NZ234614A - Fusion protein containing parts of p24 and gp41 proteins of hiv-1 - Google Patents

Description

234614 Priority Date(s): Complete Specification Filed:"5^".^.',^?*.. Class: f^CD.

Aki ■ Publication Date: ?. ?. Mm.

P.O. Journal, No: .liC^^VfrV NEW ZEALAND PATENTS ACT. 1953 Divided from No. 226026 No.: Date: Under the provisions of Regulation 23 (1) the ...C£.<M0£& Specification has been ante-dated to 19 COMPLETE SPECIFICATION FUSION CONSTRUCTS FOR USE IN ASSAY Initials \28 JANI99I*) ybi We. THE WELLCOME FOUNDATION LIMITED, a British Company, of Unicorn House, 160 Euston Road, London NW1 2BP, England, hereby declare the invention for which >4X/ we pray that a patent may be granted to rtf^/us, and the method by which it is to be performed, to be particularly described in and by the following statement:- 234 6 1 4 The present invention relates to fusion constructs, their preparation and their use in assaying for anti-HIV-1 antibody and as a vaccine.

A variety of assays have been proposed for anti-HIV-1 antibody. However, there are problems with false positive and false negative results arising. HIV-1 provokes in particular two types of antibody. These are anti-p24 against the gag protein and anti-gp41 against the env protein. We have now prepared a specific fusion construct to which both anti-p24 and anti-gp41 bind. This enables accurate and sensitive assays to be carried out without the risk of false positive or false negative results.

Accordingly, the present invention provides a protein of the sequence: 20 MetAsnSe rProAspThrGlyHi sSerSerGlnValSerGlnAsnTyrProIleValGln pl8 > P24> 40 AsnlleGlnGlyGlnMetValHi sGlnAlalieSe rProArgThrLeuAsnAlaTrpVal 50 60 234 6 14 LysValValGluGluLysAlaPheSe rProGluValIleP roMetPheSerAlaLeuSer 70 -80 GluGlyAlaThrProGlnAspLeuAsnThrMe tLeuAsnThrValGlyGlyHi sGlnAla 90 100 AlaMe tGlnMetLeuLysGluThrlleAsnGluGluAlaAlaGluTrpAspAr gValHi s 110 120 ProValHisAlaGlyProIleAlaProGlyGlnMetArgGluProArgGlySerAspIle 130 140 AlaGlyThrThrSerThrLeuGlnGluGlnlleGlyTrpMetThrAsnAsnProProIle 150 160 ProValGlyGluIleTyrLysArgTrpIleileLeuGlyLeuAsnLysIleValArgMet 170 180 TyrSerP roThrSe rlleLeuAspIleArgGlnGlyProLysGluProPheArgAspTyr 190 200 ValAspArgPheTyrLysThrLeuArgAlaGluGlnAlaSe rGlnGluValLysAsnTrp 210 220 Me tTh rGluTh rLeuLeuValGlnAsnAlaAsnP roAspCysLysTh r11eLeuLysAla 230 2 4 0 LeuGlyProAlaAlaThrLeuGluGluMetMetThrAlaCysGlnGlyValGlyGlyPro 250 260 AsnSerProArgGlnLeuLeuSe rGlylleValGlnGlnGlnAsnAsnLeuLeuArgAla gp41> 270 280 11eGluAlaGlnGlnHi sLeuLeuGlnLeuThrVaiTrpGlyIleLysGlnLeuGlnAla 290 300 ArgIleLeuAlaValGluArgTyrLeuLysAspGlnGlnLeuLeuGlylleTrpGlyCys 310 320 SerGlyLysLeuIleCysThrTh rAlaValProTrpAsnAlaSe rTrpSe rAsnLysSe r 330 3 4 0 LeuGluGlnl1eTrpAsnAsnMe tTh rTrpMe tGluTrpAspA rgGluIleAsnAsnTyr 350 360 ThrSerLeuIleHisSerLeuIleGluGluSerGlnAsnGlnGlnGluLysAsnGluGln 370 GluLeuLeuGluLeuAspLysTrpAlaSe rLeuTrpAsnTrpPheAsnGlyAspP ro; optionally modified by one or more amino acid substitutions, 234 6 14 insertions and/or deletions and/or by an extension at either or both ends provided that a protein having such a modified sequence is capable of binding to both anti-p24 and anti-gp41 and there is a degree of homology of at least 75% between the modified and the unmodified sequences.

The unmodified sequence is basically a fusion of parts of the p24 and gp41 proteins of the CBL-1 isolate of HIV-1 (WO 86/04423). These parts correspond to amino acids 121 to 356 and 542 to 674 respectively, following a similar numbering system to that of Meusing ejt a_l, Nature, 313, 450-458 (1985). The start of these parts is shown above at amino acids 17 and 244 respectively. Amino acids 5 to 16 above are derived from the pl8 protein. Amino acids 1 to 4, 241 to 243 and 377 to 379 above are derived from the expression vector from which the fusion construct was obtained and from DNA manipulations.

The sequence may be modified by one or more amino acid substitutions, insertions and/or deletions. These may occur anywhere in the sequence but especially in the parts of the sequence which are not derived from the p24 and gp41 proteins. In the case of substitutions, one or more of the amino acids of the unmodified sequence may be substituted by one or more other amino acid which preserves the physicochemical character of the sequence, i.e. in terms of charge density, hydrophilicity/hydrophobicity, size and configuration. For example, Ser may be replaced by Thr and vice versa, Glu may be replaced by Asp and vice versa and 23 4 6 1 4 Gln may be replaced by Asn and vice versa. The Ser residue at amino acid 10 may be replaced by Asn.

The sequence may also be extended on one or both gpl) ends. This may be no more than the provision of an additional carboxy-terminal Cys residue. However, the sequence may be extended by up to 50 amino acid residues at either or both ends. Up to 40 amino acids, for example up O to 20 amino acids, may therefore be added to the amino-terminus and/or carboxy-terminus of the unmodified sequence. The amino-terminal amino acid, however, will normally be Met due to the translational start codon of the nucleic acid sequence from which the protein is expressed. This is unless the protein has been expressed fused at its amino-terminus to a carrier protein and the fusion protein has been cleaved to release the protein of the invention.

The sequence may be modified by introducing corresponding changes into the DNA sequence encoding the unmodified protein. This may be achieved by any appropriate 1^^ technique, including restriction of the sequence with an endonuclease, insertion of linkers, use of an exonuclease and/or a polymerase and site-directed mutagenesis techniques. Whether the modified DNA sequence encodes a modified protein to which both anti-p24 and anti-gp41 are capable of binding can be readily determined. The modified sequence is cloned into an appropriate plasmid, a host cell is transformed with the plasmid and the protein that is expressed is tested for its ability to bind anti-p24 and 234 6 14 o anti-gp41. Also, there must be a degree of homology of at least 75%, for example of 0 5% or more or of 90% or more, between the amino acid sequences of the modified and unmodified proteins.

A protein of the invention is prepared by a process compri sing: (i) transforming a host cell with a vector which O incorporates a gene encoding the protein and which is capable, in the host cell, of expressing the protein; (ii) culturing the transformed host cell so that the protein is expressed; and (iii) recovering the protein.

The gene encoding the protein is preferably constructed in two parts, an amino-terminal part incorporating the DNA sequence encoding the p24 amino acid residues and a carboxy-terminal part incorporating the DNA sequence encoding the gp41 amino acid residues. The two parts are then fused together and inserted in an expression vector. The gene is provided with appropriate transcriptional regulatory sequences and translational start and stop codons. The gene is therefore inserted in the expression vector in the correct reading frame with respect (£3^ to an ATG start codon under the control of a promoter.

Any suitable expression vector may be employed, for example a plasmid or a viral vector such as a recombinant baculovirus or recombinant vaccinia virus. The host transformed with the expression vector may be eucaryotic or 234 6 14 o o procaryotic, for example unicellular microorganisms or mammalian cells. As a unicellular eucaryotic host, Saccharomyces cerevisiae, S. kluveromyces and S. pombe may be mentioned. Strains of bacteria such as E. coli, B. subtilis or B. thermophilus may be used as procaryotic hosts. E. coli is preferred. The transformed host is cultured and the protein that is expressed is recovered.

A protein of the invention can be used in assays for anti-HIV-1 antibody, in particular for anti-p24 and/or anti-gp41. A test sample of any appropriate physiological fluid may be used in the assay, for example urine, plasma, blood or serum. The assay method comprises contacting a test sample with a protein of the invention and determining whether any anti-p24 and/or anti-gp41 binds to the protein. For this purpose, a test kit may be provided comprising a protein of the invention and means for determining whether any anti-p24 and/or anti-gp41 there may be in a test sample binds to the protein.

A variety of assay formats may be employed. The W protein can be used to selectively capture anti-p24 and/or anti-gp41 from solution, to selectively label anti-p24 and/or anti-gp41 already captured, or to both capture and label. In addition the protein may be used in a variety of homogeneous assay formats in which the antibodies which react with the protein are detected in solution with no separation of phases. The protein can also be used for HIV-1 antigen detection.

The types of assay in which the protein is used to o 234614 capture antibodies from solution involve immobilization of the protein onto a solid surface. This surface should be capable of being washed in some way. The sort of surfaces which may be used are polymers of various types (moulded into microtitre wells; beads; dipsticks of various types; aspiration tips? electrodes; and optical devices), particles (for example latex; stabilized blood, bacterial or fungal cells; spores; gold or other metallic sols; and proteinaceous colloids; with the usual size of the particle being from 0.1 to 5 microns), membranes (for example nitrocellulose; paper; cellulose acetate; and high porosity/high surface area membranes of an organic or inorganic material).

The attachment of the protein to the surfaces can be by passive adsorption (for which it is ideally suited by virtue of its hydrophobic nature) from a solution of optimum composition which may include surfactants, solvents, salts, chaotropes; or by active chemical bonding. Active bonding may be through a variety of reactive or activatible functional groups which may be attached to the surface (for example condensing agents; active esters, halides, anhydrides; amino, hydroxyl, or carboxyl groups; sulphydryl groups; carbonyl groups; diazo groups; unsaturated groups). Alternatively the active bonding may be through another protein (itself attached to the surface passively or through active bonding), e.g. through polyclonal or monoclonal antibody directed against any or all of the epitopes presented by the protein, or through a carrier protein such as albumin or casein, to which the protein may be chemically bonded by any of a variety of methods and which may confer advantages because of isoelectric point, charge, hydrophilicity or other physico-chemical property. The protein may also be attached to the surface (usually but not necessarily a membrane) following electrophoretic separation of a reaction mixture e.g. an immune precipitation.

After reacting the surface bearing the protein with a solution contain antibody of interest and removing the excess of the sample where necessary by any of a variety of means (washing, centrifugation, filtration, magnetism, capilliary action), the captured antibody is detected by any means which will give a detectable signal. For example, this may be achieved by use of a labelled molecule or particle as defined above which will react with the captured antibody (for example protein A or protein G and the like; anti-species or anti-immunoglobulin-sub-type; rheumatoid factor; antibodies to any of the epitopes contained in the protein and used in a competitive or blocking fashion; or any molecule containing the epitopes of the protein, including the protein itself and other proteins and peptides derived directly or indirectly from HIV-1).

The detectable signal may be optical or radio-active or physico-chemical, provided by directly labelling the molecule referred to with for example a dye, radiolabel, electroactive species, magnetically resonant species or 234 6 14 fluorophore; or indirectly by labelling the molecule or particle with an enzyme itself capable of giving rise to a measurable change of any sort. Alternatively the detectable signal may be due to, for example, agglutination, diffraction effect or birefringent effect occurring if any of the surfaces referred to are particles.

Those types of assay in which the protein is used to label an already captured antibody require some form of labelling of the protein which will allow it to be detected. The labelling can be direct, by chemically or passively attaching for example a radio-, magnetic resonant-, particle or enzyme label to the protein; or indirect by attaching any form of label to a molecule which will itself react with the protein, e.g. antibody to any of the epitopes of the protein, with subsequent reaction of the labelled molcule with the protein. The chemistry of bonding a label to the protein can be directly through a moiety already present in the protein, such as an amino or sulphydryl or through an inserted group such as a maleimide. Capture of the antibody may be on any of the surfaces already mentioned, by any reagent, including passive or activated adsorption, which will result in specific antibody or immune complexes being bound. In particular capture of the antibody could be by anti-species or anti-immunoglobulin-sub-type, by rheumatoid factor, proteins A, G and the like, or by any molecule containing any of the epitopes of the protein as described above. 234614 -li- For those assays in which the protein is used to provide a measure of HIV-1 antigen in a sample, the-protein may be labelled in any of the ways described above, and used n in either a competitive binding fashion so that its binding by any specific molecule on any of the surfaces exemplified above is blocked by antigen in the sample, or in a non-competitive fashion when antigen in the sample is bound specifically or non-specifically to any of the surfaces above, in turn binds a specific bi- or poly-valent molecule (e.g. an antibody) and the remaining valances of the molecule are used to capture the labelled protein.

In general in homogeneous assays the protein and an antibody are labelled, so that, when the antibody reacts with the protein in free solution, the two labels interact, for example to allow non-radiative transfer of energy captured by one label to the other label, with appropriate detection of the excited second label or quenched first label (e.g. by fluorimetry, magnetic resonance or enzyme measurement). Addition of either antigen or antibody in a sample results in restriction of the interaction of the labelled pair, and so to a different level of signal in the detector.

The protein of the invention may be used in a sandwich immunoassay for anti-p24 or anti-gp41 in which recombinant antigens, which present the same antigenic sequence but which have been expressed in different organisms, are present on either side of the sandwich. The most common expression systems for cloning antigens are those in which the antigens are expressed in E. coli. However, E. coli occurs noticeably as part of the gut flora. Many human sera therefore contain antibodies to E. coli.

When antigens produced in E. coli are used in sandwich immunoassays for antibody, there is the possibility that some individuals may react with contaminating bacterial material. Such a reaction may give a false positive result for that particular individual. This type of false positive may be minimised by mixing material from E. coli with a test sample.

We have now devised a new way of overcoming the problem of false positives due to antibodies to contaminating material. In essence, immunologically identical antigens are engineered in different organisms. Cloned antigen from different organisms is therefore provided on either side of the "sandwich". Antibody it is not wished to detect in a test sample may bind to contaminating material associated with one of the cloned antigens but, because they are from different sources, not to contaminants associated with both cloned antigens. False positives can thus be avoided. Importantly, it is not then necessary to ensure that the cloned antigens are absolutely free of contaminating proteins. The specificity of the assay is greatly improved. The preparation of the antigens is simplified.

In essence an immunoassay in this format for anti-p24 2 3 4 6 14 <s» o and/or anti~gp41 antibody comprises: (i) contacting a solid phase, on which is immobilised a first recombinant peptide which presents an antigenic sequence to which the antibody is capable of binding, with a test sample; (ii) contacting the solid phase with a second recombinant peptide which presents the said antigenic sequence, which is labelled and which was expressed in a different organism than the first recombinant peptide; and (iii) determining whether the test sample contained any said antibody.

The first recombinant peptide and/or the second recombinant peptide is a recombinant protein according to the invention. The first and second recombinant peptides need not be identical in structure. It is permissible for one to be longer than the other, for example. They must both present the antigenic sequence to which the anti-p24 and/or anti-gp41 it is wished to detect are/is capable of binding, however. In other words, each must include the same antibody binding site(s).

Two recombinant proteins of different sequences according to the invention or two recombinant proteins of the same sequence according to the invention may therefore by used. Alternatively, only one of the recombinant peptides is a protein according to the invention. In these circumstances, however, the other peptide and the recombinant protein according to the invention must present 234 6 14 a common epitope. The other recombinant peptide may be a qaq/env fusion or just a gag or an env peptide.

Detection of anti-p24 and/or anti-gp41 depends on the fact that antibodies have at least two antigen combining sites, two for IgG, IgA and IgE and five for IgM. Taking IgG as an example, one of the antigen combining sites binds with the first recombinant peptide which is immobilised on a solid phase. The labelled second recombinant peptide binds with the second antigen combining site of IgG.

The recombinant peptides in this assay format are engineered in different organisms. Preferably, the genus at least of each organism is different. More preferably, each organism is of a different family. For example, different bacterial expression systems may be used. Alternatively, one antigen cloned in a bacterial expression system and one cloned in yeast or in insect or mammalian cells may be used. Most preferably, one antigen is cloned in a procaryotic organism whilst the other is cloned in a eucaryotic system. Examples of suitable hosts for cloning include B. subtilis, E. coli, S t reptomyces, insect cells, yeast and mammalian cells. A baculovirus expression system or a vaccinia virus expression system, in which peptides are expressed in insect and mammalian cells respectively, are preferred eucaryotic alternatives. E. coli is a preferred procaryotic host.

A protein of the invention may also be used as a vaccine against HIV-1. For this purpose, the protein may be formulated in a pharmaceutical composition with a 934614 pharmaceutically acceptable carrier or diluent. The protein may be presented as an injectable formulation. Suitable diluents include Water for Injections and isotonic saline solution. The composition is administered parenterally, for example intravenously, intramuscularly or subcutaneously. An effective amount is given to a human. Typically a dose offrom 10 to 200ug is given parenterally.

The following Examples illustrate the invention. In the accompanying drawings: Figure 1 shows the preferred env sequence as incorporated in pDM322 in Example 1; Figure 2 shows the preferred gag sequence as incorporated in pDM614 in Example 1; Figure 3 shows the DNA sequence of the gag/env protein of Example 1 with vector-related sequences are shown in bold; Figure 4 shows the construction of pFOHc in Example 3; Figure 5 shows the shuttle vector pvFOHC, the sequence of which contains two in-phase initiation codons separated by the FMDV VPl 142-160 sequence and six amino acids of the authentic HB "pre-core" sequence; Figure 6 presents the results of the sandwich ELISA in Example 3; and Figure 7 shows the sucrose gradient profile of core reactive material obtained in Example 3.

Example 1. E. coli expression constructions A lambda gtlO (DNA Cloning vol. 1, Editor - D.M. Glover, IRL Press 1985) cDNA library was constructed from / *+nNT0FF<CE / :,2<fcTf99o I 2 3 4 6 14 poly(A+)-RNA from CEM cells infected with the British isolate of HIV (CBL-1) by standard techniques (Moledular • Cloning (1982) Maniatis et al_ Cold Spring Harbor Press). CD Cells of a leukaemic T-cell line designated CCRF-CEM which harbour CBL-1 have been deposited at ECACC, Porton Down, GB on 11 January 1985 under deposit number 85 01 1101.

A lambda recombinant containing the entire envelope CD gene was identified and the EcoRI insert fragment was subcloned into the plasmid vector pUC8 to produce the plasmid pDP4 . The exact location of the envelope gene within pDP4 was determined by comparison with published data following restriction enzyme analysis and partial DNA sequencing (Meusing £t. £1^ (1985) Nature 313 4 50-458, Wain-Hobson e_t £l (1985) Cell £0 9-17, Sanchez-Pescador £t al (1985) Science 227 484-492).

We expressed parts of the HIV genome in E. coli by fusing fragments to the lacZ gene. The expression vector chosen was pXY460. This is an Open Reading Frame vector where the strong tac promoter drives a mutated lacZ gene in which the initiating ATG and the coding sequence for p-galactosidase are out of frame. The vector is derived from pXY410 (Winther et a_l J. Immunol. 13 6 ( 1986 ) 1835-1840 ). There are restriction sites for EcoRI, Smal and BamHI at the start of the lacZ gene. Any DNA fragments inserted here which can restore the reading frame will produce fusion proteins consisting of the sequence encoded by the insert attached to f3-galactosidase. 234614 o EcoRI BamHI . . . ATG AAT TCC C GGG GAT CC lacZ Smal Vector pXY460, harboured in E. coli HB101, was deposited at 0NCIMB, Aberdeen, GB on 26 July 1988 under accession number ..

NCIB 40039. a) Envelope The region of the HIV-1 env gene coding for amino acids 542-674 was cloned into pXY460 in the following manner. A DNA seguence containing a BamHI restriction site was introduced into the gene at nucleotide position 8276. The result of this was to add on two amino acids (Gly-Asp) at position 674 and to introduce a BamHI site in-frame with the pXY460 vector. 674 BamHI env... . TGG TTT AAC GGG GAT CC ...Trp Phe Asn Gly Asp Pro ...

There is a site for the enzyme Haelll at position 7875 (amino acid 541) which is in-frame with the Smal site in pXY460. The 403bp Haelll/BamHI fragment was purified by elution from an agarose gel slice and ligated with Smal/BamHl digested pXY460. The ligated DNA was transformed into E. coli strain TGI and recombinants were selected as o blue colonies on L-Amp-Xgal agar plates. Individual transformants were characterised by restriction enzyme digestion and one of those with the predicted pattern of fragments was called pDM322. The env sequence incorporated in pDM322 is shown in Figure 1. b) Gag (core) Th e region of the HIV-1 gag gene which encodes amino acids 121-356 was cloned into pXY460 in the following manner.

There is a site for the enzyme Neil (CC(C/G)GG] at position 1180 (amino acid 356) within the gag gene.

Neil ... GGA GGA CCC GGC Gly Gly Pro Gly The 5'overhang generated by this enzyme was filled in by incubation with the Klenow fragment of E. coli DNA polymerase I in the presence of only dCTP. This end was then in-frame with the Smal site in pXY460. Similarly the PvuII site at position 468 (amino acid 121) gave a blunt end in-frame with the Smal site in pXY460.

PvuII ... GCA G CA GCT GAC Ala A la Ala Asp The PvuII/Ncil(blunted) fragment (710bp) was purified by agarose gel electrophoresis and elution from a gel slice and ligated with Smal-digested, phosphatased pXY460. The DNA was transformed into E. coli strain TGI and recombinants selected as blue colonies on L-Amp-Xgal plates. Individual transformants were characterised by restriction enzyme digestion and one of those which had the predicted pattern of fragments was called pDM614. Transformants were also characterised by their immunoreactivity with sera from AIDS patients. The gag sequence incorporated in pDM614 is shown in Figure 2. 0 14 c) Gag/env The HIV-1 sequences expressed individually in-pDM322 and pDM614 were combined together in the following manner. In pDM614 a Smal site was retained at the 3' end of the gag insert because the filled-in Neil site represents "half" a Smal site (CCCGGG). When the EcoRI site of pDM322 was filled-in using the Klenow fragment of E. coli DNA polymerase I, the resulting blunt end was in-frame with the Smal site of pDM614.

DM324 DM322 gag . . . GCA GGA CCC AAT TCC CCC AGA CAA . . . env Gly Gly Pro Asn Ser Pro Arg Gin The plasmid DM322 was digested with EcoRI, filled-in and then digested with BamHI; the resulting 410bp fragment was gel-purified. The plasmid pDM614 was digested with Smal and BamHI and gel purified. The two fragments were ligated together and transformed into E. coli TGI. The blue colonies selected on L-Amp-Xgal plates were analysed by restriction enzyme digestion. One transformant which had the predicted pattern of digestion fragments was called pDM624. The 1120bp EcoRI/BamHI fragment from pDM624 was also transferred to the plasmid pXY46X, which was derived from pXY460 by deletion of all of the lacZ gene and the insertion of in-frame termination codons next to the BamHI site, to produce pT)M626. The DNA sequence of the gag/env fusion of pDM626 is shown in Figure 3. d) Antigen production 234614 The recombinant E. coli strains were plated onto selective media (L-Amp or L-Amp-Xgal) and single colonies, used to inoculate overnight cultures (<300ml). Portions of o these inocula were added to larger volumes of medium (up to ''•w' 3 litres) in fermentation vessels. The growth of the cultures was monitored and when the OD, „ „ reached 2.0-2.5 6 0 0 the inducer IPTG (isopropyl-|3-D-thiogalactopyranoside) was iO' added to a final concentration of 5//g/ml. The bacteria were then grown for a further 2-4 hours to allow induction of the recombinant proteins.

The cells were harvested by centrifugation and resuspended in buffer (25mM tris-Cl, pHB.O, 1 mM EDTA, 0.2% Nonidet P40) to a final concentration of 100 OD, „ units. 6 0 0 Extracts were prepared from the cells by one of two methods. i. Lysozyme and PMSF were added to the resuspended cells to final concentrations of 1 mg/ml and 1 mmol respectively and incubated overnight at 4°C. MgSO,, (2mM) and DNasel (40 jjg/ml) were added and incubation ("~\ continued at 4°C. More PMSF was added to 2 mM and EDTA was adjusted to 5 mM. The extract was clarified by centrifugation at 15000g for 20 minutes. The supernatant was decanted and retained at -70°C. ii. The resuspended cells were passed through a French-pressure cell at an operating pressure of 12-15000 psi. PMSF and EDTA were added to the lysate to final concentrations of 2 mM and 5mM respectively. The extract was clarified by centrifugation at 15000g for 20 minutes.

CI 234614 n G The supernatant was decanted and stored at -70°C. The passages through the pressure cell may be repeated to increase the release of antigen from the cells.

The env/0-gal and gag/(3-gal fusion proteins were purified by affinity chromatography making use of the g-galactosidase enzyme protein engineered into them, either on an anti-galactosidase affinity column or on a substrate-affinity column. Following this procedure the antigens may be further purified on a size exclusion column, and are then homogeneous by analytical gel-electrophoresis.

The gag/env fusion protein was purified from the insoluble fraction of lysed cells by differential extraction with a chaotropic agent (urea). The fraction soluble between 3 and 8H urea was further purified by chromatography on a phenyl Sepharose column with elution in 8M urea. The overall yield was greater than 70% of antigenic activity and greater than 80% of the protein in bands immunologically reacting as fusion proteins subsequent to electrophoresis (SDS-PAGE) and Western blotting.

Example 2. Baculovirus expression constructions The baculovirus Autographa californica nuclear polyhedrosis virus (AcNPV) has been developed into a useful helper-independent eukaryotic expression vector (Rohrnann (1986) J. Gen. Virol. 6_7 1499-1513 , Kuroda £t a_l ( 1986) EMBO J. _5 No. 6 1359-1365 ). The vector makes use of the polyhedrin gene of the virus which has two attractive properties. It is highly expressed late in the virus life-cycle. It is not essential for virus growth. Foreign genes are introduced into the polyhedrin gene so that they are expressed under the control of its promoter and other regulatory elements and the polyhedrin gene is itself inactivated. The AcNPV genome is very large and cannot be used for direct cloning; instead a transfer vector must be used. The transfer vector contains the polyhedrin gene and some flanking sequences, in an E. coli plasmid such as pUC8, with convenient restriction sites introduced into the gene. When the transfer vector is introduced into insect cells (e.g. Spodopte ra f rugipe rda, FP9) which are also infected with AcNPV recombination can occur. The recombinants, which occur at a frequency of less than 1%, have replaced the wild-type polyhedrin gene with one which now expresses the foreign gene; such viruses are polyhedrin -minus and can be identified by their plaque morphology. We have expressed a truncated form of HIV-1 envelope protein and the gag/env protein in this system. a) Envelope The Dral site (AATATT) at position 6190 is 65bp upstream of the start of the HIV-1 env gene. This site was converted to BamHI by cloning into the vector pUC9 to produce pDMlOO.

... TAGGAAAATATTAAGACA ....-> . . . .GGATCCCCATTAAGAGA Dral BamHI There is a BamHI site within the env gene at position 8505 which truncates the coding sequence at amino acid 752. 750 BamHI ... GTG AAC GGA TCC TTA Val Asn Gly Ser Leu The 2.32Kb BamHI fragment from pDMlOO, purified by agarose gel electrophoresis as above, was ligated with the so-called baculovirus transfer vector pAc373 (Kuroda et al) and transformed into E. coli strain HB101. A transformant which had the env insert in the correct orientation for expression from the polyhedrin promoter was identified by restriction enzyme mapping. This was then transfected into SP9 cells infected with AcNPV. Polyhedrin -minus viruses were identified and plaque purified by standard techniques (Summers and Smith "Manual of Methods for Baculovirus Vectors and Insect Culture Procedures" (1986) Texas Agricultural Experimental Station; "Current Topics in Microbiology and Immunology", Number 131, The Molecular Biology of Baculoviruses, Year: 1986, Editos: W. Doerfler and P. Bohm, Publishers: Springer Verlag). b) Gag/env The 1120bp EcoRI/BamHI fragment of pDM626 cannot be expressed directly using the pAc373 vector because it does not have its own initiating ATG codon. The EcoRI site at the 5' end of the fragment was modified by the addition of a synthetic oligonucleotide which provided an ATG and converted the EcoRI site to that for Bglll.

' AATTCAT AGATCT ATG 3' 3' GTA TCTAGA TACTTAA 5' Bglll The gag/env coding sequence was cloned into 234 614 BamHI-digested pAc373 as a Bglll/BamHI fragment and recombinant AcNPV generated as described above, c) preparation of antigens The recombinant AcNPV were used to infect fresh cultures of SP9 cells at a multiplicity of infection of 1-10. The infected cells were maintained in stirred flasks at 28°C for 36-72 hours, at which time the cells were harvested and lysed at 2-5 x 101 cells/ml with 1% Nonidet NP40 to release the antigen. The antigen was clarified by centrifugation at 15000gav for 20 minutes at 46C. The supernatant was stored at -70°C until required. The antigen may be purified by affinity chromatography on lentil lectin columns. Antigens prepared in this fashion are significantly purified over the cell lysate but are not analytically pure.

Example 3: Vaccinia virus expression constructions Two clones were used to construct fusion proteins composed of the major antigenic epitope of foot-and-mouth disease virus (FMDV) fused to the amino-terminus of hepatitis B core antigen {HBcAg). One clone representing HBcAg was obtained from Dr. P. Kighfield (pWRL 3123). This clone had been modified at the NH2 terminus such that it could be expressed in bacteria as a fusion protein to the E. coli protein TRP E. pWRL 3123, harbored in E. coli HB101, was deposited at NCIB, Aberdeen, GB on 6 March 1987 under accession number NCIB 12423.

A second clone representing FMDV VPl 142-160 234614 sequences from Oj Kaufbeuren linked to the amino terminus of P-galactosidase was obtained from Dr. M. Winther (pWRL 201) (Winther et J. Immunol. 136, 1835, 1986). Restriction maps of each clone are shown in Figure 4. As can be seen in Figure 4, the junction between the FMDV sequence and the p-galactosidase comprises a BamHI restriction site. The strategy undertaken therefore involved the fusion of the FMDV sequence and the HBcAg sequence through this BamHI site .

The initial stage in the construction therefore involved insertion of a synthetic oligonucleotide linker for BamHI at the 5' end of the HBcAg gene of pWRL 3123. The site used for insertion of the linker was the Narl site at position 290. However a second Narl site at position 1230 was also present in this plasmid. The plasmid was therefore partially digested with Narl so that a population of plasmid molecules which had been cut at only one Narl site could be observed by agarose gel electrophoresis and purified. After flush ending the Narl sites using the Klenow fragment of DNA polymerase I, a synthetic oligonucleotide linker representing a BamHI site was ligated into the partial Narl digest and the resulting plasmids were used to transform E. coli. Clones were then analysed for the presence of a BamHI linker in the correct Narl site by restriction mapping.

One such clone, designated pEB208, was isolated and DNA prepared. The length of the BamHI linker had been 234614 specifically chosen so that, when ligated to the FMDV portion of pWRL201 (Fig 4), the translational reading frame would be continuous and a fusion protein could be produced. Concomitant with the insertion of the BamHI linker, the Narl site into which it had been inserted was destroyed. It was therefore possible to remove the HBcAg sequence from pEB208 by BamHI - Narl digestion whereupon a DNA fragment of 940 bases was produced. Similarly a BamHI - Narl fragment from pWRL201 of approximately 3.5 kilobases purified. These two fragments were ligated together and the correct clone (pFOHc) was identified by restriction mapping.

As can be seen from Figure 4, pFOHc can be expressed in bacterial cells under the control of the tac promoter. In order to facilitate the transfer of the hybrid gene to a vaccinia virus (VV) shuttle vector, however, plasmid pFOHc was cut at the single Narl site and a second EcoRI site was introduced as a synthetic linker. This enabled the complete hybrid gene to be isolated as an EcoRI fragment.

The VV shuttle vector was pVpllk which was derived from from the vector pH3JARlA (Newton e_t aJL, Vaccines 86: New Approaches to immunisation, Cold Spring Harbor Laboratory, 303-309, 1986) by deletion of extraneous VV sequences. This shuttle vector has a W promoter (in this case pllK) inserted into W thymidine kinase (TK) gene.

This vector has a unique EcoRI site immediately following the W pllk promoter and AUG (Bertholet ej: aJL, PNAS, 82, 2096, 1985). The EcoRI site and AUG are in the same translational reading frame as the amino terminal EcoRI site of the hybrid gene in pFOHc. The FMDV-HBcAg gene was therefore inserted as the EcoP.l fragment into EcoRI cut dephosphorylated pVpllk. Clones with the hybrid gene in the correct orientation relative to the pllk promoter were identified by restriction mapping. This clone was designated pvFOHc (Figure 5).

This shuttle plasmid was then inserted into the genome of the Wyeth (US vaccine) strain of W, under the control of the pllk promoter, by homologous recombination using the flanking TK sequences. Individual progency plaques with a TK" phenotype were screened for the presence of FMDV-HBcAg DNA by dot blot hybridisation.

CV-1 Cell lysates from'wild-type (Wyeth) and recombinant (vFOHc) infected cells were screened for the presence of core antigen and for FMDV sequences by sandwich ELISA. Antigen from infected cells was bound to ELISA plates using either FMD virus particle (146S) or FMD VP1 141-160 antisera raised in rabbits. Each trapped antigen was then assessed for the presence of either HBc, FMD 146S or FMD VPl 142-160 epitopes by binding with the respective guinea pig antisera and development with anti guinea pig peroxidase conjugate. The results are shown in Figure 6. As can be seen in Fig 6, a protein recombinant from (vFOHc) infected cell lysates was trapped with anti-FMDV 141-160 antiserum and this protein could then react with anti HBc, anti-FMDV 141-160 and FMDV antivirion serum in a sandwich ' / v \ i \~2 3 JAN 1993 ■'] 234614 ELISA.

Furthermore, this protein could be purified by ultracentrifugation suggesting that it was particulate in nature. This was illustrated more clearly when the products of centrifugation were sedimented on a sucrose density gradient and fractions we re re-assayed for the presence of core antigen by ELISA. Cell lysates from recombinant (vFOHc) vaccinia virus infected cells or bacteria expressing native core antigen were fractionated on 15-45% sucrose gradients. Fractions were assayed foe the presence of core reactive material by indirect sandwich ELISA using human anticore antiserum as trapping antibody and guinea pig HBc antigen antiserum for detection. The results are shown in Figure 7. The position at which FMD virus sediments is also indicated. Fig 7 shows that a peak of HBcAg reactive material was observed in a position similar to that observed when core particles expressed in bacteria were centrifuged in parallel. Thus it appears that the presence of the FMDV VP1, seauence does not interfere with the particulate 1 1 2 - 1 (I 0 * nature of the core particles.

The ability of the fusion protein to self assemble into regular, 27nm core like particles was confirmed by electron microscopic examination of immune complexes formed with sucrose gradient purified material. The complexes were formed by reacting the FMDV-HBcAg particles with antiserum raised to intact foot and mouth disease virus. The complexes were adsorbed to form over coated grids and 234614 negatively stained with phosphotungstic acid. As was to be expected from the ELISA data shown in Fig 6, immune complexes were also seen after reacting the particles with antisera to HBcAg or to synthetic FMDV peptide 141-160.

Example 4: Assay using recombinant HIV antigens a) Preparation of coated microwells The antigens obtained from insect cells in Example 2 were coated at a predetermined optimum concentration onto the microwells by passive adsorption from an amine-containing buffer at pH 8. The wells were then supercoated with a solution containing high levels of a bovine protein to ensure that any remaining hydrophobic sites are fully occupied. b) preparation of conjugates 1. The purified antigens from E. coli from Example 1 were labelled with alkaline phosphatase. The alkaline phosphatase was attached to the sulphydryl radicals of the p-galactosidase using well established maleimide-sulphydryl chemi st ry. 2. The conjugates were freeze-dried in a matrix of a sugar alcohol with serum protein additives, and reconstituted before use with a diluent containing the metal cofactors of alkaline phosphatase. c ) Performing an assay 1. Sequential format. A sample to be assayed for the presence of anti-HIv antibodies was added to a microwell *04614 -Stand incubated for 30 minutes at a temperature of 45°C. The residues were washed from the microwell and then the" conjugate was added to the well and incubated for 30 minutes at a temperature of 45°C. The excess conjugate was removed by washing. Then the presence of alkaline phosphatase was detected using the cyclic amplification system described previously. Any significant amount of enzyme bound to the well indicated the presence in the sample of antibodies to the envelope proteins of HIV.

The assay was tested using 1662 sera known not to contain antibodies to the envelope proteins of HIV, and with 6 sera containing such antibodies. The results are shown in Table 1 ("Insect cell cultured antigen"), with for comparison an assay performed with the antigens derived from E.coli both as conjugate and as coating protein ( "E. coli cultured antigen"). It can be seen that the background colouration is higher in the latter case and that there occur samples which give rise to signals which can be ascribed to impurities in the antigen preparations and not to the presence of genuine anti-HIV antibodies. These samples do not give rise to signals on those assays wherein the coating and conjugated antigens are prepared from different sources.

Table 1 (a) Backgrounds generally 234614 Insect cell cultured antigen : assay background = 0.15 + 0.03 O.D. Units E. coli cultured antigen : assay background = 0.26 + 0.05 (b) False positive data (ascribable to anti-coli activity) E. coli cultured antigen : 2 false positive signals from 1662 samples Not reactive on Insect cell cultured antigen (c) Positives found as positive using Insect cell cultured antigen 6/6 Example 5: Detection of antibody against HIV by competition for binding of a labelled antibody itself directed to the gag/env protein of the invention A classic competitive enzyme immunoassay (EIA) was used. The gag/env fusion protein, designated here 626, obtained in Example 1 was coated onto microtitre wells by capture through a monoclonal antibody (TL03) directed against p24. Samples were then added to the prepared wells, and a conjugated anti-HIV added immediately. The samples were serum samples and plasma samples from blood donors and serum samples from patients with AIDS, AIDS-associated condition and other diseases. The enzyme used in the conjugate was peroxidase. After an incubation period of about 1 hour the wells were washed and substrate for the -Xlr enzyme was added. This was 3,3,5,5'-tetramethyl benzidine. Anti-HIV in the sample was ascertained by comparison with a standard taken through the procedure. The results are given in Tables 2 and 3.

Table 2: Detection of antibody to HIV in serum samples and plasma samples from blood donors Centre No. samples Non- Initially Kepeatably tested reactive reactive reactive 1 1699 1696 3 1 (0.06%) 2 1783 1780 3 2 (0.11%) 3 2037 2034 3 1 (0.05%) 4 1908 1904 4 2* (0.10-0.16%) 975 974 1 0 Total 8402 8388 14 6 (0.17%) (0.07%) * Only 3/4 samples retested Table 3: Reactivity of sera from patients with AIDS, AIDS associated condition and other diseases.

Clinical No. of Antibody Confirmed Group Samples Positive Antibody Positive* AIDS 59 59 59b AIDS related complex 62 62 62 AIDS associated conditions'1 97 97 97 High riskd 426 272+ 271 Diseases unrelated to AIDS8 67 1++ 0 234614 m C> Miscellaneousf 80 0 0 * Confirmation was by Western Blot and/or at least two alternative immunoassays (except b below). b Samples from 30 AIDS patients were confirmed with one alternative immunoassay. c Includes patients with Persistent Generalised Lymphadenopathy, Kaposi's sarcoma, opportunistic infections and patients known to be KIV antibody positive. d Patients in established risk groups, * Patients with acute viral diseases, autoimmune disease, neoplas ia. f Includes samples from healthy individuals and patients with undefined conditions.

+ The discrepant sample (from an IV drug abuser) gave an uninterpretable Western blot.

++ The discrepant sample was grossly haemolysed.

Example 6: Detection of antibody against HIV by labelling with conjugated anti-globulin * s :! i 2 8 JANi99iy v <-• 234614 Th i s is the standard method of detecting antibodies to HIV. 626 from Example 1 was coated onto microwells by passive adsorption. Samples of 50ul of serum, pre-diluted 1/100, were added to the wells. After an incubation of about 30 minutes the samples were washed out of the wells and 50ul conjugated anti-human globulin added. The enzyme used in the conjugate was peroxidase. After incubation for a further 30 minutes approximately, the wells were again washed and enzyme substrate added. The presence of anti-HIV was detected by comparison with a standard taken through the procedure. The results are given in-Table 4.

Table 4: Indirect anti-globulin test Number Negatives Number Number Positives Number Tested Negative Tested Positive 55 55 25 25 Example 1: Detection of antibody against HIV by use of labelled 626 626 from Example 1 was coated passively onto microwells as in Example 6. Undiluted 250ul samples were added to the prepared wells. Conjugated 626 (50ul) was added immediately after samples. After an incubation for about an hour the wells were washed and substrate added.

The enzyme used was alkaline phosphatase. The results are shown in Table 5.

Table 5: Direct sandwich assay Number Number Number Number Negatives Negative Positives Positive Tested Tested Sera 810* 809 58 58 234614 Plasma 175* 169 1 false positive sera and 6 false positive plasmas (not all repeatably false positive) Example 8: Detection of captured antibody against HIV using 626 labelled with a particle A standard agglutination test was effected. Latex particles of a diameter of 0.2 micron precoating were coated passively with 626 obtained in Example 1. Samples of sera or plasma were mixed with the latex either using an apparatus or by stirring on a surface. The presence of antibodies to HIV (which cause agglutination of the particles) was ascertained by visual inspection, or by appropriate instrumentation, a few minutes after mixing the reagents. The results are shown in Table 6.

Table 6: Latex Agglutination: Assay Number Number Number Number Negatives Negative Positives Positive Tested Tested Sera 480 479 80 80 Plasma 50 50 Example 9: Use of 626 to label anti-HIV antibody captured by anti-globulin Anti-human globulin was coated passively onto microwells. Samples each of 50ul of undiluted serum were added to the prepared wells. 626 from Example 1 labelled with alkaline phosphatase was added immediately. After a c o q O 3!° -x- incubation of about one hour the wells were washed and substrate added. The presence of anti-HIV antibody was ascertained by comparison with a standard. The results are shown in Tables 7 and 8.

Table 7: Anti-human capture assay Number Number Number Negatives Negative Positives Tested Tested 16 16 15* includes 3 weak positives not detected Table 8: Anti-human capture assay Number Number Number Negatives Negative Positives Tested Tested 138* 137 46" 1 false positive very weak positive not detected Number Posi tive 12 Number Posi tive 45 37 C.yJ 'f D

Claims (8)

WHAT WE CLAIM IS:

1. A protein of the sequence: 10 20 MetAsnSerProAspThrGlyHisSerSerGlnValSerGlnAsnTyrProIleValGln pl8> p24> 30 40 AsnlleGlnGlyGlnMe tValHisGlnAlaIleSerProArgThrLeuAsnAlaTrpVal 50 60 LysValValGluGluLysAlaPheSerProGluVallleProMetPheSerAlaLeuSer 70 80 GluGlyAlaThrProGlnAspLeuAsnThrMetLeuAsnThrValGlyGlyHisGlnAla 90 100 AlaMetGlnMetLeuLysGluThrIleAsnGluGluAlaAlaGluTrpAspArgValHis 110 120 ProValHisAlaGlyProIleAlaProGlyGlnMe tArgGluProArgGlySerAspIle 130 140 AlaGlyThrThrSe rThrLeuGlnGluGlnlleGlyTrpMetThrAsnAsnProProIle 150 160 ProValGlyGluIleTyrLysArgTrpIlelleLeuGlyLeuAsnLysIleValArgMet 170 180 Ty rSe rPr oThrSe rIleLeuAspIleArgGlnGlyProLysGluProPheArgAspTyr 190 200 ValAspArgPheTyrLysTh rLeuArgAlaGluGlnAlaSerGlnGluValLysAsnTrp 210 220 MetThrGluThrLeuLeuValGlnAsnAlaAsnProAspCysLysThrlleLeuLysAla 230 240 LeuGlyProAlaAlaThrLeuGluGluMetMetTh rAlaCysGlnGlyValGlyGlyPro 23 4 6 1 250 260 AsnSerProArgGlnLeuLeuSerGlyIleValGlnGlnGlnAsnAsnLeuLeuArgAla O gp41> 270 280 IleGluAlaGlnGlnHisLeuLeuGlnLeuThrValTrpGlyIleLysGlnLeuGlnAla 290 300 ArgIleLeuAlaValGluArgTyr LeuLysAspGlnGlnLeuLeuGlyIleTrpGlyCys O 310 320 SerGlyLysLeuIleCysThrThr AlaValProTrpAsnAlaSerTrpSerAsnLysSer 330 340 LeuGluGlnileTrpAsnAsnMetThrTrpMetGluTrpAspArgGlu HeAsnAsnTyr 350 360 ThrSerLeuIleHisSerLeuIleGluGluSerGlnAsnGlnGlnGluLysAsnGluGln 370 GluLeu LeuGluLeuAspLysTrpAlaSerLeuTrpAsnTrpPheAsnGlyAspPro; optionally modified by one or more amino acid substitutions, insertions and/or deletions and/or by an extension at either or both ends provided that a protein having such a modified sequence is capable of binding to both anti-p24 and anti-gp41 and there is a degree of homology of at least 75% between the modified and the unmodified sequences.

2. A process for the preparation of a recombinant protein, which process comprises (i) transforming a host cell with a vector which incorporates a gene encoding a protein as claimed in claim 1 and which is capable, in the host cell, of expressing the protein; (ii) culturing the transformed host cell so that the protein is expressed; and U 234614 -39- (iii) recovering the protein.

3. A process according to claim 2 substantially as hereinbefore described in Example 1, 2 or 3.

4. A protein as claimed in claim 1 when produced by a process as claimed in claim 2 or 3.

5. An assay for anti-p24 and/or anti-gp41 HIV-1 antibody, which assay comprises contacting a test sample with a protein as claimed in claim 1 and determining whether any of the said antibody binds to the protein.

6. An assay according to claim 5, substantially as hereinbefore described in any one of Examples 5 to 9.

7. A test kit for use in an assay for anti-p24 and/or anti-gp41 HIV-1 antibody, which kit comprises a protein as claimed in claim 1 and means for determining whether any of the said antibody in a test sample binds to the protein.

8. A pharmaceutical composition comprising a pharmaceutically acceptable carrier or diluent and, as active ingredient, a protein as claimed in claim 1.