showed that even after double dilution into the cuvette, NaN₃ in the absence of H₂0₂ retained ability to inhibit hemoglobin peroxidatic activity approximately 55%. Therefore, the surviving peroxidatic activity for hemoglobin after incubation in H₂0₂ and NaN₃ was artifactually low by approximately 55%, if one wished to assess catalytic suicide during incubation independent of inhibition during assay. A correction for this effect was estimated by dividing the measured surviving activity by 0.45. Comparison of H₂0₂-free HL-60 lysis solutions with and without 1 mM NaN₃ showed that the double-dilution protocol sufficed to prevent azide interference with peroxidatic assay of myeloperoxidase. NH₂OH greatly increases the sensitivity of peroxidatic assays to myeloperoxidase (see Example I), and halide ion further enhances this effect.^' Therefore NH₂OH and KBr were included in the HL-60 indicator medium to prevent ςonfusion of HN₂OH effects on catalytic inactivation and on assay. Table 3 shows that both non-peroxide catalase inhibitors, hydroxylamine (actually the hydroxylammonium ion at pH 5) and sodium azide, dramatically improve the ability of H₂0₂ to inactivate hemoglobin and myeloperoxidase in test samples which also contain catalase. In the whole-blood dilutions, the suicide reaction could be followed visually by a color change from red to green to yellow-brown, completely absent if HN₂OH and NaN₃ were absent. In addition, omission of catalase inhibitor also resulted in visible effervescence in H₂0₂- containing lysis solutions, presumably representing 0₂ produced by H₂0₂ dismutation. The percentage of surviving activity seen with HL-60 cells diluted into H₂0₂ without catalase inhibitors varied as a function of cell storage. Fresher cells showed less suicide inactivation and more effervescence presumably because of greater catalase activity. Suicide inactivation of myeloperoxidase (a haloperoxidase) was greatly increased by inclusion of chloride or bromide ion in the lysis solution, with bromide being slightly more effective than chloride. Suicide inactivation of hemoglobin and myeloperoxidase by H₂0₂ appears to be slightly more efficient with NaN₃ than with NH₂OΞ when the two catalase inhibitors are used at the same concentration. However, NaN₃ exists significantly as hydrazoic acid (pKa 4.72) at the pH values used here for lysis and assay. Given the solubility and toxicity of HN₃, there is some preference for NH₂OH*ΞCl (pKa 6.00) in preparing safe and stable formulations at low pH. In addition, hydroxylamine is less inhibitory than azide to peroxidatic catalysts in low-pH assays, so that it is less likely to interfere with peroxidatic assay than is azide.

Table 3

Effects of Non-peroxide Catalase Inhibitors on H₂0 Suicide Inactivation of Hemoglobin and Myeloperoxidase

% Control Activit Remainin at 11 Minutes

Catalase Inhibitor:

Peroxidatic Catalyst Myeloperoxidase (HL-60 cells) Hemoglobin (whole blood) 1/40 x 1/10 dilution

1/400 dilution 93 0.9

* Corrected for 55% inhibition of hemoglobin peroxidatic activity by azide in the cuvette

EXAMPLE V

Effect of Chelators on the Peroxidatic Activity of Fe(III)

A peroxidatic activity indicator solution was prepared containing IO^"2 M Na citrate, 0.4 mM TMB, 3.0 mM H₂0₂, pH 4.0. Stronger chelators than citrate were omitted from this indicator solution to avoid interference with the chelators used to prepare the catalyst stock solutions. Stock 0.5 M FeCl₃ was mixed with water and 10^"1 M chelator stock solutions, pH 4-6, to prepare catalyst stock solutions which were 5 mM in FeCl₃ and 50 mM in chelator. The following chelators were processed in this manner: hydroxyethylethylene diamine triacetic acid (HEDTA), nitrilotriacetic acid (NTA), N,N'-ethylenediamine diacetic acid (EDDA), ethylene glycol-bis(beta- aminoethylether)N,N,N',N--tetraacetic acid (EGTA), N,N- dihydroxyethyl glycine (DΞG) ,3,3' ,3"-nitrilotripropionic acid (NTP), DTPA, citrate, TTHA, EDADP-alpha, EDTA, CDTA, EDADP-beta, and DFA. One chelator, EDDHA, was kept in an 0.05 M stock solution at pH 8.4, and was mixed with 0.5 M FeCl₃ without added water to create a catalyst stock solution which was 5.0 mM in Fe(III) and slightly less than 50 mM in EDDHA. Because acidification by FeCl₃ reduced EDDHA solubility, this stock solution had to be prepared immediately before use, as EDDHA precipitated from it after standing for a few hours at room temperature. All chelators except DFA and EDDHA were reagent-grade material, supplied by Sigma Chemical Company or Mallinckrodt, Inc. or in one case (DHG), Eastern Chemical Company. EDDHA, commercially supplied (Aldrich Chemical Co.) as a tan solid which gave at least 6 peaks by high-pressure hydrophobic interaction chromatography, was repurified by a variation of the procedure of Rogers [(1973) Infection and Immunity 7, 445-456, incorporated herein by reference, differing from the published method in that (a) the EDDHA was dissolved without boiling, adding concentrated HCl dropwise until complete dissolution was obtained; (b) IO^"3 M DTPA was added before neutralization; and (c) one volume equivalent of acetone, instead of 7.5 equivalents, was added. Repurified EDDHA showed only three trace impurities by high-pressure hydrophobic interaction chromatography monitored at 300 and 450 nm. DFA, supplied by CIBA-GEIGϊ under the registered trademark, Desferal, was used without further purification.

The peroxidatic activity of these chelated Fe(III) preparations was measured by diluting the stock catalyst 1/101, 1/51, or 1/26 into the peroxidatic activity indicator solution at 24C and monitoring the increase in Ag₅₂ (1 cm path length) for 5 or 10 minutes in an Hewlett- Packard spectrophotometer, HP8450A. The final concentrations in the cuvette were 10^"^ M for Fe(III) and IO^"3 M for added chelator (in addition to the 10~² M citrate used to buffer the indicator solution). These reactions were performed in disposable polystyrene cuvettes, which contained low enough levels of transition metal ions to give background activities (no added catalyst) of 1-4 x IO^"3 Ag₅₂/5 min. The background activity was checked before almost each reaction; if it exceeded 6 x 10~³ Ag₅₂/5 min. the cuvette was replaced. Glass/quartz cuvettes were unacceptable for measuring the low activities reported here, because they carried too high a level of catalytically active transition metal ions.

Table 4 summarizes the results. It shows that Fe(III) complexes of seven chelators gave significantly higher peroxidatic activities than citrate, taken here to represent the standard because it commonly is used to buffer solutions in the 4-6 pH range. Clearly, these seven chelators, three of which (HEDTA, EGTA, and DTPA) are at least hexadentate, form Fe(III) complexes with enhanced peroxidatic activity. At a 1:10 mole ratio of FeCl₃ to chelator, the activity differences among chelators are unlikely to be due to trace transition metal ion impurities in the chelators themselves. On the other hand, Fe(III) complexes of six chelators, EDADP-alpha, EDTA, and especially CDTA, EDADP-beta, EDDHA, and DFA, showed substantially lower peroxidatic activity than the citrate complex. These six chelators, and especially the last four, make attractive candidates for pre-treatment of test samples in assays where transition metal ions are not the analyte but might interfere with peroxidatic detection of the analyte. Such an analyte might be hemoglobin, itself a peroxidatic catalyst, or might be the subject of a peroxidase-linked specific binding assay. Table 4 shows that chelators are not generically suited for suppressing transition-metal-ion interference. Only some chelators show the claimed improvement, and some show surprising activation of peroxidatic activity. It is expected that screening of other chelators not listed in Table 4 would yield some which are less effective and some which are significantly more effective than citrate in suppressing the peroxidatic activity of Fe(III).

In one experiment, the Fe(III) 'HEDTA catalyst was added to the peroxidatic activity indicator solution after the latter had incubated five minutes with the totally inactive Fe(III)'DFA catalyst. Each catalyst contributed IO"³ M chelator and 10^"4 M Fe(III) to the indicator solution. Even though DFA serves to inactivate Fe(III) in this peroxidatic reaction, Fe(III)'HEDTA addition resulted in a significant initial reaction rate, 76% of that seen in the absence of DFA. This activity dropped to zero with a half-time of 45 seconds. The kinetics of activity loss should report on the dissociation of Fe(III) from Fe(III) 'HEDTA, a necessary step before DFA can bind Fe(III) to form an inactive complex. This experiment demonstrates the importance of incubating the test sample with chelator before assay rather than relying on chelator in the indicator solution to suppress the peroxidatic activity of transition metal ions. If transition metal ions are bound to moderately strong chelators, such as HEDTA, and the resulting complexes have measurable peroxidatic activity, suppression by an inactivating chelator, such as DFA, EDDHA, EDADP-alpha, CDTA, EDADP-beta, or EDTA, during the peroxidatic signal-generating reaction may be too slow to block transition-metal-ion interference.

Table 4

Effect of Chelator on Fe(III) Peroxidatic Activity at pH 4

Agg₂/5 min After Background Chelator Correction*

HEDTA 1.67 NTA 0.62

EDDA 0.12

EGTA 0.093

DΞG 0.088

NTP 0.064 DTPA 0.051 citrate 0.044

TTHA 0.026

EDADP-beta 0.017

EDTA 0.015 CDTA 0.007

EDADP-alpha 0.005

EDDHA 0.002

DFA^' 0.000

* rates for 10^"4 M chelated Fe(III) in the assay solution; background rates ranged from 0.001 to 0.004 EXAMPLE VI

Comparative Effects of Suicide Substrate Formulations on Six Peroxidatic Catalysts

The peroxidatic activity indicator solution was that of Example II. The six peroxidatic catalyst preparations were the following: 11 ug/ml HRP in IO^"3 M NaP0₄, 10^"1 M NaCl, pH 6.0; 8.5 mg/ml horse metmyoglobin in IO"² M Na citrate, pH 5.0; 3.2 mg/ml human methemoglobin in 10~² M Na citrate, pH 5.0; EDTA-stabilized whole human blood; IO^"2 M hematin in water augmented with enough NaOH to dissolve the hematin completely; and HL-60 cells at a density of 5 x 10^ cells/ml in SPG transport medium. They were procured and standardized as described in Example I. Four suicide substrate formulations were prepared: (1) 10 mM Na citrate, 1 mM Na EDTA, 1 mM H₂0₂, 1 mM NH₂OH ^• HCl, 0.05 mM 4-chloronaphthol, pH 5.0; (2) 10 mM Na 'citrate, 1 mM EDTA, 0.1M NaCl, 10 mM H₂0₂, 10 mM NH₂OH'HCl, pH 5.0; (3) 10 mM Na citrate, 1 mM NaEDTA, 40 mM t-butyl hydroperoxide, pH 5.0; and (4) 10 mM Na citrate, 1 mM NaEDTA, 20 mM cumene hydroperoxide, pH 5.0. The control (non-suicidal) diluent for the peroxidatic catalysts was 10~² M Na citrate, pH 5.0. At zero time in a suicide inactivation experiment, a peroxidatic catalyst preparation was diluted into a suicide substrate formulation or into control diluent. The dilution factors were 1/251 for HRP, 1/10 for metmyoglobin and methemoglobin, 1/301 for whole blood, 1/20 for hematin, and 1/14 for HL-60 cells. After 11 minutes incubation at room temperature (24-26C), 10 ul of diluted catalyst were assayed as in Example II. Percentage of control activity remaining at 11 minutes was calculated for each combination of peroxidatic catalyst and suicide substrate by division by the activity measured for a control dilution of the same catalyst, where activity was measured as the total increase in Ag₅₂/5 min. The results are summarized in Table 5. They suggest the following conclusions.

(a) 4-chloronaphthol plus H₂0₂ (protected by NH₂OH) are especially specific for inactivating HRP, destroying over 99% of its activity in 11 minutes under temperature and concentration conditions which spare 16-20% of hemoglobin and methemoglobin activity and 60-70% of myeloperoxidase and hematin activity. Metmyoglobin actually is activated somewhat. The suicide reagent concentrations are extraordinarily low.

(b) A ten-fold higher concentration of H₂0₂ (protected by a ten-fold higher concentration of HN₂OH) without 4-chloronaphthol spares over 50% of HRP activity and almost 90% of hematin activity, while inactivating hemoglobin, methemoglobin, and myeloperoxidse by 94-97%. Metmyoglobin, while more resistant-than hemoglobin, is significantly inactivated by the higher peroxide concentration. If NaCl had been omitted from this suicide substrate formulation, myeloperoxidase would have been spared to a significant degree, while the other catalysts would have shown similar inactivation behavior.

(c) Cumene hydroperoxide and t-butyl hydroperoxide show very similar suicide specificity, sparing HRP almost completely, activating myeloperoxidase somewhat, inactivating metmyoglobin 70-80%, and inactivating hemoglobin and methemoglobin 94-97%.

These results are characteristic of the incubation time and temperature and the reagent conditions used here. Greater inactivation was obtained by using longer times, higher temperatures, and/or higher concentrations. Mixtures of these three specificity classes of suicide substrate broadened inactivation specificity as would be predicted from Table 5. The greatest possible breadth usually is desirable for the first aspect of this invention. For the second aspect of the invention, the narrower specificity of the individual formulations is preferred. These data show how 4-chloronaphthol and H₂0₂ should essentially eliminate plant peroxidase activity in stool specimens while significantly sparing hemoglobin and methemoglobin (the targeted analytes), whereas the organic hydroperoxidases should practically eliminate hemoglobin and methemoglobin while almost completely sparing plant peroxidase activity. These particular suicide conditions are more equivocal in eliminating hematin and metmyoglobin interference. Myeloperoxidase is unlikely to interfere with occult blood tests, but it groups with HRP in its sensitivity to the organic hydroperoxides.

Table 5

Peroxidatic Suicide¹ Specificities of Substrate Formulations

Suicide Substrate: 4-chloronaphthol H₂0₂ + NaCl* cumene t-butyl

^{+ H}2°2* hydroperoxide hydroperoxide % Control Activit Remainin at 11 Minutes

VO oo

10

15

also containing NH₂0H

Example VII

Reprobing of a Western Transfer

Legionella, Chlamydia and Mycoplasma positive- control materials are prepared by suspending 10° colony forming units (CFϋ) of each organism per ml of PBS (3.0 mM KC1, 137.0 mM NaCl, 1.5 mM KH₂P0₄, 8.0 mM Na₂HP0₄). Chlamydia positive-control material is prepared by suspending IO⁶ plaque forming units/ml of PBS.

Monoclonal antibodies (Mabs) specific to the three infectious agents are obtained from commercial vendors. For example, Chemicon, 100 Lomita Street, El Segundo, CA 90245 supplies purified immunological reagents.

The monoclonal antibodies are conjugated with horseradish peroxidase (HRP) by means commonly practiced in the art. For example, the method for coupling IgG to HRP using maleimide and a thiol is described in Ishikawa, et al. (1983), J. Immunoassay, 4_:209-327, incorporated herein by reference.

Transfers are performed by a modification of the procedure of Burnette, et al. (1981), Anal. Biochem. 112:195-203, incorporated herein by reference. Sputum specimens and throat swabs as well as Legionella, Mycoplasma and Chlamydia positive control samples are brought to a concentration of 2% sodium dodecyl sulfate, 0.1 M dithiothreitol, 10% glycerol, and 0.1 M Tris-HCl, pH 6.8, with 0.01% bromophenol blue and heated to 100C for 5 to 10 minutes to completely liquify the sample.

After the heat treatment, these samples are applied to a 10 to 16% gradient polyacrylamide 14 x 14 cm slab gel, 0.1 % SDS, 0.1 M Tris-HCl, pH 8.85, that has a 4% stacking gel in 0.1% SDS and 0.1 M Tris-HCl, pH 6.8, and electrophoresed at a constant current of 25 mA (Laemmli, U.K. (1970), Nature, 227:680, incorporated herein by reference).

Antigens are transferred to a 0.45 micron nitrocellulose membrane in 20% methanol, 0.025 M Tris-HCl, 0.192 M glycine, pH 9.0 buffer under constant voltage of 40 volts for one hour. Nitrocellulose is blocked with 1 M glycine, 1% ovalbumin and 5% dry milk for 15 minutes at room temperature. After washing with PBS containing 0.1% Tween 20, the membrane is probed for one hour at 37C with the anti-Legionella HRP labeled Mab (0.1 ug/ml, 0.5 ml/cm² of membrane). Unbound Mab is washed off using PBS plus 0.1% Tween 20 (PBST).

Direct detection of the membrane bound probe is done by the color development method of Sheldon, et al. (1986) PNAS, j53_:9085, incorporated herein by reference. The membrane is washed with 2 x with 100 ml of CDB-C solution (CDB-C = 237 mM NaCl, 2.7 mM KC1, 1.5 mM KH₂P0₄, 8.0 mM Na₂HP0₄, pH 7.4, 5% v/v Triton X-100, 1 M urea, 1% dextran sulfate) for two minutes at room temperature. Color is developed by replacing the CDB-C solution with 100 ml of CDB-D solution (CDB-D solution = 100 mM sodium citrate, pH 5.0, 0.1 mg/ml 3,3' ,5,5'-tetramethylbenzidine), and adding 50 ul of 3% H₂0₂. After 30 minutes incubation in this peroxidatic activity indicator solution, the membrane is agitated in four 100 ml changes of deionized water and photographed.

To inactivate bound HRP the membrane is incubated in a solution containing 10~⁴ M 4-chloronaphthol, 10~³ M hydrogen peroxide, 10~² M Na citrate, 10~³ M EDTA, pH 5.0 for five minutes at room temperature. The membrane is next washed once in 0.5 M sodium acetate and three times in 100 ml of PBST. Next, the membrane is reprobed for one hour at 37C with anti-Chlamydia HRP-labeled Mab at the concentra¬ tion and amount used with the first probe. Unbound material is washed off and bound HRP detected in the manner described above for the initial probing.

After photographing the second signal, the membrane is incubated with the peroxidase inactivating solution, washed as described above, and reprobed a third time with anti-Mycoplasma HRP-labeled Mab again using the same concentration and amounts as the initial probing.

By comparing the photographs from all three probings it is possible to discern the presence of proteins from each pathogen individually on the same membrane.

The invention has been described in detail with particular reference to preferred embodiments thereof, but it will be understood that variations can be effected within the spirit and scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. A method for inactivating peroxidatic catalysts in a test sample comprising contacting said test sample with an effective concentration of an organic hydroperoxide and incubating the sample at an effective temperature for at least about one minute.

2. The method of claim 1, wherein said organic hydroperoxide is a tertiary alkyl hydroperoxide.

3. The method of claim 2, wherein said tertiary alkyl hydroperoxide is selected from the group consisting of cumene hydroperoxide (cumOOH), t-butyl hydroperoxide (t-BuOOH) and di-isopropylbenzene hydroperoxide.

4. The method of claim 1, wherein the concentration of said organic hydroperoxide contacting the test sample is between about 10~³ M and about 1 M and the temperature for incubating said sample is between about 0C and about 60C.

5. The method of claim 1 that further comprises contacting said test sample with effectives amounts of a non-peroxide catalase inhibitor and a hydroperoxide selected from the group consisting of a hydrogen peroxide and urea hydrogen peroxide.

6. The method of claim 5, wherein said non- peroxide catalase inhibitor is hydroxylamine.

7. The method of claim 5, wherein the concentration of said hydroperoxide in the test sample is between about 10~³ M and about 1 M, and the concentration of said non-peroxide catalase inhibitor in the test sample is between about IO^"4 M and about IO^"1 M.

8. The method of claim 5 further comprising contacting said test sample with a halide ion selected from the group consisting of chloride and bromide ion at a concentration of between about IO"³ M and about 1 M.

9. The method of claim 1, further comprising contacting said test sample with an effective chelating agent at a concentration of between about 10~⁵ M and about IO^"2 M.

10. The method of claim 9, wherein the chelating agent is taken from the group consisting of ethylenediamine-di(o-hydroxyphenyl-acetic acid)- (EDDHA), ethylenediamine tetracetic acid (EDTA), trans-1,2- diaminocyclohexane-N,N,N'N'-tetraacetic acid (CDTA), deferoxamine-B (DFA), ethylenediamine-N,N'-diacetic acid- N,N'-di-alpha-propionic acid (EDADP-alpha), ethylenediamine-N,N'-diacetic acid-N, '-di-beta-propionic acid (EDADP-beta), ethylenediamine-tetra-alpha-propionic acid (EDTP-alpha), ethylenediamine-beta-propionic acid (EDTP-beta), N,N^•-bis(2-hydroxybenzyl)ethylenediamine diacetic acid (HBED), and pyrophosphate (PP^).

11. The method of inactivating peroxidatic catalysts in a test sample comprising contacting said test sample with effective concentrations of a non-peroxide catalase inhibitor and a hydroperoxide selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide and incubating the sample at an effective temperature for at least about one minute.

12. The method of claim 11, wherein said non- peroxide catalase inhibitor is hydroxylamine.

13. The method of claim 11, wherein the concentration of the hydroperoxide contacting the test sample is between about 10~³ M and about 1 M, the concentration of said non-peroxide catalase inhibitor contacting the test sample is between about 10"⁴ M and about 10^"1 M, and the temperature for incubating said sample is between about 0C and about 60C.

14. The method of claim 11 further comprising contacting said test sample with a halide ion selected from the group consisting of chloride and bromide ion at a concentration of between about 10~³ M and about 1 M.

15. The method of claim 11 further comprising contacting said test sample with an effective chelating agent at a concentration between about 10~⁵ M and about IO^"2 M.

16. The method of claim 15, wherein the chelating agent is selected from the group consisting of EDDHA, EDTA, CDTA, DFA, EDADP-alpha, EDADP-beta, EDTP- alpha, EDTP-beta, HBED, and PP^.

17. The method of inactivating peroxidatic catalysts in a test sample, comprising contacting said test sample with an effective concentration of 4-chloronaphthol and a hydroperoxide selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide, and incubating the sample at an effective temperature for at least about one minute.

18. The method of claim 17, wherein the concentration of hydroperoxide contacting the test sample is between about 2 x 10~⁵ M and about 1 M, the concentration of 4-chloronaphthol contacting the test sample is between about IO^"5 M and about IO^"2 M, and the temperature for incubating said sample is between about 0C and about 60C.

19. The method of claim 17, further comprising contacting said test sample with an effective concentration of a non-peroxide catalase inhibitor.

20. The method of claim 19, wherein said non- peroxide catalase inhibitor is hydroxylamine.

21. The method of claim 19, wherein the non- peroxide catalase inhibitor concentration in the test sample is between about 10~⁴ M and about 10^"1 M.

22. The method of claim 17 further comprising contacting said test sample with a halide ion selected from the group consisting of chloride and bromide ion at a concentration between about IO^"3 M and about 1 M.

23. The method of claim 17 further comprising contacting said test sample with an organic hydroperoxide at a concentration between about IO^"3 M and about 1 M.

24. The method of claim 23, wherein said organic hydroperoxide is a tertiary alkyl hydroperoxide.

25. The method of claim 17 further comprising contacting said test sample with an effective chelating agent at a concentration between about 10~⁵ M and about 10^{" 2} M.

26. The method of claim 25, wherein said chelating agent is selected from the group consisting of EDDHA, EDTA, CDTA, DFA, EDADP-alpha, EDADP-beta, EDTP- alpha, EDTP-beta, HBED, and PP^.

27. A composition of matter comprising a test sample contacting an organic hydroperoxide, said organic hydroperoxide having a concentration greater than about 10^{" 3} M.

28. The composition of claim 27, wherein said organic hydroperoxide is a tertiary alkyl hydroperoxide.

29. The composition of claim 28, wherein said tertiary alkyl hydroperoxide is selected from the group consisting of cumOOH, t-BuOOH and di-isopropylbenzene hydroperoxide.

30. The composition of claim 27, further comprising an effective chelating agent at a concentration greater than about 10^"5M.

31. The composition of claim 30, wherein said chelating agent is selected from the group consisting of EDDHA, EDTA, CDTA, DFA, EDADP-alpha, EDADP-beta, EDTP- alpha, EDTP-beta, HBED, and PP^.

32. A test kit for performing a peroxidase- linked analyte-specific binding assay comprising a composition for reducing peroxidatic activity in a test sampl*e, said composition comprising an organic hydroperoxide, and instructions for the method of claim 1.

33. A composition of matter comprising a hydroperoxide selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide at a concentration greater than about 10~³ M, and a non-peroxide catalase inhibitor at a concentration greater than about

IO^"4 M.

34. The composition of claim 33, wherein the non-peroxide catalase inhibitor is hydroxylamine.

35. The composition of claim 33 further comprising a halide ion selected from the group consisting of chloride and bromide ion at a concentration greater than about 10~³ M.

36. The composition of claim 33 further comprising an organic hydroperoxide at a concentration greater than about 10~³ M.

37. The composition of claim 36, wherein said organic hydroperoxide is a tertiary alkyl hydroperoxide.

38. The composition of claim 37 wherein said tertiary alkyl hydroperoxide is selected from the group consisting of cumOOH, t-BuOOH and di-isopropylbenzene hydroperoxide.

39. The composition of claim 33 further comprising 4-chloronaphthol at a concentration greater than about 10~⁵ M.

40. The composition of claim 33, further comprising an effective chelating agent at a concentration greater than about 10~⁵M.

41. The composition of claim 40, wherein said chelating agent is selected from the group consisting EDDHA, EDTA, CDTA, DFA, EDADP-alpha, EDADP-beta, EDTP- alpha, EDTP-beta, HBED, and ^

42. A test kit for performing a peroxidase- linked analyte-specific binding assay containing a composition for reducing peroxidatic activity in a test sample, said composition comprising a non-peroxide catalase inhibitor and a hydroperoxide selected from the group consisting of hydrogen peroxide.

43. A composition of matter comprising a test sample contacting a hydrogen peroxide at a concentration greater than about 2 x 10~⁵ M and 4-chloronaphthol at a concentration greater than about 10~⁵ M.

44. The composition of claim 43, further comprising an effective chelator at a concentration greater than about IO^"5 M.

45. The composition of claim 44 wherein said chelating agent is selected from the group consisting of EDDHA, EDTA, CDTA, DFA, EDADP-alpha, EDADP-beta, EDTP- alpha, EDTP-beta, HBED, and PP^

46. The composition of claim 43 further comprising a halide ion selected from the group consisting of chloride and bromide ion at a concentration greater than about 10~³ M.

47. The composition of claim 43, further comprising an organic hydroperoxide at a concentration greater than about IO^"3 M.

48. The composition of claim 47, wherein said organic hydroperoxide is a tertiary alkyl hydroperoxide.

49. The composition of claim 48, wherein said tertiary alkyl hydroperoxide is cumene hydroperoxide or t-butyl hydroperoxide.

50. A test kit for performing a peroxidase- linked analyte-specific binding assay containing a composition for reducing peroxidatic activity in a test sample, said composition comprising hydrogen peroxide and 4-chloronaphthol.

51. A method of detecting the presence of hemoglobin as a test for blood or hemolysis in a test sample comprising the steps:

(a) contacting a first portion of said test sample with an organic hydroperoxide and incubating at an effective temperature for at least one minute; (b) assaying said treated first portion and an untreated second portion of said test sample with a chromogenic electron donor and an hydroperoxide to measure the remaining peroxidatic activity in each; and

(c) comparing the peroxidatic activity of said treated first portion and said untreated second portion, wherein said presence of hemoglobin is indicated by a higher level of peroxidatic activity in said untreated second portion.

52. The method of claim 51, wherein said organic hydroperoxide is a tertiary alkyl hydroperoxide.

53. The method of claim 52, wherein said tertiary alkyl hydroperoxide is selected from the group consisting of cumene hydroperoxide, t-butyl hydroperoxide and di-isopropylbenzene hydroperoxide.

54. The method of claim 51, wherein the concentration of said organic hydroperoxide contacting said treated first portion is greater than about 10~³ M and the incubation temperature is between about 0C and about 60C.

55. The method of claim 51, wherein an effective chelating agent is contacted with both portions from said test sample, said chelating agent having a concentration greater than about IO^"5 M.

56. The method of claim 55, wherein the chelating agent is selected from the group consisting EDDHA, EDTA, CDTA, DFA, EDADP-alpha, EDADP-beta, EDTP- alpha, EDTP-beta, HBED, and PP_j_.

57. The method of claim 51 wherein said test sample is selected from the group consisting of stool, urine, gastric juice, blood serum, cerebrospinal fluid, synovial fluid and blood plasma.

58. A kit for testing for the presence of blood or hemoglobin in a test sample comprising:

(a) one or more hydroperoxides, wherein at least one is an organic hydroperoxide;

(b) a chromogenic electron donor; and

(c) instructions based on the method of claim 51.

59. A method for detecting the presence of hemoglobin as a test for blood or hemoglobin in a test sample comprising the steps: (a) contacting said test sample with 4- chloronaphthol and a hydroperoxide selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide at an effective temperature for at least about one minute; and

(b) assaying the treated sample with a chromogenic electron donor and an hydroperoxide to measure peroxidatic activity wherein said hemoglobin is indicated by the presence of said peroxidatic activity.

60. The method of claim 59, wherein the concentration of said hydroperoxide is greater than about 2 x 10~⁵ M, the concentration of said 4-chloronaphthol is greater than about IO^"5 M, and the incubation temperature is between about 0C and about 60C.

61. The method of claim 59, wherein in step (a) said sample is contacted with chloride or bromide ion at a concentration greater than about IO^"3 M.

62. The method of claim 59, wherein in step (a) said sample is contacted with a non-peroxide catalase inhibitor at a concentration greater than about IO^"4 M.

63. The method of claim 62, wherein said non- peroxide catalase inhibitor is hydroxylamine.

64. The method of claim 59, wherein in step (a) said sample is contacted with effective chelating agent at a concentration greater than about 10^"5 M.

65. The method of claim 64, wherein said chelating agent is selected from the group consisting of EDDHA, EDTA, CDTA, DFA, EDADP-aiphό, £ ADP-beta, EDTP- alpha, EDTP-beta, HBED, and PP^

66. The method of claim 59, wherein the test sample is selected from the group consisting of stool, urine, gastric juice, blood serum and blood plasma.

67. A kit for testing the presence of blood or hemoglobin in a test sample comprising:

(a) hydrogen peroxide or urea hydrogen peroxide;

(b) 4-chloronaphthol;

(c) a chromogenic electron donor other than 4-chloronaphthol; and (d) instructions based on the method of claim 59.

68. A method for detecting the presence of hemoglobin as a test for blood or hemolysis in a test sample comprising:

(a) recording the spectrophotometric kinetic trace obtained when a peroxidatic activity indicator solution comprising 3,3'5,5'-tetramethylbenzidine (TMB) and hydrogen peroxide is contacted with a portion of the test sample.

(b) analyzing the kinetic trace for the presence of a transient exponential loss of catalytic activity with a rate-relaxation half-time about the same as that measured in the same peroxidatic activity indicator solution for a pure sample of hemoglobin or methemoglobin, wherein said hemoglobin is indicated by a half-time at least about three times larger than that for pure metmyoglobin and no larger than about one-third of the value measured for pure horseradish peroxidase (HRP) or hematin.

69. The method of claim 68, wherein the pH of the peroxidatic activity indicator solution is about - .

70. The method of claim 68, wherein the pH is greater than 4 and an effective concentration of an effective organic cosolvent is included in the indicator solution, such that the rate-relaxation half-times of hemoglobin and methemoglobin are at least about three times larger than that of metmyoglobin and no larger than about one third of the value measured for HRP or hematin.

71. The method of claim 70, wherein said organic cosolvent is epsilon-caprolactam.

72. The method of claim 71, wherein said indicator solution pH is about 5 and the cosolvent concentration is about 10% by mass.

73. The method of claim 68, wherein said indicator solution pH is greater than 4 and an effective amount of an effective detergent is used, such that the rate-relaxation half-time of hemoglobin or methemoglobin is at least about three times larger than that of metmyoglobin and no larger than about one third of the value measured for HRP or hematin.

74. The method of claim 73, wherein said detergent is selected from the group consisting of sodium dodecyl sulfate, Zwittergent 3-12, Neodol 25-3S, cetyl trimethylammonium bromide, and octyl-beta-D- thioglucopyranoside.

75. The method of claim 74, wherein the pH is about 5 and said detergent concentration in mass percent is about 0.0013 for sodium dodecyl sulfate, 0.068 for Zwittergent 3-12, 0.00068 for Neodol 25-3S, 0.0023 for cetyl trimethylammoniuπ- bromide, or 0.19 for octyl-beta-D- thioglucopyranoside.

76. The method of claim 68, wherein

(a) said kinetic trace is curve-fitted by a nonlinear least-squares method to an equation of the form, y=A+Bt+C(1-e)~^Dt+E(l-e^"Ft)+G(l-e~^Ht); (b) the magnitude of the fitted parameter, E, is compared to the magnitude of its standard deviation; and

(c) the presence of hemoglobin is defined as being indicated by a value of E significantly larger than its standard deviation.

77. The method of claim 76, wherein the parameters, D, F, and H, are fixed at the values of the respective rate-relaxation rate constants measured for pure samples of metmyoglobin, hemoglobin or methemoglobin, and HRP or hematin.

78. The method of claim 76, wherein (a) said kinetic trace for the peroxidatic activity of said test sample is edited to exclude time points within the interval required for the kinetic trace of an independently assayed sample of pure metmyoglobin to substantially complete its exponential decay, and (b) the parameter, C, is fixed at a value of 0.

79. An indicator solution for peroxidatic activity comprising hydroxylamine, a hydroperoxide and a chemical selected from the group consisting of a chromogenic or fluorogenic electron donor.

80. The solution of claim 79, wherein said chromogenic electron donor is selected from the group consisting of 3,3',5,5'-tetramethylbenzidine (TMB), benzidine, diaminobenzidi e (DAB), 2,2'-azino-di(3-ethyl- benzthiazoline sulfonόte) (ABTS), o-phenylenediamine (oPD), 3,3'dimethylbenzidine (o-tolidine), 3,3'- dimethyloxybenzidine (o-DAD), 2,7-diaminofluorene (DAF), and aminoethyl carbazole (AEC).

81. The solution of claim 79 wherein the concentration of said hydroxylamine is between about 10~⁴ M and 10^"1 M, the concentration of said chromogenic or fluorogenic electron donor is greater than about 10^"^ M, and the concentration of said hydroperoxide is greater than about 10~⁴ M.

82. The solution of claim 79, wherein said hydroperoxide is selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide.

83. The solution of claim 79, wherein said chromogenic electron donor is selected from the group consisting of benzidine, TMB, DAB, o-tolidine, o-DAD, DAF, further comprising a chelating agent taken from the group consisting of DFA, EDTA, EDDHA, EDADP-alpha, EDADP-beta, CDTA, EDTP-alpha, EDTP-beta, HBED, and PP^, wherein the chelator concentration is between about 10~⁵ M and about IO^"2 M.

84. The solution of claim 79, wherein the pH is between about 3.5 and about 5.5.

85. A method of stopping the assay of the peroxidatic activity of a plant peroxidase, comprising contacting said peroxidase with an effective amount of 4- chloronaphthol in the presence of an effective amount of hydrogen peroxide or urea hydrogen peroxide for an interval of at least about one minute.

86. The method of claim 85, wherein the plant peroxidase is horseradish peroxidase.

87. The method of claim 85, wherein the 4-chloronaphthol concentration in contact with said peroxidase is between about IO^"5 M and 10~³ M and said hydrogen peroxide or urea hydrogen peroxide concentration in contact with said peroxidase is between about 2 x IO^"5 M and about 10~² M.

88. The method of claim 85, wherein an aqueous mixture of said 4-chloronaphthol and said hydrogen peroxide is contacted with said peroxidase, said peroxidase being bound to a solid phase from which a peroxidatic activity indicator solution has been removed.

89. A method of stopping' an assay for the , peroxidatic activity of a peroxidatic catalyst selected from the group consisting of hemoglobin and methemoglobin comprising contacting said peroxidatic catalysts with a stopping reagent selected from the group consisting of a tertiary alkyl hydroperoxide, a combination of hydrogen peroxide and non-peroxide catalase inhibitor, and a combination of urea hydrogen peroxide and a non-peroxide catalase inhibitor.

90. The method of claim 89, wherein said tertiary alkyl hydroperoxide is selected from the group consisting of t-BuOOH and cumOOH and di-isopropylbenzene hydroperoxide.

91. The method of claim 89, wherein the non- peroxide catalase inhibitor is hydroxylamine.

92. The method of claim 89, wherein said stopping reagent is a tertiary alkyl hydroperoxide at a concentration between about IO^"3 M to about 10^"1 M.

93. The method of claim 89, wherein the

5 concentration of said hydrogen peroxide or urea hydrogen peroxide is between about 10~³ M and about 10^"1 M, and the concentration of said non-peroxide catalase inhibitor is between about IO^"4 M and about IO"² M.

94. The method of claim 89, wherein said assay ° involves sequentially the binding of said peroxidatic catalyst to a solid phase; the contacting of a peroxidatic activity indicator solution with said solid phase; and the contacting of said stopping reagent with said solid phase.

95. The method of claim 89, wherein said assay 5 involves sequentially reacting said peroxidatic catalyst with a peroxidatic activity indicator solution to produce a colored oxidation product of a chromogenic electron donor, binding said product to a solid phase, removing said peroxidatic catalyst and contacting said solid phase with ₀ said stopping reagent.

96. A method of analyzing for the presence of a plurality of analytes contained in a test sample, said method comprising the steps:

(a) contacting said test sample with a solid 5 phase wherein said analytes, if present in the test sample, are bound to the solid phase;

(b) contacting the solid phase produced in step (a) with a probe comprising a binding moiety conjugated to one or more molecules of a plant peroxidase, wherein said 0 binding moiety specifically binds a unique epitope of said plurality of analytes; (c) contacting the solid phase produced in step

(b) with an aqueous solvent to remove unbound probe;

(d) contacting the solid phase produced in step

(c) with a peroxidatic activity indicator solution comprising a chromogenic electron donor in combination with a hydroperoxide to cause the oxidation of said chromogenic electron donor to a chromophoric product;

(e) detecting the amount of said chromophoric product produced in step (d); (f) contacting the solid phase produced in step

(d) with a peroxidatic activity stopping solution, comprising 4-chloronaphthol in combination with a hydroperoxide selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide, for an interval of between about 1 minute and about 10 minutes;

(g) contacting the solid phase produced in step (f) with an aqueous solution to remove unreacted stopping solution; and

(h) repeating the sequence of steps (b) through (g), each time using a probe containing a binding moiety which binds specifically to a different epitope from any that is recognized by a probe used earlier in the method, wherein steps (f) and (g) may be omitted from the sequence only after the last probe has been used.

97. The method of claim 96, wherein said plant peroxidase is horseradish peroxidase.

98. The method of claim 96, wherein said chromophoric product is soluble in said indicator solution, and said indicator solution is removed from said solid phase after step (d) and before step (f).

99. The i.iethod of claim 96, wherein said peroxidatic activity indicator solution comprises TMB in combination with a hydroperoxide selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide, wherein the chromophoric product of TMB oxidation is immobilized on said solid phase as an insoluble salt or in a complex ion also containing a polymeric anion, and wherein said chromophoric oxidation product is dissolved and removed from said solid phase by contacting said solid phase with an aqueous salt solution after step (e) and before step (h).

100. The method of claim 99, wherein said peroxidatic activity indicator solution further comprises the anionic conjugate base of an aromatic an unsaturated aliphatic acid.

101. The method of claim 99, wherein said polymeric anion is selected from the group consisting of dextran sulfate, polyphosphate, polyanethole sulfonate, carboxymethyl cellulose, polyacrylic acid, sulfoethyl cellulose, and polystyrene chemically modified to contain sulfate, sulfonate, or carboxylate groups, wherein said polymeric anion is bound to said solid phase at any point before step (d) through the process of first contacting said solid phase with an aqueous solution of said polymeric anion for an interval of at least about one minute, followed by contacting said solid phase one or more times with an aqueous solvent effective in removing unbound polymeric anion from said solid phase.

102. The method of claim 99, wherein said aqueous salt solution used to dissolve said chromophoric product of TMB oxidation contains a salt selected from the group consisting of ammonium acetate, sodium acetate. dipotassium monohydrogen phosphate, monosodium dihydrogen phosphate, potassium chloride and sodium chloride, said salt is having a concentration between than about 0.05 M and about 1 M.

103. The method of claim 96, wherein said peroxidatic activity stopping solution comprises 4-chloronaphthol at a concentration between about 10~⁵ M and about 10~³ M and a hydroperoxide selected from the group consisting of hydrogen peroxide and urea hydrogen peroxide at a concentration between about 2 x 10~⁵ M and about IO^"2 M.

104. The method of claim 96, wherein said analytes are antigens and said binding moieties are antibodies or antibody fragments.

105. The method of claim 96, wherein said analytes are antibodies and said binding moieties are antigens which bind specifically to those antibodies.

106. The method of claim 96, wherein said analytes are carbohydrates and said binding moieties are lectins which bind specifically to those carbohydrates.

107. The method of claim 96, wherein said analytes are sequences of DNA or RNA and said binding moieties contain sequences of DNA or RNA complementary by Watson-Crick base-pairing to those analyte sequences.

108. The method of claim 96, wherein the probe of step (b) is replaced by a pair of probes, one of which is a primary binding moiety specific for said epitopes and the other is a secondary binding moiety which binds specifically to said primary binding moieties, wherein said binding moiety is attached to one or more molecules of a plant peroxidase.

109. The method of claim 108, wherein said analytes are antigens and said primary probes are antibodies or antibody fragments.

110. The method of claim 109, wherein said secondary probe is an antibody or antibody fragment.

111. The method of claim 109, wherein said primary probe is an antibody and said secondary probe is Protein A or Protein G from Staphylococcus species.

112. The method of claim 108, wherein said primary probe is covalently attached to one or more biotin molecules and said secondary probe is selected from the group consisting of avidin, streptavidin, and an antibody or antibody fragment which binds specifically to biotin.

113. The method of claim 108, wherein said primary probe is covalently attached to one or more molecules selected from the group consisting of avidin, streptavidin, and an antibody fragment which binds specifically to biotin, and wherein said secondary probe is biotin.

114. The method of claim 108, wherein said primary probe is covalently attached to one or more molecules of fluorescein and said secondary probe is an antibody or antibody fragment which binds specifically to fluorescein.

115. The method of claim 108, wherein said primary probe is covalently attached to one or more molecules of an antibody or antibody fragment which binds specifically to fluorescein and said secondary probe is fluorescein.

116. A method for inactivating peroxidatic catalysts in a test sample, comprising contacting said test sample with an amount of a chelator effective to reduce the peroxidatic activity of ferric ion by at least about 90%, for an interval of at least about one minute.

117. The method of claim 116, wherein said chelator is selected from the group consisting of DFA, EDTA, EDDHA, EDADP-beta, EDTP-alpha, HBED, CDTA, EDTP- alpha, EDTP-beta and PP^, and wherein the chelator concentration lies between about 10~-⁵ M and about 10~²,M.