WO1994017412A1

WO1994017412A1 - Assay method

Info

Publication number: WO1994017412A1
Application number: PCT/GB1994/000142
Authority: WO
Inventors: Claire Elizabeth Lewis; James O'donnell Mcgee; Anthony Meager
Original assignee: Isis Innovation Limited; National Institute For Biological Standards And Control
Priority date: 1993-01-27
Filing date: 1994-01-26
Publication date: 1994-08-04

Abstract

An assay method uses cells (16) or liposomes containing a predetermined reagent to amplify the signal generated by an analyte (10). The analyte is incubated with the cells (16) or liposomes to form analyte-modified cells or liposomes, which are separated from unmodified cells or liposomes and lysed to release the reagent into solution. The reagent is assayed as a measure of the analyte (10). Preferably the cells (16) are red blood cells and the reagent is haemoglobin (26). The amplification results in an assay of improved sensitivity.

Description

ASSAY METHOD

There is a continuing need for improved assays for the detection and quantitation of analytes of biological origin. Various competition assay formats are well known, involving the use of labelled reagents designed to emit radioactive signals, or the use of enzymes and their substrates to show colour change, chemiluminescence or fluorescence. Signalling systems based on enzymes are popular and have the advantage of avoiding the use of radioactive species. One example of an assay format is the enzyme-linked immunosorbant assay (ELISA) . Assays of this kind are typically capable of detecting and quantitating analyte concentrations down to the picogram range (i.e. > 10^~12 g/ml) .

This invention aims to provide assays of improved sensitivity. In one aspect it makes use of the phenomenon of complement-induced lysis of cells.

This phenomenon has been known since the 1930s as part of the body's immune system. The complex mixture of proteins called complement is capable of lysing cells when, and only when, these cells carry antigen/antibody complexes on their surfaces. Cells without the antigen/antibody complex are not lysed.

Use of this phenomenon was made in the 1970s in the Haemolytic Plaque Assay¹. In this, lymphocytes are taken from animals that have been immunised against sheep red blood cells. These lymphocytes are incubated in a monolayer with sheep red blood cells (SRBC) and complement. Anti-SRBC antibody secreted by a lymphocyte binds to SRBCs in the vicinity, which are then lysed by the complement resulting in the formation of a clear spot or plaque in an opaque layer of SRBC. A corresponding Reverse Haemolytic Plaque Assay (RHPA) was developed in the 1970s². This involves incubating antigen-secreting cells in a monolayer with protein A-coated SRBCs in the presence of a specific antiserum and complement. Since the size of the haemolytic plaques formed in these assays is directly proportional to the amount of product secreted per single cell, the techniques have been used to quantify the amount of a product secreted by individual cells. An immunoassay for an antigenic analyte has been described³'⁴, which involves the use of red blood cells coated with an antibody to the analyte. A sample containing the analyte is incubated with the coated red blood cells and with complement, which lysis those red blood cells to which molecules of analyte have become attached, releasing the contents of those cells for further analysis. A disadvantage of the method is that complement-induced lysis is not in practice confined to those cells to which analyte has become attached. As a result, the assay is not very specific or sensitive. The present invention provides a method of assaying for an analyte, which method uses cells or liposomes containing a predetermined reagent and comprising the steps of i) Incubating the cells or liposomes with the analyte so as to bind molecules of the analyte to the surface of the cells or liposomes, thereby providing analyte-modified cells or analyte-modified liposomes, ii) Separating the analyte-modified cells or analyte-modified liposomes from cells or liposomes not so modified, iii) Lysing the analyte-modified cells or analyte- modified liposomes to thereby release the reagent into solution, iv) Assaying the reagent released into solution as a measure of the analyte. The analyte may be any molecule of biological interest. Examples are such proteins as enzymes, cytokines, hormones, etc., lipids, polysaccharides, nucleic acids, steroids, and drugs. Some of these analytes may need to be labelled with a reporter molecule to which an antibody can be raised. When the analyte is a hapten or antigen, step i) may be performed by incubating the analyte, with its associated antibody and with cells or liposomes whose surfaces are coated with protein A, these two reagents being used together or sequentially. Alternatively, the antibodies may be conjugated directly to the surfaces of the cells or liposomes, thus avoiding the use of protein A. Alternatively, use may be made of the biotin- avidin or biotin-streptavidin system. These can have the advantage of avoiding any risk of interference by endogenous antigen or antibody in the (body fluid) sample. Thus for example, an analyte (an antigen or hapten) may be incubated with its biotin-labelled antibody; and with streptavidin-labelled cells or liposomes. Or the analyte may be incubated with its streptavidin-labelled antibody; and with biotin- labelled cells or liposomes. Or a double antibody system may be used (as in the protocol below) in which a universal second antibody is e.g. biotin-labelled.

The cells or liposomes are used as amplification vehicles. They contain substantial quantities of a predetermined reagent, which after being released into solution, is readily assayed. Thus, cells or liposomes of various types may be artificially loaded up with a reagent; such as a β- emitting radionuclide or one which emits Auger electrons and is detectable only in solution; or an enzyme such as peroxidase or alkaline phosphatase, or a component of the enzyme system. Such loading may be effected in known manner. The surfaces of the cells or liposomes may be coated with protein A, so as to be capable of being lysed by complement, when an antigen/antibody complex is present. In aspects of the invention described and exemplified below, the cells are red blood cells and the predetermined reagent is haemoglobin. This represents a preferred embodiment. But other cells (or liposomes) that can be lysed, by complement or otherwise are possible; and other reagents, naturally present or artificially introduced into them, can be provided for assay in solution in step iii) .

For example it is possible to use large vesicles formed from fragments of red blood cell membranes (resulting from lysis of the red blood cells) . These are readily formed and can be reconstituted around suitable reagents such as enzymes e.g. peroxidase, alkaline phosphatase or β- glucoronidase. Following lysis of these vesicles, the released enzyme is readily detected by addition of a substrate and other reagents in known manner. The term "cells" is hereinafter used, where the context permits, to include liposomes and vesicles of the kind described. The analyte may be present in a fluid, for example a body fluid such as serum or urine, or another fluid such as cell culture supernatant, water or milk. In this case, a sample of the fluid may be incubated with the antibody to the analyte, and with protein A- coated red blood cells, all in solution. The protein A binds to the Fc portions of the antibodies, thereby providing analyte-modified cells in solution having the structure:- cell - protein A - antibody - analyte Use of the biotin-streptavidin system may provide analyte-modified cells in solution having the structure:- cell - streptavidin - biotin - antibody - analyte.

In step ii) the analyte-modified cells or analyte-modified liposomes are separated from cells or liposomes not so modified. Separation may be achieved, for example, by precipitating out of solution either the analyte-modified cells or liposomes or alternatively the unmodified cells or liposomes. Techniques for effecting such selective precipitation are well known in the art.

Alternatively, the analyte to be assayed may be immobilised, e.g. by being bound to antibodies which are bound either to a membrane filter or to particles or to the wall of a plastic or a glass reaction well. In this case, the solid-phase analyte may be incubated with protein A-coated cells, thereby providing solid- phase analyte-modified cells of the above structure. Cells that have not been modified may be removed by washing. This technique has the advantage of avoiding possible interference by other components, e.g. endogenous antibodies or complement, present in a fluid sample.

Alternatively again, the biotin-avidin or biotin-streptavidin system described above may be used with the analyte in an immobilised state.

When the analyte is a nucleic acid, the following protocol is possible. The target nucleic acid analyte is immobilised in any desired manner, e.g. on a nylon membrane. A biotin-labelled probe, complementary to a sequence of the target analyte, is caused to hybridise to the immobilised target. Streptavidin-coated red blood cells are then used to create an immobilised complex having the linear structure Streptavidin-labelled red blood cells - biotin-labelled probe - immobilised target sequence. The unreacted streptavidin-coated red blood cells are removed by washing. The immobilised red blood cells are lysed to release haemoglobin (or other reagent) into solution. In step iii) the analyte-modified cells are lysed to release the reagent contained therein into solution. This may involve incubating the analyte- modified cells with a solution of complement, under condition to lyse the cells and release the reagent contained therein into solution. The term complement is here used to include, not only the naturally occurring system of proteins, but any system which is capable of lysing the cells, by the classical pathway or the alternative pathway, when but only when the surfaces of those cells carry linkages that include antibody-antigen complexes. Enough complement should be used, and the reaction conditions should be such as, to ensure substantial or complete lysis of the analyte- modified cells. Other forms of lysis are possible. A simple freeze-thaw technique can then be used to lyse the cells and release their contents into solution. This alternative has the advantage that the use of complement is not required. Alternatively lysis can be effected by any other chemical or physical technique, such as for example by the addition of saponin.

When the analyte is a hapten or antigen, the amount of antibody used in step i) needs to be at least sufficient to bind with all the analyte present, but is otherwise not critical. Protein A-coated red blood or other cells are easily prepared by published methods. Generally, sheep red blood cells are used, although this is not critical. Cells labelled with biotin or avidin or streptavidin are also available or readily made by standard methods. The concentration of cells also appears not critical, but i<= preferably from 1% to 25 % ( v/v ) .

It appears that complement-induced lysis of a cell can only take place when there are several antigen-antibody complexes attached to the surface of the cell. The minimum required number of such complexes is not known. For example, lysis of one red blood cell carrying 100 antigen-antibody complexes (and thereby carrying 100 analyte molecules) on its surface, would typically release about 100 million molecules of haemoglobin into solution. These figures suggest a possible amplification of the analyte signal by a factor of about a million.

As few as 2-10 complexes on the surface of the cells could, in theory, induce complement-mediated lysis of each cell. Even if more complexes are required since the antibodies used in this method are usually polyclonal ones (and, therefore, capable of binding to multiple sites on each analyte) , it is possible that a single analyte molecule may be bound to more than one antibody molecule (which could theoretically be on the same and/or separate cells) . This means that if the concentration of cells is sufficiently high to allow such 'sharing^' of analyte- antibody molecules by cells to occur, the number of analyte molecules might not be a limiting factor.

Diluting red blood cells (RBC's) from 1% to 25% (v/v) (equival. to 2.5 to 62.5 x 10⁸ cells/ml) has been tried successfully and the assay seems to be more reproducible at the higher rather than the lower concentrations of RBC^'s.

In a variant of the method, steps i), ii) and iii) can be repeated two or even more times. Thus, the reagent released into solution in step iii), is used as the analyte in a repeat of steps i) and ii) to cause lysis of a fresh set of cells and the release into solution of a larger amount of a second reagent (which may be the same as or different from the first one) . In this way further amplification is achieved, and the sensitivity of the assay method is further improved. Step iv) involves assaying the reagent released into solution. This assay may be performed by any of a variety of well established techniques, depending on the nature of the reagent. For haemoglobin, the enzyme-linked immunosorbant assay (ELISA) is one example. The ELISA is based on the 'antibody sandwich principle^' involving the following steps. First, a range of haemoglobin standards and test samples are incubated with a solid-phase monoclonal antibody usually precoated on microtitre wells. The solid-phase bound protein is then incubated with a polyclonal antibody raised against a different epitope on the haemoglobin; a third antibody - a polyclonal raised against the second polyclonal antibody - can be used to amplify the reaction. The last of these polyclonal antibodies is usually conjugated directly to an enzyme (e.g. peroxidase or alkaline-phosphatase) ; or to biotin which is then subsequently detected using a streptavidin-enzyme conjugate. The enzyme substrate and a chromogen are then added and the resulting colour development is directly proportional to the amount of haemoglobin present in the sample and standards. Absorbance values are read using a conventional ELISA reader and the concentration of haemoglobin in test samples is then determined by comparing their respective absorbances with those obtained for the standards plotted on a standard curve.

The method of this invention can be performed as a qualitative assay for the analyte, by detecting the presence or absence of reagent released into solution. Alternatively, the method can be performed as a quantitative assay for analyte, by measuring the concentration of reagent released into solution, and comparing that measured concentration to a standard curve obtained using samples containing known concentrations of analyte. Reference is directed to the accompanying drawings in which:

Figure 1 is a diagrammatic summary of the method of the invention; and

Each of Figures 2 and 3 is a graph of haemoglobin concentration against analyte (in this case the cytokine, TNFα) concentration.

Referring to Figure 1 , a fluid sample containing molecules of an antigenic analyte 10 is incubated with antibodies 12 to the analyte, resulting in the formation of an analyte/antibody complex 14. Red blood cells 16 whose surface carries protein A 18 are added to the incubation mixture, thereby providing analyte-modified red-blood cells 20. Addition of complement 22 punctures the surfaces of the red blood cells 24, releasing haemoglobin 26 into solution. The haemoglobin is then assayed, e.g. by ELISA or radioimmunoassa .

Example 1 This Example shows the use of the assay method of this invention, in which a haemolytic amplification step is performed in the liquid phase, to measure sub-fe togram quantities of the cytokine, TNFα, in culture fluid. However, it could be used to measure any protein in solution, as well as any other molecule to which an antibody can be raised.

In this Example, the analyte (human TNF α) is used in an immobilised state; and is bound by a biotin-streptavidin link to red blood cells; which are lysed by freezing and thawing. A. Solid Phase Amplification

(a) A 96-well immunoabsorbant ELISA plastic plate (available commercially) is coated with monoclonal mouse anti-analyte diluted to 1/200 in 0.1 M sodium bicarbonate buffer (pH 9.6) for 2 h at 37 ^*C. 150 μl of 1% (w/v) bovine serum albumin (BSA) in PBS is then added for 30 mins. at room temperature.

(b) The wells are emptied onto blotting paper and washed x4 with 200 μl of 0.1% Tween 20 in PBS.

(c) A range of known concentrations of analyte solutions made up in 1% BSA/PBS (i.e. the standards) and the sample containing unknown quantities of the analyte are added to the wells for 2 h at 37'C. (d) After washing, as in (b) , to remove unbound analyte 100 μl of a polyclonal (rabbit) anti-analyte, diluted to 1/200-1/500 in 1% BSA/PBS, is added to each well for 2 h at 37^*C.

(e) After washing, as in (b) but using PBS alone (i.e. no Tween 20), 100 μl of biotinylated donkey or goat anti-rabbit IgG, diluted to 1/200 - 1/1000 in 1% BSA/PBS, is added to each well for 1h at 37^*C.

(f) After washing, as in (b) , 100 μl of 2% streptavidin-coated sheep erythrocytes (RBC's) is added to each well for 30 rain at 37^*C [NB RBC's are coated with streptavidin in an identical manner to that outlined above for coating RBC^'s with protein A. These can be stored at 4^*C in culture medium or a preservative called Alsever^'s solution for 4-6 weeks] . (g) After washing as in (b) , but using PBS alone (i.e. no Tween 20), either:

50 μl of PBS alone is added to each well and the 96 well plates frozen at -70"C and then thawed rapidly at 37'C to lyse the RBC^'s attached to the floor of the wells. A multichannel pipette is then used to mix the cont°nts in each well thoroughly in the PBS , or

50 μl of PBS containing 0.1% saponin to lyse the RBC's.

(h) RBC's contain a high concentration of both haemoglobin and peroxidase.

B. Haemoglobin ELISA:

(i) Supernatants from the standard Hb solutions and unknown samples are incubated for 2 h at 37"C in plastic wells previously coated with a monoclonal anti- sheep Hb (moAB) ,

(j) After washing to remove all but the Hb bound to the moAB, a polyclonal (rabbit) anti-sheep Hb is added to the wells for 2 h at 37^*C, (k) After washing, a biotinylated donkey anti- rabbit IgG is added to the wells for 1 h at 37'C. (1) After washing, a streptavidin-peroxidase complex is added to the wells for 30 min at 37"C. (m) After washing the peroxidase is then detected using phenylenediane (and product; yellow/orange colour which takes 5-10 mins. to develop) .

(n) The optical density of the solution in each well is then read using a 'Titertek Multiskan^' densitometer (this or a similar machine is currently being used in many laboratories, using conventional ELISA kits) .

The Hb measurements for each of the standard solutions of the protein (in this case, TNFα) are plotted. This standard curve is then used to convert the Hb levels in test samples to the concentration of TNFα originally present.

Following lysis of RBC^'s (step g) , the haemoglobin content in each well is measured either by the haemoglobin ELISA previously outlined (see Figure 2), or directly by measuring the absorbance of the wells in an ELISA reader at 405 nm (see Figure 3) . Alternatively, the peroxidase content in each well can be quantified by adding a small quantity of the substrate, peroxide, and the chromagen, OPD. Within 4- 6 mins., the resultant yellow/brown colour reaction in each well is measured using an ELISA plate reader set at 492 nm.

This new assay has the enhanced sensitivity afforded by the marked amplification of the original protein signal by the Hb released in the liquid-phase RHPA. The data illustrate that the method detects as little as 10^~18 g/ml of TNFα in culture fluid. This contrasts with the sensitivity of conventional ELISA's (currently available commercially) which can reliably measure in the picogram range (i.e. > 10^~12 g/ml).

It should also be noted that this new method currently takes approximately 9 h, and can thus be performed in one working day. Furthermore, the equipment needed is identical to that used in conventional ELISA's, thereby obviating the need for laboratories already using ELISA technology to spend money on anything but the required reagents.

The invention is expected to be useful for assaying cytokines, hormones such as growth hormone and thyroxine, and drugs (both therapeutic drugs and drugs of abuse) . The method is expected to be applicable to tests performed on senior company executives and on athletes.

References

1. Jerne, N. K. et al .. Transplant Rev., 18, 130 (1974) .

2. Molinaro, G. A. & Dray, S., Nature, 248, 515 (1974) .

3. EP 132556A.

4. US 4,820,634.

Claims

1. A method of assaying for an analyte, which method uses cells or liposomes containing a predetermined reagent and comprising the steps of i) Incubating the cells or liposomes with the analyte so as to bind molecules of the analyte to the surface of the cells or liposomes, thereby providing analyte-modified cells or analyte-modified liposomes, ii) Separating the analyte-modified cells or analyte-modified liposomes from cells or liposomes not so modified, iii) Lysing the analyte-modified cells or analyte- modified liposomes to thereby release the reagent into solution, iv) Assaying the reagent released into solution as a measure of the analyte.

2. A method as claimed in claim 1, wherein red blood cells are used as the cells or liposomes.

3. A method as claimed in claim 2, wherein the predetermined reagent is haemoglobin.

4. A method as claimed in any one of claims 1 to 3, wherein the molecules of the analyte are bound to the surface of the cells or liposomes by linkages that include immune complexes.

5. A method as claimed in any one of claims 1 to 4, wherein there are used in step i) cells or liposomes whose surfaces are coated with avidin or streptavidin or biotin.

6. A method as claimed in any one of claims 1 to 5, wherein the analyte is an antigen or hapten, and step i) involves the use of an antibody to the analyte.

7. A method as claimed in any one of claims 1 to 6, wherein the analyte is initially present in solution in a fluid sample.

8. A method as claimed in any one of claims 1 to 7, wherein the analyte is used in an immobilised state, and step i) involves providing the analyte-modified cells or analyte-modified liposomes in an immobilised state.

9. A method as claimed in claim 8, wherein the immobilised analyte-modified cells or immobilised analyte-modified liposomes are lysed in step iii) by freezing and then thawing them.

10. A method as claimed in any one of claims 1 to

9, wherein the reagent is assayed in step iv) by means of an enzyme-linked immunosorbant assay.

11. A method as claimed in any one of claims 1 to

10, wherein in step iv) the concentration of the reagent is determined as a measure of the starting analyte concentration.