NO300462B1 - Monoclonal antibodies and peptides that can be used to diagnose HIV infections - Google Patents

Description

The present invention generally relates to in vitro diagnosis of viral infections. More specifically, the invention relates to monoclonal antibodies that can

reversed in the in vitro diagnosis of Human Immunodeficiency Virus (HIV) infections.

The infectious agent responsible for Acquired Immunodeficiency Syndrome (AIDS) and its prodromal phases, AIDS-related complex (ARC) and lymphadenopathy syndrome (LAS), is a previously unknown lymphotrophic retrovirus. The virus has alternately been called LAV, HTLV-III, ARV and more recently

HIV.

As the spread of HIV reaches pandemic proportions, the treatment of infected individuals and the prevention of transmission to uninfected individuals at risk of exposure become of paramount importance. Different therapeutic strategies have been directed at different stages of the virus' life cycle, and these are discussed in Mitsuya and Broder, 1987, Nature 325:773. In one method, antibodies are used that bind to the virus and inhibit virus replication either by interfering with the virus' penetration of host cells, or by means of other mechanisms. Once the virus component or components suspected of antibody intervention have been identified, it is hoped that antibody titres sufficient to neutralize the virus's infectivity will be able to be elicited by vaccination or by passive administration of immunoglobulins or monoclonal antibodies with the desired specificity.

The glycoproteins that surround most retroviruses are thought to react with receptor molecules on the surface of susceptible cells, whereby the infectivity of the virus towards certain hosts is determined. Antibodies that bind to the glycoproteins can block the virus's interaction with the cell receptors, as they neutralize the virus's infectivity. See generally The Molecular Biology of Tumor Viruses,

534 (J. Tooze, published 1973) and RNA Tumor Viruses, 226, 236

(R. Weiss et al., published 1982). See also Gonzalez-Scarano

et al., 1982, Virology 120:42 (La Crosse Virus),

Matsuno and Inouye, 1983, Infect. Immune. 39:155 (Neonatal Calf Diarrhea Virus) and Mathews et al., 1982, J. Immunol., 129:2763 (Encephalomyelitis virus).

The general structure of HIV is a ribonucleoprotein core surrounded by a lipid-containing envelope, which the virus acquires from the infected host cell's membrane during cell detachment. Embedded in the envelope and protruding from there are the virally encoded glycoproteins. HIV's envelope glycoproteins are initially synthesized in the infected cell as a precursor molecule of 150,000-160,000 daltons (gp150 or gp160) which is then processed in the cell to an N-terminal fragment of 110,000-120,000 daltons (gp10 or gp120) for development of the external glycoprotein, and a C-terminal fragment of 41,000-46,000 daltons (gp41) representing the transmembrane envelope glycoprotein.

For the reasons discussed above, HlV's gp110 glycoprotein has been the subject of extensive research as a potential target for interrupting the viral life cycle.

Sera from HIV-infected individuals have been shown to neutralize HIV in vitro, and antibodies that bind to purified gp11O are found in the sera. Robert-Guroff et al., 1985, Nature 316:72, Weiss et al., 1985, Nature 316:69 and Mathews et al., 1986, Proe. Nati. Acad. Pollock. U.S.A., 83:9709. Purified and recombinant gp110 stimulated production of neutralizing serum antibodies when used for immunization of animals, Robey et al., 1986, Proe. Nati. Acad. Pollock. U.S.A., 83:7023, Lasky et al., 1986, Science 233:209 and a human,

Zagury et al., 1986, Nature 326:249. Binding of the gp110 molecule to the CD4-(T4) receptor has also been shown, and monoclonal antibodies recognizing certain epitopes of the CD4 receptor have been shown to block HIV binding, syncytia formation and infectivity. McDougal et al.,

1986, Science 231:382. Putney et al. (1986, Science 234:1392) elicited neutralizing serum antibodies in animals after immunization with a recombinant fusion protein containing the carboxyl-terminal half of the gp110 molecule, and

further showed that glycosylation of the envelope protein is unnecessary for a neutralizing antibody reaction.

A subunit vaccine against AIDS using the HIV gp110 molecule or parts thereof may thus be desirable. Subunit vaccines are an alternative to vaccines prepared from inactivated or weakened viruses. Inactivated vaccines are of concern because of the risk that not all virus particles have been killed, and weakened viruses may be able to mutate and regain their disease-causing ability. In the case of subunit vaccines, only those parts of the virus that contain antigens or epitopes capable of eliciting immune reactions, i.e. neutralizing antibodies, ADCC and cytotoxic T-cell reaction, are used to immunize the host. A significant advantage of subunit vaccines is that irrelevant virus material is excluded.

Virus subunits for use in a vaccine can be elicited in several different ways. For example, the envelope glycoprotein can be expressed and purified from a bacterial host although this molecule will lack most post-translational modifications (such as glycosylation) or other processing. Such a modification can be achieved using a eukaryotic expression system such as yeast or cultured mammalian cells. Viral genes have been introduced into mammalian cells using vaccinia virus as a vector. See e.g. Mackett, M., et al., 1982, Proe. Nat. Acad. Pollock.

USA 79-7415, Panicali, D. and Paoletti, E., 1982, Proe. Nat. Acad. Pollock. USA 79:4927. Recombinant vaccinia virus can be constructed by the method of Hu et al., Nature 320:537 (1986) or Chakrabarti et al., Nature 320:535

(1986). In these systems, viral glycoproteins prepared from cells infected with recombinant vaccinia are appropriately glycosylated and can be transported to the cell surface for extrusion and final isolation.

An important step in the preparation of a subunit vaccine is the appropriate purification of the desired glycoprotein from the complex mixture in the expression system. Several methods can be used for this cleaning. These include, but are not limited to, preparative polyacrylamide gel electrophoresis, gel permeation chromatography and various chromatographic methods (ie ion exchange, reverse phase, immunoaffinity, hydrophobic interaction) and others. Most of these methods are used in various combinations to form essentially pure preparations (Kleid, D.G., et al., 1981, Science 214:1125, Cabradilla, CD. et al., 1986, Biotechnology 4:128, Dowbenko, D.J., 1985, Proe. Nat. Acad. Sci. USA 82:7748).

For the production of subunit vaccines, there is a need for methods whereby the number of steps required to achieve maximum purification of a given viral antigen from a complex expression mixture is reduced. The effective separation of the antigens from extraneous components can be achieved using immunoaffinity chromatography. This technique, which is also known as immunoadsorption, consists in principle of selective adsorption of an antigen to a solid support to which a specific antibody is covalently bound. The selectively adsorbed antigen is then eluted from such an antibody affinity adsorbent by changing e.g. the buffer's pH and/or ionic strength.

Polyclonal antibodies obtained from animals immunized with the desired antigen or from naturally infected individuals (see e.g. Lasky et al., supra), have frequently been used as immunoadsorbents, but these reagents are generally fraught with significant disadvantages, such that ( i) not all of the antibodies bound to the insoluble support are specific for the molecule of interest, necessitating further purification, (ii) yields of the desired antigen are often low, and (iii) antibody affinities often vary from one preparation to the other, which requires modification of the elution procedures. These difficulties could be avoided by using monoclonal antibodies specific for the desired viral antigen to be used in a subunit vaccine preparation, instead of polyclonal antibodies.

Murine monoclonal antibodies that bind HIV antigens have been described. Several groups have reported monoclonal antibodies specific for nuclear protein p25 (see, e.g., di Marzo Veronese, et al., 1985, Proe. Nat. Acad. Sei. USA 82:5199 and Chassagne, J., et al., 1986, J. Immunol. 136:1442). Monoclonal antibodies specific for the membrane glycoprotein gp41 are also have been described (see, e.g., di Marzo Veronese, et al., 1985, Science 229:1402).

Within the field, there is a continuing need for monoclonal antibodies that are specific for epitopes within well-defined regions of the major envelope glycoprotein gp110. Monoclonal antibodies that bind to these areas and induce a reduction or elimination of HlV's replication and transmission capacity would thus have significant therapeutic and prophylactic utility. The monoclonal antibodies could also be used for purification of the desired region of gp110 from disrupted virus or recombinant expression systems for use in e.g. vaccines. Moreover, the region containing the epitope(s) recognized by the monoclonal antibodies could be chemically synthesized, whereby the difficulties arising from the purification and administration of larger fragments of the gp110 molecule could be avoided. The present invention fulfills these and other similar needs.

Peptides which are able to immunologically mimic neutralizing epitopes of HIV proteins, nucleic acid probes which encode such peptides, and monoclonal antibodies which can react with such peptides as well as other peptides which interfere with HIV infectivity are thus provided. These hitherto unknown materials can be found e.g. use in diagnostic analyzes for the detection of HIV infections, and in therapeutic prescriptions for the treatment of or vaccination against such infections.

The present invention relates to the inhibition of the formation or the cellular transmission of infectious HIV in a host. More specifically, peptides that mimic a neutralizing region of HIV and monoclonal antibodies that can react with such a region are used for the diagnosis of HIV infections. In this regard, the term "neutralizing region" indicates those parts of HIV, in particular HIV proteins, that contain amino acid segments that define one or more epitopes that can react with antibodies, which either individually or in combination with other antibodies are capable of to neutralize HIV infections. Suitable assays for neutralization are well known and may include reduction of HIV infections in T cell lines, reduction of plaque-forming units of VSV(HIV) pseudotypes carrying HIV envelope glycoproteins, tests and syncytial inhibition and virion receptor binding tests. The neutralizing activity can be compared with the antibody reactivity by immunochemical tests such as immunofluorescence, immunoprint and radioimmunoprecipitation analysis.

The invention thus relates to a monoclonal antibody, which antibody is characterized in that it neutralizes HIV by specifically reacting with a gp110 neutralizing epitope within the HIV region from amino acid 301-336 of LAV which has the sequence II (29a)

and homologues thereof. The invention also relates to cell lines characterized by being HIV-gpllO-1, ATCC HB9175, HIV-gpllO-2, ATCC HB9176, HIV-gpllO-3, ATCC HB9177, HIV-gpllO-4, ATCC HB9405, HIV-gpllO -5, ATCC HB9406, HIV-gp110-6, ATCC HB9404, HIV-p25-6, ATCC HB9409, HIV-p25-7 and ATCC HB9410. The invention finally relates to the use of monoclonal antibodies in the in vitro diagnosis of HIV. (a) the amino acid sequences of HIV isolates and LAVDnn can be aligned to achieve maximum homology between the two sequences; (b) peptides comprising HIV isolates amino acid- sequences corresponding to the localization of LAV_._. - peptides

BRU

which immunologically mimic LAVr)_>T proteins, can be identified. The thus identified peptides comprising HIV isolate amino acid sequences will typically immunologically mimic corresponding HIV isolate proteins.

This method can be applied to HIV strains that have not yet been discovered. As new strains of HIV are identified,

can e.g. their envelope and core amino acid sequences are aligned with LAVBRUlg to obtain maximum homology. The methods by which the sequences are aligned are known to those skilled in the art. When the sequences are aligned, it is desired to maintain as much homology as possible between cysteine residues. The amino acid sequence of the new HIV strain or species that corresponds to the localization of the peptides specifically described here can be synthesized and used in accordance with the invention.

Another method for determining sequences of a homologous region in other HIV strains is described by Scharf et al., Science (1986) 233:1076. This uses two oligo-nucleotide primers that bind to conserved sequences outside the sequence region of interest, and contains different restriction positions in each primer. DNA from HIV strains can then be amplified in vitro, after which the resulting oligonucleotides can be cloned into vectors for sequence analysis and incorporated into a vaccine as a cassette representing a particular epitope from the HIV strain.

It is not necessary that the epitopes contained in such sequences are cross-reactive with antibodies for all strains or species of HIV. Peptides that comprise immunological epitopes that distinguish one species or serogroup from another can be used for the identification of particular species or serogroups, and can in fact help to identify individuals infected with one or more species or serogroups of HIV .

The peptides of interest will preferably originate from the gp110 region of the virus. Of particular interest in this area are peptides encoded within the env open reading frame which extends from about base pair (bp) 6667 to about 6774 in the LAVBRU~ isolate. Thus, various homologous regions of other HIV isolates include the homologous sequences obtained from the Los Alamos Data Bank (with the exception of LAV2) as indicated in Table I.

Among other peptides that are suitable for the development of or screening for monoclonal antibodies are those encoded in the open env reading frame from about bp 7246 to about 7317 in LAVBRU- Such antibodies and reactive peptides are particularly useful for immunoassays.

In the gag region of the LAVBRU isolate, the p25 amino acid sequences from about 278 to 319 and 315 to 363 are additional neutralizing regions of HIV. The person skilled in the art will understand that additional neutralizing regions of HIV can be identified on the basis of the instructions given here, in particular combinations of monoclonal antibodies that react with different HIV epitopes will exhibit neutralizing activity.

The code for peptide I, also called peptide 29, is located in the open env reading frame from about amino acid residue 308 to about 328, and the peptide will have the following amino acid sequence where oligopeptides included in the following sequence will include linear epitopes within such a sequence:

wherein Y and Y', when present, each denote sequences of up to approx. 20 amino acids. When Y and/or Y' are present, they can e.g. comprise one or more amino acids from sequences flanking amino acid residue No. 308 to 328 in the HIV envelope sequence or any part of these flanking sequences. For example, Y can comprise the entire LAVT,r)TT-i3.RU envelope amino acid sequence from about residue no. 301 to 307 or parts thereof, and Y<1> can comprise the entire LAVD_.TT-vøp-dKU amino acid sequence from about residue no. 329 to No. 336, or parts thereof, as follows: Alternatively, truncated sequences of peptides according to the invention can be prepared. In this regard, the following sequences of peptide 29 may be particularly useful: wherein Y and/or Y<1>, when present, each represent sequences of up to twenty amino acid residues.

wherein Y and Y', when present, each represent sequences of up to twenty amino acid residues.

In another embodiment, homologous regions of the ARV-2 isolate of particular interest are encoded in the open reading frame from about amino acid residue no. 306 to about no. 323, and will typically have the following amino acid sequence where oligopeptides included in the following amino acid sequence will include linear epitopes with such a sequence:

in which Y and Y', when present, each represent up to approx. twenty amino acid residues or more. When Y and/or Y<1> are present, they may comprise one or more amino acid residues from sequences flanking amino acid residue #306 to 323 of the ARV-2 envelope sequence, or any part of these flanking sequences. In particular, Y may comprise the entire HIV envelope amino acid sequence from about No. 299 to 306 or parts thereof, Y<1> may comprise the entire HIV envelope amino acid sequence from residue No. 324 to 333.

Alternatively, truncated sequences of peptide V can be prepared. In this regard, the following sequences may be particularly valuable:

wherein Y and/or Y', when present, each represent sequences of up to twenty or more amino acid residues.

A further example comprises homologous regions of the LAV-2 isolate, e.g. which is encoded in the open env reading frame from amino acid residue no. 311 to 330 and will typically have the following sequence:

wherein Y and/or Y<1>, when present, each represent sequences of up to twenty or more amino acid residues. (See Nature 326:662 (1987)).

The monoclonal antibodies are able to selectively at extremely high titers (from IO<2> to IO<4> to IO<7>

or more) to recognize neutralizing regions contained in a predetermined sequence of envelope glycoprotein gpllO or p25, their protein precursors, biologically expressed, recombinant fusion proteins and synthetic peptides containing one or more epitopes in the predetermined sequence region of gpllO or p25. The hybrid cells in question have an identifiable chromosome in which the germline DNA has been rearranged to code for an antibody with a binding site for an epitope for gp110 or p25 that is common to certain or all HIV clinical isolates. These monoclonal antibodies can

used for diagnosis and for the identification of other cross-reactive antibodies such as blocking antibodies.

Production of monoclonal antibodies

Production of monoclonal antibodies can be accomplished by immortalizing the expression of nucleic acid sequences encoding antibodies specific for HIV by introducing such sequences, typically cDNA, encoding the antibody, into a host that can be grown in culture. The immortalized cell line can be a mammalian cell line which has been transformed via oncogenesis, by transfection, mutation or the like. Among such cells are myeloma lines, lymphoma lines or other cell lines capable of supporting the expression and secretion of the antibody in vitro. The antibody can be a naturally occurring immunoglobulin from a mammal, which is produced by transformation of a lymphocyte, in particular a splenocyte, by means of a virus or by fusion of the lymphocyte with a neoplastic cell, e.g. a myeloma during the formation of a hybrid cell line. The splenocyte will typically originate from an animal immunized against the HIV virus or a fragment thereof containing an epitope site.

Immunization regimens are well known and can vary significantly, but still be effective. See Goding, Monoclonal Antibodies: Principles and Practice, Academic Press, 2nd ed. (1986). Broken virus, synthetic peptides and bacterial fusion proteins containing antigenic fragments of the gp110 or p25 molecule can be used as immunogens. The immunogen of broken virus, peptides or recombinant proteins will preferably be enriched with regard to proteins or fragments thereof containing the epitopes to which antibody-producing B cells or splenocytes are desired. More specifically, solutions containing broken virus lysates or extracts or supernatants of biologically expressed recombinant proteins or broken expression vectors can be enriched if desired

with regard to glycoproteins using methods

such as e.g. polyacrylamide gel electrophoresis. Lecithin-

affinity purification is a preferred and appropriate method for the purification of gp110 and other glycoproteins, e.g. affinity purification using lentil lectin. The degree to which the glycoproteins are purified from the solutions for use as an immunogen can vary widely, ie from less than 50%, usually or at least 75% to 95%, preferably 95% to 99%, and preferably to absolute homogeneity.

Once the proteins have been purified to the desired degree, they can be suspended or diluted in a suitable physiological vehicle for immunization, or they can be coupled to an adjuvant. A preferred technique includes e.g. adsorption of the proteins and their fragments on lentil-lectin agarose or another macromolecular carrier for injection. Immunogenic amounts of antigenic preparations enriched for HIV proteins including gp110 glycoprotein and p25 core protein, or antigenic portions thereof, are usually injected at concentrations ranging from 1 µg to 20 mg/kg host. The administration can take place by injection, e.g. intramuscularly, peritoneally, subcutaneously, intravenously, etc. The administration can take place one or more times, usually at intervals of 1 to 4 weeks. Immunized animals are monitored for antibody production against the desired antigens, after which the spleens are removed, and splenic B lymphocytes are isolated and fused with a myeloma cell line or transformed. The transformation or fusion can be carried out in a conventional manner, and the fusion technique is described in a large number of patents, e.g.

US Patent Nos. 4,172,124, 4,350,683, 4,363,799, 4,381,292 and 4,423,147. See also Kennett et al., Monoclonal Antibodies (1980) and the references therein, and Goding, supra.

The immortalized cell lines can be cloned and screened according to conventional techniques, and the antibodies in the cell supernatants capable of binding to the desired gp110 or p25 HIV virus proteins, recombinant fusion proteins or synthetic peptides containing the desired epitope region can be detected. The appropriate immortalized cell lines can then be cultured in vitro or injected into the peritoneal cavity of a suitable host to produce ascitic fluid. By virtue of having certain antibodies that are known to be specific for epitopes contained e.g. in the regions encoded by the LAVBRU gene region from about bp6688 to about bp6750 (encoding peptide 29) or from about bp7246 to about 7317 (encoding peptide 36) (bp numbering according to Wain-Hobson et al., Cell 44:9, 1985), the supernatants can be screened in competition with the relevant monoclonal antibodies by a competitive analysis. Thus, additional immortalized hybridoma cell lines with the desired binding characteristics can be readily prepared from a variety of sources based on the availability of antibodies specific for the particular antigen. Alternatively, these cell lines can be fused with other neoplastic B cells where such other B cells can serve as recipients for genomic DNA that codes for the antibody.

Although neoplastic B cells from rodents, and in particular of murine origin, are preferred, other mammalian species can be used such as lagomorphs, cattle, sheep, horses, pigs, poultry or the like. These animals can be easily immunized and their lymphocytes, and in particular the splenocytes, can be obtained for fusions.

The monoclonal antibody secreted by the transformed cell lines or hybrid cell lines can be of any of the immunoglobulin classes or sub-classes such as IgM, IgD, IgA, IgG^_4 or IgE. As IgG is the most commonly used isotype in diagnostic analyses, it is most often preferred. The monoclonal antibodies can be used intact or as fragments, such as Fv,

Fab, F(ab')2" but is usually used intact.

To avoid the possible antigenicity of a monoclonal antibody derived from a non-human animal in a human host, chimeric antibodies can be constructed in which the antigen-binding fragment of an immunoglobulin molecule

(variable region) is connected by a peptide bond with i

at least a part of another protein that is not recognized as foreign by humans, such as the expelled part of a human immunoglobulin molecule. This can be achieved by fusing animal variable region exons with human kappa or gamma constant region exons. Various techniques are known to those skilled in the art, such as those described in PCT 86/01533, EP171496 and EP173494.

Use of monoclonal antibodies for immunoaffinity purification

Monoclonal antibodies that are specific for polypeptides containing gp110 or other antigenic determinants, in particular the antigenic determinants obtained from biologically expressed recombinant fusion proteins or lysates or extracts of cultured HIV, are particularly advantageous to use in purification regulations. In general, the antibodies will have affinity association constants in the order of 10 8 to 10 12 M. Such antibodies can be used to purify the recombined fusion proteins from the culture medium from the recombinant expression system if the expressed protein is secreted, or from the components of the broken, biological expression system if they are not secreted. In general, the monoclonal antibodies capable of reacting with gp110 or other antigenic determinants are attached to or immobilized on a substrate or support. The solution containing the HIV antigenic determinants is then brought into contact with the immobilized antibody under conditions suitable for the formation of immune complexes between the antibody and the polypeptides containing the gp110 antigenic determinants. Unbound material is separated from the bound immune complexes, which complexes or gp110 antigenic fragments are then separated from the carrier.

The monoclonal antibodies will typically be roughly purified from ascites fluid or cell culture supernatants before binding to a carrier. Such procedures are well known to those skilled in the art and may include fractionation with high concentration neutral salts. Other methods such as DEAE chromatography, gel filtration chromatography, preparative gel electrophoresis or protein A affinity chromatography can also be used to purify the monoclonal antibody before it is used as an immunoadsorbent.

The carrier to which the monoclonal antibodies are immobilized should have the following general characteristics: (a) weak interaction between proteins in general to minimize non-specific binding, (b) good flow characteristics that allow passage of high molecular weight materials, (c) be in possession of chemical groups that can be activated or modified to allow chemical binding of the monoclonal antibody, (d) be physically and chemically stable under the conditions used for binding the monoclonal antibody, and (e) be stable against the conditions and components that are included in the buffers that are necessary for adsorption and elution of the antigen.

Some generally applicable carriers are agarose, derivatized polystyrenes, polysaccharides, polyacrylamide beads, activated cellulose, glass and the like. There are different chemical methods for binding antibodies to substrate carriers. See generally Cuatrecasas, P., Advances in Enzymology, 36:29 (1972). The antibodies according to the invention can be linked directly to the carrier or via a linking or spacer arm.

General conditions for the immobilization of monoclonal antibodies to chromatographic supports are well known to those skilled in the art. See e.g. Tijssen, P., 1985, Practice and Theory of Enzyme Immunoassay. Current coupling procedures will depend somewhat on the characteristics of the antibody to be coupled and its type. Monoclonal antibodies have characteristics as usual

is uniform from batch to batch, whereby it is possible that such conditions can be optimised. Attachment typically takes place by means of covalent bonds.

A suspension of extracts or lysates of HIV virus, the supernatant from a cultured biological expression system or a suspension of the broken cells is then added to the separation matrix. The mixture is incubated under such conditions and for a period of time sufficient for immune complex formation to occur, usually at least 30 minutes and more usually from 2 to 24 hours. The immune complexes containing polypeptides with antigenic parts of gp110 are then separated from the mixture. Typically, the mixture is removed e.g. by elution, and the bound immune complexes are washed thoroughly with adsorption buffer. The immune complexes can then be eluted from the separation matrix using an eluent compatible with the particular carrier used. Such eluents are well known to those skilled in the art. The polypeptides containing gp110 or other antigenic parts can also be selectively removed. For example, peptides containing an epitope recognized by the antibodies can be used to compete for the antibody binding position, providing an alternative elution technique that can be performed under mild elution conditions. The selectively adsorbed polypeptide containing the gp110 antigen can be eluted from an antibody affinity adsorbent by changing the buffer's pH and/or ionic strength. Chaotropic agents can also be used when removing the bound antigen. The choice of chaotropic agent, its concentration and other elution conditions, depend on the characteristics of the antibody-antigen interaction, but once these are determined, they should not be subject to the changes usually required in polyclonal affinity separation systems.

The eluted material may require adjustment to a physiological pH if buffers with a low or high pH value or ionic strength are used to separate the bound gp110 antigens from the separation matrix. Dialysis or gel filtration chromatography may also be necessary to remove excess salts used in the eluent to enable reconstitution of gp110 or polypeptides containing antigenic fragments of gp110 to native conformation.

The methods provide, for example, substantially purified gp110 or polypeptides containing antigenic fragments thereof, produced either naturally with infected cell cultures or with recombined expression systems of bacteria, yeast or cultured insect or mammalian cells. gp110 and the fragments or other purified proteins will typically be greater than 50% pure, more commonly at least 75% pure and frequently greater than 95 to 99% pure. These molecules can then find use for many different purposes.

The HIV gpllO proteins, polypeptides contain-

and the antigenic fragments thereof or other proteins that are substantially purified by the method can be used for many different purposes, here-

under in AIDS subunit vaccine formulations, in which the immunogen comprises an effective dose of antigenic determinants of e.g. gp110 or a neutralizing region thereof. Other components of the formulation may include the antigenic parts of HIV proteins or glycoproteins that stimulate the production of antibodies (preferably neutralizing antibodies) in an immunized host, which antibodies are able to protect against subsequent infection with HIV.

Diagnostic uses of monoclonal antibodies Monoclonal antibodies according to the invention are used for diagnostic purposes. For such purposes they may be either marked or unmarked. Diagnostic analyzes typically entail detection of the formation of a complex via the binding of the monoclonal antibody to an HIV antigen. When unlabelled, the antibodies can e.g. find application for agglutination analyses. Unlabeled antibodies can also be used in combination with other labeled antibodies (second or secondary antibodies) that can react with the monoclonal antibody, such as antibodies that are specific for immunoglobulin. Alternatively, the monoclonal antibodies can be labeled directly. Widely different markers can be used, such as radionuclides, fluorescence-producing agents, enzymes, enzyme substrates, enzyme cofactors, enzyme inhibitors, ligands (in particular haptens), etc. Numerous types of immunoassays are available, and some of the assays are described, for example, in US patent documents no. 3,817. 827, 3,850,752, 3,901,654, 3,935,074, 3,984,533, 3,996,345, 4,034,074 and 4,098,876.

Generally, the monoclonal antibodies are used

and the peptides by enzyme immunoassays where e.g.

the relevant antibodies or secondary antibodies from another species are conjugated to an enzyme. When a biological sample containing HIV antigens, such as human blood serum, saliva, seminal fluid, vaginal secretions or virus-infected cell culture suspension is combined with the antibodies in question, binding occurs between the antibodies and the molecules displaying the desired epitope. Such proteins or virus particles can then be separated from the unbound reagents, and a secondary antibody (labelled with an enzyme) added. The presence of antibody-enzyme conjugate specifically bound to the antigen is then detected. Other conventional techniques well known to those skilled in the art may also be used.

Kits or equipment for use can also be delivered

with the antibodies in question upon detection of HIV infection, or of the presence of HIV antigen. The monoclonal antibody preparations in question can thus be delivered, usually in lyophilized form, either alone or in connection with additional antibodies that are specific

received for other HIV epitopes. The antibodies which may be conjugated to a marker or unconjugated are included

in the kits or equipment together with buffers such as Tris, phosphate, carbonate etc, stabilizers, biocides, inert proteins, e.g. bovine serum albumin or similar. Generally, these materials will be present in an amount of less than approx. 5% by weight based on the amount of active antibody, and usually present in a total amount of at least approx. 0.001% by weight, again based on the antibody concentration. It will often be desirable to include an inert dissolving agent or excipient to dilute the active ingredients where the excipient may be present in an amount from approx. 1 to 99% by weight of the total preparation. When another or secondary antibody capable of binding to the monoclonal antibody is used, this will usually be contained in a separate container. The second antibody is typically conjugated to a marker and formulated in an analogous manner to the antibody formulations described above.

Detection of gp110 or p25 antigens, or the whole virus in various biological samples can find application in the diagnosis of an ongoing infection with the HIV virus. Biological samples may include, but are not limited to, blood serum, saliva, seminal fluid, tissue biopsy samples (from brain, skin, lymph glands, spleen, etc), cell culture supernatants, broken, eukaryotic and bacterial expression systems and the like. It is tested for the presence of virus by incubating the monoclonal antibody with the biological sample under conditions conducive to immune complex formation, followed by detection of complex formation. In one embodiment, complexation is detected using a secondary antibody capable of binding to the monoclonal antibody that is typically conjugated to a marker and formulated in an analogous manner to the antibody formulations described above. In another embodiment, the monoclonal antibody is bound to a solid support which is then brought into contact with a biological sample. After an incubation step, labeled monoclonal antibody is added to detect the bound antigen.

Production and use of synthetic peptides

The new peptides mimic, among other things, immunological protein epitopes for which it is coded by the HIV retrovirus,

in particular epitopes encoded in the env or gag regions of the virus genome which code for resp. gp110 and p25. In order to accommodate variations from strain to strain among different isolates, adjustments can be made for conservative substitutions and selection among the alternatives where non-conservative substitutions are involved. These peptides can be used as immunogens to inhibit or eliminate HIV antigen production in vitro for the detection of the virus or of antibodies against the virus in a physiological sample. Depending on the nature of the prescription, the peptides may be conjugated to a carrier or other compounds, labeled or unlabeled, bound to a solid surface or the like.

In one embodiment, peptides of interest are derived from the gp110 region of the virus. Of particular interest is the area in the open env reading frame which extends from approx. base pairs (bp) 6688 to approx. 6750, and approx. from base pair 7246 to approx. 7317.

The peptides of interest, including blocking peptides, will include at least 5, sometimes 6, sometimes 8, sometimes 12, sometimes 21, usually fewer than about 50,

and more generally fewer than approx. 35 and preferably fewer than approx. 25 amino acids that are included in a sequence for which an HIV retrovirus codes. The peptide will advantageously be as small as possible while still maintaining substantially the entire immunoreactivity or antiviral activity of the larger peptide. In some cases, it may be desirable to join two or more non-overlapping oligopeptides to form a single peptide structure, or to use them at the same time as individual peptides which separately or together give equivalent sensitivity to the starting point.

The peptide can be modified by introducing conservative substitutions in the peptide, usually less than 20 numerical percent and more generally less than 10 numerical percent of the amino acids are replaced. In those situations where regions are found to be polymorphic, it may be desirable to vary one or more particular amino acids to more effectively mimic the deviating epitopes of the different retrovirus strains. In many cases, methionine can be replaced with norleucine (Nor) to provide chemical stability.

It should be noted that the peptide used need not be identical to any particular HIV polypeptide sequence as long as the compound in question is able to compete immunologically with proteins from at least one of the HIV retrovirus strains. The peptide in question can therefore be subject to various changes such as insertions, omissions or substitutions, either conservative or non-conservative, where such changes can provide certain practical advantages. Conservative substitutions mean substitutions within groups such as gly, ala; val, ile, leu; aspen, glu; donkey, gin; see, thr; bright, angry; phe, bull; and nor, met. Generally, the sequence will not differ by more than 20% from the sequence of at least one HIV retrovirus strain, except where additional amino acids may be added at one or both ends, for the purpose of providing an "arm" with which the peptide can be conveniently immobilized. The arms will usually be at least one amino acid long and may be 50 or more, more often 1 to 10 amino acids long.

The peptide in which the amino acid sequence is modified by substitution, addition or deletion of amino acid residues should mainly retain the entire immunoreactivity or antiviral activity of the unmodified peptides, which can conveniently be measured using various analysis techniques described here. The d-isomer form of one or more amino acids can, if desired, be used to modify biological properties such as activity, degradation rate, etc.

Also, 1, 2 or more amino acids can be added to the end of an oligopeptide or peptide to enable easy linking of peptides to each other, coupling to a carrier or a larger peptide, for reasons discussed later, to modify the peptide or oligopeptide's physical or chemical properties or the like.

Amino acids such as tyrosine, cysteine, lysine, glutamic or aspartic acid or the like can be introduced at the C- or N-terminus of the peptide or oligopeptide to provide a valuable functionality for linking. Cysteine is particularly preferred to facilitate covalent linkage to other peptides, or to form polymers by oxidation.

The peptide or oligopeptide sequences may also differ from the natural sequence in that the sequence has been modified by terminal aminoacylation, e.g. acetylation or thioglycolic acid amidation, carboxy terminal amidation, e.g. with ammonia or methylamine, forming stability, increased hydrophobicity for concatenation with or binding to a support or another molecule, or for polymerization.

For example, it is a preferred embodiment when Y or Y' is present in the above-mentioned peptides I-VIII and IX-XV, when Y or Y' comprises one or more cysteine residues or a combination of one or more cysteine residues with spacer amino acids. Glycine is a particularly preferred spacer amino acid. Preferred peptides for use in oxidative polymerization are those in which Y or Y' denotes at least two cysteine residues. When two cysteine residues are present at the same end of the peptide, it is a preferred embodiment when the cysteine residues are separated from each other by one or two spacer amino acids, preferably glycine. The presence of cysteine residues may allow the formation of dimers of the peptide and/or an increase of the resulting peptide's hydrophobicity, facilitating immobilization of the peptide in solid phase or immobilized assay systems. Of particular interest is the use of the mercaptan group in cysteines or thioglycolic acids used for acylation of terminal amino groups or the like to link two of the peptides or oligopeptides, or combinations thereof, via a disulfide bond or a longer bond to form polymers containing a number of epitopes. Such polymers have the advantage that they give an increased immunological reaction. Where different peptides are used to build up the polymer, they have the additional ability to elicit antibodies that immunoreact with several antigenic determinants on different HIV isolates.

In order to achieve the formation of antigenic polymers (synthetic multimers), compounds with bis-haloacetyl groups, nitroaryl halides or the like can be used, where the reagents are specific for thio groups. The bond between the two mercapto groups in the different peptides or oligopeptides can thus be a single bond or a linking group of at least 2, usually at least 4 and no more than approx. 16, and usually no more than approx. 14 carbon atoms.

The peptide in question can be used bound to a soluble macromolecular (eg not less than 5kDal) carrier. The carrier can conveniently be a polyamino acid, either naturally occurring or synthetic, against which it is unlikely that antibodies will be found in human serum. Examples of such carriers are poly-L-lysine, albu-shell hemocyanin, thyroglobulin, albumins such as bovine serum albumin, tetanus toxoid, etc. The choice of carrier depends primarily on the intended final use of the antigen and on suitability and availability.

With such conjugates, there will be at least one molecule of at least one peptide in question per macromolecule, and no more than approx. 1 per 0.5 kDal, usually no more than approx. 1

per 2kDal of the macromolecule. One or more different peptides can be linked to the same macromolecule.

The linking method is conventional, using such reagents as p-maleimidobenzoic acid, p-methyl-dithiobenzoic acid, maleic anhydride, succinic anhydride, glutaraldehyde, etc. The linking can take place at the N-end, the C-end or in a position between the ends of the molecule. The peptide in question can be derivatized by concatenation, can be concatenated while bound to a carrier or the like.

Various analysis regulations, which are well known to those skilled in the art, can be used to detect the presence of either antibodies against retrovirus proteins or the retrovirus proteins themselves. Of particular interest is the use of the peptide as the labeled reagent, where the marker provides a detectable signal or binding of the peptide to a surface either directly or indirectly, where antibody against the peptide in the sample will be bound to the peptide on the surface. The presence of human antibody bound to the peptide can then be detected using a xenogeneic antibody specific for human immunoglobulin, normally both human IgM and IgG, or a labeled protein specific for immune complexes, e.g. Rf factor of S. aureus protein A.

Illustrative of an analysis technique is the use of a sample container, e.g. wells in microwell plates, where the relevant polypeptide or conjugates thereof are adsorbed to the container bottom and/or walls, either covalently or non-covalently. The sample, normal human blood or serum diluted in a suitable buffered medium, is added to the container and a sufficient period of time is allowed to elapse for the polypeptide or peptides to complex with any cognate antibodies in the sample. The supernatant is removed and the container is washed to remove non-specifically bound proteins. A labeled, specific binding protein that specifically binds to the complex, such as xenogenic antiserum against human immunoglobulin, is used for detection.

The peptide can be produced in many different ways. Because of its relatively short length, the peptide can be synthesized in solution, or on a solid support according to conventional techniques. Today, there are various automatic synthesizers on the market that can be used in accordance with known regulations. See e.g. Steward and Young, Solid Phase Peptide Synthesid, 2nd ed., Pierce Chemical Co., 1984, and Tam et al., J. Am. Chem. Soc. (1983) 105:6442.

Alternatively, hybrid DNA technology can be used where a synthetic gene can be produced using single strands that code for the polypeptide, or mainly complementary strands thereof, where the single strands overlap and can be brought together in a normalizing medium to hybridize. The hybridized strands can then be ligated to form the complete gene,

and by choosing appropriate terminals, the gene can be introduced into expression vectors readily available today. See e.g. Maniatis et al., Molecular Cloning, A Laboratory Manual, CSH, Cold Spring Harbor Laboratory, 1982. The region of the viral genome encoding the peptide can also be cloned by conventional recombinant DNA techniques and expressed (see Maniatis, supra).

Among the DNA coding sequences from the LAV,,-.,- and ARV 2 isolates of HIV that can be used for expression of the peptides are the following:

Fragments of a sequence can be used for the expression of peptide fragments, conservative base changes can be made where the modified codon(s) code for the same amino acid(s), or non-conservative changes in the coding sequence can be made where the resulting amino acid can be a conservative or a non- conservative change of the amino acid sequence as previously discussed.

The coding sequence can be extended at either the 5' or 3' end or both to extend the peptide while retaining its epitope site(s). The extension may provide an arm for co-chaining, e.g. with a marker such as an enzyme, to join two of the peptides or all the peptides in the same chain, to provide antigenic activity, suitable restriction positions for cloning or the like.

The DNA sequence itself, fragments thereof or larger sequences, usually of at least 15 bases, preferably at least 18 bases, can be used as nucleotide probes for the detection of retroviral RNA or proviral DNA, or for the identification of homologous regions for cloning or sequence analysis. Numerous techniques are described such as the Grunstein-Hogness technique, the Southern technique, the Northern technique, the dot-blot technique, improvements thereof as well as other methods, e.g. as described in US Patent No. 4,358,535.

The invention is explained in more detail by means of the following examples.

Example 1

Development and characterization of monoclonal antibodies

Example 1 describes the development of hybrid cell lines that produce monoclonal antibodies specific for the HIV envelope glycoproteins. This procedure makes use of lectin-purified extracts of LAVD_.T7 attached

BRU

to lentilectin agarose as immunogen. The monoclonal antibodies that are then developed with the hybrid cell lines are characterized by their ability to immunoprint and radioimmunoprecipitate gp110 from purified LAV, and as biologically expressed recombinant fusion protein. The monoclonal antibodies that bind to epitopes on gpllO are also reactive by ELISA with broken, whole virus, fusion proteins and synthetic peptides, and react with whole virus by indirect

fluorescence assays.

The procedures for development of the monoclonal antibody-producing hybrid cell lines, and characterization of the antibodies, were as follows.

LAV virus purified from infected CEM cells

(A.T.T.C. No. CRL8904) was dissolved in 50 mM Tris, pH 7.4, 0.15 M NaCl, 1.0% aprotinin, 2.0% "Nonidet" P-40 (NP-40)

(octylphenoxypolyethoxyethanol). The extract was clarified twice by centrifugation and adjusted to 0.5% NP-40

by adding three volumes of digestion buffer without NP-40. Lentillectin "Sepharose" was prewashed in digestion buffer without NP-40 and then equilibrated in adsorption buffer (50 mM Tris, pH 7.4, 0.15 M NaCl, 1.0% aprotinin, 0.5% NP-40) . Purified virus extract was adsorbed with lentilectin "Sepharose" for 42 hours at 4°C. Unadsorbed material was removed by washing with excess adsorption buffer. Elution of adsorbed material was carried out with 0.2 M alphamethylmannoside in adsorption buffer. The eluent was dialyzed against PBS to remove the sugar, and the material was reabsorbed onto lentilectin "Sepharose".

The glycoprotein-lentillectin "Sepharose" complex was used to immunize BALB/c mice by three intraperitoneal injections without adjuvant, given 2-3 weeks apart. The spleens were removed from immunized mice showing circulating antibody against HIV glycoproteins by immunoprinting, RIP and/or ELISA.

For the development of cell lines, the procedures of Kohler and Milstein (Nature 256:495 (1985)) with those of Goldstein, L.C., et al., (Infect. Immun. 38:273) were generally used

(1982)) proposed modifications. B lymphocytes from

spleens from the immunized mice were fused with NS-1 myeloma cells using 40 wt/vol% polyethylene glycol. After fusion, the cell mixture was resuspended in HAT medium (RPMI - 1640 medium supplemented with 15% fetal calf--4 -7

serum, 1 x 10 M hypoxanthum, 4 x 10 M ammoptenn, and 1.6 x 10 — 5 M thymidine) to select on the basis of the growth of hybrid cells, and were then dispensed into 96-well microculture plates at a concentration of 1 to 3 x 106 cells/ml

and was incubated at 37°C in a humidified atmosphere containing 6% CC>2' The cultures were fed by replacing half of the supernatant with fresh HAT medium. The wells were observed under an inverted microscope for signs of cell proliferation, and when the cells were at sufficient density, the supernatants were tested for anti-LAV antibody.

Wells containing hybrid cells producing antibody to LAV were identified by ELISA that measured binding to either purified, fully digested virus or biologically expressed fusion proteins. ELISA assays with digested virus were performed on LAV EIA plates. Plates were incubated with cell culture fluids at 37°C for 45 minutes and then washed three times with 0.05 Tween-20 in phosphate-buffered saline (PBS-Tween).

Peroxidase-goat anti-mouse IgG (1:2,000 dilution

in PBS-"Tween") was added 100 pl per well, and the plates were incubated at 37°C for 45 minutes and washed as above. Substrate (0.025 M citric acid, 0.05 M dibasic sodium phosphate, pH 5.0, containing 14 mg o-phenylenediamine and 10 µl 30% hydrogen peroxide per 50 ml) was added, and the plates were incubated in the dark at room temperature for 30 minutes. The reaction was stopped with 3 N sulfuric acid, and colorimetric reactions were read quantitatively with an automated microplate reader. Wells that gave positive results were subcloned by limiting dilution, retested for specificity and then expanded.

The monoclonal antibodies secreted by the resulting hybrid cell lines were further characterized for specificity and reactivity by immunoblotting, immunoprecipitation and ELISA with disrupted LAV virus, recombinant LAV fusion proteins and synthetic LAV peptides. All antibodies were determined to be of the IgG-^ isotype. The cell lines HIV-gpllO-1, HIV-gpllO-2 and HIV-gpllO-3 have, prior to the submission of the present application, been deposited at the American Type Culture Collection under the following deposit numbers

HB 9175, HB 9176 and HB 9177.

Recombinant fusion proteins tested for reactivity have previously been designated ENV2, ENV3, ENV4, and ENV5. Protein ENV2 is expressed by pENV2 (A.T.C.C. No. 53071) which is a region of LAV from base pair (bp) 6598 to bp 7178 (numbering according to Wain-Hobson et al., Cell 44:9 (1985)), ENV3 is expressed by pENV3 (A.T.C.C. No. 53072) comprising the LAV region from bp 7178 to bp 7698, ENV4 is expressed by pENV4 (A.T.C.C. No. 53073)

and includes pb 7698 to bp 8572, and ENV5 is expressed by pENV5 (A.T.C.C. no. 53074) which includes the LAV region bp 5889 to bp 7698. The production of the recombinant fusion proteins is described in more detail in the applicant's US patent application no. 721,237.

Collection of synthetic peptides

Peptide I (29) and VIII (110-2-2) were mounted on a benzhydrylamine (polystyrene/divinylbenzene) resin. Peptide V

(177) was mounted on a t-butyloxycarbonyl-(Boe)-ethylbenzyl-cysteine-phenylacetamidomethyl (PAM) polystyrene/divinylbenzene resin. Symmetric anhydride couplings were carried out in an Applied Biosystems 430A synthesizer. Cysteine was added as the first residue in both peptides.

Dicyclohexylcarbodiimide linkages in the presence of hydroxylbenzotriazole were used for asparagine and glutamine. Benzyl-based side chain protection and Boe alpha-amine protection were used. Other routinely used side chain protections were Boe-(formyl)-tryptophan, Boc-methionine sulfoxyd, Boc-(tosyl)-arginine, Boe-(methylbenzyl)-cysteine, Boc-(tosyl)-histidine, Boe-(chlorobenzyloxycarbonyl)-lysine and Boe -(bromo-benzyloxycarbonyl)-tyrosine.

When the peptides were radiolabelled, this happened by acetylation of the amino end with <3>H-acetic acid and dicyclohexylcarbodiimide in excess.

Deprotection and cleavage of the peptide from the resin occurred according to the "low-high" HF protocol according to Tam (Tam et al., supra). Extraction from the resin took place with 5% acetic acid, and the extract was subjected to gel filtration chromatography in 5% acetic acid*

Among synthetic HIV peptides tested for reactivity with the monoclonal antibodies were peptide 29,

36 and 39. Peptide 2 9 is encoded by the LAVBRU genome region from approx.

bp 6688 to bp 6750, peptide 36 is encoded by the region from approx.

bp 7246 to bp 7317, and peptide 39 is encoded by the region from approx. bp 7516 to bp 7593. Peptides 36 and 39 are described in more detail in US patent 4,629,783.

The blocking peptides IX-XV were placed essentially as described above on a methyl-benzhydrylamine (polystyrene/divinylbenzene) resin. Symmetric anhydride couplings were performed on an Applied Biosystems 430A synthesizer. Dicyclohexylcarbodiimide couplings in the presence of hydroxylbenzotriazole were used for asparagine. For protection, benzyl-based side chain and boc-alpha-amine protection was used, while Boe-(bromobenzyloxycarbonyl) was especially used for tyrosine side chains. Any acetylation was carried out using acetic anhydride or glacial acetic acid and dicyclohexylcarbodiimide. Deprotection and cleavage of the peptide from the resin was performed by "high" HF standard regulation (Stewart et al., supra). Extraction from the resin spikes was performed with 50% acetic acid, and the extract was then subjected to gel filtration chromatography in 20% acetic acid. If desired, high performance liquid chromatography was performed on a "Vydac" C18 column using a 0.1% trifluoroacetic acid, acetonitrile gradient.

Immunoprinting

Characterization by immunoblotting was performed on clone supernatants or ascites fluid using purified LAV virus and recombinant fusion proteins as antigens. The antigens were first separated by polyacrylamide gradient gel electrophoresis (7.0-15.0%) and were transferred to nitrocellulose membrane (NCM) by electrophoresis for 4 hours at 25 µl 25 mM sodium phosphate (pH 7.0). After transfer, NCM was blocked to prevent non-specific cross-reactions by incubation in PBS-"Tween" or "Blotto" (5% nonfat dry milk in PBS) at room temperature for 1 hour. NCM

was incubated with cell culture supernatant or ascites fluid diluted in PBS-"Tween" at room temperature for 1 hour and was rinsed with three portions of PBS-"Tween". In the second stage,

NCM incubated with goat anti-mouse IgG-horseradish peroxidase diluted in PBS-"Tween" for one hour at room temperature. This incubation was followed by washing with PBS-Tween and immersion in horseradish peroxidase color development solution for 20 minutes. The reaction was stopped by immersion in deionized water. Monoclonal antibody reactivity was compared to a positive control serum that reacted with purified, fragmented virus or expressed fusion protein. The results showed that all antibodies bound to gp110 and its precursor molecule gp150 using preparations of broken virus. Antibodies 110-1 and 110-2 also recognized the fusion protein ENV3, while antibodies 110-3, 110-4, 110-5

and 110-6 formed immunocomplexes with ENV2.

Immunoprecipitation

Virus extracts for radioimmunoprecipitation were prepared from CEM cells infected with the LAV,3„T isolate of HIV adapted to lytic growth by continuous transfection. When earlier cytopathic effects were clearly seen, the cells were transferred to labeling media containing ]-methionine

(0.05 mCi/ml) or <3>[H]-glucosamine (0.025 mCi/ml), were then incubated for 24 hours until most of the cells were lysed releasing the virus into the culture supernatant. Viruses were pelleted (1 h at 100,000 x g) from the cell-free supernatant, and detergent extracts were prepared in P-RIPA buffer (phosphate-buffered saline containing 1.05% "Triton" X-100, 1.0% deoxycholate, 0.1% SDS and 1% aprotinin). Similar extracts were prepared from the supernatants of uninfected CEM cells.

Immunoprecipitation assays were performed with 100 µl of virus extract incubated with 100 µl of culture supernatant from the hybrid cell lines for one hour on ice. 4 microliters of rabbit anti-mouse Ig was added to each sample and incubated for 30 minutes. 100 µl of immunoprecipitin resuspended in P-RIPA buffer containing 1.0% ovalbumin was added to each sample and incubated for an additional 30 minutes. The bound complexes were washed and separated by SDS-polyacrylamide gel electrophoresis (15.0% acrylamide, DATD gel). After electrophoresis, the gels were fixed, soaked in Enhance, dried and exposed to Kodak XR-5 film. A positive reference serum immunoprecipitating all HIV viral proteins was reacted with virus-infected and mock-infected CEM cell supernatants as positive and negative controls.

The results showed that all six monoclonal antibodies specifically immunoprecipitated gp110 and gp150.

Enzyme-linked immunoadsorbance assay

In order to map the gp110 epitopes that were recognized by the monoclonal antibodies according to the invention, culture supernatants from hybrid cell lines or ascites fluid were further characterized by reactivity in ELISAs with biologically expressed fusion proteins and synthetic peptides. The procedures were the same as described above, with the exception that fusion proteins or synthetic peptides replaced purified virus as antigen adsorbed to the surface of the microtiter wells.

When peptides were used as antigen, the plating procedure was as follows. Lyophilized peptide was dissolved in 6M guanidine HCl. Just prior to plating in the 96-well plates, the guanidine solution was diluted in 0.05 M carbonate/bicarbona buffer (pH 9.6) to a final peptide concentration of up to 100 µg/ml. A 50 µl volume of the diluted peptide was placed in each microtiter well, after which the plates were incubated overnight at 4°C. Excess peptide solution was "shaken off", the plates were blocked with Blotto, and the procedure described above was followed for the remainder of the ELISA.

In a similar way, recombinant protein was diluted to a final concentration of approx. 2 µg/ml in 0.05 M carbonate/bicarbonate buffer (pH 9.6) before the same procedure was followed.

The results are shown in Table II. Monoclonal antibodies produced by the cell lines HIV gpllO-1 and HIV gpllO-2 reacted with ENV3, ENV5, peptide 36 and fragmented virus. Antibodies from the cell lines HIV-gpllO-3, HIV-gpllO-4, HIV-gpllO-5 and HIV-gpllO-6 reacted with both ENV2 and peptide 29 as broken virus.

The results in Table II showed that the monoclonal antibodies 110-1 and 110-2 recognized an antigenic determinant encoded by a DNA sequence in the pENV3 region, more specifically the region of the HIV genome defined by an amino acid sequence in peptide 36. The monoclonal antibodies gpllO-1 and 110-2 thus bind to a peptide region of gpllO, for which it is encoded within bp7246 to bp7317, which the formation of immune complexes with peptide 36

and ENV3 shows. This region of the HIV genome has previously been identified as conserved, ie there is little change in the DNA sequence in the region that peptide 36 codes for among different virus isolates from different geographical locations. See Starcich et al., Cell 46:637 (1986). In contrast, the monoclonal antibodies gp110-3,

-4, -5 and -6 to HIV peptides defined by the area for which peptide 29 from bp 6688 to approx. bp 6750 codes. The region defined by peptide 29 in gp110 has been identified as containing several nucleotide substitutions among different virus isolates. Monoclonal antibodies that selectively bind gp110 polypeptides containing conserved epi-

peaks such as antibody 110-1 and 110-2, may have increased applicability in many different circumstances, such as in affinity chromatography etc. In an ELISA analysis, peptide 110-2-2 also reacted with sera from the individual from which LAV- 2 was isolated.

Indirect immunofluorescence assay

Indirect immunofluorescence assays using monoclonal antibodies directed against the HIV gp110 antigen were performed on acetone-fixed and live cells. Acetone-fixed preparations prepared from LAV-infected CEM cells were incubated with diluted culture supernatant or ascites fluid at 37°C for 1 h, while live cells were incubated with culture supernatant or ascites fluid at 4°C for 1 h, before the cells were plated on slides and acetone-fixed. For both methods, fluorescein isothiocyanate-labeled anti-mouse IgG was used to detect cells with the reactive gp110 antigen. Monoclonal antibody HIV-gp110-1 gave positive results using either live or acetone-fixed LAV-infected cells.

Example II

Immunoaffinity separation of gpllO using monoclonal antibody

Monoclonal antibodies against the HIV gp110 antigen can be used to mainly purify bacterially expressed recombinant fusion proteins by immunoaffinity separation procedures. If the expressed protein is secreted by the bacterium, the protein can be isolated from the culture supernatant. If the protein is not secreted, breaking the bacterial cells may be necessary.

The construction of plasmid pENV-5 (A.T.C.C. no. 53074) is described in the applicant's US patent application no. 721,237. Plasmid pENV-5 encodes a major portion of the carboxyl terminus of gp110 and a portion of the amino terminus of gp41 from LAV introduced into the trp expression vector. E. coli C600 transformed with this vector expresses, but does not secrete, the gp110 fusion pro-

the tea.

E. coli C600 containing the pENV-5 plasmid is grown

in medium containing tryptophan (20 µg/ml) and ampicillin (100 µg/ml) overnight at 37°C under aeration. The overnight cultures are then inoculated at 1:100 in fresh minimal medium containing ampicillin (100 µg/ml) but no tryptophan. These cultures are grown under aeration for 2-3 hours (up to early logarithmic phase) at 37°C. The inducer 3-B-indolacrylic acid is added to a final concentration of 20 ug/ml from a freshly prepared supply of 20 ug/ml in 95% ethanol. Induced cultures are then grown at 37°C under aeration for 4 to 5 hours, then pelleted and optionally frozen. Protein yields from pENV-5 are typically less than 1 mg/liter.

The pelleted bacterial cells are lysed using P-RIPA buffer (PBS containing 1% "Triton" X-100,

1% deoxycholate, 0.1% sodium dodecyl sulfate and 1% aprotinin) which will lyse E. coli cells. The suspension can be sonicated to shear DNA and RNA, followed by centrifugation to remove particulate matter. A dilution or concentration step may then be necessary to standardize the protein concentration.

Monoclonal antibody HIV-gp110-1 is first precipitated from ascites fluid or cell culture supernatants at room temperature or cold with (NH4)2SO4 or Na2SO4 solutions buffered to pH 7.3 to a final saturation of, respectively. 33 and 18%. Precipitated proteins are removed by centrifugation and redissolved in PBS and precipitated once more with 33% (NH4)2SO4 or 12-15% Na2SO4. This step can be repeated as needed. The pellet is dissolved again in PBS, and excess salts are removed by gel filtration through a desalting matrix, or by thorough dialysis against PBS.

Purified monoclonal 110-1 antibody can then be coupled to cyanogen bromide-activated Sepharose. The necessary friend-— 3

sufficient amount of gel is swollen in 10 M HCl solution on a glass filter (1 g of freeze-dried material gives a final volume of approx. 3.5 ml) and washed for 15 minutes with the same solution, and anti-

the substance is added immediately afterwards. In general, the coupling reaction proceeds most efficiently in a pH range of 8-10, but a lower pH can be used if necessary for reasons of antibody stability. The antibody should be dissolved in PBS or a carbonate/bicarbonate or borate buffer of high ionic strength with 150 mM NaCl. The suspension of activated "Sepharose" and antibody is gently stirred at room temperature for 2-4 hours or overnight at 4°C, and then washed with coupling buffer on a coarse sintered glass filter. Any remaining active groups are blocked by treatment with 1.0 M ethanolamine at pH 8 for 2 hours. The final antibody "Sepharose" product is then washed successively with buffer solutions of high and low pH (borate buffer, 0.1 M, pH 8.5, 1 M NaCl and acetate buffer, respectively,

0.1 M pH 4.0, 1 M NaCl) four or five times. This washing removes traces of non-covalently adsorbed materials.

The finished immunoaffinity separation matrix is stored below 8°C in the presence of a suitable bacteriostatic agent such as 0.01% azide.

Addition of the expressed protein suspension to the immunoaffinity separation matrix results in selective removal of the gp110 antigen. The mixture is allowed to react for 2-24 hours, preferably 12-18 hours, under slow stirring or rocking. A column format can also be used where the immunoaffinity matrix is poured into a column, brought to equilibrium, and the expressed protein suspension is slowly added to the column. After the protein suspension is added, flow through should be stopped to allow maximum immune complex formation.

Unbound material is washed out or separated by thorough washing with absorption buffer. A coarse sintered glass filter with vacuum or column flow can be used. The bound material is then eluted using low or high pH buffers (acetate buffer, pH 4.0 or borate buffer, pH 8.56) or a chaotropic agent.

Example III

Immunoaffinity purification of recombinant gpllO from a mammalian expression system

The monoclonal antibodies according to the invention find use in immunoaffinity purification of recombinant fusion gp110 expressed by mammalian cells. Mammalian cells are infected with recombinant vaccinia (Mackett et al., J. Virol. 49:857 (1984)) containing sequences encoding at least the portion of gp11O that is antigenic and elicits neutralizing antibodies.

The recombinant vaccinia is constructed according to the method described in US patent application no. 842,984. Briefly, sequences encoding the HIV envelope glycoprotein are introduced into a plasmid vector (pGS20) downstream of a vaccinia transcriptional control element. This chimeric gene is flanked by sequences coding for the viral thymidine kinase (TK) gene.

Chimeric plasmid vectors containing the vaccinia virus promoter ligated to the LAV envelope gene are used for transformation of E. coli strain MC1000. Introduction of the chimeric LAV-env sequences into the vaccinia virus genome is carried out by in vivo recombination made possible by the fact that the chimeric genes in pv-env5 plasmids are flanked by vaccinia virus sequences that code for the TK gene. This plasmid is then introduced into cells pre-infected with wild-type vaccinia virus, and recombination is allowed to take place between the TK sequences on the plasmid and the homologous sequences in the vaccinia virus genome, whereby the chimeric gene is introduced. African Green Monkey kidney cells (strain BSC-40, a line derived from BSC-1 cells, A.T.C.C. No. CCL26) are used as the host in the expression system.

Confluent BSC-40 cells are infected to a multiplicity of infection of 10 with recombinant vaccinia virus. The infection is allowed to proceed for 12 hours, after which the cells are harvested, washed once with PBS and collected by centrifugation. Cell pellets are resuspended in lysis buffer (1.0% NP 40, 2.5% sodium deoxycholate, 0.1 M NaCl, 0.01 M Tris-HCl, pH 7.4, 1 mM EDTA), after which the lysate is clarified by centrifugation. Immunoaffinity separation of the expressed recombinant fusion protein is performed as described above, for the bacterial expression system, using the monoclonal antibody gp110-1. The proteins produced with this expression system are much more similar to naturally produced HIV gp110 due to the processing and glycosylation provided by the mammalian cells.

Microorganism deposition data

The following microorganisms forming part of the invention were deposited at the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852, U.S.A.

Data regarding the deposits are as follows:

Scientific

Hybridomas HB 9175, HB 9176 and HB 9177 were tested and found to be viable on 26 August 1986. The other hybridomas were tested and found to be viable on 4 May 1987.

Claims (6)

1. Monoclonal antibody, characterized in that it neutralizes HIV by specifically reacting with a gpllO neutralizing epitope within the HIV region from amino acid 301-336 of LAV having the sequence II (29a) II (29a) Cys-Thr-Arg-Pro-Asn-Asn-Asn-Thr-Arg-Lys-Ser-Ile-Arg-Ile-Gln-Arg-Gly-Pro-Gly-Arg-Ala-Phe-Val-Thr-Ile- Gly-Lys-Ile-Gly-Asn-Met-Arg-Gln-Ala-His-Cys and homologues thereof. 2. Monoclonal antibody according to claim 1, characterized in that it binds to a neutralizing gp110 epitope located within LAVBru amino acid sequence 308 to 328, or homologues thereof. 3. Monoclonal antibody according to claim 1, characterized in that it is obtained from a cell line which is HIV-gpllO-3, HIV-gpllO-4, HIV-gpllO-5 or HIV-gpllO-6, with deposit no. ATCC HB9177, HB9405, HB9406 and HB9404, or competes with a monoclonal antibody obtained from at least one of these cell lines for binding to gp110. 4. Cell lines, characterized in that they are HIV-gpllO-1, ATCC HB9175, HIV-gpllO-2, ATCC HB9176, HIV-gpllO-3, ATCC HB9177, HIV-gpllO-4, ATCC HB9405, HIV-gpllO-5, ATCC HB9406, HIV-gp110-6, ATCC HB9404, HIV-p25-6, ATCC HB9409, HIV-p25-7 and ATCC HB9410. 5. Use of monoclonal antibody according to claims 1-3 for diagnosis with respect to the presence of HIV in a biological sample, by incubating a monoclonal antibody capable of reacting with HIV gp110 together with a biological sample, and detecting the presence of immune complexes formed between the monoclonal antibody and the antigenic determinant in the biological sample, and on the basis of this determines the presence or absence of HIV in the sample. 6. Use according to claim 5, wherein the monoclonal antibody is capable of reacting with a neutralizing region of gp110.