The present invention relates to a method of assaying the formation of multi-protein complexes (such as the FVII/TF complex) , optionally in the presence of test substances, such as peptides.

Antibodies are frequently used by those skilled in protein chemistry for the isolation, purification and/or identification of particular proteins of interest. In this field the availability of monoclonal antibodies has led to great improvements in all techniques relying on the formation of a specific antibody-antigen complex. Monoclonal antibody production (for example by fusing a single antibody-producing cell with a tumour cell to form a stable hybridoma) allows the manufacture of multiple copies of antibodies having identical specificity.

Various assay methods have been developed in which specific antibodies selectively bind the protein of interest. One such well-known assay method is the "solid phase immunoassay" in which the antibody to the protein of interest is immobilised on a polymeric support, for example polyvinylchloride. The antigen-containing test substance is then applied, and any excess antigen is washed off after formation of the antibody-antigen complex. A second antibody (specific for a different site on the antigen) is then added. This second antibody is labelled, for example with a radioactive or fluorescent label, and the amount of label bound is determined. Where the second antibody is labelled with an enzyme (such as alkaline phosphatase) which converts a substrate into a coloured or fluorescent product, the assay is known as an "enzyme-linked immunosorbent assay" or ELISA. The

amount of coloured or fluorescent product formed is a measure of the amount of antigen bound. This particular assay is extremely sensitive and can detect antigen in amounts of less than a nanogram due to the amplifying effect of the enzymic label.

The present invention provides an assay for the formation of multi-protein complexes. Additionally, the present invention provides an assay to determine the ability of the test substrates such as peptides or other additives to affect the formation of a multi-protein complex.

The assay according to the invention comprises the following steps:

i) reacting a first protein of a multi-protein complex with an immobilised first antibody specific therefor which does not interfere with complex formation;

ii) optionally adding further proteins which form part of the multi-protein complex;

iii) optionally adding a test substance;

iv) adding the remaining protein(s) required for formation of the multi-protein complex;

v) adding a labelled second antibody specific to a protein added in step (iv) ; and

vi) detecting and optionally determining the amount of the second antibody immobilised as an indication of multi-protein complex formation.

Although not specifically mentioned above, washing steps are advantageously performed after each of the stages

listed above, to remove any excess reagent present. Water or, more preferably, buffer may be used as the washing medium.

Conveniently the assay of the invention is carried out at a pH in the range of from 7.0 to 7.5, preferably from 7.2 to 7.3.

Thus, the assay of the present invention can be used to determine whether or to what degree a naturally produced multi-protein complex is formed by an individual. In this way any malfunction in formation of a multi-protein complex, for example due to a genetic disorder or physiological disturbance, can be ascertained. Measurement of the amount of label thus correlates directly to the ability of that patient to form the multi-protein complex. Thus, in this embodiment the invention provides an assay as hereinbefore described, in which the said remaining protein added in step (iv) is contained in a body fluid of a patient whereby formation of the labelled multi-protein complex indicates the presence in the body fluid of the said remaining protein in a form capable of forming said complex.

In the other main embodiment of the invention, test substances of interest are investigated by use of the assay described above. The amount of multi-protein complex formed (determined via the amount of label retained) in the presence and absence of the test substance indicates whether and to what degree the test substance affects the formation of the multi-protein complex. Thus, in this embodiment the invention provides an assay as hereinbefore described, in which the test substance added in step (iii) is a substance which may affect formation of said multi-protein complex whereby the ability of the test substance to affect

complex formation can be assessed.

Conveniently, the first antibody is immobilised on a solid support. Suitable solid supports include beads, tubes, plates, dishes, trays, columns etc. Any free protein binding sites remaining on the support should advantageously be blocked with a blocking protein prior to commencement of the assay. Suitable blocking proteins would be well-known to those skilled in the art and should be selected for their non-involvement with the multi-protein complex under test. Any excess blocking protein should be removed by washing.

Any multi-protein complex can be used in the assay method of the present invention. By "multi-protein complex" is meant an association or combination of at least two different independent proteins which forms spontaneously under suitable conditions. Formation of the multi-protein complex may be reversible. Thus, examples of a multi-enzyme complex includes enzyme-substrate complexes, enzyme-inhibitor complexes or a complex of two or more proteins in a functional (e.g. enzymatic) complex, for example on a cell membrane. One example of a multi-protein complex occurs in the extrinsic pathway of the blood clotting cascade, where Factor VII or Vila (FVII or FVIIa) combines with tissue factor (TF) . The FVIIa/TF complex is responsible for catalysing the conversion of Factor X to it's active form in the next stage of the cascade.

The use of monoclonal antibodies in the assay of the present invention are, of course, greatly preferred. Either or both of the first and second antibodies may be monoclonal. Monoclonal antibodies to a wide variety of proteins are commercially available and the manufacture of monoclonal antibodies to a particular protein is within the ability of the skilled man.

In order to determine the amount of second antibody bound by the assay, that antibody must be labelled. Excess second antibody is removed by washing prior to measurement. The second antibody may be labelled in any known manner and several suitable labels are available.

Thus, the second antibody may be labelled with a chromophore or fluorophore such as dansylchloride or with a radioactive atom such as ¹²⁵ I . Other labels include enzymes, such as an alkaline phosphatase or peroxidase which catalyse the formation of detectable products, and colloidal metals such as gold.

The present invention also provides the use of an antibody to a component of a multi-protein complex in the assay described above.

Further, the present invention provides a kit for determining the effect, ie enhancement or inhibition, of a test substance on the formation of a multi-protein complex, said kit comprising means for forming a, preferably immobilised, first antibody-first protein conjugate and separately thereto a second protein of the multi-protein complex, and a labelled second antibody specific therefor. The kit may also include buffer or water for washing between each step of the assay procedure. Preferably the antibodies provided in the kit are monoclonal .

Where the label provided on the second antibody is an enzyme, the kit according to the invention may also comprise the substrate for the particular enzyme used. Optionally a reagent to halt the reaction catalyzed by the enzyme-label may additionally be provided.

The present invention also provides a method for the in vitro diagnosis of abnormal formation of multi-protein

complexes, said method comprising assaying a body fluid (or extract or diluent thereof) with the assay described above. Body fluids suitable for the assay procedure include blood, serum, urine, saliva, plasma etc.

Figure 1 is a schematic diagram of an assay complex according to the invention, involving the multi-protein complex of FVII/TF.

Figure 2 indicates the binding of FVII to immobilised rTF (recombinant tissue factor) at different coating concentrations as a function of FVII concentration.

Figure 3 illustrates the requirement for calcium ions in formation of the FVII/TF multi-protein complex (filled diamonds) . Open squares represent controls without tissue factor.

Figure 4 illustrates how formation of the FVII/TF multi- protein complex is affected by addition of increasing concentrations of a monoclonal antibody which binds to FVII in competition with TF.

Figure 5 illustrates how formation of the FVII/TF multi-protein complex is affected by addition of various synthetic peptides.

Figure 6 shows FVII from normal plasma at various dilutions binding to immobilised rTF.

The invention is further illustrated by reference to the following, non-limiting, examples:

Example 1: Assay method with the multi-protein complex FVII/TF

COATING OF ELISA PLATES

96 well immunoassay plates (Nunc Maxisorp cat# 442404) were coated with a monoclonal antibody to human tissue factor (American Diagnostica cat# 4503) as follows: a solution of the antibody at 5 μg IgG/ml was made up in 0.05 M Tris. HC1 pH 9.5 and 100 μl was pipetted into each well. The wells were then covered with a plastic sealer, and the plates were incubated for 15 hours at 37°C. They were then washed twice with 250 μl/well of 10 mM Tris.HCl pH 7.3 containing 5 mM CaCl ₂ and 0.05% Tween 20 (hereafter referred to as TCCT) .

BLOCKING OF ELISA PLATES

The non-specific binding of the wells was blocked by allowing 150 μl/well of 1 mg/ml bovine serum albumin (BSA) in TCCT to stand in the wells for 1 hour at 23°.

TISSUE FACTOR LAYER

The washed and blocked wells were then treated for two hours with 100 μl/well of a 0.25 μg/ml solution of recombinant human tissue factor (rTF: American Diagnostica cat# 4500) in TCCT containing 1 mg/ml BSA. After this the wells were again washed three times with 250 μl/well TCCT.

COMPETITION BINDING OF FACTOR VII WITH INHIBITORS

The peptide inhibitors were added next as dilutions in TCCT/BSA) to the treated wells (75 μl//well) . FVII solution (0.1 μg/ml in TCCT/BSA) was then added (75 μl/well) on top of the inhibitor solution, and the

mixture was allowed to stand in the wells for 2 hours. The wells were then washed three times with TCCT.

ASSAY OF BOUND FVII

To measure the bound FVII, lOOμl of a solution of peroxidase-conjugated rabbit antibodies to human FVII (from Stago FVII.Ag kit cat# 00412) was added to each well and allowed to stand for 2 hours. The wells were washed again three times with TCCT and then 100 μl peroxidase substrate solution was added to each well as for the Stago FVII.Ag assay.

Characterisation of the ELISA assay showed that the colour yield was a function of both the rTF concentration used to coat the wells and also the concentration of FVII ligand (Fig. 2) . Binding of FVII was strongly calcium ion dependent (Fig. 3) , as already documented in the literature, and inhibited by a monoclonal antibody which binds to FVII in competition with TF (Fig. 4) .

Example 2 : Effect of synthetic peptides on formation of FVII/TF multi-protein complex.

The following synthetic peptides were individually assayed as test substances to ascertain their effect on the formation of the FVII/TF multi-protein complex:

FVII-1 -GKIPILEKRNA- 136-146

FVII-3 -VGHFGV- 371-376

FVII-4 -SDHTGTKRSCR- 103-113

FVII-5 -CVNENGGCEQYC- 91-102

FVII-4B -SDHTGTKRS- 103-111

FVII-7 -CASSPCQNGGSC- 50-61

FVII-10 -NGGCEQYCSD- 95-104

FVII-11 -DKIKNWRNLIA- 196-206

The results are shown in Figure 5.

Of the FVII peptides tested, the most inhibition of FVII binding to rTF was obtained with cyclic peptides representing the ring structures present in the FVII growth factor domains. Some activity was also present in a linear peptide of residues 136-146 from a region close to the site of cleavage which activates FVII.

Example 3 : Assay of human plasma for ability to form the multi-protein complex FVII/TF

The binding assay of Example 1 was used to test the binding ability of FVII from human plasma. The assay was performed using normal plasma, plasma immunodepleted of FVII (Diagnostica Stago) and plasma from a patient having a homozygous mutation at residue 100 (glutamine to arginine) in the second growth factor (GF) domain of FVII. The plasma samples were diluted in TCCT buffer containing 50μm p-nitrophenyl guanidinobenzoate to inhibit all serine proteases of the clotting pathway.

The results, illustrated in Figure 6, show that the deficient FVII in the plasma of patients can be detected.