The United States Government may have certain rights in the present invention pursuant to Grant No. AI24684 awarded by the National Institutes of Health.

The invention relates generally to recombinant expression vectors useful for host cell expression of readily identifiable fusion proteins. These vectors are useful in a wide range of applications, e .g. , in rapid and convenient techniques for the preparation of antigenic fusion proteins or peptides, and for isolating specific gene sequences which encode these products.

Protein export from bacterial host cells using recombinant expression vectors has been explored as a method of synthesizing relatively large quantities of biologically active polypeptides. Methods typically utilize expression vectors which have been introduced into a suitable host bacterial cell, such as Escherichia coli . However, toxic effects on the host, expression of undesired protein products and production of nonfunctional proteins are some of the problems encountered with the use of expression vectors. Special expression vectors have been developed for production of proteins such as human growth factor (Yano and Muri, 1990) . However, most expression vectors are limited in the range of DNA segments that may be accepted. Moreover, most vectors will not efficiently express desired gene products unless various elements such as start codons and signal peptide encoding sequences are present in the inserted DNA. This limits successfully exported products to those derived from incorporated DNA encoding the required elements.

Expression vectors have been used to produce desired protein products in the form of fusion proteins. These products are heterologous proteins combining a protein of interest with another protein partner that has a readily identifiable activity, usually enzymatic. The utility of the partner in some cases is in the "leader" and "promoter" segments of the partner gene, enabling exportation of the fusion protein into the periplas and, often of more importance, providing a soluble, active form of the desired protein product.

Plasmid vectors have been constructed for transforming E. coll host cells in which portions of genes specifying both periplasmic and outer membrane proteins were fused to modified alkaline phosphatase

(phoA) genes (Wright, et al . , 1983). With these vectors, expression of alkaline phosphatase occurred only when successful fusions were made. Moreover, in the constructs described by Wright et al . , no signal sequence was provided, therefore, a signal sequence had to be present in the genes used to make the fusions. This would significantly limit the number of successful DNA fusions from sheared DNA for example, as many of the longer and possibly valuable sequences would not be expressed if initially sheared from their signal sequences.

Other prokaryotic vectors capable of producing alkaline phosphatase fused hybrid polypetides have been described. In the Piatak et al . (1990) vector, an active ricin A protein was produced in a prokaryotic host by linking the ricin A coding sequence with a terminated leader DNA sequence from alkaline phosphatase (Piatak, et al . , 1990). The desired protein coding sequence was out of reading frame with the leader sequence and provided a ricin A gene product only when properly fused. However,

the protein was produced intracellularly, a fused alkaline phosphatase product was not obtained, and the protein was not secreted.

In a secretion-expression system described by Petro et al . (1987), a promotor, secretion signal sequence, multiple cloning site and phoA gene were combined in a vector that expresses active alkaline phosphatase until DNA fragments are inserted into the cloning site. Once inserted, the DNA uncoupled the phoA gene from the signal sequence. When grown on alkaline phosphatase substrate- laced agar plates, bacterial host cells harboring "unsuccessful" clones which had not incorporated the foreign DNA were blue when alkaline phosphatase was expressed. On the other hand, hosts that expressed the foreign incorporated DNA did not express alkaline phosphatase, yielding ordinary white colonies. This sort of "negative" selection will segregate some successful foreign DNA fusions but will not distinguish spontaneous mutations, and, of more importance, will not provide either the detectability of the fused alkaline phosphatase or exportation properties.

European patent 0352839 (Yano and Mural, 1990) describes construction of a plasmid that was used to express hEGF (human epidermal growth factor) from an E. coli host cell. Like other specific use plasmids, this vector had a promoter, _. the desired gene (cloned DNA) and an alkaline phosphatase signal peptide gene. The expressed product was not fused with alkaline phosphatase, so detection and isolation depended on standard protein product isolation separation from the medium.

Gene fusions are utilized for a number of purposes, including studies on protein secretion and folding (Smith

et al . , 1987), methods of simplifying protein purifications (Piatak et al . , 1990), and development of specific immunogens (Chondra, 1990) . One popular use of gene fusion products is the production of bifunctional hybrid proteins which are useful in enzyme immunoassays (EIA) as "capture" agents. For example, an antibody is coupled with another protein that is readily detectable. The antibody then binds to specific antigens and is quantitated by virtue of the coupled "reporter" protein. Bifunctional hybrid proteins have included, in addition to antibodies. Protein A, Protein G and streptavidin conjugated to reporter proteins such as alkaline phosphatase or horseradish peroxidase. Preparation of these reagents involves purification of each protein followed by covalent coupling and purification of.the fusion product. Such processes are time-consuming, expensive and necessarily limit the number of hybrids that can be screened.

Despite the value of gene fusion products as tools for basic and applied science, current approaches suffer from a number of limitations. In particular, the production of a specific gene-fusion product by current strategies either 1) relies on prior DNA sequence information to facilitate correct reading frame alignment of target and reporter genes, or 2) requires specific antibody to the target gene product to identify the correctly expressed fusion in a population of predominant failures. In practice, these.limitations restrict the diversity of gene fusions that may be generated and thus characterized. Generally there is a lack of control over gene expression, resulting in selection of, for example, unwanted spontaneous mutants.

The present invention addresses one or more of the foregoing or other problems by providing a general

recombinant vector design that allows overexpression and secretion of a recombinant fusion protein, accepts DNA from a wide variety of sources, and utilizes reading frame restoration between two otherwise silent genetic elements. The invention will generally allow the rapid identification of successful gene fusions after random insertion of a desired DNA without need to screen a large number of independent clones.

The cloning vector of the present invention is designed to allow the expression of a fusion protein from a host cell. The most general structure of the cloning vector includes, in order, an inducible promoter sequence, a multiple cloning site, and a reporter gene. The reporter gene product is capable of being exported into host cell periplas ic space and is also capable of expression as a fusion product with a protein product of DNA inserted into the multiple cloning site with a proper reading frame. The inducible promoter sequence is upstream of the other DNA segments. The cloning vector does not express a gene product until a foreign DNA is inserted into the multiple cloning site with a correct reading frame. Thus neither the product of the reporter gene segment nor the product of inserted DNA will be separately expressed. If DNA insertion into the multiple cloning site is in the correct translational frame, the resulting fusion product is a bipartite hybrid protein composed of a reporter protein and the gene product encoded by the inserted DNA. This cloning vector is useful for expression of fusion proteins derived from DNA having segments encoding signal peptide or equivalent function. Generally this is eukaryotic DNA but could be other DNA with a signal peptide element.

In further embodiments, the incorporation of a signal peptide coding sequence in the cloning vector with

an inducible promoter sequence, multiple cloning site and reporter gene sequence allows expression of a wide range of DNAs, including eukaryotic DNA that may lack a signal sequence segment as well as prokaryotic DNA. The signal peptide sequence is located downstream of the inducible promoter sequence and is not expressed until the DNA encoding a target or desired protein is positioned within the multiple cloning site. This positioning allows a translational reading to be established so that a protein fusion product- is expressed with the reporter gene product. Neither the signal peptide nor the reporter gene product will be expressed individually. Thus those gene segments are silent until a DNA inserts into the multiple cloning site to connect the signal and reporter gene segments.

In still further embodiments, the cloning vector optionally includes a "tag" gene positioned between the signal peptide coding sequence and the multiple cloning site. By "tag" is meant a gene coding for a polypeptide that when expressed as a fusion polypeptide will elicit an antigenic response when challenged with antibodies directed to its gene product. The product of this tag gene is thus an antigenic peptide attached to the amino terminus of the fusion product of the reporter gene and inserted DNA. The presence of the tag therefore provides an easy means of separation or detection with antibodies specific for the protein product of the tag gene. In practice of the invention, the amino-ter inal tags are typically fairly short, for example, a thirteen amino acid sequence such as Ser-Thr-Gln-Ser-Asn-Lys-Lys-Asp- Pro-Leu-Glu-Ser-Thr. Although longer sequences of antigenic polypeptide tags may be used, the shorter ones are preferred as this will decrease the amount of reading required in order to express the gene product. Longer gene sequences frequently exhibit lower production rates

because mRNA reading efficiency decreases with length of reading frame.

The inducible promoter sequence used typically does not include a translation initiation site but includes only a bacteriophage transcription element. For the prokaryotic expression system, the promoter transcription element is typically from bacteriophage T7 but could be from other bacteriophage including T3, SP6 or λ. A preferred bacteriophage promoter transcription element is found in T7 bacteriophage and has the base sequence TTAATACGACTCACTAT.

A preferred signal peptide gene is a segment from the estB gene. The signal peptide encoded by this gene has the following amino acid sequence: Met-Lys-Lys-Asn- Ile-Ala-Phe-Leu-Leu-Ala-Ser-Met-Phe-Val-Phe-Ser-Ile-Ala- Thr-Asn-Ala-Tyr-Ala. Minor changes in amino acid sequence arising from modification of the base sequence of the gene could be made without impairing the function of this signal peptide. The signal peptide allows a newly synthesized protein to be transported through the inner membrane during which time the signal peptide is cleaved from the amino-terminus of the polypeptide. Protein secretion requires a signal peptide; otherwise, proteins remain in the cytoplasmic compartment.

A reporter gene is located downstream of the multiple cloning site. The reporter gene product forms a fusion protein with the gene product of the inserted DNA. The reporter gene may encode a phosphatase, a peroxidase, luciferase, or β-lactamase. In a preferred embodiment, the reporter gene is the phoA gene of E. coli encoding alkaline phosphatase. In any event, the reporter gene will preferentially encode a protein or peptide capable of being exported; otherwise, protein fusions produced

ay remain in the host cell cytoplasm. Unless exported to the periplasm, fusion proteins are generally not accessible for easy detection and are subject to proteolysis by cytoplasmic proteases. Exported polypeptide gene fusion products are detectable by methods suited to the particular product of the reporter gene, for example, colorimetric, fluorescence or enzymatic analysis.

The multiple cloning site within the cloning vector is not limited to any particular base sequence. While there are no limitations on length, it is preferable to limit the multiple cloning site length so that its incorporation as part of the gene fusion will not add unduly to the length of the resulting fusion protein. Of course the entire cloning site is not always incorporated, depending on which site is utilized for introduction of target DNA into the vector. Regardless of its length or composition, the multiple cloning site should be devoid of translational stop codons in all three reading frames to insure that all possible reading frames of target DNA can be used. Thus one or more restriction sites may be incorporated into the sequence. In a preferred embodiment where a signal peptide coding sequence is incorporated, cleavage sites for endonucleases Xbal, Sail, and Pstl are present. A typical multiple cloning site without stop codons and possessing cleavage sites for these restriction endonucleases may have the base sequence AGATCCTCTAGAGTCGACCTGCAGC. In preferred cloning vectors where a signal peptide encoding segment is not included, cleavage sites recognized by restriction endonucleases -Ωnal, Smal, BaiziHI, Sail and Pstl are present. Other restriction sites may be utilized, however, depending on the composition of the multiple cloning site.

In another aspect of the invention the cloning vector optionally includes a gene that allows identification of host cells transfected with the cloning vector. In preferred embodiments this will be a gene conferring antibiotic resistance on the transfected cell. This gene segment is typically located in another site on the vector and will be expressed independently of gene fusion products. There are several genes that could be used for conferring antibiotic resistance. In a preferred embodiment, the bla gene encoding β-lactamase and encoding antibiotic resistance is incorporated into the cloning vector. This gene confers an ampicillin resistance on the host microorganism.

In the usual practice of insertion procedures, DNA termini resulting from endonuclease cleavage are modified to accept DNA. This is preferably accomplished using DNA polymerase I Klenow fragment. In a preferred method, the DNA intended for insertion is made blunt-ended by treating with the Klenow fragment and a mixture of the four deoxy nucleotide triphosphates. Not all fragments will be restored to the proper reading frame by this procedure. However, to insure that all possible reading frames are restored, the target DNA is preferably digested partially with a frequent-cutting restriction endonuclease such as one recognizing a 4 base-pair sequence and the overhang sequentially filled in four separate reactions with one base at a time. Treatment with mung bean or S, nuclease then cleaves the remaining single-stranded overhang. This will effectively restore every possible reading frame of the target DNA segment.

When the cloning vector includes a signal peptide sequence, primary gene fusion constructs may be prepared from many different sources of DNA including viral, eukaryotic, prokaryotic, or even synthetic DNAs. As used

herein, primary gene fusion constructs refer to the plasmid vectors containing inserted target DNA fused in the same translational frame with a signal peptide and/or reporter gene segment(s) . Genomic DNA, if fragmented so that stop codons are removed, can also be inserted into the multiple cloning site and a gene fusion product expressed. The presence of stop codons does not necessarily preclude translation of desired fused genes, provided stop codons fail to effectively terminate translation. In fact virtually any transcribable and translatable DNA segments may be inserted into the multiple cloning site of the vectors of the invention, for example suitable DNA segments prepared by shearing or fragmenting the DNA of interest.

Generally, fusion proteins from eukaryotic DNA will not be effectively expressed from a cloning vector unless a signal peptide coding sequence is incorporated into the vector. Such a vector will also include a ribosomal binding (Shine-Dalgano) site, which is a short segment adjacent to the inducible promoter sequence. A preferred Shine-Dalgano sequence is 5'-GGAGGA-3'. Ribosomal binding sites allow ribosomes to load onto mRNA and initiate translation.

It can be seen that several variations of cloning vectors may be produced by the aforementioned methods of vector construction. One particular embodiment, for example, is a cloning vector encoding T7 transcriptional promoter, Shine-Dalgano translational promoter, a prokaryotic secretory signal peptide and an im unoaffinity tag. p-ANTIGEN-2 is a particular example constructed with the φlO promoter, an ampicillin resistance gene and an immunoaffinity tag with the particular base sequences shown in Figure 1. This expression vector is useful for accepting prokaryotic

DNA, for example, influenza viral DNA segments or HIV-1 DNA segments.

Yet another particular embodiment is p-ANTIGEN-3. This example of an expression plasmid may be used to express eukaryotic DNA. Its base sequence is shown in Figure 2 and unlike p-ANTIGEN-2 embodiments, lacks a signal peptide sequence. This embodiment includes the øio promoter and a leaderless and promoterless bacterial alkaline phosphatase gene.

In still further aspects, the invention concerns a method for detecting antigenic proteins or peptides. The method includes first preparing one of the above- described cloning vectors, inserting DNA segment(s) into the multiple cloning site of such vector, and transforming it into a host cell. Vector-containing host cells are selected and expression of protein fusion constructs induced. Colonies which produce an antigenic protein or peptide fused with the expression product of the reporter gene are then identified. The cloning vector may be designed to accept either eukaryotic or prokaryotic DNA depending on whether a signal peptide sequence is included.

The invention also contemplates that antibodies specific for a desired antigenic protein or peptide may be obtained from animal or human serum. This is accomplished by exposing the animal to a disease-causing agent, by injecting with a selected vaccine and allowing antibodies to develop or the like. Antibodies may then be isolated from the serum and used in antigen capture methods with the antibodies fixed to a matrix. Exemplary matrices include plastic microtiter plates, nitrocellulose filters or other accepted immobilization surfaces.

In yet another aspect of the invention, primary gene fusions may be modified so that polypeptides are produced without a fusion partner, that is, without the reporter gene. A fusion partner refers to one of the proteins encoded by two genes in a cloning vector where a single protein product is expressed. The partner refers to the protein used as the reporter protein and is permanently incorporated into the originally designed vector. In order to obtain a target protein without the reporter protein, the steps involve digesting the primary gene fusion construct in the cloning vector with a restriction endonuclease to remove the reporter gene segment. The digested gene fusion construct is then treated with DNA ligase to reseal the cloning vector. The resultant cloning vector may be transformed into a suitable whole cell such as one containing an inducible T7 RNA polymerase gene. The desired polypeptide then may be expressed without the fusion partner. This method is particularly useful for obtaining polypeptide gene products desirable as antigens or vaccine candidates.

The host cell used in this method is typically E. coll , although other bacterial hosts may be used, such as Salmonella typhimurium, Erwina carotavora, Klebsiella pneumoniae and like microorganisms.

As part of the invention, kits useful for the expression of fusion proteins are also envisioned comprising separate containers each having suitably aliquoted reagents for performing the foregoing methods. For example, the containers may include one or more vectors prepared in accordance with claim 1, particular examples being pANTIGEN-2 or pANTIGEN-3. Suitable containers might be vials made of plastic or glass, various tubes such as test tubes, metal cylinders, ceramic cups or the like. Containers may be prepared with a wide range of suitable ali uots, depending on

applications and on the scale of the preparation. Generally this will be an amount that is conveniently handled so as to minimize handling and subsequent volumetric manipulations. Most practitioners will prefer to select suitable restriction endonucleases such as Kpnl , Smal , BamRI , Xmal , Sail , and Pstl from common supplies usually on hand; however, such restriction endonucleases could also be optionally included in a kit preparation.

Vectors supplied in kit form are preferably supplied in lyophilized form, although such DNA fragments may also be taken up in a suitable solvent such as ethanol, glycols or the like and supplied as suspensions. For most applications, it would be desirable to remove solvent which for ethanol, for example, is a relatively simple matter of evaporation.

Figure 1 shows a partial DNA and deduced amino acid sequence of pANTIGEN-2. Base sequences 1-15 encode the T7 promoter. A signal peptide is encoded by nucleotides 61-130 and nucleotides 131-152 code for an antigen tag. Part of the alkaline phosphatase gene is shown beginning at base 162.

Figure 2 shows a partial DNA and deduced amino acid sequence of pANTIGEN-3. Bases 1-15 encode the T7 promoter. The N-terminal sequence of alkaline phosphatase is shown beginning at base number 39.

Figure 3 is a Western blot of HIV-1 env-phoA gene fusion products with a monoclonal antibody to E. coli alkaline phosphatase. Panel A is a Western blot using a monoclonal antibody specific for bacterial alkaline phosphatase. Panel B is an autoradiogram of the same blot shown in Panel A, demonstrating T7 mediated

exclusive metabolic radiolabeling. Lane 1: prestained markers; lane 2: bacterial alkaline phosphatase; and lanes 2-7: clones 6, 12, 15, 19, 20 and 29 respectively.

Figure 4 is a schematic map of selected HIV-1 gpl20- phoA fusion clones.

Figure 5 is a schematic representation of the enzyme immunoconjugate assay. A indicates wells precoated with the immuno-absorbent anti-F _c antibody. B shows addition of test antibody, c shows addition of epitope-enzyme conjugate and 4 indicates identification of bound conjugate by cleavage of chromogenic substrate.

Figure 6 shows results of an enzyme immuno-conjugate assay on selected gpl20-phoA clones and HIV+ serum samples.

Figure 7 shows an N-terminal tag capture of pANTIGEN-2 expressed Env-PhoA fusion proteins.

The present invention provides a general design for a cloning vector capable of producing one or more desired proteins fused to a marker protein. The marker or "reporter" protein is encoded by a gene which is part of the vector design, but the protein segment fused to the reporter may derive from any of an enormous array of DNA segments incorporated into the vector.

Cloning vectors of the invention are constructed so that a multiple cloning site for target DNA is positioned between a reporter gene and a signal peptide coding segment. The signal coding segment will act as a signal sequence for a DNA that inserts into the cloning site. Thus, a wide range of DNA segments inserted with the proper reading frame will be produced as a tripartite

fusion protein with the signal peptide fused to the N- terminal end of the target protein which in turn is fused to the N-terminal end of the reporter gene product. The final expressed product will appear in the periplasm of the host cell without the signal peptide which is lost during exportation of the gene fusion product from the cytoplasm.

The expressed gene fusion product is readily detected by taking advantage of the properties of the reporter gene product. This product is typically a protein with enzymatic function but could be a protein with other distinct properties such as reactivity toward a fluorophor or antigenic determinants. Detection is most convenient if distinct colonies are visualized, for example, colored colonies on bacteriological media in the presence of a chromogenic indicator that reacts with the product of the reporter gene. PhoA works well as a reporter gene. Its gene product, alkaline phosphatase, is expressed when fused with a gene product from the DNA in the cloning site. The alkaline phosphatase is active, once exported into the periplasm of E. coli and is detectable with an indicator that stains blue in the presence of active enzyme.

Incorporation of a signal peptide coding sequence into the cloning vector contributes to its versatility. Most cloning vectors are limited in the DNA that can be incorporated into the multiple cloning site. If a signal peptide coding sequence is not part of the desired gene DNA sequence inserted into the cloning site, no useful gene fusions will be obtained because exportation will be not possible. By incorporating a signal peptide coding sequence into the vector, virtually any length or type of DNA that can be successfully inserted into the site with the correct reading frame will be expressed. The signal

peptide encoded by the signal peptide coding sequence enables export of the fusion protein through the cytoplasmic membrane. It is removed during export of the fusion protein into host cell periplasmic space. This increases stability of the expressed protein by avoiding proteases commonly found in the cytoplasm.

The multiple cloning site, as designed, recognized particular endonucleases, but it will be appreciated that other multiple cloning sites would work as well, provided there are no translational stop codons in any of the possible reading frames.

A leaderless, promoterless reporter gene is utilized to prevent expression of the reporter gene in the absence of foreign DNA inserted in the same reading frame as the reporter and/or signal peptide coding segment. In addition, the reporter is inactive until exported from the cytoplasm of the host bacterium. Thus only correctly aligned gene fusions give rise to an easily detectable phenotype. When alkaline phosphatase is the reporter gene, bacterial colonies expressing alkaline phosphatase turn blue on bacteriological medium containing the alkaline phosphatase chro ogenic substrate 5-bromo-3- indolyl-phosphate. Since protein fusions are expressed only when the T7 RNA polymerase is induced, the host cell is not subject to inappropriate overproduction, that is, overproduction sufficient to interfere with cell function or continued production of gene fusion products. Although the alkaline phosphatase reporter gene is highly preferred, other such genes could be incorporated into the vector. Selection should be based on ease of detectability of the gene product, ability to stabilize the target DNA polypeptide product and product activity only when transported out of the cytoplasm.

In some cases, it may be desirable to express the target polypeptide without the fusion partner. Thus consideration should be given to restriction endonucleases that may be used to remove the reporter gene from the fusion vector without removing target DNA. In the case of p-ANTIGEN-2, the reporter protein is readily removed by cleavage of the fusion construct with Pstl fllowed by resealing the target DNA-containing vector with DNA ligase. This method will work for any target DNA which lacks Pstl sites. Additional strategies based on a similar protocol could be devised based on the target DNA sequence and/or restriction map.

The DNA for insertion into the multiple cloning site may be derived from numerous sources, for example, cDNA or genomic DNA, DNA from bacteria, yeast, viruses or higher organisms. When transformed into bacterial host cells, eukaryotic gene sequences with intervening sequences are generally not expressible because the bacterial host cell cannot remove the intervening sequences. Thus if a particular eukaryotic gene product is desired, the DNA for that product is almost always a cDNA or a synthetic gene.

In order to increase the efficiency of recovering gene fusions from a fusion constructionn protocol, target DNA can be prepared by two different methods, both of which yield target DNA termini which are random with respect to translational reading frame. The first method is to simply shear the DNA by timed sonic disruption precalibrated with respect to DNA concentration and time to yield fragments of the desired length (500-200 base pairs, for example) . DNA treated in this fashion is incubated with DNA polymerase and the four nucleotide triphosphates (dNTPs) to repair any uneven ends. The blunt-ended DNA fragments are then treated with alkaline

phosphatase to remove the 5 ' phosphate group from each DNA strand. This treatment prevents individual target DNA molecules from ligating with one another during subsequent ligation iwth vector DNA. The vector DNA is prepared by digestion with Sail fllowed by DNA polymerase I and 4dNTPs to fill the restriction enzyme generated overhang. Vector DNA is then ready for mixing and ligation with DNA fragments.

A second method for preparing randomly ended DNA fragments is to partially digest the DNA with a restrictiion enzyme such as Sau3A. In addition to cutting DNA frequently, the recognition sequence for Sau3A and the remaining overhang are both GATC. The overhang generated by partial digestion of target DNA can then be sequentially ^* filled. After each addition the remaining DNA is treated with either SI or mung bean nuclease to remove the remaining overhang. The reaction mixtures are then pooled, treated with alkaline phosphatase and combined into the vector.

Antigen-enzyme protein fusions generated in a bacterial host may be used for antibody detection. Potential immunogens produced as protein fusions from bacterial host cells do not have to be purified.

Transformed host cell colonies with productive fusions may be quickly screened because of the presence of the readily detectable fusion partner. For example, antigen capture with immobilized IgG or IgM polyclonal or monoclonal antibodies allows binding with the appropriate antigen. The fusion partner, preferably alkaline phosphatase, may then be directly detected by a convenient assay. Traditional methods of antigen detection using labeled antigens prepared from radiolabeled amino acids in the host cell require overnight incubation for autoradiography. However, a

reporter protein colorimetric assay using, for example, alkaline phosphatase, can be performed in accordance with the present invention immediately after binding with the capture agent.

The following examples are intended to illustrate the practice of the present invention and are not intended to be limiting. Although the invention is here demonstrated with DNA from HIV and from influenza virus, virtually any cloned gene could be used. The assay is particularly useful when the product of a cloned gene of interest is poorly characterized. The ability to generate and rapidly screen monoclonal antibodies to the gene product is a major advance in characterization of the protein product.

EXAMPLE 1

Construction of pANTIGEN-1

A leaderless and promoterless bacterial alkaline phosphatase {phoA) gene fragment was removed and purified from the plasmid pCH2 by Pstl restriction enzyme digestion and agarose gel electrophoresis, respectively. The purified phoA fragment was inserted into pUC19

(previously digested with Pstl) by ligation with T4 DNA ligase to form the plasmid pANTIGEN-1. The orientation of inserted fragment was determined by restriction analysis using Pstl; PstI-PvuII; and Xbal-EcoKL to demonstrate that the N-terminal coding region of the phoA gene was proximal to the Pstl-PvuII; and _?JaI-__"coRI to demonstrate that the N-terminal coding region of the phoA gene was proximal to the Pstl site of the pUC19 multiple cloning site.

p-ANTIGEN-1 has been deposited with the American Type Culture Collection, Accession Number 68501.

EXAMPLE 2

Construction of pANTIGEN-2

A DNA fragment encoding a T7 transcriptional promoter, a Shine-Dalgano translational promoter, a prokaryotic secretory signal peptide, and an immunoaffinity purification tag were liberated from the plasmid pUDl by PvuII-BIgTI restriction enzyme digestion. The purified fragment was ligated into Sjπal-BamHI digested pANTIGEN-1 using T4 DNA ligase to form the plasmid pANTIGEN-2. Correct fragment insertion was verified by restriction analysis using SalX-Nsil, and Xbal-Saσl and by DNA sequencing (Figure 1) . The signal peptide and phoA coding regions were verified by DNA sequence analysis to be in separate reading frames and free of inappropriate stop codons.

p-ANTIGEN-2 has been deposited with the American Type Culture Collection under ATCC Accession Number 68502.

EXAMPLE 3

Construction of pANTIGEN-3

A DNA fragment encoding a bacteriophage T7 promoter was liberated from the plasmid pT7-5 by PvuII-BaraHI digestion, purified by agarose electrophoresis and ligated into Sinal-Bau-HI digested pANTIGEN-1 using T4 DNA ligase. The resulting plasmid was then linearized by Xbal-Sall digestion, and the overhang was filled by

Klenow fragment in presence of the four deoxynucleotide

triphosphates (dNTPs) ; the fragment was then recircularized with T4 DNA ligase to form the plasmid pANTIGEN-3. The plasmid construction was verified by DNA sequence analysis using a primer specific to the bacteriophage T7 promoter sequence (Figure 2) .

pANTIGEN-3 has been deposited with the American Type Culture Collection under ATCC Accession Number 68503.

EXAMPLE 4

Construction and Expression of Eukaryotic Gene Fusions Usinσ PANTIGEN-2

The HIV-1 gpl20 encoding region was ligated into expression plasmid pANTIGEN-2. To accomplish this, pANTIGEN-2 was linearized by Sail digestion, the overhang removed by ung bean nuclease digestion, and dephosphorylated with calf intestinal phosphatase. The gpl20 encoding region was then liberated from plasmid pBHIO (Hahn et al . , 1984) by Xhol-Sall digestion and the 3.1 Kb fragment purified. The env coding region then purified and further fragmented by a combination of mechanical shearing (rapid and repeated pipetting) and Sau3A partial digestion.

The resultant fragments were then split into four tubes for serial end repair to restore all possible reading frames. To one tube, dCTP and the Klenow fragment of DNA polymerase I was added to complement the first base of a Sau3A overhang (GATC) . The remaining overhang was removed by mung bean nuclease. To a second tube, the first and second complementary bases of the overhang (dCTP and dTTP) were incorporated as just described before mung bean nuclease treatment.

Similarly, three and four base repair reactions were performed in tubes 3 and 4 respectively.

Following end repair, each tube was heated to 55°C to inactivate the polymerase, and the DNA fragments recovered by precipitation in ethanol. The fragments were then dissolved in water and combined. Thus, the combined tubes contained all of the possible reading frames in both orientations of the DNA fragments.

The env encoding fragments were then ligated into the blunt-ended Sail site of pANTIGEN-2. Ligation mixtures were precipitated in ethanol, rehydrated and used to transform competent HB101 (pGPI-2) . Transformants were grown on L plates containing carbenicillin (Cb) , kanamycin (Km) , and 5-bromo-4-chloro- 3-indoylphosphate (XP) at 30°C until colonies reached approximately 1 mm in diameter. The plates were then incubated at 42°C for 1 hr to induce the T7 RNA polymerase encoded by pGPl-2. Under these conditions colonies expressing alkaline phosphatase (AP) fusion proteins turned bright blue due the cleavage of XP by AP. Blue colonies were isolated and further characterized.

Characterization of env-phoA Fusion Protein

Five ml cultures of AP ⁺ clones were grown in Luria broth (LB) containing Cb and Km at 30°C. After overnight growth, 0.2 ml of each culture was transferred to sterile 13 x 100 mm tubes containing 5 ml M9 medium and centrifuged at 5,000 x g for 10 min. The bacterial pellets were resuspended in 1 ml M9 supplemented with 0.1% 18 amino acids (minus cysteine and methionine) and incubated with vigorous shaking at 30°C for 60 min. The temperature was then shifted to 42°C for 15 min before the addition of rifampicin (Rf) at a final concentration

of 200μg/ml. Following an additional 10 min at 42°C, the cultures were shifted to 30°C for 20 min before 10 μCi ³⁵ S labeled methionine and cysteine was added. Radiolabeling proceeded at 30°C for 5 min, after which time the culture was chilled on ice (10 min) and the cells harvested by centrifugation. Cell pellets were boiled for 15 min in electrophoresis loading buffer containing 2- mercaptoethanol (5% vol/vol) and fractionated by SDS- PAGE. The separated proteins were electrophoretically transferred to nitrocellulose and probed with monoclonal antibody to bacterial alkaline phosphatase. Antibody reactive fusion proteins were stained with goat anti- mouse-enzyme conjugate and insoluble substrates. Metabolically radiolabeled proteins were identified by exposing the nitrocellulose to X-OMAT AR film (Figure 3) .

DNA Sequence Analysis

DNA sequencing reactions were performed on plasmid DNA template using Sequenase (U.S. Biochemical Corp.,

Cleveland, OH) and premixed dideoxy- and deoxynucleotide triphosphates. Primers to initiate DNA synthesis from the T7 promoter and the 5"-terminal junction of the phoA gene were used to characterize the 5' and 3' termini of the cloned insert DNA, respectively (Figure 4) .

EXAMPLE 5

Leσionella pneumophila

Legionella pneumophila chromosomal DNA was used as a source of prokaryotic DNA. Purified chromosomal DNA was sheared by sonication, fractionated by agarose electrophoresis, and the fragments ranging in size between 0.5-2 Kb was gel purified. The fragment ends were flushed by treatment with the Klenow fragment of DNA

polymerase I and 4 dNTPs before ligation into Smal digested and dephosphorylated pANTIGEN-3. Ligation mixtures were precipitated in ethanol, rehydrated in water and used to transform competent HB101 (pGPl-2) . Transformants were grown on L (Cb, Km, XPO plates at 30°C until colonies reached approximately 1 mm in diameter. The temperature was then shifted to 42°C for 1 hr to induce expression of T7 RNA polymerase encoded by pGPl-2. Colonies expressing fusion proteins, identified as AP ⁺ blue colonies, were collected for further analysis.

DNA Sequence Analysis

DNA sequencing reactions were performed on plasmid DNA template using Sequenase and premixed dideoxy- and deoxynuσleotide triphosphates. Primers to initiate DNA synthesis from the 5'-terminal region of the phoA gene were used to characterize the fusion joint at the 3'-end of the inserted DNA segment.

Characterization of L. pneuroophila-PhoA Fusion Protein

Cultures of L. pneumophil -AP expressing strains were treated exactly as described in Example 4 to functionally verify the pANTIGEN-3 plasmid.

EXAMPLE 6

Protocol for EIC Antigen Capture Assay

Cultures (5 ml) of env-phoA and native AP expressing strains were grown to mid-log at 30 ^β C in LB (Cb-Km) . The cultures were induced at 42°C for 30 min then shifted to 37°C for 2 hr. The cells were harvested by centrifugation, washed twice in 10 ml ice-cold phosphate buffered saline (PBS) resuspended in 1 ml PBS containing

0.5% sarkosyl then lysed by two 30 sec pulses of sonic disruption using a Branson Sonicator set at 70 Watts. Insoluble cell debris was removed by high speed centrifugation (10,000 g for 60 min). The relative amount of AP activity present in each lysate was determined using a p-nitrophenyl phosphate (PNPP) hydrolysis assay.

ELISA Pro-Bind plates (Falcon 3915, Becton- Dickinson, Lincoln Park, NJ) were coated with 25 μg/ml goat anti-human Fc _c antibody (Jackson Immuno-Research Laboratories, WestGrove, PA) (diluted in PBS) at 4°C overnight. The wells were washed 3 times with PBS containing 0.05% Tween-20 (PBST) before the addition of test sear (diluted in PBS0. Following incubation at 37°C for 2 hr, the wells were washed 3 times with PBST to remove unbound antibody. Non-specific binding sites were blocked by the addition of PBS-3% gelatin (1 hr, room temperature) . Wells were washed 3 times with PBST to remove excess gelatin then antigen-AP conjugate (diluted in PBST containing 1% gelatin) was added. After 2 hrs at 37°C, the assay plates were washed with PBST (5 times) before the addition of PNPP (4 mg/ml) . Color development was stopped after 1 hr incubation at 37°C by the addition of 2 N NaOH, and the absorption at 405 nm was determined (Figure 5) .

EXAMPLE 7

N-terminal Tag Immuno-capture of pANTIGEN-2 Expressed Protein

Falcon Pro-Bind plates were coated with affinity purified anti-peptide antibody (25 μg/ml) specific for the N-terminal tag present on all pANTIGEN-2 expressed fusion proteins. Following overnight antibody

adsorption, the wells were emptied, washed with PBS, and each well blocked by addition of PBS-3% gelatin. To the precoated wells was then added an aliquot of the cell lysate described in Example 6 (adjusted to equal AP activities) . Antigen capture was carried out at 37° for 2 hr. The wells were emptied and washed 5 times with PBST before the addition of PNPP substrate. The color development was carried out at 37 ^β C for 1 hr then stopped by the addition of 2 N NaOH and the absorption at 405 nm determined (Figure 6) . Each value was standardized against a native phoA control. The relative amount of HIV-AP captured for selected clones is shown in Figure 7.

********************************

The present invention has been described in terms of particular embodiments found by the inventors to comprise preferred modes of practice of the invention. It will be appreciated by those of skill in the art that in light of the present disclosure numerous modifications and changes can be made in the particular embodiments exemplified without departing from the intended scope of the invention. All such modifications are intended to be included within the scope of the claims.

REFERENCES

The references listed below are incorporated herein by reference to the extent that they supplement, explain, provide a background for or teach methodology, techniques and/or compositions employed herein.

1. Chondra, J., U.S. Patent No. 4,973,551, November 27, 1990.

2. Petro, J., Jackson, J. and Putney, S., Abst. Pap. Am. Chem. Soc. , 194 Meeting, 1987).

3. Piatak, Jr., M. , Laird, W.J. and Lane, J.A., U.S. Patent No. 4,948,729, August 14, 1990.

4. Smith, H., Bron, S., EE, J.V. and Venema, G. , J. Bacteriol. 169, 3321-3328 (1987).

5. Wright, A., Hoffman, C. and Fishman, Y., J. Cell Biochem. Supp. 7B, 346, 1983.

6. Yano, J. and Murai, M. , European Patent Application No. 352,839, Jan. 31, 1990.

Hahn, et al , Nature 112, 166 (1984)