AU684715B2 - Reagent for agglutination assays - Google Patents

Description

AOJP DATE 10/03/94 PCT NUMBER PCT/AU93/00228 Iill111ll Il lll ll lll ll 111l AU9340523 INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREAIY (PC 0 (51) International Patent Classification 5 (11) International Publication Number: WO 93/24630 C12N 15/13, 15/70, C07K 15/12 Al C07K 15/28, GOIN 33/563, 33/577 (43) International Publication Date: 9 December 1993 (09.12.93) (21) International Application Number: PCT/AU93/00228 (74) Agent: SANTER, Vivien, Bedford; Griffith Hack Co., 509 St. Kilda Road, Melbourne, VIC 3004 (AU), (22) International Filing Date: 19 May 1993 (19.05.93) (81) Designated States: AT, AU, BB, BG, BR, CA, CH, CZ, Priority data: DE, DK, ES, Fl, GB, HU, JP, KP, KR, KZ, LK, LU, PL 2551 22 May 1992 (22.05.92) AU ?MG, MN, MW, NL, NO, NZ, PL, PT, RO, RU, SD, SE, SK, UA, US, VN, European patent (AT, BE, CH, Di, DK, ES, FR, GB, GR, IE, IT LU, MC, NL, PT, SE), (71) Applicant (for all designated States except US): AGEN LI- OAPI patent (BF, BJ, CF, CG, Cl, CM, GA, GN, ML, MITED [AU/AU]; 11 Durbell Street, Acacia Ridge, MR, NE, SN, TD, TG).

QLD 4110 (AU).

(72) Inventors; and Published Inventors/Applicants (for US only) LILLEY, Glenn, 1Wlth international search report.

George [AU/AU]; Lot 2, Munts Road, Upper Plenty, VIC 3756 HUDSON, Peter, John [AU/AU]; 6 Cabena Street, Donvale, VIC 3111 HILLYARD, Carmel, Judith [AU/AU]; 21 Gulliver Street, Paddington, QLD 4064 (AU).

(54) Title: REAGENT FOR AGGLUTINATION ASSAYS (57) Abstract The invention provides a bifunctional recombinant protein comprising a particle-binding antibody or antibody fragment (PBM), and an analyte-binding moiety or molecule (ABM). Preferably, the particle-binding antibody or antibody fragment is an erythrocyte-binding antibody or antibody fragment (EBM), and/or the ABM is selected from the group consisting of an antigenic peptide from an immunodominant region of an env protein of HIV-I or HIV-2, a gag protein of HIV-1 or HIV-2, and an immunodominant region of the surface antigen of Hepatitis B. Alternatively, the ABM is a single chain Fv region of an antibody direct-i ed against an antigen selected from the group consisting of Hepatitis B surface antigen, D-dimer and canine heartworm antigen.

In a particularly preferred embodiment, the EBM is a single chain Fv region of an anti-erythrocyte antibody, such as an anti-glycophorin antibody. The protein of the invention is particularly suitablr for use in agglutination immunoassays. Methods for producing the protein of the invention, DNA sequences, and assay methods and reagent kits therefor are also provided.

i"; WO 93/24630 WO 9324630PCT'/A U93/00228

I

REAGENT FOR AGGLUTINATION ASSAYS The present invention relates to a reagent for use in agglutination assays, and in particular whole blood agglutination assays. The invention also relates to a method for detecting an antigen, antibody or other analyte in a sample using the reagent, and to a kit containing the reagent. The invention describes the use of recombinant DNA methods in B. ccli to produce the key reagents for this assay.

This application claims priority from Australian Provisional Patent Application No. PL 2551, the entire disclosure of which is herein incorporated by reference.

Background of the Invention Immunoassays and analogous specific binding assays are now very well-known and widely used in a variety of biomedical and other fields. The most commonly used immunoassays utilise complex detection systems involving radioisotopes or enzymes, and suffer from the disadvantage that the assay procedure is lengthy and involved, and requires expensive instrumentation. Radioijmmunoassays further suffer from the disadvantage of the radioactive hazard presented by the isotopes. Agglutination immunoassays, using erythrocytes or latex particles as the detection agent, have been proposed as an alternative.

Immunoassays and agglutination immunoassays are described in our International Patent Application WO 91/04492, entitled "Agglutination Assay". in particular, our U.S. Patents No.

4,894,347 and No. 5,086,002 describe an agglutination immunoassay designed for use with whole blood samples, in which the endogenous erythrocytes are used as indicating particles, and in which an agglutination reagent comprising an erythrocyte binding molecule conjugated either to an' analyte-binding molecule or to an analyte analogue is used.

WO 93/24630 PCT/AU93/00228 2 Non-specific agglutination is avoided if the erythrocyte binding molecule recognises an abundant, well-distributed erythrocyte membrane constituent such as glycophorin. WO 91/04492 describes an autologous agglutination assay of improved sensitivity. The entire disclosures of U.S. Patent No. 4,894,347 and International Patent Application No. WO 91/04492 are also incorporated herein by reference.

Conventional immunoassays, an3 some agglutination assays, require the isolation of serum or plasma, which in turn usually requires electrical power and specialised equipment, and consequently is very difficult under field conditions or in remote or under-developed areas. It is therefore highly desirable to be able to use a test system which can utilise whole blood, and which requires a minimum of sophisticated apparatus. Test systems for use under field conditions should be stable, rapid, reliable and specific, and should provide a clear-cut demarcation between positive and negative results. In order to be cost effective, such a system should require the minimum number of reagents, which in themselves should be easy to produce. The system described in US-4,894,347 and WO 91/04492, which is marketed as SimpliRED, MicroRED and VetRED tests (Trade Marks of Agen td), meets the requirements of simplicity and ease of use under difficult conditions, and requires a minimum of equipment. It is however, expensive to produce the reagents.

Two main types of reagent are desired for use in these immunoassays, namely antigen-antibody constructs and bispecific antibody constructs. The currently manufactured antigen-antibody reagents utilise, for example, an antigenic peptide from an immunodominant portion of HIV virus (HIV-1 or HIV-2), coupled chemically to the Fab fragment of antibody which is able to bind to glycophorin A on the red cell surface. Alternative reagents of the antigen-antibody type utilise a larger protein, rather than an immunodominant WO 93/24630 PCT/A U93/00228 3 peptide, for example hepatitis B surface antigen.

Bispecific antibody and Ab-peptide conjugate reagents are currently manufactured by a series of steps involving chemical and enzymic manipulation of antibodies; they consist of two Fab molecules with differing specificities, linked by disulphide bonds at the hinge region. The resultant bispecific F(ab) 2 molecule reacts both with an indicator reagent, such as an erythrocyte, and a circulating antigen in a blood sample.

Antibodies (Abs) and Ab fragments can be produced by recombinant DNA technology (Winter and Milstein, Nature, 1991 349 293; U.S. Patent No. 4,946,778 by Ladner et al; Australian Patent No. 612,370 by Creative Biomolecules, Inc., using either mammalian cells (Oi, V.T. et al, Proc. Natl.

Acad. Sci. USA, 1983 80 825-829) or bacteria (Boss, M.A. et al, J. Nucl. Acids Res., 1984 12 3791-3806 and also U.S.

Patent No. 4,816,397 by Boss et al; Cabilly, Proc. Natl.

Acad. Sci. USA, 1984 81 3273-3277 and European Patent No.

125,023 by Genentech Inc. and City of Hope). In the Fab region the combination of two chains (heavy and light) provides six variable surface loops at the extremity of the molecule. These loops in the outer domain (Fv) are termed complementarity-determining-regions (CDRs), and provide the specificity of binding of the Ab to its antigenic target.

Binding function is localised to the variable domains of the antibody molecule, which are located at the amino terminal end of both the heavy and light chains. The variable regions remain noncovalently associated (as V1VL dimers, termed Fv regions) even after proteolytic cleavage from the native antibody molecule, and retain much of their antigen recognition and binding capabilities (see, for example, Inbar et al, Proc. Natl. Acad. Sci. USA, 1972 69 2659-2662; Hochman et al, Biochem., 1973 12 1130-1135 and siochem., 1976 15 2706-2710; Sharon and Givol, Biochem., 1976 WO 93/24630 PCT/AU93/00228 4 1591-1594; Rosenblatt and Haber, Biochem., 1978 17 3877- 3882; Ehrlich et al, Biochem., 1980 19 4091-4096). Methods of manufacturing two-chain Fv substantially free of constant region using recombinant DNA techniques are disclosed in US- 4,642,334 and corresponding published specification EP- 088,994.

We have now found that by using recombinant DNA technology, it is possible to improve and to broaden significantly the applicability of the assay described in US- 4,894,347 and WO 91/04492. These references teach the application of either a bispecific antibody F(ab), fragment, one half of which binds to erythrocytes and the other to the analyte, or an Fab fragment of the red cell binding antibody attached to a specific peptide. The reagents are manufactured in a series of steps, firstly by digesting the purified mouse antibodies with an enzyme to remove the Fc region, then reduction to Fab, blocking and conjugation.

Each stage in the process, but more importantly the entire bioprocess, can be simplified by the use of genetically engineered reagents. For example, oligonucleotide synthesis can provide the gene fragments that encode the various Cterminal peptide tails that constitute analyte specificity.

The red cell binding molecule, providing it has sufficient affinity, may be an Fv fragment rather than a complete Fab.

Alternatively single chain scFv, or smaller domain structures, can be engineered which may have advantages for product stability and yield. Further improvements to the reagents include the elimination of mouse constant domains, with resulting increased specificity, and improved solubility properties. We have been able to prbduce bifunctional recombinant proteins comprising an antibody variable region domain, together with either a second antigen-recognising domain or an antigenic region.

Expression systems that are available for the WO 93/24630 PCT/A U93/00228 production of antibody fragments include E. coli and alternative prokaryotes, yeast, baculoviral vectors and mammalian cells. We have developed novel E. coli secretion vectors which (Power et al, Gene 1992 113 95-99) now allow the expression, to exceptionally high levels, of the V,/V,/scFv domains of anti-neuraminidase Abs. Downstream processing has been addressed in the context of high-level Ab-domain production. A number of denaturation/ renaturation regimes have been tested, and molecular "flags" incorporated into the expressed antibody domains to aid in purification and conformational assessment. Current physical tests for protein conformation and binding affinity include ELISA, fluorescence quenching, circular dichroism, airfuge centrifugation, and biosensor applications.

We have surprisingly found that the activity of the complementarity determinants at the very ends of IgG Fab arms is maintained, even after 75% of the supporting molecular structure normally present in IgG molecules is removed. The affinity of the antigen-binding domains is not significantly affected; nor does the removal of the supporting molecular structure appear to decrease the stability of the molecule.

The synthesis of antibody variable region domains in recombinant organisms has the potential to enable the production of reagents which might otherwise be impossible to manufacture, such as constructs using large recombinant proteins, many of which are usually produced as insoluble molecules for solid phase assays. Often an antigen will not produce a single immunodominant response in an infected host, and several epitopes from an antigen are necessary to detect circulating antibodies. In such a case, several peptides as a recombinant construct with Ab or fragments avoid the need for the multiple Fab-peptide conjugates. The use of multiple conjugates requires large amounts of blocking reagents to avoid non-specific agglutination resulting from interaction WO 93/24630 PCT/AU93/00228 6 of the Fab constant regions. Thus, the ability to express the bifunctional molecules in a recombinant host dramatically decreases manufacturing costs for reagents which otherwise would require complex chemical synthesis and additional blocking reagents.

Summary of the Invention The invention provides an assay, utilizing a series of reagents produced by recombinant DNA technology, that is useful for the detection of drugs, hormones, steroids, antibodies, and other molecules in a biological fluid, particularly in blood. Technology for this assay and these reagents is taught which provides a sensitive assay and a means to produce the key reagents as recombinant antibody molecules, including single chain antibody molecules, in E.

coli, or in other expression systems known to the person skilled in the art.

According to one aspect of the invention, there is provided a bifunctional recombinant protein comprising a particle-binding antibody or antibody fragment (PBM), and an analyte-binding moiety or molecule (ABM).

The analyte may be an antigen or an antibody.

Preferably the particle binding antibody or antibody fragment is an erythrocyte binding antibody or antibody fragment

(EBM).

Preferably the ABM is selected from the group consisting of an antigenic peptide from an immunodominant Sregion of the env gp41 protein of HIV-1 or HIV-2, and one of the gag proteins, and an immunodominant region from the, surface antigen of Hepatitis B surface antigen. The specific ABMs may be produced by expression from gene fragments, for example, from synthesised oligonucleotides, that encode peptides which constitute the analyte specificity.

In an alternative embodiment, the ABM is a single WO 93/24630 PCT/A U93/00228 7 chain Fv region of an antibody directed against an antigen selected from the group consisting of Hepatitis B surface antigen, D-dimer and canine heartworm antigen.

Preferably the EBM is a single chain Fv region of an anti-erythrocyte antibody, more preferably an antiglycophorin antibody.

The use of single chain Fv region in the construct presents the advantage that the constant region of the antibody is almost completely removed, and consequently there is less opportunity for interference by heterophile antibody in the final assay, and the manufacture of a complete reagent is more efficient that would be the case if no blocker antibody were used.

We have found that the orientation of the ABM in relation to the EBM is critical to the sensitivity and specificity of the final product.

In a particularly preferred embodiment, the EBM is the single chain Fv domain of the anti-glycophorin A monoclonal antibody produced by the hybridoma cell line G26.4.1C3/86, which is described in US-4,894,347, and W091/04492. A sample of this cell line was deposited under the Budapest Treaty at the American Type Culture Collection (12301 Parklawn Drive, Rockville MD, 20852) on 7 September 1988, and received the ATCC accession number HB9893.

In a second aspect of the invention, there is provided a DNA sequence encoding as a single transcriptional unit a particle-binding moiety operatively linked to an analyte-binding moiety, as well as expression vectors and host cells comprising said sequence.

A third aspect of the invention provides assay methods and kits utilising the recombinant protein of the invention.

Although the use of an Fv region of an antibody is preferred, it should be clearly understood that the invention WO 93/24630 PCT/AU93/00228 8 includes within its scope the use of Fav, F(ab), or VI fragments of antibodies.

The host cell may be any of those currently used by those skilled in the art of expression in recombinant organisms, and is preferably E. coli. However, it will be clearly understood that other hosts, such as other bacteria, yeasts or insect, mammalian or plant cells may be used. The E. coli expression vectors described herein are novel, particularly with respect to the design of protease-resistant 'tails' with the unique features required by the diagnostic test. We have optimised the induction regime and fermentation conditions for high-yielding production.

The DNA encoding both the erythrocyte binding activity and the specific analyte binding activity may be located on a DNA element capable of replication and the expression of the genes for the bifunctional reagents. This DNA element may be a plasmid or any equivalent DNA element capable of replication and expression in an appropriate host.

The portion of the bifunctional reagent which has specific analyte binding activity may be encoded by DNA which has been produced from cells and tissues by any of the standard techniques known in the art for the amplification of DNA, such as the polymerase chain reaction, the ligase chain reaction, or isothermal amplification.

The use of recombinant bifunctional reagents provides the following advantages: 1. Simplification of current production procedures: No chemical coupling through disulphide bonds is necessary.

The bifunctional fusion protein is made as a single polypeptide chain.

2. Any two of a wide range of analyte-binding molecules can be incorporated into a bifunctional single WO 93/24630 PCT/AU93/00228 9 polypeptide chain.

3. Ease of manipulation to produce modified bifunctional single polypeptide chain by mutation of the DNA.

4. No batch to batch variation.

5. Expression from host cells produced large amounts of soluble polypeptide.

6. Ease of identification, isolation and purification.

Less expensive to produce.

7. Increased scope of bifunctional reagents.

8. No dependence on high levels of protein production from hybridomas.

9. Recombinant DNA techniques make infinite permutations possible.

Detailed Description of the Invention In the agglutination assay of this invention, a recombinant reagent is provided which is derived from cloned DNA coding for the erythrocyte binding antibody, which as a result of genetic manipulation is fused to an analyte binding molecule encoded by the gene or gene fragment for the specific analyte binding molecule without substantially changing the binding characteristics of the binding portion.

The reagent is non-agglutinating when incubated with endogenous erythrocytes in the absence of the analyte.

The invention will be described in detail by way of reference only to the following non-limiting examples, and to the drawings in which: Figure 1 illustrates the sequence of the IgG (1C3/86) gamma chain derived from clone gammal.l.la. The nucleotide and deduced amino acid sequence (mature sequence shown in bold type and single letter code) of 1C3/86 IgG gamma-l..lla are shown; Figure 2 illustrates the sequence of the IgG WO 93/24630 PCIT/AU93/00228 (1C3/86) kappa chain derived by PCR amplification and clone K4AC1/C2. The nucleotide and deduced amino acid sequence (shown in bold type and single letter code) of mature 1C3/86 IgG kappa chain (sequence is a composite of that determined from clones K4AC1/C2 and the gene amplified directly from mRNA by polymerase chain reaction) are shown; Figure 3 illustrates the strategy for the amplification of 1C3/86 gamma and kappa gene variable domains and the construction of the scFv in expression vector pPOW. PCR primer-template combinations used to amplify various antibody fragments are shown.

Figure 4 illustrates the strategy for the amplification and cloning of 1C3/86 scFv in expression vector pHFA. PCR primer-template combinations used to amplify various antibody fragments are shown.

Figure 5 illustrates the strategies for amplification and cloning of scFv's with combined FLAG and HIV immunodominant peptide epitopes in the expression vector pHFA,,. PCR primer-template combinations used to amplify various antibody fragments are shown.

Figure 6 illustrates the vectors pPOW, pHFA and pHFAS,, used for the construction and expression of the 1C3/86 scFv (described in figures 3,4 and 5) with pertinent cloning sites. Ampr; ampicillin resistance gene, ColEl or Ori; E.coli origin of replication M13 ORI; M13 phage origin of replication, Gene3; gene 3 phage surface protein gene, Amber; amber stop codon,(TAG) fD; transcription terminator, placZ; lacZ promoter, c1857; lambda heat labile repressor gene, P, and P 1 lambda phage right and left promoters, FLAG; gene for epitope recognised by M2-anti-flag IgG and pelB; gene for pectate lyase signal sequence.

Figure 7 illustrates the protein sequences of peptide epitopes which may be generated by PCR reaction and linked in the reaction or added by recombinant DNA WO 93/24630 PCT/AU93/00228 11 techniques.

Figure 8 illustrates the activity of the recombinant protein in ELISA assays.

The anti-glycophorin A monoclonal antibody 1C3/86 was selected as a model antibody. The gene encoding 1C3/86 IgG was cloned into an Escherichia coli host, and the nucleotide sequence of the antibody was determined.

Synthetic oligonucleotide primers were designed in order to enable the variable domains of the antibody to be cloned, linked together to form a single chain Fv domain (scFv), into various expression vectors. Various peptide epitopes were added to the C-terminus of the scFv molecule.

Example 1 Isolation and Characterisation of Genes Encoding Antibody Fragments A strategy utilising the polymerase chain reaction (PCR) to identify segments of the genes encoding the antibody and to add linkers and peptide epitopes to those segments to form single chain, antibody-based reagents was adopted.

a) Messenger RNA (mRNA) was prepared from a monoclonal cell line (1C3/86), referred to above, which produced antierythrocyte IgGs which bound with high affinity to RBCs but did not produce auto-agglutination.

From this mRNA template, single and double stranded complementary DNA (ss- and ds-cDNA respectively) were synthesised. The ds-cDNA was cloned into lambda-gtlO arms and packaged into a phage library. The heavy chain clone gamma-M/1.1 (Tyler et al, Proc. Natl. Acad. Sci., 1982 79 2008-2012) and the light chain clone pH76-kappa-10 (Adams et al, Biochem., 1980 19 2711-2719) were used to source ds-DNA inserts for the screening of the gtl0 library. Positive clones were amplified, and the positive insert cDNA sub- WO 93/24630 PCT/AU93/00228 12 cloned into pUC18. As a result, a near full-length gamma clone (gamma-l.l.la) was identified, the nucleotide sequence was determined and from this the protein sequence was deduced (Figure The sequences of a partial kappa clone (kappa- 4AC1) which encoded the 3' end of the variable domain and full constant domain were determined in a similar fashion.

To determine the nucleotide sequence of the 1C3/3 kappa light chain at the 5' end, the following approach was adopted. A mixed N-terminal sequence (see below) was first determined for the intact 1C3/86 Ig in an Applied Biosystems sequencer.

mixed sequence D/E I/V V/R M/L S/L Q/E S/S P/G S/G (automatic sequencer) From the mixed amino acid sequence above and the sequence deduced from a gamma heavy chain clone as follows: gamma chain E V R L L E S G G (clone 1.1.la) the N-terminus of the variable region of the kappa light chain, not present in gtlO library clones, was determined to be; kappa chain D I V M S Q S P S (deduced by difference) From this sequence for the N-terminus of the kappa chain above an approximation of the 5' light chain variable region was compiled by applying common usage triplet codes found in IgG genes (see oligonucleotide N960 in Table 1 below). The light chain variable region gene was then amplified by PCR using the redundant, forward (sense) primer WO 93/24630 PCT/A U93/00228 13 N960 and the reverse (antisense) primer N852 (see Table 1), which was based on the kappa constant region beginning at nucleotide 337 (see Figure as described by Chiang et al, Biotechniques, 1989 7 360-366. The amplification reaction yielded a single product, which when cloned and sequenced showed a coding sequence consistent with a kappa light chain and identical at the 3' end with the overlapping kappa clone K4AC1. The sequences derived from PCR and gtI library enabled the compilation of the sequence shown in Figure 2.

Table 1 Forward N 907 (sense) oligonucleotides: GGG GTC GCG GAG GTG AGG CTT CTC CCC GCC AGA CGT/C GAT/C ATT/C GTG/C ATG N 960 N 978 N 979 N1237 CCC ACG GTC ACC GTC TCA GGA GGA GGA GGT TCA GGA GGA GGA GGT GAC ATC GTC ATG AAA AAA GCG GCC CAG AGG CTT CTC GAG TCT GGA GGT GGC CCG CTG AAA CTC TCC ATG GCG GAG GTG AGG GGC CCG G GCC TCC GGT GGT GGT GGT TCG GGT GGT GGT GGT TCG CCG GCC ATG GCC GAG GTG N1479 GTA CAA CCT GGA GGA TCT N1617 CTT CTT GAG TCT GGA GGT NSfils CAT GCC ATG ACT CGC GGC CCA GCC GGC CAT GGC C(C/G)A GGT (A/G)CT GCA G(C/G)A GTC (A/T)GG WO 93/24630 WO 93/24630PCI/A U93/00228 Reverse complementary oligonucleotides: *N852 CC GAA TTC GAT GGATAC AGC CCO 908 GAC GGC CAG GAT ACG GCC CAG AGT N 909 GCA GCC CCA OAT GCC CAG CAG ATA ACG TTC GAC GOC N 911 GAC GGC CAG GAT ACG CCG GGT GCC N 976 GGG GAA TTC TTA AGA CGC COT GOT GCA GAT N1294 GAC GGC CAG GAT ACG TTT N1296 GAC CGC COT OGT GCA GAT GCC CCA GAT CC N1645 AAA AAA CO CGG GAA TTC N1646 AAA AAA CCG CGO GA7A TTC AGT TOG TGC AGC ATC OGC GGA GAC GOT GAC

CAG

CAO

TTT

CTG CTO ATC TTT OAT ACO AAT CTC GAG CTT ATT CGO GAC CC

ATC

CAG

TTA

TTA

ATC

TTT

AGA

ACA

ATC

0CC

COO

CAC

ATC

AGA OCA

ATT

CTG

NV7CFORK0'20 eqpuimolar mixture of: GAG TCA TTC TOO 0CC COO COG TTT GAG TCA TTC TOO GOC CCC CCC TTT GAG TCA TTC TGC GC CGO CCC TTT GAG TCA TTC TGC GOC COO CCO TTT OAT TTC CAG CTT GOT GCC TAT TTC CAC OTT GOT CCC TAT TTC CAA OTT TGT CCC CAG CTC CAG CTT CGT CCC WO 93/24630 PCT/AU93/00228 Primers are 5' to 3' (left to right). Forward (sense) oligonucleotides translate to the amino acid sequence of the expressed protein segment whereas reverse primers need to be reversed and complemented.

"'ab"Making antibodies in bacteria and on phage", EMBO Practical Course Manual, IRBM, Pomezia, Italy b) A single chain antibody fragment (scFv) was constructed from the 1C3/86 molecule as follows: i) Amplification and cloning of the heavy-chain variable domain.

1. The amplification of genes and synthesis of DNA sequences in these genes for cloning were performed by application of the polymerase chain reaction (PCR) as follows. A typical reaction (100 1 volume) contained 1-10 ng of template DNA, 1-2 U of thermostable DNA polymerase, 1 of a mixed A,C,G and T deoxynucleotide (dNTP solution) with each base at a concentration of 2 mM, 5 1 of each terminal primer (10 pMolar each) and, where used, 1 1 of internal primers (0.05-0.1 pM), Mg++ to a final concentration of 1-5 mM, a reaction buffer appropriate for the particular polymerase chosen (supplied by manufacturer), and water to 100 1. The reactants were mixed and overlayed with paraffin oil (Sigma biochemicals) and subjected to 25-30 cycles in a thermal cycler (Corbett Research, Australia). The general strategy for each of the examples in Figures 3, 4 and consisted of a denaturation step at 93 0 C (usually 1 minute), an annealing step between 50 and 65 0 C for 1 minute and an extension step at 72 0 C for 2 minutes. Annealing temperatures were adjusted as required to give final product.

2. Oligonucleotide primers (Table 1) were synthesized to amplify the variable domain (Vh) from the WO 93/24630 PCT/AU93/00228 16 heavy chain cDNA clone gamma-1.1.la, and to add a Thai restriction site at the 5' end (N907) and a Bst E2 peptide epitope -Eco R1 sequence at the 3' end (N908/N909/N1296/N976), a described in Figure 3A. The product was digested with Tha 1 and Eco Ri, and cloned into the Mac 1/Eco R1-digested expression vector pPOW (Power et al, Gene., 1992 113 95-99) (Figure 6A), and transformed into E coli strain TG-1 (Gibson 1984, "Studies on the Epstein-Barr virus genome", Ph.D. t' is, Cambridge University, England) 3. Transformed E coli were screened for the presence of plasmids carrying the V, gene fragment and selected clones (hereafter referred to as pPOW1C3Vhn) were sequenced to check the integrity of the cloning procedure.

These clones are identified in PL 2551 as pP1C3Vh ii) Amplification and cloning of the light chain variable domain and construction of composite single-chain antibody (scFv) reagents 1. Oligonucleotide primers were synthesized to simultaneously amplify (as in i(1) above) and add to the cloned light chain gene, in a PCR amplification reaction, a Bst E2 site and a sequence coding for a linker (amino acid sequence -(GGGGS),- at the 5' end (N978/N979) and a peptide editope-Eco R1 sequence at the 3' end (N911/N909/N1296/N976) as described in Figure 3B.

2 The V, product described in Figure 3B was digested with Bst E2 and Eco R1 and cloned into the Bst E2/Eco R1 digested plasmid construct, pPOW1C3scVha n above (Figure 3A).

3. Transformed E. coli (TGI) were screened for the presence of the V, and V, sequences and selected clones, hereafter referred to as pPOWlC3scFva, (Figure 3C), were partly sequenced to check the integrity of the cloning WO 93/24630 PCT/A U93/00228 17 procedure. These clones are identified in PL 2551 as pPlC3scFv.

4. Oligonucleotide NSfi15 and NVKFORNOT (Table 1) were used to add Sfi 1 and Not 1 restriction sites (by PCR amplification) to the 5' and 3' ends respectively of the scFv gene construct in pPOWIC3scFv,,,( see Figure 4) in this amplification, the gp41 HIV1 epitope was removed. The PCR product was digested with these restriction enzymes and cloned into the likewise restricted vector pHFA (see Figure 6B) which contains the alternative octapeptide FLAG tag (Figure 7A) pHFA is the parent of the vector pHEN (Hoogenboom et al, 1991) The construct was then transferred into the E. coli strain HB2151, a strain in which the nucleotide sequence TAG (amber mutation) is recognised as a stop codon Clones, referred to as pHFAIC3scFvn, were identified by hybridization, were sequenced, and were tested for expression 'of a scFv-peptide fusion as evidenced by reactivity to the M2-anti FLAG antibody (IBI Corp., USA).

The HIV1 and HIV2 epitopes (Figures 7B and 7C) -were added back to the scFv to give plasmid constructs pHFAAlC3scFvrn ~m and pHFAElC3scFv ,ar (Figures 5A and respectively). In this procedure, the FLAG epitope and alternative HIV epitopes were added to the scFv gene in pHFA described in ii)4 above by PCR amplification. The sequence changes to the Vh gene, introduced in pHFA constructs, were returned to the native and a BamHl restriction site adjacent to the 5' end of the gene was removed by the use of the forward oligonucleotides N1479, N1617 and N1237. At the 3'end of the FLAG sequence terminating the scFv gene construction in pHFA, oligonucleotides N1294, N909, N1296, N976 and N1645 introduced a HIV1 epitope, Eco RI and Sac 2 sites. In a similar fashion the HIV2 epitope and restriction siteb were added with 3' oligonucleotides N1297, N1311, N1310 and N1646 (Table PCR products were WO 93/24630 PCT/AU93/00228 18 restricted with Sfi 1 and Sac 2 were cloned into the likewise restricted vector pHFAAc (Figure 6C), a derivative of pHFA.

In this procedure, the FLAG sequence in the vector was deleted and replaced with the FLAG sequence of the PCR construct but was between the scFv and the HIV epitopes a TAA stop codon was included so that in suppressor or nonsuppressor cell lines, translation would terminate after the HIV epitope.

c) Expression of recombinant scFv Recombinant E. coli were grown in 10mls of 2X-YT medium (10 gm yeast extract, 15 gm tryptone, 5 gm NaCI per litre) overnight at 30 0 C in the case of pPOW constructs, and at 370 in the case of pHFA constructs. Overnight cultures were diluted to an OD(, 0 of 0.05 into 100ml of fresh medium glucose was included in the case of pHFA constructs) and grown to mid-log phase (OD 0 0.5-0.9).

Cultures of pHFA plasmids were induced upon the addition of isopropyl- -D-thiogalactopyranoside; (IPTG; Sigma 15502) to a concentration of ImM, and growth continued at 30 0 C as required.

Cultures of pPOW were induced by raising the temperature of the medium to 42oC for 15 minutes, after which the incubation was continued at 37 0 C for 2-4 hours.

Levels of recombinant proteins in the 3. coli periplasmic space and the culture supernatant in each case were assayed by ELISA, Western blots of SDS-PAGE gels, and by the agglutination assay described below.

Example 2 Activity and Expression Levels of Recombinant scFv i) Western analysis Periplasmic proteins were isolated by suspending the E. coli in 25% w/v sucrose/10 mM Tris-HC1 (pH 7.5) and 16 mM EDTA. Cells were then collected by centrifugation and WO 93/24630 PCr/A U93/00228 19 resuspended in ice-cold water. The particulate material and the soluble fractions were analysed by SDS-PAGE followed by Western blot. Active expression was assessed by the presence of a product of apparent M, 30 Kd. Mouse antibodies directed against the C-terminal FLAG peptide (M2 anti FLAG) or HIV epitopes (IBI or 2B4), included in scFv constructs, were used as primary antibodies in this analysis, and were detected with horse-radish peroxidase bound to goat anti-mouse IgG in the normal manner.

ii) ELISA assay Supernatants from E. coli cultures and purified scFv reagents were assayed by ELISA. The assay was performed as follows: 1. "Nuclon" plates were coated overnight with 100 1 of 10 g/ml human glycophorin-A (Sigma) in PBS.

2. Washed plate wells 3X with PBS.

3. 'Block with 200 1 of 2% skim milk in PBS for 2h.

4. Wash 3X with PBS.

Add 20 1 10% skim milk in PBS and 80 1 culture fluid or purified antibody in PBS and incubate at 0 C for 20 mins.

6. Wash 3X with PBS/0.05% 7. Wash 3X with PBS 8. Add 100 1 of 2 g/ml anti-tag antibody in PBS/2%(W/V) skim milk powder and incubate for 60 mins at 20 0

C

9. Repeat steps 6 and 7 Add 100 1 of 1-2 g/ml HRP-goat anti-mouse IgG antibody in PBS/2%(W/V) skim milk powder and incubate for mins at 11. Wash as in steps 6 and 7 12. Add 100 1 of activated ABTS solution (see below) and develop for 30 mins at WO 93/24630 PCT/A U93/00228 13. Quench by adding 50 1 of of 3.2 g/1 of sodium fluoride and read at 405nm ABTS (2,2 azino di (3-ethyl)benzthiazoline sulfonic acid) solution (25 ng/ml): 0.25g ABTS 10 ml H,0 Store in a- dark bottle at 4 0 C and dilute 1:50 in citrate buffer for use.

Citrate buffer 0.1M pH 4: 2.58g citric acid 2.18g Na 2

HPO,

make up to 200ml with distilled water and adjust pH to 4.

Activated ABTS: add 10 1 of 30% hydrogen peroxide to 10ml of diluted ABTS solution.

iii) Agglutination assay Aliquots of culture isolates or supernatants 500 1) which contained the scFv were mixed with 10 1 of whole blood for volumes greater than 100 1, the mixture was gently mixed for 15 min and the sensitised cells collected by centrifugation and re suspended in PBS). Simultaneously, 1 of a second antibody (25 g/ml) directed against the Cterminal epitope of the scFv was added and the mixture stirred briefly with a plastic rod. The level of the recombinant scFv was assessed against negative and positive controls by the rate and degree of agglutination over a two minute period.

Example 3 Functional Epitopes linked to scFv Antibody Epitopes of the surface protein gp41 from HIV and HIV2 virus types may be combined with epitopes from surface protein or p24 core protein or substituted for the M2-FLAG epitope in scFv constructs or added to the scFv-M2 WO 93/24630 PCT/AU93/00228 21 FLAG construct, thereby producing various bifunctionAl reagents capable of binding erythrocytes and serum antibodies which may be present in patient's serum. The sequences of the M2-FLAG, HIV1 and HIV2 epitopes are shown in Figure 7.

Example 4 Expression of Active scFv Antibody When cultured under the conditions described in Example 1, the host cells expressed scFv antibody protein, which was transported through the host cell membranes to the periplasmic space and culture supernatant.

Example Activity and Specificity of the Recombinant scFv Antibody The recombinant 1C3/86 scFv-FLAG efficiently agglutinates erythrocytes in an assay which uses monoclonal antibody directed against the M2-FLAG epitope (IBI Corp, as the cross-linking moiety, with activity comparable to that shown by the prior art SimpliRED assay, in which a synthetically produced conjugate of the HIV-1 gp41 immunodominant epitope and 1C3/86 Fab was used as the reagent, and antibody 1B1/114 was used as a known positive sample. Constructs with either HIV-1 or HIV-2 sequences were effective also in mediation of agglutination when respective monoclonal antibodies 1BI or 2A6 and 2B4 (for HIVI and HIV2 respectively) were included in the assay.

Example 6 Affinity of the Recombinant Antibody The scFv antibody recognises glycophorin A with comparable affinity to Fab, as judged by ELISA assay. These results are illustrated in Figure 8.

WO 93/24630 PCT/AU93/00228 22 Example 7 Anti-human red cell, single chain Fv fragment linked to HIV-1 peptide with "flag" peptide (scFvflagHIV-l) was tested with 29 HIV-1 confirmed seropositive samples and 22 seronegatives. All of these samples were correctly identified in agglutination tests, and the results were in agreement with those obtained using a chemically-constructed Fab-peptide conjugate. Similarly, scFvflagHIV-2 was tested with 18 confirmed seropositive samples; 22 seronegtives and all samples were correctly detected, which was also in agreement with results obtained with a chemical construct.

The results are summarised in Table 2.

Table 2 REAGENT SENSITIVITY SPECIFICITY HIV-1 Seropositive HIV-1/2 Seronegitive scFvflagHIV-1 -100% (29/29) 100% (22/22) Chemical HIV-1 100% (29/29) 100% (22/22) HIV-2 Seropositive HIV-1/2 Seronegative scFvflagHIV-2 100% (18/18) 100% (22/22) Chemical HIV-2 100% (18/18) 100% (22/22) Example 8 A Hepatitis B surface antigen binding fragment may be substituted for the HIV-binding peptide, thereby producing a bifunctional reagent which has the capacity to bind a different analyte, in this case Hepatitis Bland in so doing to agglutinate the erythrocytes.

References cited herein are listed on the following pages.

It will be clearly understood that the invention in its general aspects is not limited to the specific details referred to hereinabove.

WO 93/24630 PCT/AU93/00228 References 1. G. Winter and C. Milstein, Nature, 1991 349 293 2. M. Neuberger et al, Nature 1985 314 268 3. P. Jones et al, Nature, 1986 321 522 4. M. Verhoeyen et al, Science, 1988 239 1534 L. Riechmann et al, Nature, 1988 332 232 6. J. Huston et al, Proc. Natl. Acad. Sci. USA, 1988 5879 7. R. Bird et al, Science, 1988 242 423 8. D. Capon et al, Nature, 1989 337 525 9. R. Byrne et al, Nature, 1990 344 667 R. Lerner et al, Sci. Amer., 1988 258 58 11. W. Huse et al, Science, 1989 246 1275 12. S. Ward et al, Nature, 1989 341 544 13. A. Amit et al, Science, 1986 233 747 14. A. Hartmann et al, J. Immunol., 1988 141 932 P. Colman et al, Nature, 1987 326 358 16. P. Alizani et al, Ann Rev. Immunol., 1988 6 555 17. C. Chothia et al, Nature, 1989 342 877 18. J. Varhese et al, Nature, 1983 303 19. B. Tulip et al, Cold Spring Harbor Symposium. 1990 J. McCafferty et al, Nature, 1990 348 552 21. H. Hoogenboom et al, Nucleic Acids Res., 1991, 19 4133-4137 22. P. Hudson, Today's Life Science, 1990 2(5) 38 23. B. Power et al, Gene, 1992 113 95-99 24. Tyler et al, Proc. Natl. Acad. Sci., 1982 79 2008- 2012 Adams et al, Biochemistry, 1980 19 2711-2719 26. Chiang et a2, Biotechniques, 1989 7 360-366 27. Power et al, Gene, 1992 113 95-99 28. Gibson 1984, "Studies on the Epstein-Barr virus genome", Ph.D. thesis, Cambridge University, England WO' 93/24630 PCr/A U93/00228 29. "Making antibodies in bacteria and on phage", Emo Practical Course Manual, IRBM, Pomezia, Italy

Claims (23)

1. A bifunctional recombinant protein comprising a particle-binding antibody or antibody fragment (PBM), and an analyte-binding moiety or molecule (ABM).

2. A bifunctional recombinant protein according to claim 1 in which the particle-binding antibody or antibody fragment is an erythrocyte-binding antibody or antibody fragment (EBM).

3. A bifunctional recombinant protein according to claim 1 or claim 2 in which the ABM is selected from the group consisting of an antigenic peptide from an immunodominant region of an env protein.of HIV-1 or HIV-2, a gag protein of HIV-1 or HIV-2, and an immunodominant region of the surface antigen of Hepatitis B.

4. A bifunctional recombinant protein according to claim 1 or claim 2 in which the ABM is a single chain Fv region of an antibody directed against an antigen selected from the group consisting of Hepatitis B surface antigen, D- dimer and canine heartworm antigen.

A bifunctional recombinant protein according to any one of the preceding claims in which the EBM is a single chain Fv region of an anti-erythrocyte antibody.

6. A bifunctional recombinant protein according to claim 5 wherein the anti-erythrocyte antibody is an anti- glycophorin antibody. WO 93/24630 PCT/AU93/00228 26

7. A bifunctional recombinant protein according to claim 6 wherein the EBM is the single chain Fv domain of the anti-glycophorin A monoclonal antibody produced by the hybridoma G26.4.1C3/86 (ATCC number HB9893).

8. A DNA sequence encoding as a single transcriptional unit a particle-binding antibody or antibody fragment (PBM) operatively linked to an analyte-binding moiety or molecule (ABM).

9. An expression vector comprising a DNA sequence according to claim 8.

10. A host cell comprising a DNA sequence according to claim 8.

11. A host cell according to claim lo which is Escherichia coil.

12. A DNA element capable of replication and expression, comprising a DNA sequence according to claim 8.

13. A DNA element according to claim 12 whic, is a plasmid.

14. A specific binding assay for detection of an analyte, comprising as a detection agent a bifunctional recombinant protein according to any one of claims 1 to 7. A specific binding assay according to claim 14 which is an immunoassay.

WO 93/24630 PCf/AU93/00228 27

16. A specific binding assay according to claim 14 which is an agglutination immunoassay.

17. A kit of reagents adapted for use in a specific binding assay according to any one of claims 14 to 16.

18. A method of preparing a bifunctional recombinant protein according to any one of claims 1 to 7 comprising the step of utilising a DNA sequence according to claim 8.

19. A method of preparing a bifunctional recombinant protein according to claim 1, comprising the steps of: a) preparing a DNA sequence encoding a complementarity determining region of an antibody specific for a particle; b) Preparing a DNA sequence encoding an analyte- binding protein; c) operatively linking the DNA sequences from step a) and step b) under the control of transcriptional and translational regulators; d) transferring the product of step c) into a host organism; e) permitting the host organism to express the DNA sequences; and f) recovering the protein.

A method according to claim 19 in which the antibody specific for a particle is an anti-erythrocyte antibody.

21. A method according to claim 20 in which the anti- erythrocyte antibody is an anti-glycophorin antibody. WO 93/24630 PCT/AU93/00228 28

22. A method according to any one of claims 19 to 21 in which the analyte-binding protein is selected from the group consisting of an antigenic peptide from an immunodominant region of an env protein of HIV-1 or HIV-2, a gag protein of HIV-1 or HIV-2, and an immunodominant region of the surface antigen of Hepatitis B.

23. A method according to any one of claims 19 to 21 in which the analyte-binding protein is an antibody specific for an antigen which is selected from the group consisting of Hepatitis B surface antigen, D-dimer, and canine heartworm antigen.