AP81A - Agglutination assay - Google Patents

Abstract

In a novel

Description

TECHNICAL FIELD

Fie present invention relates to a reagent and a method for detecting igen, antibody or other analyte in human or animal blood by ocyte agglutination. The invention also concerns a kit containing agent and processes of preparation of the reagents. '

BACKGROUND APT

Assaying blood samples for a particular antigen or antibody has ionally involved the step of separating the cellular components from rum components of the blood by centrifucarion and/or clotting, prior a 7.

This presents several potential problems. Firstly, such an assay is not suited to testing being conducted undei* field conditions. In many veterinary situations a quick test in the field is more desirable tnan the alternative of transporting samples to laboratories for separation ,·.ηΡ assay. Also, veterinary surgeons who do not have access to a centrifuge frequently need to assay blood samples for the presence of infectious

Further, assays being used for the detection of i nil.’ i i ». > L 7 c. H vi agents such as heartworn:. diseases in Third World countries presens • ;.···,·,· cost are of the essence.

fu;;ondiy, in ce-tain pathologic sendit’ons, se sa.reins uecernes difficult. Blood taken from patient as Haldenstrom¹s macroglobulinemia is Gift such a n c

I; i '.'h on wne;

eea ra t i on or t:i¹ : sufferint: csuq^: : 000 cion;

i fractions makinc an assav wnicn can pe conducted on whole ir> s ue

Thirdly, blood samples are often used for testing for the presence of high!7 contagious and potentially dangerous disease states. In these cases it ; preferable that as little handling and processing of the samples as e is undertaken in order to minimize the risk to personnel ing Die assay. Further, certain conditions [iiaitt e-connfer finger-prick assays highly desirable, rily he suited to performance on whole blood, immunoassays have revolutionized human diagnostic and veterinary ;;ed:c;ne since trie introduction of techniques such as the rad i o i imiiiinoa s .a v first reported by Yalow and Eerson <1959) Nature Del, 1G4C, and tin? enzyme i iiiaiurion s s ay or LIA which was first reported by Engvall and Perlman (1971 ) Immunco hem 8, 671 and Van Hecman and Schuurs (1371) FEES Letters j6, EEC. Whilst such assays are based on ant i body-an t i gen interactions trie concur over-i ne

I.ii / I Ο'.Κ.

Che provi: ion oh Such assart must

I 8 0 0 0 0 dV

BAD ORIGINAL &

detection systems utilized are usually complex. The reagents used are generally enzyme or radiolabelled antigens, antibodies or complexes thereof which require either incubation with specific substrates and measurement of a color end-point either visually or by means of a colorimeter or measurement of radioactive decay with radiation counters to detect the presence of the analyte being tested. These assays also involve several washing steps. Most immunoassays for the detection of analytes in blood are currently of this nature. Thus, whilst these assays are sensitive, they are lengthy and involve procedures which may require expensive instrumentation, for detection of the analyte under test.

An alternative to these assays is provided by immunoassays of the type described by Gupta, et al., (1985) Journal of Immunological Methods 80 177-187. These are immunoassays in which erythrocytes and anti-erythrocyte antibodies are used in the indicator system. In these assays exogenous erythrocytes such as sheep erythrocytes are used.

In recent years it has been possible to attach antibodies to latex beads, thus providing a rapid agglutination assay. This, however, still entails the separation of the serum/plasma phase from the cellular phase and consequently requires the use of a centrifuge or filtration system. Latex agglutination assays are described in Castelan et al., J.Clin. Pathol (1968), 21, 638; and Singer & Plotz AM. J. Med. [1956 (Dec)], 888.

Both direct and indirect agglutination immunoassays are well known in the art. In these assays, the agglutination of particles to which antigen or antibody is bound is used to indicate the presence of absence of the corresponding antibody or antigen. A variety of particles, including particles of latex, charcoal, kaolinite, or bentonite, as well as both microbial and red blood cells, have been used as agglutinatable carriers. See Mochida, US 4,308,026. The use of erythrocytes as indicator particles is strongly criticised by Patel, US 3,882,225, who says that it is diffuclt to standardize indicator erythrocytes.

Molinaro, US 4,130,634 describes an assay for an antigen which employs antibody-coated red blood cells. Molinaro emphasizes that the method used to couple the antibody to the erythrocyte must not destroy the reactivity of the antibody. He makes it clear that antibodies which are specific for the erythrocyte are not useful for his assay. He does mention, however, the possiblity of using a hybrid antibody with one binding site specific for the antigen and the other specific for the red

- 2 BAD ORIGINAL

Λ .'

15. 03/S3

161.

NO.479

P006/022 •V>

blood cell.

Chang, L

S 4,433,059 disclose which two antibodies are covalent to alter their by an indicate s an agglutination Immunoassay reagent in y linked tai 1-to-tail, i.e., so as not specificity. One entlbody Is specific for an antigen borne r substance, such a:

preferably univalent to avoid non:

an erythrocyte. This antibody is pecific agglutination. The other antibody is di the assay, frt valent and is specific for the analyte. In preparation for sh erythrocytes are antibody conjcgate-coated RQCs arc

Chang does not contemplate the as' non-autoagglutinating antl-RBC an

Chu, US 1 ec11n-antIger

4,493.793 discloses covalently coupled coated with the conjugate. The double then Incubated with the test serum, aying of whole blood samples using a ibody and endogenous erythrocytes, the construction of a lectin-antibody or conjugate. His Table I (incorporated by reference) sets forth the carbohycrate specificities of several lectins.

He does not tc elther the 1ec ach coupl1 ng such a tin or the antibody

Other tai 1-to-tal 1 Imrnuno

4.676,9G0 sets conjugate to an erythrocyte through receptor.

logical conjugates are known. Sogal, US forth the construction of a tai1-to-tal1 conjugate of a target cell surface antigen-specific antibody and of a cytotoxic effector cell rcceptorspecific antibody.

incorporated by reference, are des

Several cross-11nk1 ng methods, cribed. This conjugate is intended for use in Immunotherapy, In that it vill cause the cellular immune system of tiie patient tc lyse the target cell. The target cell would not, of course, be an erythrocyte endogenous to th

LI , US of an analyteof the first a e host.

e production of a tail-to-tail conjugate ,C6l,444 suggests th binding antibody anc of an antibody specific for the idiotype ntibody. This conjugate was to be used in conjunction with an insolub111 zed analyte-binding antibody in an immunoassay.

Hardlaw, US 4,695,553 teach^ universal erythrocyte antigen as a interface between red blood cells whole blood.

s use of a monoclonal antibody against a RBC agglutinating agent to clarify the and white blood cells In centrlfuyed ie prefers use of antibodies against glycophorin or against H antigen, but also mentions the possibility of using a mixture of lectins. Guesdon, US 4^666,637 discusses the use of anti-red blood cell antibodies or of lectins for the purpose of erythroadsorptlon. Bigbee, [Molecular describes the production and testing of glycophorin A. The general concept of

Immunology, 20

1353-1362 (1983)] four monoclonal antibodies against using In an imnunoassay an antibodly which reacts with an antigenic

BAD ORIGINAL

15/03- Ο:

NO. 479

Ρ007/022 determinant shared among all member (mlcroorganlsms A number ) Is set forth In Me s of a class of analytes of Interest Laughlin, US 4,683,196.

of patents deal wi tlS antibodies useful In blood typing.

See, e.g., Llo^d, US 4,678,747; Graham, Jr., US 4,358,436; Liu, US

4,550,017; Steplewski, US 4,607,00'); Lennox, WO83/O3477. These antibodies blood typing because they bind to antigens found only In le for the purpose of this Invention, 4t mixtures thereof) which bind to are useful for certain blood dell populations, wh Is desirable t$ use antibodies (or erythrocytes.

essentially al

Zuk, US 4,594,327 recognizes the desirability of performing an immunoassay directly on whole blood samples. In his method, the simple is contacted with both an Insolubt 112 with a red blood cell binding agen lectin. The analyte-specific Immu ?d, analyte-specific Immunoroagefit and 1 such as a RBC-specIf 1c anti bod) or a loreagent and the RBC binding agint are not coupled together, and the assaL· disclosed is not an agglutination assay.

The prob em, in an agglutination Immunoassay, of nonspecific agglutination of erythrocytes by aitl-erythrocyte antibodies endoginous the blood sample, was noted by Czl smas, US 3,639,558. He proposed eliminating all naturally occurring antigenic sites on the particle by coating the pa *Ucle with protein.

Theof1loaoulos, US 4,342,566

US 4,331,647 a fragments of a The construct! Auditore-Hargr 4,474,393. Mo antibodies or e of Interest as de

Duermeyer, US 4,292,403 and Goldenberg, nonstrating the use of specific binding itibodles as substitjtes for Intact antibodies In astays.

on of heterobifuncti eaves, US 4,446.233; :hida, US 4,200,436 jlndlng fragments th pnal antibodies is taught by

Paulus, US 4,444,878; and Reading, US discloses the use of monovalent ?reof in certain Immunoassays. Forrest,

US 4,659,878 mentions that monovalent antibodies cannot form dimerj or mor<

extensive complexes with the antig jrfering with the bi support.

DESCRIPTION capable of int a solid phase ?n; such aggregates were said to be nding of the antigen-antibody complex

OF THE INVENTION (0

The pres?nt Inventors recognized that there was a need for a method which can be used In the laboratory and in the field, particularly In Third act of medical testing facilities for ytes In whole blood. As indicated

World Countries where there Is analysis of di above, earlier plasma and are ferent types of ana methods require separation of the blood cells from jer therefore difficult urn or and in many cases Impossible to

BAD ORIGINAL l 8 0 0 0 0 dV

15--ΰ?·'83

16:2?

110. 47?

Ρ0ΟΘ/Ο22 hImplement In th» field.

jcyte-bindlng molecules are coupled to specific molecules, then the

If orythr analyte-binding bind both pndog?nous erythrocytes, resulting conjugate could be used to and analytes present in a blood sample. The present Invention results from the finding that when such a complex Is exposed to a bl^od sample, agglutination of the erythrocytes endogehous to that sample will serve as an Indicator of the presence of the relcv&nt analyte (usually, an antigen or antibody) due to cross-linking of erythrocytes with the analyte.

Advantages of endogenous RBCs

1) Simplifies current assay procedures no need to centrifuge sample; whole blood, collected In the mQ presence of a suitable a or plasma.

for samples from patients with Infectious diseases, such as AIDS or hepatitis, there Is miinimal sample handling, appropriate for mass screening programs as conducted by the 1-lorld Health Organization in third world countries, whose fac,1111es are 11m,ted.

there Is only a single reagent, which .table In the present of a bacteriostatic agent such as 0.01 X (w/v) sodium azide.

can be used as a field test by veterinary practitioners, when the appropriate animal red cells are used for immunization to nticoagulant, Is used Instead of serum the assay is very robust prodjuee species the ml nu the compl lance.

specific

MAb.

agglutination occurs in less test is very fast tes.

nethod can be used tb monitor therapeutic drugs and patient :han three also has possible use as an over-the-counter self testing as sab-.

the only equipment needed is a mixing stick., glass or plastic idp, lancet and possib ly a mlcrocapl11 ary,

11) Adva. itaqes over exogenous erythrocytes Include:

pr etreatmenf of erythhocytes. Patent 4.433,059 uses blood negative cells, which have been spun down, reacted with s-30 minutes and washed 3x In P3E.

Pateht 4,668,547 uses shpep red blood, cells, which have been

BAD ORIGINAL £

16:.

NO.479

PC09/C6 washed and resuspended lr PBS.

place no pr ’nave compl to in nece from on a solid support etreatment of samplt to be heat inactlva

After the reaction, which takes the cells are then fixed, s. Patent 4,433,059 notes that samples ed to avoid Interference due to ement. Rabbit serutb and bovine albumin must also be added n1ml2e other non-specific reactions. None of this Is a

sary with the present system, where undiluted whole’blood ;ted directly with reagont. Thia ;d monoclonal antibody to prevent any patients may be rea reagent contains unrelat antb-mouse reactions, wh'lch may occur.

Thus, It s possible to dlspeise with' the cumbersome separation of cells from serum and with the sensl tlzatlon and fixing of exogenous erythrocytes In assays. The en Another n ended for use as Indicator particles in agglutination logenous erythrocyte ovel aspect of app11 is the selection of an erythrocyte s are sensitized by the reagent, cants' agglutination reagents and assays binding molecule such that Incubit'on of conjugate with erythrocytes unless analyte molecules are tprmed herein The eryth and especial 1y, endogenous erythrocytes will not cause agglutination of the s also 'non-au present -- such erythrocyte binding to-agglutinating.

-ocyte binding molecule is preferably a monoclonal afitibody in the human svs ten believed that tnls antibody Is nonan anti-glycophorIn antibody, it 1 autoagg1utinating for sterlc reasons:

AP 0 0 0 0 8 1 either the binding sites of the intact antibody are able only to blhd adjacent cp1 top sites can bind App11 cant determlnants es on the same erythrocyte or only one of the two bidding to glycophorin at one time.

’s assay can detect using two conjugates, small antigens without repeating one bearing an analyte-specific binding molecule and the other, a binding molecule specific for a new epitope formed by the binding of the first conjugate to the analyte. This allows cross-linking in the presence of the antigen to be measured

Brief Descript ion of the Drawlngs

Fig. 1 1s a schematic representation of erythrocytes showing positive with antibody complexes In the presence and negative agglutination results and absence of antigen respectively

Fig

Is a schematic representation of erythrocytes showing positive and negative agglutination results antigen In the antibodies.

presence and absence with a complex of antibody and an , respectively of ant,-antigen

BAD ORIGINAL £

Vb i

Fig 3. is a schematic representation depicting (a) erythrocyte agglutination and (b) inhibition of erythrocyte agglutination due to presence of analyte or antigen.

Fig 4. is a schematic representation depicting mechanisms of agglutination/nonagglutination in connection with an overlapping antigen assay.

DETAILED DESCRIPTION OF THE INVENTION

In the agglutination assay of this invention, a reagent is provided which comprises an erythrocyte binding portion provided by an erythrocyte binding molecule attached to an analyte binding portion provided by an analyte binding molecule, or to an analyte analogue, without substantially changing the binding characteristics of the binding portions. The reagent is non-agglutinating when incubated with endogenous erythrocytes in the absence of the analyte.

Erythrocyte Binding Molecules

Erythrocyte membranes contain various antigenic surface constituents, including proteins, glycoproteins, glycolipids and lipoproteins. Antibodies which recognize these constituents may be prepared by conventional techniques using the membrane, or the purified constituents thereof, as immunogens. These antibodies may be monoclonal or polyclonal in nature. Either the intact antibody, or specific binding fragments thereof, may be used as erythrocyte binding molecules (EBM). The antibody or antibody fragment may be polyvalent, divalent or univalent.

In addition, glycoproteins, glycolipids and other carbohydrate structures on the surface of erythrocytes are recognized by chemicals known as lectins, which have an affinity for carbohydrates. These lectins may also be used as EBMs. Other receptor molecules with specific affinity for the erythrocyte surface also may be used. These could also include molecules with an affinity for the lipid bilayer of the membrane. Examples of such molecules are: protamine, the membrane binding portion of the bee venom, melittin, and other very basic peptides.

The preferred EBMs of the present invention will recognize erythrocyte membrane constituents found on all, or nearly all erythrocytes, so that erythrocytes endogenous to the blood sample may be used as the agglutinating particles. Such constituents include the so-called public antigens.

Erythrocyte membranes are lipid bilavers with a variety of proteins either on the surface or with a hydrophobic portion allowing the protein to

BAD ORIGINAL &

5/09/33

13:23

NO.479

P011/022 anchor 1n or pass through the membraJne, and may have part of the molecule inside the cell.

Glycophorin A Is an example of membrane. The blood group speciftcl glycolipid moieties, which are attac a molecule which traverses the cell ty is conferred by carbohydrate Or hed to membrane proteins. It 1S^: thus important that ajn EBM should recognize either the protein part of a membrane glycoprotein constituent, vhich is common to all erythrocytes of Γ particular species or another commor* structure. The ability of a bVvalent EBM to agglutinate red cells will depend on steric factors, such as the mobi 11ty of the lipid b1layer. The prote molecule and the position of the binding site above the ns of erythrocyte membranes include:- glycophorin A (MN, En“, Wr°), glycophorin B <Ss, 'Ν'. IJ) and the minor constituents, integral membrane protein 1 (Rhesus¹, membrane attached glycoproteid C4 (Chido & Rodger: ), integral membrane glycoprotein (anion channel), glycoliplds (Lewis), glycosphingol1olds (ABH, 11. P, Tk), ankyrin, spectrin, protejn 4-1, F-actin. [Tpe associated blood group factors are parentheses.3

The following publications ark Incorporated by reference:

The red cell membrane. S.B. Siohet St E. Beutler. in: Hematology, 3rd ed. Eds: Williams, Beutler, E *slev & Llchtman, 1983.

(Review o ell erythrocyte membrane antigens)

ΔΡ0 0 0 0 8 1

The red oil membrane skeleton. V.T. Marches!. Blood, 61. l-ll, 1 983.

(Review o the skeleton protejins)

An especi illy preferred EBM ik one recognizing glycophorin. When erythrocyte sla oglycopeptides are extracted from membranes, the main fraction (approximately 751 of tota

1) is glycophorin. This molecult comprises 131 anlno acids with 16 oligosaccharide chains. Thus, this is *

¥ an abundant moiety agglutinating the red cells, pure form comme’Clal ly, e.g., which could allow It is from

Fractionation Membrane H. Fu When the ?f the Major Slalogi thmayr, M, Tomita Si ?rythrocyte binding antibody attachment without also readily available in a relatively Hgma Chemical Company. (See 'copeptldes of the Human Red Cell

V.T. Marchesl. BBRC 6$, 1975, 113-122) molecule Is multivalent, as in the case of a normal antibody, It is desirable that the molecule recognize ah erythrocyte memprane constituent which Is abundant and well-distributed, and the binding cells is inhlbl site should be ln s jch a position that crosslinking between ted by steric hindrance, thereby avoiding premature red cell

BAD ORIGINAL £

ζ'' ν

agglutination.

It is preferable, but not necessary, that a single EBM be used that recognizes essentially all erythrocytes. Several EBMs may be used, either in the same or in separate conjugates, each of which recognizes a particular group of erythrocytes, but which in aggregate recognize essentially all erythrocytes.

While it is preferable that the EBM recognize a natural surface constituent of the erythrocyte, it is possible to coat erythrocytes with a ligand recognized by the EBM, or to treat the erythrocytes so as to expose a normally cryptic ligand.

Analvtes

This invention is not limited to the detection of any particular analyte. The analyte may be a substance normally found in blood, such as a blood protein or a hormone, or it may be a foreign substance, such as a drug (including both therapeutic drugs and drugs of abuse), or an organism, such as a virus (by recognizing a virus coat protein) bacterium, protozoan, fungus, or multicellular parasite (e.g., heartworm).

The analyte may have repeating epitopes, recognizable by one analyte binding molecule, or unique epitopes, where a mixture of analyte binding molecules is necessary. However, analytes which can only be bound by one ABM at a time, may also be detected.

Analvte Binding Molecule

The analyte binding molecule may be any substance having a preferential affinity for the analyte, including monoclonal or polyclonal antibodies lectins, enzymes, or other binding proteins or substances (or binding fragments thereof). Where the analyte is an antigen, the ABM is usually an antibody. Where the analyte is an antibody, the ASM is usually an antigen recognized by that antibody. When the analyte to be detecred has no repeating epitopes, two or more ABMs are required with different specificities for the analyse. The reagent in this case will be either a mixture of Ξ3Μ bound to A3M1 and E3M bound to A3M2, or E3M with both A3Ms attached.

The analyte binding molecule need not bind the analyte directly. For example, in an assay for growth hormone, one ABM may be directed against the thyroxine binding protein if the latter is known to be present in the sample, or is separately provided.

Coupling of E3M and ABM

The

3M and the A3M may be coupled together directly or indirectly,

TI u /1^0* r

3/09/-53

IS: i'9

NO.4T9

P013/022 and by covalent or non-covalent meins (or a combination thereof). Where multiple EBMs cr ABMs are used, EBMs or A8Ms may be coupled together, with one or more ABhs coupled directly summarizes somt o an EBM. The following table of the covalent coupling methods Known In the art.

Heteroblfunctlcna1

SPOP (N-Succl rjlml dy1-3,2-(pyridyl |j1 thio)propionate) Neurath,

MBS et al., 1981, J. VI -ol., Meth., 3, 155-1 65.

(m-mal e1ri1doben2oyl-N-hydrox /succinimide ester) et al . . 1976, J. Blochem.-, 79, 223-236.

K1tagaw,

SIAB (N-succ1iilmidyl-4-1odoacetyl kml nobenzoate)

Weitman, etai.. 1 983., Bio.

Selective B1functional

Techniques. 1, 148-1 52.

P-i sothlicyanatobenzoy1 chi ori de

US Paten Bifunctional

Other .

680 338

BSOCOES

B1 s L2-( sibccinlml dooxy carbonyl) oxy )e thyl ] s u 1 phone Zarling, et al.. 1980, J. Imnunol., 124, 913-920 BS

Bi s(sulpiosuccinim1dyl)suberkte

Staros,

982, Biochemistry, El, 3950-3955

Glutaraldehyde

Avrameas

1969, Immunochem.

6, 43.

Perlodat^ Oxidation

Nakane ahd Kawol, 1 974, J. Hfistochem. Cytochem., 22 , 1084-Ί0ί1

3. CarbodlImide Of the ft

Is preferred.

The EBM And the ABM may also by (a) attaching biotin to one and (b) attaching regoing methods of covalent coupling, conjugation with SPOP be coupled noncovalently, for eiample, avidin (or strepavldln) to the Other), in anti-antibody to pne, which then binds the other, (c) attaching Prot?1n A to one, which and (d) attaching a sugar to one a

It shoul 1 be unoerstood that then binds the F_c portion of the other, id a corresponding lectin to the other.

1n coupling the EBM and the ABM, the

BAD ORIGINAL

AP 0 0 0 0 8 1 s' ί

binding characteristics should be changed as little as possible. It may be advantageous to provide a spacer moiety betwen the EBM and the ABM to reduce steric hindrance.

The ABM/ABM conjugate may be a chimeric antibody. One method of constructing such a conjugate is the following:

(a) preparing Ffab)^ fragments of a selected antibody by pepsin digestion;

(b) reducing and treating the fragments with Ellman's reagent to produce Fab' fragments of the selected antibody;

(c) thiolysing a selected specific antibody or a selected erythrocyte antibody; and (d) coupling the thioylated Fab' fragment to the Ellman's reagent treated Fab¹ fragment to produce a chimeric anti-erythrocyte antibody-antigen specific antibody conjugate.

Another method is set forth below:

(a) treating an anti-erythrocyte monoclonal antibody-producing hybridoma and an antigen specific monoclonal antibody-producing hybridoma with a distinct site-specific irreversible inhibitor of macromolecular biosynthesis;

(b) fusing the two different monoclonal antibody-producing hybridomas with polyethylene glycol;

(c) cloning the fused cells either in soft agarose or by limited dilution;

(d) selecting cloned heterohybridomas secreting chimeric anti-erythrocyte antibody-antigen specific antibody with a screening assay appropriate to the antibodies;

(e) purifying the antibody product by affinity purification to free it from non-hybrid antibodies.

Preferably the inhibitor is selected from the group consisting of emetine, actinomycin D, hydroxyurea, ouabain, cycloheximide, edine and sparsomycin.

The chimeric antibody may be two half-molecules, one with specificity for erythrocytes (the EBM) and the other with specificity for the analyte (the ABM). In this case the disulfide bonds of the antibody couple the ABM to the EBM to form the conjugate. Alternatively, the two half-molecules may be specific for the same or different epitopes of the analyte. In this second case, the chimeric antibody is really two ABMs and must be coupled to an EBM to form a tripartite conjugate. Tripartite conjugates may be formed by other means, such as attaching the EBM and two ABMs to a macromolecular spacer.

BAD ORIGINAL

The simplest agglutination reagent contemplated Is one comprising a single conjugate of one ESM to one ABM. This reagent is suitable for the detection of antigens with repeating epitopes.

Antigenic analytes large enough to allow simultaneous binding of two antibody molecules, but which lack repeating epitopes, are known. They include many peptide and protein hormones. For agglutination to occur, the antigen must Interact with the reagent so that at least some molecules of antigen act as a bridge between proximate erythrocytes. For assaying such analytes. It is preferably to employ a reagent comprising two or inor® distinct conjugates, i.e., ABM1/EBM + ABM2/EBM where ABM1 and ABM2 bind to different, non-overlapping epitopes of the analyte. One might instead use. a more complex single conjugate, ABM1/ABM2/EBM, where the special conformation is unlikely to favor the binding of both ABMs on the same conjugate molecule to the same analyte molecule.

Erythrocyte Agglutination Assay

Both direct and indirect agglutination assays are known in the art.

In the conventional direct assay for an antigen, red cells are coated with antibody, and reacted with the sample. Multifunctional antigens act as bridges between the coated red blood cells, creating an agglutinate. In the conventional indirect assay, red cells are coated with antigen, and contacted with both a soluble antibody and with sample. Sample antigen competitively Inhibits the binding of the sensitized red cells by the antibody, and hence the agglutination. It is also possible to additionally use an antibody - sensitized RBC. See Mochida, U.S. 4,308,026.

The reagent of the present invention may be used in either a direct or an indirect agglutination assay format. However, unlike conventional

E assays, it is not necessary to precoat erythrocytes with antibody or antigen. Rather, the reagent may be added to a blood sample containing endogenous erythrocytes, whereupon it will sensitize the cells, rendfring them able to bind sample analyte (a direct assay) or to compete with sample analyte for a soluble analtye-binding molecule (an indirect assay).

For some small circulating molecules such an synthetic or natural steriods, i.e., digoxin, theophylline, etc., or drugs of abuse, i.e., phenobarbital, cannabinoids, opioids, etc., the analyte in question may be too small to provide the two necessary antigenic epitopes for antibody binding (or other epitopes for recognition by other binding mol ecu 1 e.s.) to allow cross-linking and subsequent erythrocyte agglutination.

BAD ORIGINAL

For the assay of small molecules, as in drug monitoring or indeed for any other antigens, an agglutination inhibition assay is preferred. In this case, a two stage test is expected. The first stage would be addition of a reagent consisting of the analyte or analyte analogue coupled to the non-agglutinating EBM, and the second state would be addition of an unconjugated ABM. (The two stages may be reversed). If analyte is present in the blood sample, the specific binding of the ABM to the EBM-analyte analogue conjugate will be inhibited, leading to a loss of agglutination. Otherwise, agglutination occurs.

The term analyte analogue includes both the analyte, and any substance also specifically bound by the ABM when such binding is competively inhibited by the analyte. The analyte analogue may be anti-idiotypic antibody raised against the’antigen-binding site of an £ analyte-binding antibody.

For the detection of such small molecules by direct agglutination assay, at least two specific monoclonal antibodies could be used. One monoclonal antibody which is capable of binding directly to the small circulating antigen would be coupled to the erythrocyte binding molecule.

^rjC> The second (secondary) monoclonal antibody would be incubated with the c

j above conjugate and the analyte and would be capable of binding to a new

Λ antigenic determinant comprised of an overlapping region of the first ~ monoclonal antibody and the antigen that exists only when the first monoclonal antibody binds antigen. Thus, the second monoclonal antibody acts as the erythrocyte bridge, finally causes cross-linking between different red cells allowing agglutination to occur. This method, of course, is not restricted to monoepitopic analytes.

Because of spacial conformation, it may be difficult for a single secondary antibody molecule to bind simultaneously to two conjugate: analyte complexes. Thus, it may be preferable to conjugate the secondary antibody with an erythrocyte binding molecule.

In stating that a sample is to be incubated with a plurality of reagents it is to be understood that the contact may be simultaneous or sequential, without limitation to any particular order or duration of contact.

EXAMPLE 1 - Preparation of erthyrocyte binding molecule (anti-glycophorin antibody)

Immunization and Screening Procedure

Mice were immunized with human red blood cells and monoclonal

BAD ORIGINAL 0, antibodies produced by fusing the spleen cells of immunized animals with mouse myeloma cells. The antibodies were screened by both spin agglutination assay and enzyme immunoassay, where glycophorin was bound to a microtitre plate. Spin agglutination was performed by a modification of Hyatt & Street, Aust. J. Med. Lab. Sci, 4 48-50. 50 μΐ of cell culture supernatant was mixed with 50 μΐ of a 17. red blood cell suspension in a microtitre plate. For this example, antibodies which bound glycophorin, but did not agglutinate, were selected. The reaction of monoclonal antibody and glycophorin was determined by enzyme immunoassay. Microglates were coated with 10 micrograms/ml human glycophorin [Sigma Cat. No. G 7389] and washed, then incubated with serial dilutions of monoclonal antibody. After further washing, the presence of bound antibody was determined by the addition of enzyme labelled anti-mouse antibodies followed by the addition of substrate. The.titre was determined to the largest, di 1ution of

Λ monoclonal, which gave an A420 reading greater than 0.1 OD units above background. ~ ··

Of 384 wells, 40 primary clones were chosen. These gave either a positive spin agglutination assay, a response to glycophorin on EIA or both.

EIA	Sp i n	aaalutination	Number of clones
Negat i ve		Posi tive	4
Pos i t i ve		Pos i t i ve	20
Pos i ti ve		Negative	16
Subsequent	absorption studies were performed to	confirm that the
antibodies recognized a glycophorin domain exposed on	the red cel 1 surface.
The results	of the screening assays on ascitic	fluid are listed belov
	Asci tic Fluid Ti tre
	Spin		Red Ce1Γ
Clone	Agglutination	Glycoohorin EIA	Absorption
			Test
RAT 1D3/167	512000	<1000	Pos i t i ve
RAT 3D6/5	6400	1024000	Pos i ti ve
RAT 1C3/86	<1000	1024000	Pos i ti ve
RAT 3B1/172	256000	2000	Pos i t i ve
RAT 3D3/22	4000	1024000	Pos i ti ve
RAT 3D5/61	128000	1024000	Pos i t i ve' ' ‘

non ft 0 0 8 1

BAD ORIGINAL

4

RAT 1A2/187	<1000	256000	Pos i ti ve
RAT 2A2/187	<1000	128000	Positive
RAT 1A3/129	<1000	12800	Weak
RAT 1C4/5	<1000	128000	Positive
RAT 4C3/13	<1000	128000	Positive
RAT 3B1/70	<1000	517000	Pos i t i ve

f 8 0 ft ft n cia

RAT IC3/86 has been deposited under the Budapest Treaty, with the designation G 26.4.IC3/86, ATCC HB9893 at the American Type Culture Collection, 12301 Parklawn Drive, Rockville MD, 20852, USA on 7 September 1988. Purification of RAT 1C3/86

Monoclonal.antibodies were purified to homogeneity from ascitic fluids by chromatography on hydroxylapatite (Stanker, et al., J. Immunol. Methods 76, 157, 1985).

EXAMPLE 2: Preparation of HIV peptide - Ab conjugate < The spread of the human immuno deficiency virus (HIV-1) has become a major global health problem. At present there is no recognised cure or vaccine for this disease. The diagnosis of infected Individuals is a major factor in attempts to curtail the spread of the virus. Moreover, the need to prevent blood product contamination and protect health care personnel has increased the demand for simple, rapid, inexpensive and specific tests for the presence of anti-HIV antibodies.

We have made use of the patient's own red cells to provide a potential detection system for anti-HIV antibodies. This has been accomplished by selecting a non-agglutinating monoclonal antibody to human red blood cells. Chemically cross-linking this antibody with a synthetic HIV peptide antigen permitted specific agglutination of patients' red cells in the presence of antibodies to this antigen. The synthetic peptide antigen derived from gp41 of HIV-1 (residues 579-602), was chosen on the basis of the Welling procedure, FEBS LETT. 188: 215 (1985) and corresponds with the region identified as a major epitope recognised by antibodies from approximately 98% of AIDS patients. [Wang et al Proc. Nat. Acad. Sci. USA 83: 6259 (1986)).

Synthetic peptides were synthesized using the Merrifield procedure [RS Hodges and RB Merrifield, Anal. Biochem. 65, 241 (1975)) with the aid of an Applied Biosystems Model 430 synthesizer using double coupling cycles supplied by the manufacturer. The N-t-butyloxycarbonyl amino acid derivatives were obtained from the Protein Research Foundation (Osaka,

BAD ORIGINAL

Japan). Side chain protection was the same as supplied by Applied Biosystems with the exception of arginine for which the omega-NC^ derivative was used. Chain assembly was monitored using ninhydrin [V Sarin et al Anal. Biochem. 117, 147 (1981)]. The assembled peptides were simultaneously cleaved and deprotected using anhydrous HF containing 10% anisole (v/v) [JM Stewart and JD Young, Solid Phase Peptide Synthesis, pp44 and 66, WH Freeman, San Francisco (1966)]. The crude peptide was precipitated with diethylether, washed with ethylacetate, and extracted with 60% acetonitrile in 0.1% trifluoroacetic acid (v/v). Synthetic peptides were purified by preparation reverse phase chromotography (Amicon C-jθ resin, 250A pore size 25 x 400mm), eluting with a gradient of 1 ,000ml, to 60% acetonitrile in 0.1% trifluoroacetic acid. The synthetic peptide was approximately 95% pure as judged by analytical reversed phase HPLC and by quantitative amino acid analysis following acid hydrolysis.

1. SPDP labelling of the erythrocyte binding Ab (RAT 1C3/86)

To 0.25 ml of 13.8 mg/ml RAT 1C3/86 was added 12.5 pi of 2 mg/ml SPDP in dimethyl formamide and the reaction was allowed to proceed for 1 hour at 25°C. Unreacted SPDP was removed by gel filtration on Sephadex G 25 and the level of SPDP labelling (1.4 moles/mole) was determined.

2. Reduction of peptide 3.2

Peptide 3.2 (sequence RILAVERYLKDQQLLGIWGCSGK, corresponding to residues 579-601 of the major coat protein of HIV 1) was dissolved in 1 ml of lOOmM Tris HC1, ImM EDTA pH 8.0 and reacted with 10 pi of

2-mercaptoethanol for 45 minutes at 40°C. The reaction was terminated by the addition of 4 drops of trifluoroacetic acid (TFA) and 1 ml of aqueous 0.1% TFA. The mixture was applied to a Sep-pak (Waters) C 18 cartridge that had been treated with 20 ml of 60% acetonitrile, and equilibrated with 0.1% TFA. The reduced peptide was cycled through the Sep-pak twice before washing with 20 ml 0.1% TFA. The reduced peptide was eluted from the «

Sep-pak with 2 x 2 ml of 60% acetonitrile, 0.1% TFA. The sample was rotary* evaporated to dryness prior to coupling.

3. Conjugation

The peptide was dissolved in 0.2 ml of a buffer containing lOOmM potassium phosphate, lOOmM sodium chloride and 4M guanidine HC1 pH 7.4 and mixed with 2.2 mg of SPDP labelled antibody in the same buffer, but without guanidine HC1. The flask was incubated overnight at 25°C.

The degree of substitution of the antibody influenced the solubility of the conjugate; 20 moles peptide per mole of conjugate became insoluble.

*onη n 0 8 1

- 16 BAD ORIGINAL &

The range 5-7 moles of peptide per mole antibody was optimal. The capacity of the conjugate to bind red blood cells was monitored using the agglutination test with rabbit antimouse antibody and with HIV positive whole blood.

4. Gel Filtration Chromatography

Unreacted peptide and SPDP by-products were removed by gel filtration on a Superose 6 column (Pharmacia) in phosphate buffered saline and antibody containing fractions were pooled and stored at 4°C after addition of 0.01% sodium azide as a preservative.

5. Preparation of reagent for assay

Two volumes of conjugates were mixed with one volume of a lOmg/ml solution of an unrelated monoclonal antibody (Bruce 5) prepared as described in Bundesen, et al., Vet. Immun., Immunopath. 8, 245-260, 1985.

Assay procedure

For assay, 10 μΐ of heparinized whole blood was placed on a glass slide. 30 μΐ of reagent was added and mixed. The slide was rocked for up to three minutes and the presence of absence of agglutination noted.

Nine independent peptide/antibody conjugates were prepared and found ·> to be active in agglutinating seropositive patient’s red blood cells.

Active conjugate was also prepared using m-Maleimidobenzoyl-N-hydroxysuccinimide ester as the cross-linking reagent.

Results were at least comparable in accuracy to those observed with

S* an enzyme immunoassay using a similar antigen (Table 1).

Comparative testing of blood samples was by means of ELISA for the purpose of confirming positives and negatives obtained with the erythrocyte assay.

Control blood samples comprised ELISA negative blood samples and ELISA positive samples from infected patients. HIV positive patients were confirmed western blot positive by the Victorian State Reference Laboratory. Fairfield hospital patients were negative either by western blot or EIA (Abbott Laboratories). Blood donors were tested by EIA (Genetic Systems). False positive or negative values are given in parentheses and were verified by EIA or western blot analysis.

- 17 BAD ORIGINAL ft table ί

Au tolog ous Red Coll 'Agglutination Test

HIV + 7 ·> p a L i e η I s

Fair rield iljipi tai patients Heal th·/ blood donors

Agglut i na	Lion; »	ΕΙΛ	Test
Test	1 ί
+ve	-ve	+ ve	-ve -
42	< ί)	43	0
(3)	63 I	0	66 .*
(1)	873	(2)	872
ty of the	1 test 'a	series of

*. I . I s^·., v W 'w b I V. y s, V * synthetic peptides corresponding to other re'gions of tffe HIV—1 envelope jroteins was tested for (Table 2 ) residues u72-591, which their capacity’to inhibit agglutination reaction ho unrelated peptide competed and the synthetic gp41 fragment, is missing the ^ssential carboxytermina1 epitope

I

The inhibition of agglutination in confirming the occasional weak region, did not inhibit agglutination, with free synthetic antigen was useful positive samples. If addition of synthetic peptide had^- failed to inhibit agglutination it would hjive been indicative of a false positive related to the anti-red iolood cell mtibody.

Synthetic peptide (0.125 nig/ιτιΐ > v/as added to the conjugated antibody I I whole blood. jThe agglutination test wtis performed icn sequences are underlined.

TABLE 2 prior to the addition of as described above. Com

ADA ft ft 0 ft 1

Specificity of peptide inhibition of agglutination

Adc/ed Synthetic Peptide (sequence) inhibition of Aggl utir.at ion gp41 (5/9-601) ,up41 (572-591) gp!20 (.192-200) gp!20 (105-117) up 120 (101-1 IS) gp!20 ( IO5-I29)·/¹ none ;

I ;

RILaVe.RYLKDCOLLGINGC5GK

GIKQLARILAVERYLKADOQ asttJtnyt

HEDIISLHDQSLK i

VCQMiirDI ISLHDOSLKP ' HEDII5LHSQSLKPAVKLTP LCVSY

100%

0%

0%

0%

0%

0%

TLH/lOOc

- 18 BAD ORIGINAL

EXAMPLE 3 - Preparation of Chimeric Antibodies (anti-glycophorin/ anti-human D-dlmer) and use in assay for D-dlmer

Monoclonal antibodies RAT 1C3/86 (anti-human red blood cell) and DD—1C3/108 (anti-human D-dlmer as described by Rylatt, et al., 1983, Thrombosis Res., 31, 767-778) were digested with pepsin essentially as described by Hackman, et al., 1981. Immunology, 1_5, 429-436, and purified by chromotrography on a TSK-3000 SW column, 2 mg RAT 1C3/86 was digested for 45 minutes with 17. w/w pepsin in a buffer containing 0.1M acetic acid, 70mM sodium chloride pH 3.5. Meanwhile, 2 mg DD-1C3/108 was digested with

17. w/w pepsin for 2 hours In the same ouffer. The reactions were terminated by the addition of 1.5M Tris to raise the pH to 8. The F(ab)₂ fragments were purified by gel filtration chromatography on a TSK-3000 SW column.

Reduction of the F(ab)₂. and subsequent blocking of the Fab f~ fragment, was carried out as described by Brennan, et al ., 1985, Science

229, 81-83. A 3 mg/ml F(ab>£ preparation was treated with ImM mercaptoethylamine, in the presence of lOmM sodium arsenite, for 16 hours at 25°C. The Fab fragments were stabilized by reaction with 5.5'-dithiobis

-» (2-nitrobenzoic acid) (Ellman's reagent) for 3 hours at 25°C. The Fab ¹⁵ fragment was then purified by gel filtration chromatography on a TSK-3OOO

SW column.

The thiol form of DD—1C3/1O8 was regenerated by reaction with lOmM Λ mercaptoethylamine for 30 minutes at 25^eC. Excess reagent was removed by Φ gel filtration chromatography on a TSK-3000 SH column. A mixture of the thiol DD-1C3/108 and the Ellman's reagent treated RAT 1C3/86 was incubated for 16 hours'at 25’C as described by Brennan, et al .· Finally, the chimeric antibody was purified by further gel filtration chromatography on a TSK-3000 SW column.

Preparation of reagent

Two volumes of 0.1 mg/ml chimeric antibody was mixed with one volume of 7.5 mg/ml unrelated monoclonal antibody (Bruce 5).

Assay procedure

For assay, 10 pi of heparinized whole blood was placed on a glass slide. 30 μΐ of reagent was added and mixed. The slide was rocked for three minutes and in the presence of D-dimer agglutination was observed.

EXAMPLE 4 - Preparation of SPDP-coniuqated Di goxl n/Anti-qlycophorin antibody conjugate

Preparation of Diqoxin/Ab conjugate 1) Preparation of periodate oxidized digoxin.

BAD ORIGINAL &

ml lOOmM sodium periodate was added slowly, dropwise, to 40 mg of digoxin (Sigma), suspended in 2 ml 95% ethanol and the reaction was allowed to continue for 30 min at 37°C. The reaction was stopped by the addition of 60 pi of 1M ethandiol. Finally, the Schiff's base intermediate was stabilized by the addition of 2 ml 40mM cystine (30 min: 37°C) and subsequent reaction with 1 ml of 15 mg/ml sodium borohydride (16h: 25°C).

2) Reduction of cystine/digoxin conjugate.

ml of cystine/digoxin conjugate was reduced by the addition of 40 pi mercaptoethanol (40 min: 37°C) and the product purified by chromatography on a Waters Sep-Pak C 18 cartridge as described for the reduction of peptide in Example 1. After rotary evaporation, the sample was reacted with SPDP labelled RAT 1C3/86, which had been labelled with 5 propyldithiopyridine groups/antibody as described in Example 1 (16h: 25°C). Finally, the digoxin/antibody conjugate was purified by gel filtration chromatography on Superose 12.

EXAMPLE 5 - Preparation of HIV peptide/anti-glycophor in F(ab ^conjugate

An alternative reagent for the detection of anti-HIV antibodies uses an F(ab)£ derivative of the anti-glycophorin antibody.

RAT 1C3/86 (2 mg/ml in 70mM acetate lOOmM sodium chloride pH 3.5) was digested with 10 pg/ml pepsin (Sigma P6887) for 40 min at 37°C and the reaction terminated by the addition of 1/10 vol 1.5M Tris base. After overnight dialysis into a buffer containing 5mM sodium phosphate pH 8.0, the antibody fragment was purified by ion-exchange chromatography on DEAE cellulose on a 5-300mM gradient of sodium phosphate pH 8.0.

SPDP labelling of the Ffab)? fragment, reduction of the peptide 3.2, conjugation of peptide 3.2 to Ffab)? RAT 1C3/86, purification of the peptide conjugate, preparation of reagent for assay and testing procedure were carried out as described for the whole antibody conjugate.

F(ab)^ conjugates of other erythrocyte or analyte-binding antibodies may similarly be prepared and may be coupled to molecules other than the HIV peptide.

EXAMPLE 6 - Preparation of HIV peptide/anti-glycophorin Fab' conjugate

Another alternative reagent employs a univalent Fab' fragment as the

EBM.

Fiab)^ RAT 1C3/86 in phosphate buffered saline was incubated with lOmM mercaptoethanol for lh at room temperature. Then 15mM iodoacetamide was added and the reaction allowed to proceed for 15 min in the dark.

AP00008 1

- 20 BAD ORIGINAL ft

Finally, the reaction wa: terminated by the by dialysis Jnto 100 vol of phosphate buffered sal 1nL ' 1

SPD? labelling of .he s-carboxyme'thylated Fab', reduction of the i

peptide 3.2. conjugation of peptide 3.2 to the Fab'-TNB (thionitrobenzoyl) RAT fragment of RAT 1C3/|55, purification of the pept 1 de' conjugate, preparation of reagent for assay and testing procedure were carried out as described for the whole mtibody conjugate.

EXAMPLE 7 - Prepara11 on of Mel 1ttin as alternative E8M ~l I i

A peptide from the bee venom, melittln (CVl TTGLPALISWIKRKRQQ), was

A'.

, 'r

This peptide binds to the erythrocyte surface without lysing the cell (deGraJo KF, Kezdy F3, Kaiser ET. J Am Chem Soc 1981; M3; 679-SD. Ihe peptide was synthesised sy the Merrifield procedure (Hodges, Merrifield. Anal Biocnem 1 975; 6_5; 241). i J ⁴

One advantage of using me!a ttin as the EBM 1s that it and a peptide-type ASM may be ^synthesized as a single unit. !

EXAMPLE S - Use oj Av i di n-S lot.i J 11 nkaqe to couple EBM and ABM

I J i

The A5M and EBM need not be covalently coupled. One alternative is avidin-biotin linkage. r

Preparation of 31otin-1abel1ed melittlA

The mol ittin peptide (10 mg) wasjreduced with mercaptoethanol as described for the 3.2 peptide in Example 1. After rotary evaporation, the sample was resuspended ijn 2 ml 0.1M Tris-HCl, 5mM EDTA’pH S.O and reacted with 1 ml dimethylsulfoxide (DMSO), containing 4.3 mg [

N-lodoacetyl-N-biotinyltexylenediaminei (Pierce). This was allowed to react • I for 15 minutes at room temperature andthe biotinylated derivative

I ’ separated from byproducts on a Sephadex GIO column.

Preparation of Avidin-labelled peptide¹3.2 coupled to avjdin in the same manner as it was

The 3.2 peptide b coupled to antibody in Example 1 Assay

Sub-agglutinating doses of the avidin-labelled peptide are added 'to the red cells. Notes

It will be understood that the a'yidinated and biotinylated molecules may be interchanged.

Claims (5)

1. An agglutination reagent for direct detection ot an analyte in a blood sample, comprising an erythrocyte binding molecule coupled to an analyte binding molecule, wherein the erythrocyte binding molecule is coupled to the analyte binding molecule in such a manner that the binding characteristics of the binding molecules are not altered by

5 the coupling, and the erythrocyte binding molecule is capable of binding erythrocytes endogenous to the blood sample but does not agglutinate endogenous erythrocytes in the absence of analyte.

2. The reagent of claim 1, wherein said erythrocyte binding molecule is selected frqm the group consisting of: a non-univalent anti-erythrocyte antibody; a non-univalent io fragment of a non-univalent anti-erythrocyte antibody; a fragment of a non-univalent anti-erythrocyte antibody; an F(ab)2 fragment of an anti-erythrocyte antibody; an Fab' fragment of an anti-erythrocyte antibody; a peptide having an affinity for the erythrocyte membrane but incapable of lysing erythrocytes; a peptide having an affinity for the erythrocyte membrane but incapable of lysing erythrocytes and not derived from an is antibody or lectin; a nonlytic, erythrocyte-binding fragment of protamine which does not by itself agglutinate erythrocytes; mellitin; a specific binding fragment of mellitin; a lectin; a specific binding fragment of a lectin; an anti-glycophorin antibody; a specific binding fragment of an anti-glycophorin antibody; an anti-glycophorin A antibody and a specific binding fragment of an anti-glycophorin A antibody.

20 3. The reagent of claim 1, wherein the erythrocyte binding molecule comprises an anti-erythrocyte antibody, or F(ab)2 or Fab' fragment thereof and said antibody or F(ab)2 or Fab' fragment thereof is raised against glycophorin A, glycophorin B, integral membrane protein 1, membrane attached glycoprotein C4, integral membrane glycoprotein, ankyrin, spectrin, glycophorin, glycolipid, glycosphingolipid, protein 4-1

25 or F-actin.

4. The reagent of claim 1, wherein said erythrocyte binding molecule comprises a non-univalent anti-erythrocyte antibody produced by cell line G26.4.IC3/86 accorded ATCC Deposit No. HB9893 or an F(ab)2, or Fab' fragment of said antibody.

5. The reagent of of any one of claims 1 to 3, wherein said erythrocyte binding

3o molecule comprises a univalent fragment derived from an anti-erythrocyte antibody and said analyte binding molecule comprises a univalent fragment derived from an antianalyte antibody.

6. The reagent of any one of claims 1 to 3, wherein the analyte comprises an antibody to HIV.

35 7. The reagent of any one of claims 1 to 3, wherein the analyte binding molecule comprises a HIV-1 peptide, or hepatitis virus peptide or digoxin or antibody F(ab>2 or Fab' fragment thereof raised against human D-dimer, or hepatitis virus or an antiidiotypic antibody.

8. The reagent of claim 7 wherein the HIV-1 peptide comprises gp 41 peptide.

ΔΡ 0 0 0 0 8 1

BAD ORIGINAL £ —ι

9. The reagent of any one of claims 1 to 8, wherein the erythrocyte binding molecule is attached to the analyte binding molecule by means of covalent coupling.

10. The reagent of any one of claims 1 to

4, wherein the reagent comprises a plurality of different types of conjugates, wherein the analyte binding molecule of each type of

5 conjugate binds to a different epitope of the analyte.

11. The reagent of claim 1, wherein the erythrocyte binding molecule comprises a peptide having an affinity for the erythrocyte membrane but incapable of lysing erythrocytes and not derived from an antibody or lectin.

12. The reagent of claim 11, wherein the peptide comprises a bee venom-like peptide, io 13. The reagent of claim 1, wherein the erythrocyte binding molecule but not the analyte binding molecule binds a surface protein or glycoprotein.

14. The reagent of claim 1, wherein the erythrocyte binding molecule comprises a peptide having an affinity for the erythrocyte membrane and the analyte binding molecule is a peptide or protein, and said erythrocyte binding molecule and analyte is binding molecule are conjugated by a simple peptide bond.

15. The reagent of claim 14 in which the erythrocyte binding molecule corresponds to a nonlytic, erythrocyte-binding fragment of mellitin which does not by itself agglutinate erythrocytes.

16. The reagent of claim 1 which comprises a heterobifunctional antibody or a

20 heterobifunctional binding fragment of an antibody, said antibody comprising an erythrocyte binding molecule which binds erythrocytes but not the analyte, and a binding molecule binding the analyte but not erythrocytes, the erythrocyte binding molecule being conjugated to the analyte binding molecule by one or more disulfide bonds and not by heterobifunctional coupling agent, said reagent being prepared by forming a

25 heterobifunctional hybrid of a homobifunctional erythrocyte-binding antibody and a homobifunctional analyte-binding antibody, wherein said homobifunctional erythrocytebinding antibody does not auto-agglutinate erythrocytes.

17. The reagent of claim 16 in which the erythrocyte binding molecule binds glycophorin.

30 18. The reagent of claim 17 in which the erythrocyte binding molecule comprising an antibody for glycophorin or a specific binding fragment thereof.

19. The reagent of claim 16 in which the analyte-binding molecule binds human D-dimer.

20. The reagent of claim 16 which comprises a heterobifunctional F(ab)₂.

35 21. The reagent of claim 1 in which the analyte-binding molecule binds an antigen associated with heartworm.

22. The reagent of claim 1, further comprising a secondary antibody or specific binding fragment thereof which specifically recognizes an overlapping epitope formed by the binding of said analyte binding molecule to said analyte.

BAD ORIGINAL a

23. The reagent of claim 22, wherein said secondary antibody is attached to an erythrocyte binding molecule.

24. The reagent of any one of claims 1 to 23, wherein said erythrocytes are human erythrocytes.

5 25. An agglutination reagent for indirect detection of an analyte in a blood sample, comprising an erythrocyte binding molecule coupled to an analyte analogue wherein said erythrocyte binding molecule is coupled to said analyte analogue in such a manner that binding characteristics of the erythrocyte binding molecule are not altered by the coupling, and said erythrocyte binding molecule is capable of binding erythrocytes io endogenous to the sample but does not agglutinate endogenous erythrocytes in the absence of an analyte binding reagent.

26. The reagent of claim 25, wherein said erythrocyte binding molecule is selected from the group consisting of: a non-univalent anti-erythrocyte antibody; a non-univalent fragment of a non-univalent anti-erythrocyte antibody; a fragment of a non-univalent

15 anti-erythrocyte antibody; an F(ab)₂ fragment of an anti-erythrocyte antibody; an Fab' fragment of an anti-erythrocyte antibody; a peptide having an affinity for the erythrocyte membrane but incapable of lysing erythrocytes; a peptide having an affinity for the erythrocyte membrane but incapable of lysing erythrocytes and not derived from an antibody or lectin; a nonlytic, erythrocyte-binding fragment of protamine which does not

20 by itself agglutinate erythrocytes; mellitin; a specific binding fragment of mellitin; a lectin; a specific binding fragment of a lectin; an anti-glycophorin antibody; a specific binding fragment of an anti-glycophorin antibody; an anti-glycophorin A antibody and a specific binding fragment of an anti-glycophorin A antibody.

27. The reagent of claim 26, wherein the erythrocyte binding molecule comprises an

25 anti-erythrocyte antibody, or F(ab)₂ or Fab' fragment thereof and said antibody or F(ab)₂ or Fab' fragment thereof is raised against glycophorin A, glycophorin B, integral membrane protein 1, membrane attached glycoprotein C4, integral membrane &

glycoprotein, ankyrm, spectrin, glycophorin, glycoiipid, glycosphingolipid, protein 4-1 or F-actin.

30 28. The reagent of claim 25, wherein said erythrocyte binding molecule comprises a non-univalent anti-erythrocyte antibody produced by cell line G26.4.IC3/86 accorded ATCC Deposit No. HB9893 or an F(ab)₂, or Fab' fragment of said antibody.

29. The reagent of any one of claims 25 to 28, wherein the analyte comprises an antibody to HIV.

35 30. The reagent of claim 26, wherein the erythrocyte binding molecule comprises a peptide having an affinity for the erythrocyte membrane but incapable of lysing erythrocytes and not derived from an antibody or lectin.

31. The reagent of claim 30, wherein the peptide comprises a bee venom-like peptide.

32. The reagent of any one of claims 25 to 31, wherein the erythrocyte binding

AP 0 0 0 0 8 1

BAD ORIGINAL molecule is attached to the analyte analog by means of covalent coupling.

33. The reagent of claim 26, wherein the erythrocyte binding molecule comprises a peptide having an affinity for the erythrocyte membrane, the analyte analogue is a peptide or protein, and said erythrocyte binding molecule and analyte analogue are

5 conjugated by a simple peptide bond.

34. The reagent of claim 30 or claim 33 in which the erythrocyte binding molecule corresponds to a nonlytic, erythrocyte-binding fragment of mellitin which does not by itself agglutinate erythrocytes.

35. The reagent of any one of claims 25 to 34, wherein said erythrocytes are human io erythrocytes.

36. A direct agglutination assay for the presence of an analyte in a blood sample containing erythrocytes which assay comprises mixing the sample with an agglutination reagent according to any one of claims 1 to 24 for a time sufficient for binding to occur to analyte and erythrocytes to cause agglutination; observing whether the erythrocytes

15 are agglutinated and correlating the agglutination with the amount of analyte present in the sample.

37. An indirect agglutination assay for the presence or amount of an analyte in a blood sample which comprises incubating the sample with an agglutination reagent according ' “ to any one of claims 25 to 35 and a soluble analyte binding reagent, whereby said

20 agglutination reagent competes with sample analyte for the analyte binding sites of said analyte binding reagent, wherein the incubation is for a time sufficient for analyte to <-3· prevent binding of said analyte binding reagent to said analyte analogue to prevent

43 agglutination, observing whether agglutination occurs, and determining the presence or amount of said analyte from the inverse of the degree of agglutination.

25 38. The assay of claim 36 or 37 in which the sample and the conjugate are contacted essentially only with erythrocytes endogenous to the sample.

39. An agglutination assay test kit comprising an agglutination reagent according to any one of claims 1 to 35 and a reference solution containing a known quantity of analyte.