EP0494931A4

EP0494931A4 - Monoclonal antibodies to segments of hiv-1 reverse transcriptase

Info

Publication number: EP0494931A4
Application number: EP19900915017
Authority: EP
Inventors: Stephen H. Hughes; Ammon Hizi; Stephen D. Showalter; Andrea L. Ferris
Original assignee: US Department of Commerce
Current assignee: US Department of Commerce
Priority date: 1989-10-05
Filing date: 1990-10-01
Publication date: 1992-09-09
Also published as: AU642171B2; EP0494931A1; CA2066171A1; WO1991005066A1; AU6525090A; JPH05502579A

Abstract

Hybridoma cell lines which produce monoclonal antibodies that specifically bind to antigenic recognition sites on segments of HIV-1 reverse transcriptase, the cell lines being selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248. A method of preparing a monoclonal antibody which comprises culturing hybridoma selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248 in a suitable culture medium and recovering antibody. A method of detecting segments of HIV-1 reverse transcriptase in a test sample. A diagnostic system for detecting the presence and amount of segments of HIV-1 reverse transcriptase in a test sample.

Description

MONOCLONAL ANTIBODIES TO SEGMENTS OF HIV-1 REVERSE TRANSCRIPTASE

The present invention was wholly or partially made using funds provided by the National Cancer Institute, National Institutes of Health, Department of Human Health and Services. Accordingly, the United States Government has certain rights to this invention.

This invention concerns hybridomas which produce monoclonal antibodies recognizing particular segments of HIV-1 reverse transcriptase. These monoclonal antibodies are useful in diagnosis of Acquired Immune Deficiency Syndrome (AIDS) .

BACKGROUND AND SUMMARY OF THE INVENTION

Upon infection by an AIDS virus such as HIV-l, a linear viral DNA is synthesized by viral reverse transcriptase within the host cell cytoplasm. A host RNA acts as a primer for initiation for DNA synthesis and the viral RNA is degraded by viral RNase H. A second strand of DNA is synthesized using the first DNA strand as a template.

HIV-1 reverse transcriptase is apparently capable of structural variation with accompanying antigenic variation. Particular segments of the reverse transcriptase may be conserved enabling continuity in detection of HIV-l reverse transcriptase and consistent inhibition of the HIV-1 reverse transcriptase.

An understanding of the structure of the enzyme will make its possible to design drugs to fit its active sites. Selective inhibition of either of the enzymatic functions of the HIV reverse transcriptase should specifically block replication of the virus. Treatment of AIDS patients with inhibitors of reverse transcriptase could select for variants of the enzyme partially or completely resistant to the inhibitor. Given the potential for variation, it is important not only to determine the structures of the active sites of a particular variant of the HIV-l reverse transcriptase, but also to define the possible limits that the active sites can assume and retain their proper functions. With this knowledge it may be possible to synthesize inhibitors that the enzyme cannot evade.

Although there are substantial differences in the amino acid sequences of the reverse transcriptases of different retroviruses, all these enzymes have similar enzymatic activities implying some conservation of the structure of the active sites. Closer examination of the amino acid sequences of the various known reverse transcriptases reveals substantial regions of significant amino acid homology, reinforcing the idea that the three- dimensional structures of the catalytic sites are similar.

The relatively small amounts of reverse transcriptase in the virion make large scale purification of the enzyme from virions impractical. Recombinant DNA technology enables large amounts of reverse transcriptase to be prepared in genetically engineered cells. In the present invention, HIV-1 reverse transcriptase was expressed in E. coli.

Retroviral reverse transcriptases contain two enzymatic activities, a DNA polymerase that can use either RNA or DNA as a substrate and an RNase H. These activities are associated with distinct segments of the reverse transcriptase.

HIV-1 virions contain two reverse transcriptase polypeptides that have apparent molecular weights of 66 and 51 kDa. In the experimental work leading to this invention, the

51-kDa form was found to have an apparent molecular weight of 54 kDa. The term 51 kDa is used to be consistent with the literature. These two polypeptides have identical amino- termini (N-termini) , but the larger form has additional carboxy-ter inal (C-terminal) sequences (Lightfoote, et al. (1986) J. Virol.

60.; Di Marzo Veronese, et al. (1986) Science 231) . The 66-kDa form has been shown to have both RNA-dependent DNA polymerase and RNase H enzymatic activities (Hansen, et al, (1987) J.

Biol. Chem. 262.; Larder, et al. (1987) EMBO J.

6.; Hansen, et al. (1988) EMBO J. 6., Hizi, et al. (1988) Proc. Natl. Acad. Sci. USA 85;

LeGrice, et al. (1988) J. Virol. 62). Genetic analyses and sequence comparisons have demonstrated that the RNA-dependent DNA polymerase activity is in the N terminus and that the RNase H activity is in the C terminus of the Moloney murine leukemia virus (MuLV) reverse transcriptase and that the overall organization of HIV-1 reverse transcriptase is the same as the MuLV reverse transcriptase. Although the 51-kDa form of HIV-1 reverse transcriptase contains almost precisely the amino acids that should form the RNA-dependent

DNA polymerase domain, the functio (s) of the

51-kDa form of HIVF- 1 reverse transcriptase (if any) has remained obscure. For the present invention, a highly purified preparation of the 66-kDa form of HIV- 1 reverse transcriptase isolated from a recombinant strain of E. coli was cleaved by using two highly purified preparations of HIV-l protease that derive from recombinant strains of E. coli and yeast. The HIV-l viral protease cleaves the 66-kDa form of HIV-1 reverse transcriptase to yield a product that is either extremely similar or identical to the 51-kDa form isolated from virions.

The position of the proteolytic cleavage products in the HIV-1 reverse transcriptase molecule was identified using a series of monoclonal antibodies (mAbs) that recognize specific segments of HIV-1 reverse transcriptase. The segments recognized by the mAbs are not uniformly distributed along the primary sequences of HIV-1 reverse transcriptase. A model was developed for the structure of HIV-l reverse transcriptase, and it was found, with one possible exception, that the segments recognized by the mAbs were predicted to be on the surface of the properly folded HIV- 1 reverse transcriptase.

Monoclonal antibodies are highly specific, sensitive reagents for identifying proteins or segments of proteins. Knowledge about the surface antigenic structure of human AIDS viruses has progressed with mouse (murine) monoclonal antibodies as serological probes.

In the present invention, monoclonal antibodies were used to detect and identify segments of HIV-1 reverse transcriptase resulting from digestion of the transcriptase with viral and nonviral proteases. The monoclonal antibody producing hybridoma cell lines of the present invention were formed by fusing a mouse myeloma cell line with ly pocytes from mice which were inoculated with purified reverse transcriptase.

The mice employed in the present invention were BALB/c mice providing a well proven and well recognized model for use of the present invention in humans. Four antibodies (mAb 19, 42, 50 and 51) recognize distinct segments of the HIV-l reverse transcriptase. The majority of the antibodies (mAb 21, 28, 48, 55, 58, 59 and 63) bind to the central portion of the reverse transcriptase within the RNA-dependent polymerase domain.

These mAbs enable identification of particular segments of the HIV-1 reverse transcriptase. Accordingly, one embodiment of the present invention includes hybridoma cell lines which produce monoclonal antibodies that specifically bind to antigenic recognition sites on segments of HIV-1 reverse transcriptase, the cell lines being selected from the group consisting Of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248 (respectively producing mAbs 19, 21, 28, 42, 50 and 51). This embodiment includes monoclonal antibodies produced by these cell lines.

Another embodiment of the present invention includes a method of preparing a monoclonal antibody which comprises culturing hybridoma selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248 in a suitable culture medium and recovering the antibody.

An embodiment of the present invention includes a method of detecting segments of HIV- 1 reverse transcriptase in a test sample comprising the steps of: contacting the test sample with monoclonal antibodies produced by hybridoma selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248, the contact being for a predetermined time sufficient for the antibody to specifically bind to an antigenic recognition site on at least one of the segments to produce an antibody-reverse transcriptase segment complex; and determining the presence of the antibody- reverse transcriptase segment complex. This method is provided wherein the sample may contain HIV-1 reverse transcriptase from an HIV- 1 infected human host. An embodiment of the present invention includes a diagnostic system for detecting the presence and amount of segments of HIV-l reverse transcriptase in a test sample including: at least one container wherein the test sample is contacted with an amount of monoclonal antibody produced by hybridoma selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248; and means for detecting the presence and amount of antibody- reverse transcriptase segment complex. This diagnostic system is provided, wherein the test sample may contain HIV-l reverse transcriptase from an HIV-1 infected human host. The diagnostic system may be provided as a test kit.

BRIEF DESCRIPTION OF THE FIGURES

Further objects and advantages of the present invention will be better understood by carefully reading the following description of the exemplary embodiments of this invention in conjunction with the accompanying figures, of which:

Fig. 1 shows representative Western transfers showing the reaction of two mAbs made against recombinant reverse transcriptase with recombinant reverse transcriptase and, with HIV- 1 particle lysates from three different strains;

Fig. 2 shows the deleted forms of HIV- 1 reverse transcriptase, relative to the full- length reverse transcriptase;

Figs. 3, 4, 5 show immunoblots showing the differential binding of mAbl9, mAb42 and mAb50 to native and mutant forms of HIV-1 reverse transcriptase; Fig. 6 shows a map showing binding regions of all mAbs;

Fig. 7 shows Coo assie brilliant blue- stained polyacrylamide gel showing the results of incubating purified HIV-l reverse transcriptase with purified HIV-1 protease at 37°C over a time course of 0-22 hr.

Fig. 8 shows immunoblots showing the recognition of the HIV-l reverse transcriptase proteolysis product (Fig. 7) by different mAbs (Fig. 3-5) ; and

Fig. 9 shows immunologic detection of proteolytic fragments generated by non-viral proteases.

DETAILED DESCRIPTION OF THE INVENTION

The following description is intended to illustrate this invention without limiting same in any manner especially with respect to substantially functional equivalents of cell lines described and claimed herein. The cell lines (hybridomas) disclosed in the present invention are deposited at the American Type Culture Collection, Bethesda, Md. and will be maintained in accbrdance with the Budapest Convention. They bear the following deposit numbers:

mAb ATCC #

19 HB10243

21 HB10244 28 HB10245

42 HB10246

50 HB10247

51 HB10248

Because reverse transcriptase is part of a polyprotein, the portion of the viral genome that encodes it is not bounded by initiation and termination codons. The region encoding the HIV-1 reverse transcriptase (RT) can be modified by introducing initiation and termination codons into the HIV-1 genome at positions that originally encoded the protein segment recognized by the viral protease. It is possible to introduce a termination codon at precisely the site that encodes the proteolytic recognition site at the C terminus of the HIV-l RT. However, this is not possible at the corresponding site at the N terminus. RT molecules isolated from HIV-1 virions have an N- terminal proline. Translation always initiates with a methionine so an N-terminal methionine is required. In the present invention the codon for the initiator methionine of the present DNA constructions were embedded within the recognition site for the restriction enzyme Ncol. This sequence can function as an efficient site for translational initiation in E. coli, and allows the expression of precisely the same unfused protein in the host. The Ncol/ATG expression vector was prepared. Once an appropriate genetically engineered segment with an NcoI/ATG was created, the segment was moved from one expression vector to another. If the Ncol site is retained, the first base of the second codon is specified as a G. This precludes the use of certain codons for the second amino acid (including those for proline) . The modified HIV-1 RT segment created for the present invention encodes methionine and valine as the first two amino acids, followed by the proline found at the N terminus in the virion- derived enzyme.

If the amino acid at the second position is crucial for function, synthetic DNA segments that have bases other than G a the first position of the second codon can be ligated to an NcoI/ATG vector. However, in such constructions the Ncol site is not retained and it is more difficult to move the modified segment from one vector to another. NcoI/ATG vectors can be used to express a variety of proteins including the present RT.

For the present invention, the HIV-1 RT was expressed in E. coli because there is no evidence that this enzyme is modified post transcriptionally. For expression in E. coli, the modified segment encoding the RT was introduced into the expression plasmid pUCI2N, a derivative of pUC12 that has two bases near the lacZ ATG mutated to created an Ncol site.

Introduction of the pUC12N plasmid containing HIV-1 RT into E. coli results in the synthesis of large amounts of a new 66-kDa protein, the expected size of the HIV-1 RT. Lysis of this strain with Triton X-100 releases approximately half of the 66-kDa protein in soluble form. It was desired to demonstrate directly that the active site for the HIV RNA-dependent DNA polymerase made in E. coli is essentially identical to the active site of the viral enzyme. The structure of the active site was probed with the competitive inhibitors dideoxy GTP and dideoxy TTP. The effects of these inhibitors were tested not only with the two forms of HIV-l RT (prepared from virions and from E. coli) , but also with mouse murine leukemia virus (M-MuLV) RTs. In this assay the two HIV-l RTs were indistinguishable but were clearly distinguished from the MuLV enzyme, demonstrating that the assay can discern small differences in the active sites of RNA-dependent DNA polymerases.

In an attempt to better define the portion of HIV-1 RT that is involved in forming the active site of the RNA-dependent DNA polymerase, a series of small in-frame insertions were made at various positions in the E. coli expression plasmid in the region encoding HIV-1 RT. The mutant plasmids, when introduced into E. coli, induce the synthesis of mutant forms of HIV-1 RT. A major reason for expressing the HIV-

1 RT in E. coli is to provide starting material for purifying large amounts of the enzyme. Under optimal growth conditions the bacteria contain, judged by gel electrophoresis and staining with Coomassie blue, several percent of their total protein as the HIV-l RT. The HIV-1 RT synthesized in E. coli was purified and used to generate polyclonal and monoclonal antibodies. The E. coli expression system provided a good means to map the approximate sites of the epitopes recognized by monoclonal antibodies that react well in Western transfer assays.

A nested set of amino and carboxyl terminal deletions were prepared by making deletions in the plasmid used to express the full-length HIV-1 RT. A determination was made as to which of the deleted forms of the HIV-1 RT react with individual monoclonal antibodies.

The epitopes recognized by the isolated monoclonal antibodies are not distributed uniformly along the length of the HIV-l RT.

Most of the antibodies recognize a relatively small central region of HIV-1 RT. However, monoclonal antibodies have also been obtained that react with the N and the C termini of the molecule. The monoclonal antibodies were prepared with, and originally identified as reacting with, HIV-l RT in native form. Most of the monoclonal antibodies probably recognize segments on the surface of the properly folded enzyme.

This definition of the epitopes by deletion mapping is operational. If some or all of the antibodies recognize native protein, it is possible that some or all of the binding sites for these monoclonal antibodies recognize sequences that are not contiguous in the HIV-1 RT primary sequence. Mapping the antibody binding sites using deletion mutants enabled the present operational definition of segments of the HIV-1 RT that react with the individual monoclonal antibodies. In contrast to the protein purified from^'.E. coli , the RT purified from virions is a 1-1 mixture of the 66-kDa component and a protein of approximately 51-kDa that has the same amino terminus as the 66-kDa form, but is missing a carboxyl terminal segment that comprises the RNase H domain. The 51-kDa form is, as far as in now known, unique to HIV. Other well-studied retroviruses do not contain an equivalent trunacted form of RT. Individual proteolytic digestion products of the HIV-1 RT can be unambiguously identified in Western transfer assays using monoclonal antibodies that recognize specific segments of the HIV-1 RT. This technique was used to show that, in an in vitro reaction containing only the two purified proteins, the viral protease cleaves the 66-kDa protein at a site very near or identical to the site where the cleavage that generates the 51- kDa form in virions occurs. The specificity of other, unrelated, proteases were tested. Although none of the other tested proteases have as much specificity as the HIV-1 viral protease, at least two other proteases, papain and trypsin, have preferential cleavage sites near the site used by the viral protease. The unrelated nonviral proteases are, however, less selective and make additional cleavages. The data thus shows that the cleavage in virions that produces the 51-kDa protein is made by the viral protease. The data also shows that the specificity of the cleavage event depends both on the specificity of the protease and on the structure of the 66-kDa protein. There is a region of the 66-kDa protein that is susceptible to the viral protease and to unrelated proteases. A 15-kDa (or smaller) C-terminal fragment was not detected after the in vi tro cleavage of 66-kDa HIV-1 RT with the viral protease either by Coomassie brilliant blue staining or by immunoblotting. In addition, a 15-kDa fragment was not detected in disrupted virions by immunoblotting.

In the present application, a model for the structure of the HIV-1 RT was developed based on sequence comparisons, mutational analysis and biochemical data. The proteolytic data resulting from the experiments performed for this application supported the model. In addition, the antibody binding regions are within segments of the HIV-1 RT which are believed to lie on the surface of the RT molecule.

The N terminus and the central region between amino acids 190 and 340 have been found on the surface of the properly folded RNA- dependent DNA polymerase domain. RNase H activity is part of a separate domain. Because the RNase H domain is small and is connected to the polymerase domain by a protease-sensitive segment, much of the C terminus of HIV-1 RT is on the surface.

Analysis of Segments of HIV-1 RT Recognized bv the mAbs.

Monoclonal antibodies were produced following the injection of mice with purified 66 kDa HIV-1 RT made by a recombinant strain of E. coli (Hizi, et al., (1988) Proc. Natl. Acad.

Sci. USA 85) . Antibodies that react with HIV-1

RT were initially detected with an enzyme-linked immunosorbent assay (ELISA) using purified HIV-

1 RT. For the epitope mapping experiments, a secondary screening was done to identify those mAbs that react in an im unoblot. mAbs that recognize recombinant HIV-l RT also react in immunoblots with RT obtained from virions from three different strains of HIV-1 (Fig. 1) .

Fig. 1 shows representative Western transfers showing the reaction of two mAbs made against recombinant RT with recombinant RT and with HIV particle lysates from three different strains:

A) mAb 50 binds an epitope near the C terminus of the RT (Fig. 5) . The segment of RT containing this epitope is missing from the 51- kDa form. Therefore, this antibody reacts with the 66-kDa but not with the 51-kDa protein. Lanes: 1, partially purified HIV- 1 RT from E. coli ; 2, disrupted virions of the III B strain; 3, disrupted virions of the RF strain; and 4, disrupted virions of the PH strain.

B) mAb 21 binds an epitope near the center of the RT (Fig. 6) . This epitope is present in both the 66- and 51-kDa forms. Lanes: 1, partially purified HIV-l RT from E . coli ; 2, disrupted III B virions; 3, disrupted RF virions; and 4, disrupted PH virions.

The segments recognized by the mAbs were mapped using a series of N-terminal and C- terminal deletion mutants. The plasmids used to express the deleted versions of the HIV-1 RT all derive from a plasmid that expresses the 66-kDa form of the HIV-1 RT. The bacterial strains carrying plasmids with the deletion mutations produce HIV-1 RT-related proteins of the appropriate size (Fig. 2) .

Fig. 2 shows the deleted forms of HIV- 1 RT relative to the full-length RT.

A) Mutants are named by the number of amino acids missing relative to the wild-type RT and the by end of the protein N terminus (AT) vs. C terminus (CT) from which they were lost. Construction of the plasmids producing these deleted forms is described in Materials and Methods. The relative size of the deletions is shown.

B) Coomassie brilliant blue-stained SDS-polyacrylamide gel shows the proteins (arrows) produced by E. coli (strain DH5) carrying the pUCl2N plasmid with or without an RT insert. The various strains of E . coli were grown overnight in NZY broth (25) containing 100 μg/ml ampicillin. Cells from 40- 60 μl of culture fluid were recovered by centrifugation, solubilized in SDS sample buffer and loaded on each lane. The plasmids carried by the strains were as follows. Lanes: 1, pUC12N, no insert; 2, pUC12N with full-length RT insert; 3, AT 13; 4, AT 23; 5, AT 126; 6, AT 139; 7, AT 221; 8, CT 8; 9, CT 16; 10, CT 23; 11, CT 58; 12, CT 133; and 13, CT 250.

In some cases, the specificity of the mapping was confirmed using strains of E . coli that express RT from MuLV and strains that express hybrid proteins composed of segments from MuLV and HIV-1 RT (Fig. 5) . The ability of the mAbs to recognize RT proteins produced by the various strains of E. coli was determined by immunoblot (Figs. 3-5) .

Figs. 3-5 show immunoblots showing the differential binding of mAbs to native and mutant forms of HIV-l RT. E . coli DH5 cells carrying the pUC12N plasmid with or without RT or an RT deletion mutant were grown in NZY broth containing 100 μg/ml of ampicillin, collected by centrifugation and solubilized in SDS sample buffer. Proteins from cells in 40-60 μl of culture fluid were loaded onto each lane of a polyacrylamide gel, separated and transferred to nitrocellulose paper. Transfers were probed with mAbs and binding was visualized by alkaline phosphatase staining.

Fig. 3 - mAbs 19. Lanes: 1,

PUC12N, no insert; 2, pUC12N with full-length RT insert (ρUC12N/RT) ; 3, purified Rt; 4, AT 13; 5, AT 23; 6, AT 139; 7, AT 221; 8, CT 133; and 9, CT 250.

Fig. 4 - mAb 42. Lanes: 1, pUC12N; 2, pUC12N/RT; 3, purified RT; 4, AT 13; 5, AT 23; 6, AT 139; 7 , AT 221 ; 8 , CT 250 ; 9 , CT 133 ; 10 , CT 58 ; and 11 , AT 126 .

Fig. 5 - mAb 50. Lanes: 1, pUC12N/RT; 2, pUC12 with MuLV Rt insert (Hizi, et al. (1988) Gene

66); 3, MuLV/HIV-1 RT chimera MH6- 2; 4, purified RT; 5, CT 23; 6, CT 16; 7, CT 8; 8, CT 58; 9, CT 250; 10, CT 133; and 11, AT 126. The mAbs recognize several different segments of the HIV-1 RT. These recognition sites are not uniformly distributed along the primary amino acid sequence of the protein (Fig. 6). Fig. 6 shows a map showing binding regions of 11 mAbs. four antibodies (mAb 19, 42, 50 and 51) recognize distinct segments of the HIV-1 RT. The majority of the antibodies (mAb 21, 28, 48, 55, 58, 59 and 63) bind to the central portion, of the RT within the RNA- dependent DNA polymerase domain.

While mAbs of the present invention have been developed that recognize sequences near the N- and C-termini, the majority of the present monoclonals react with a relatively small, central region of the RT. The present mAbs were elicited and screened using nondenatured HIV-1 RT. The majority of the present mAbs are believed to react with segments on the surface of the correctly folded 66-kDa form of HIV-1 RT. The immunoblots also provide information about the RT-related proteins that accumulate in the various E . coli strains. Most of the expression strains produce, in addition to the expected protein, smaller, and in some cases, larger forms that are specifically recognized by some of the mAbs. Control experiments have shown that these are not E . coli proteins (Fig. 3-5). In theory the smaller than expected forms could derive from premature termination, from internal initiation of translation, or from proteolytic degradation. Analyses with a mAb that recognizes the N terminus of "the full-length HIV-1 RT protein (Fig. 3) demonstrated that none of the smaller forms contain the sequence found at the N terminus of the normal HIV-1 RT protein. This means that, at least for the full-length clone and the C-terminal deletion mutants, the smaller forms are probably not the result of premature termination. The DNA sequence encoding HIV-1 RT contains several internal ATGs, but these are not associated with a consensus sequence of the initiation of protein synthesis in E . coli (Shine, et al. (1974) Proc. Natl. Acad. Sci. USA 71) . If the shorter forms derive from proteolytic cleavage, the cleavage is quite distinct from the proteolytic event that gives rise to the truncated form seen in virions and from the proteolytic events that produce a similar protein during purification of the HIV- 1 RT. The smaller form of RT in HIV-l virions, usually called 51 kDa, has the same N terminus as the 66-kDa protein. Although this protein is referred to as the 51-kDa form, in fact, it migrates to the position expected for a protein of 54 kDa on the SDS-polyacrylamide gels used for the present invention. By contrast, the most prominent of the small forms that accumulate in the E. coli strain expressing the full-length 66-kDa HIV-1 RT migrates as a protein of approximately 47 kDa, and has the same (or nearly the same) C terminus as the 66- kDa protein (Fig. 5) . This 47-kDa form is missing the N-terminal segment present in the 66-kDa form (Fig. 3) . At least some of the strains carrying deleted plasmids produce corresponding proteins that also lack the expected N terminus.

The larger forms of HIV-1 RT could arise from read-through. There may be read- through from plasmid sequences upstream of the lacZ ATG that is the expected start for translation. Alternatively, the termination condon inserted at the end of the RT coding regions could be suppressed.

Proteolytic Digestion of HIV-1 RT. Purified 66-kDa HIV-1 RT was digested with both viral and nonviral proteases. To investigate the cleavage of the RT by the viral protease, 5 μg of purified 66-kDa HIV-1 RT was mixed with 20 ng of cloned, purified HIV-1 protease and incubated overnight at 37°C.

Samples were taken at various times during the incubation. Aliquots of these samples were fractionated on SDS-polyacrylamide gels and visualized by staining with Coomassie brilliant blue (Fig. 7) .

Fig. 7 shows Coomassie brilliant blue- stained polyacrylamide gel showing the results of incubation purified HIV-1 RT with purified HIV-l protease at 37°C over a time course of 0- 22 hr. Each lane contains 2.7 μg of protein.

Lanes: 1, starting material (RT and protease with no incubation); 2, 1 hr; 3, 2 hr; 4, 3 hr; 5, 4 hr; 6 , 5 hr; 7, 22 hr; and 8, control, RT incubated 22 hr at 37°C without added protease. The major cleavage product has an apparent molecular weight of 54 kDa (usually called.51 kDa) . Additional aliquots were fractionated on gels, transferred to nitrocellulose paper and probed with several of the mAbs (Fig. 8) .

Fig. 8 shows immunoblots showing the recognition of the HIV-1 RT proteolysis products (Fig. 7) by different mAbs (Fig. 3-5). Each lane was loaded with 2.7 μg of protein. The proteins were separated and transferred to nitrocellulose paper. Transfers were probed with mAbs and binding visualized by alkaline phosphate staining.

A) mAb 19. Lanes: 1, control, RT incubated 22 hr at 37°C without protease; 2, 2 hr; 3, 3 hr; 4, 4 hr; 5,

5 hr; and 6, 22 hr.

B) mAb21. Lanes: 1, control, 22 hr. without protease; 2, 3 hr; 3, 4 hr; 4,

5 hr; 5, 6 hr; and 6, 22 hr.

C) mAb 50. Lanes: l, control, 22 hr without protease; 2, 3 hr; 3, 4 hr; 4, 5 hr; 5, 6 hr; and 6, 22 hr. The 51-kDa protein reacted with the N- terminal specific mAb 19 and failed to react with the C-terminal specific mAb 50. Since mAb 19 recognizes the first 13 amino acids of HIV-l RT, this demonstrates that the 51-kDa protein produced in vitro has the same N terminus as the 66-kDa protein. The 51-kDa protein from HIV-1 virions has the same apparent molecular weight and reacts identically when probed with these mAbs (compare Fig. 1 and 8) . These results show that cleavage occurs either at or very near the same site in the in vitro digestion and in virions. The 66-kDa form can be cleaved to 51- kDa form by a protease present in E. coli (Lowe, et al. (1988) Biochemistry 27). Although the HIV-1 protease made in E. coli was purified by HPLC, the possibility that a contaminating bacterial protease could have been responsible for cleaving the HIV-1 RT in vitro was investigated. To demonstrate that the HIV-1 protease was responsible for the cleavage seen in vitro, the 66-kDa form of HIV-1 RT was cleaved with a highly purified HIV-l protease synthesized by a recombinant strain of yeast. The results were similar to those obtained with the HIV-1 protease purified from E. coli . A small C-terminal fragment was not detected either in disrupted virious or in the in vitro digests with either of the mAbs that react with this portion of HIV-1 RT (Fig. 8).

The specificity of the cleavage event may reside in the viral protease, in the way the 66-kDa HIV-l RT is folded or both. The cleavage site that yields the 51-kDa form must be at or near the site cleaved by the viral protease. E. coli extracts contain a protease that also cleaves the 66-kDa form of HIV-1 RT to yield a 51-kDa form. Immunoblots probed with mAb 19 and 50 reveal that the 51-kDa, form produced by cleavage with the E. coli protease also has an intact N terminus and is missing the C terminus. Although the protease responsible for this cleavage is of E. coli origin, it has not been identified. These results show that the site (or region) cleaved by the viral protease is unusually sensitive to proteolysis. The 66-kDa protein was digested with several other proteases. Although none of the enzymes we tested show the specificity of the viral protease, both papain and trypsin cleave near the site recognized by the viral protease and at additional sites (Fig. 9) .

Fig. 9 shows immunologic detection of proteolytic fragments generated by non-viral proteases, four proteases, V8 (2.25 μg/ml) , chymotrypsin (0.75 μg/ml), papain (0.375 μg/ml) and trypsin (1.25 μg/ml) were incubated with purified RT (0.38 mg/ml) for 24, 3, 24 and 3 hr, respectively, at 23°C. The digestion products were separated on polyacrylamide gels, transferred to nitrocellulose paper and probed with three different mAbs, 19, 42 and 21.

Positive reactions were visualized by alkaline phosphatase staining. Lanes containing only pure RT have 14 μg of protein. All other lanes contain 5 μg of protein. Chromotrypsin was not found to be particularly effective in mimicking the cleavage made in virions.

Materials and Methods

Construction of the Plasmids Expressing Deleted Forms of HIV-1 RT. The plasmid used to express HIV-1 RT has been described (Hizi, et al. (1988) Proc. Natl. Acad. Sci. USA 85) . In brief, the ends of the region encoding HIV-l RT were modified using synthetic DNA segments to introduce an initiation and a termination codon at the sites in the HIV-1 genome that encode the amino acids where HIV-1 RT is normally cleaved from the polyprotein precursor. This modified segment was inserted into the expression plasmid pUC12N. The HIV-1 RT made in E. coli has two additional N-terminal amino acids, methionine and valine, when compared with the HIV-1 RT isolated from virions. All the deletions are named for the number of amino acids removed by the deletion. AT designates N-terminal deletions, CT designates C-terminal deletions.

Several C-terminal deletion mutants (CT 8, CT 16, CT 23 and CT 133) and one N-terminal deletion mutant (AT 23) were constructed. Additional C-terminal deletion mutants were constructed by cleaving the expression plasmid with Asp 718 and a second restriction endonuclease that cleaves in the HIV-1 RT coding region. The ends of the DNA were filled in using the large (Klenow) fragment of E. coli DNA polymerase I and the ends ligated together. The second restriction endonuclease used to construct CT 58 was Nsi 1, for CT 250 Pfl Ml was used. In all cases, the small DNA segment encoding the C terminus of HIV-1 RT between the sites for Asp 718 and Sal I is out of frame and does not contribute HIV-1 RT related amino acids. For some of these deletion mutants, the DNA sequence around the deletion was confirmed by directly sequencing the deleted plasmids.

N-terminal deletions were made by digesting the expression plasmid with Nco I and with a second restriction endonuclease that recognizes an internal site. To maintain the correct reading frame, an appropriate double stranded synthetic DNA segment was used to join the Nco I recognition site to the internal site. For the mutant AT 13, the second restriction endonuclease was Sma I , for AT 126 it was Ace I, for AT 139 it was Eco RV and for AT 221 it was Bst XI. In some cases, the synthetic oligonucleotides used to join the second site to Nco I encoded a short segment of HIV-1 RT sequence adjacent to the internal restriction site. The plasmids were introduced into competent DH-5 cells (Bethesda Research Labs) . Clones containing deleted plasmids were identified by digesting plasmid DNAs with appropriate restriction endonucleases.

The MuLV/HIV RT chimera (MH6-2) was constructed by ligating an Nco I to Eco RV segment from the MuLV RT expression plasmid

(Hizi and Hughes (1988) Gene 66) to the HIV-1 RT expression plasmid completely digested with Nco I and partially digested with Kpn I. The Eco RV site was jointed to the Kpn I site with synthetic oligonucleotides. The resulting chimeric protein contains 500 amino acids from MuLV RT at the N terminus and 137 amino acids from HIV-l RT at the C terminus.

Virus. Purified virus (strains IIIB, RF and

PH) was obtained from Program Resources Inc.

Production of mAbs.

Eight-week old BALB/c mice were inoculated intraperitoneally with purified RT (66 kDa form) in 0.5 ml of emulsion with complete Freund's adjuvant. Two weeks later, the animals were boosted with a 0.5 ml subcutaneous inoculation of the purified RT in incomplete Freund's adjuvant. The following week, blood was obtained from the animals by orbital bleed and the serum tested by ELISA to determine anti-RT activity. Two weeks following the second injection, the mice were given a final intravenous boost of 0.1 ml and the spleens of the mice showing the strongest reactivity were removed 4 days later.

The cell fusion protocol has been described by one of the inventors and others (Showalter, et al. Infect. Immun. (1981)). This publication is hereby incorporated by reference.

Briefly, spleens were minced, passed through sterile gauze into a conical 50-ml tube and washed twice. Cells were counted and mixed at a 5:1 ratio with actively-growing, washed NSI (NSI-1) cells which is a variant of mouse plas acytoma MOPC 21 (Kohler et al, Eur. J. Immunol. (1976)). The cell mixture was pelleted and gented resuspended over a 1-min period in 1 ml of 50% polyethylene glycol 4000 (Merck) per 1.6 x 10* lymphocytes. Two ml of RPMI-1640 containing 15% fetal bovine serum was added, followed by 10 ml of complete medium. Cells were pelleted, gently resuspended in 22 ml of complete medium and dispensed into 96-well Costar plates (~ 100 μl/well) . Cells were grown in RPMI containing hypoxanthine-aminopterin- thymidine (HAT; Gibco) for 30 days, and then maintained on medium without selection.

Beginning about day 10, media from actively-growing hybridomas was tested by ELISA for anti-RT activity. Cells from positive wells were expanded to 24-well plates (Costar) and cloned by limiting dilution in 96-well plates on a feeder layer of compatible thymocytes. Cloned cells were passaged in adult BALB/c mice primed with 0.5 m/ pristane (2,6,10,14 tetramethylpetadecane; Aldrich Chemical Co.) by intraperitoneal injection of 3 x 10* cells. The ascites fluids were harvested 7-10 days after injection, clarified and the immunoglobulin purified for further use.

Proteolysis.

Four commercially available proteases (Sigma Chemical Co., St. Louis, MO) were incubated with purified HIV-1 RT in a buffer of 25 mM Tris-Cl, 50 mM NaCl and 8 mM MgCl₃, pH 8. To digest 60 μg of RT the following amounts of the proteases were used: 0.135 μg V8 (0.074 units of activity), 0.045 μg chymotrypsin (2.7 x 10^"3 units), 0.022 μg papain (2.86 x 10^"3 units) and 0.075 μg trypsin (0.75 units). Reactions were incubated at 23°C for 3 hr or for 24 hr.

Purified RT (5 μg) was combined with 40 ng of purified viral protease in a buffer of 50 mM sodium phosphate, pH 6.5, 10 mM NaCl, 5 mM DTT and 0.05% Triton X-100. The reaction mix was incubated at 37°C for 0 hr, 1 hr, 2 hr, 3 hr, 4 hr, 5 hr, 6 hr and 22 hr. Purified RT was also incubated for 22 hr without added protease. HIV-1 protease was purified from both E . coli and yeast expression strains using a procedure including cation exchange chro atography, isoelectric focusing, reverse phase high pressure chromatography, and gel electrophoresis.

Western Transfer and Immunostaining.

For immunodetection of RT in bacterial strains, the protein from E . coli grown in 40- 60 μl of culture media was dissolved in SDS sample buffer and separated on polyacrylamide gels. Proteins were transferred to nitrocellulose paper using an ABN PolyBlot (American Bionetics, Hayward, CA) . Blocking and washing were performed using 0.3% gelatin in 137 mM NaCl, 2.68 mM KCl, 8.1 mM Na₂HP0₄ and 1.5 M KH,P0₄. Transfers were probed with a 1:50 to 1:400 dilution of mAb followed by 20 μg of alkaline phosphatase conjugated goat anti-mouse IgG (Kirkegaard and Perry, Gaithersburg, MD) .

Alkaline phosphatase activity was detected using the chromogenic substrates BCIP and NBT (BRL, Gaithersburg, MD) .

While only a few exemplary embodiments of this invention have been described in detail, those skilled in the art will recognize that there are many possible variations and modifications which may be made in the exemplary embodiments while yet retaining many of the novel and advantageous features of this invention. Accordingly, it is intended that the following claims cover all such modifications and variations.

Claims

WHAT IS CLAIMED IS:

1. Hybridoma cell lines which produce monoclonal antibodies that specifically bind to antigenic recognition sites on segments of HIV- 1 reverse transcriptase, said cell lines being selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248.

2. Monoclonal antibodies produced by the cell lines of claim 1.

3. Method of detecting segments of HIV-1 reverse transcriptase in a test sample comprising the steps of: contacting said test sample with monoclonal antibodies produced by hybridoma selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248, said contact being for a predetermined time sufficient for said antibody to specifically bind .to an antigenic recognition site on at least one of said segments to produce an antibody-reverse transcriptase segment complex; and determining the presence of said antibody-reverse transcriptase segment complex.

4. The method of claim 3 wherein said reverse transcriptase is a 66-kDa form of HIV-1 reverse transcriptase or a 66-kDa/51-kDa form of HIV-1 reverse transcriptase.

5. The method of claim 3 wherein said sample contains HIV-l reverse transcriptase from an HIV-1 infected human host. 6. The method of claim 3 wherein said determination step includes Western transfer procedures.

7. A diagnostic system for detecting the presence and amount of segments of HIV-1 reverse transcriptase in a test sample including: at least one container wherein said test sample is contacted with an amount of monoclonal antibody produced by hybridoma selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248; and means for detecting the presence and amount of antibody-reverse transcriptase segment complex.

8. The diagnostic system of claim 7 wherein said system is provided as an assay kit.

9. The diagnostic system of claim 7 wherein said means for detecting includes equipment for performing Western transfer procedures.

10. The diagnostic system of claim 7 wherein said test sample contains HIV-1 reverse transcriptase from an HIV-1 infected human host.

11. A method of preparing a monoclonal antibody which comprises culturing hybridoma selected from the group consisting of HB10243, HB10244, HB10245, HB10246, HB10247, HB10248 in a suitable culture medium and recovering the antibody.