WO1992010574A1

WO1992010574A1 - Inhibition of human immunodeficiency virus by an adeno-associated virus gene for human cells

Info

Publication number: WO1992010574A1
Application number: PCT/US1991/009008
Authority: WO
Inventors: Barrie J. Carter; Beth Ann Antoni; Arnold B. Rabson; Irving L. Miller; James P. Trempe; Nor Chejanovsky
Original assignee: The United States Of America, As Represented By The Secretary, U.S. Department Of Commerce
Priority date: 1990-12-06
Filing date: 1991-12-06
Publication date: 1992-06-25
Also published as: AU1150792A

Abstract

A recombinant plasmid is disclosed according to the invention which comprises a transcription promoter inducible by gene products expressed by a human lentivirus and a gene which is non-homologous to that human lentivirus. The non-homologous gene encodes a gene product which inhibits lentiviral growth and lentiviral protein production by interfering with the trans-acting gene regulation system of the lentivirus. The expression of the non-homologous gene is further increased by the above-mentioned transcription promoter. The invention further comprises a method of inhibiting lentiviral growth and lentiviral protein production in mammals by transfecting cells of mammals infected with the lentivirus with the above-mentioned recombinant plasmid. The invention further comprises a method of inhibiting lentiviral growth and production by administering to mammals infected with the lentivirus the Rep protein encoded by the above-mentioned non-homologous gene in a pharmaceutically acceptable carrier.

Description

INHIBITION OF HUMAN IMMUNODEFICIENCY VIRUS BY AN ADENO- ASSOCIATED VIRUS GENE FOR HUMAN CELLS Acquired immunodeficiency syndrome (AIDS) is a disease caused by the human immunodeficiency virus (HIV) and is characterized by a profound deficiency in T-cell mediated cellular immune responses due to a drastic reduction in T4⁺ T cells (Reviewed in Gallo, R.C., 1987, Sci. Am. 256:47-56; Fauci, A.S., 1988, Science 239;616- 622) . Development of AIDS almost always leads to death and generation of effective therapies against this virus are essential. HIV infection disrupts immune responses and development of AIDS vaccines is rendered extremely difficult. A complete and detailed understanding of the molecular biology of HIV growth is not yet available. Nevertheless, current evidence suggests several points in the virus life cycle at which a therapeutic agent might intervene (Mitsuya, M. and Broder, S. 1987, Nature 325:773-778; Levy, J.A. , 1988 Nature 333:519-522).

HIVl is the prototype human lentivirus representing a group of pathogenic retroviruses which induce chronic disease. The current state of our understanding (reviewed in Varmus, 1988, Gen. Dev. 2 :1055-1062; Cullen, and Greene, 1989, Cell 58:423-426) indicates that the life cycle of HIV is controlled by a complex system of trans-acting gene regulation mediated by several HIV genes. In permissive cells such as activated T-cells, infection by HIV leads to a complex cascade of gene regulation which occurs in two stages. All the HIV genes are transcribed from a single RNA transcription promoter located in the 5" terminal LTR (long terminal repeat) region. In the early stage of expression, a low level of transcription leads to expression of spliced RNAs that express several HIV genes including tat, rev and nef. These genes are trans-acting regulators which alter the

SUBSTITUTESHEET pattern of HIV gene expression and lead to a second higher level of transcription resulting in expression of the remaining HIV genes including gag-pol. env. vif, vpu. The most dominant HIV gene in the regulation scheme is tat. The Tat protein recognizes and element, TAR, in the 5' end of all HIV mRNAs and activates transcription and perhaps translation of its own gene and all other HIV genes (Hauber, et al, 1987, Proc. Natl. Acad. Sci. USA, ji4_:6363; Ja obovitz et al. , 1988, Mol. Cell. Biol. 8.:2555: Berkhout et al ., 1989 Cell 59:273). Thus, Tat represents a positive feedback trans-activation system and expression of Tat is critical both for replication (Dayton et al., 1986, Cell 44:941; Fisher et al., 1986 Nature 320:367) and for high level gene expression of HIV (Sodroski et al., 1985, Science 229:74; Arya, et al., 1985, Science 229:69; Rosen, et al., 1985, Cell 4_1:813; Cullen, 1986, Cell 46:973; Peterlin, et al., 1986, Proc. Natl. Acad. Sci. USA 83:9734, Muessing, et al., 1987, Cell 48.:691; Wright et al., 1986, Science 234:988) . Tat represents a potential target for therapeutic intervention. In the present invention we describe one embodiment of an agent designed to function intracellularly to interfere with Tat function and thus suppress growth of HIV and production of infectious HIV particles. Furthermore, in one embodiment of the invention, expression of this interfering agent is induced by the tat gene so that it could function as a negative feedback inhibitor of Tat and thus of HIV growth. The invention is based upon the concept of using another trans-acting gene product or protein to inhibit the trans-acting regulation of HIV. More specifically, the invention is based upon use of a trans-acting gene of

SUBSTITUTESHEET a heterologous virus to inhibit trans-regulation in HIV. A variety of trans-acting genes of heterologous viruses (adenoviruses, herpes virus, papova viruses, papilloma viruses) have been reported to activate HIV gene expression (for example Gendelman et al., 1986, Proc. Natl. Acad. Sci. USA. .82:9759) generally by enhancing transcription directed by the HIV-LTR. No trans-acting gene of heterologous viruses have been reported to inhibit HIV gene expression. In the present invention we demonstrate that a trans-acting gene from a human parvovirus, adeno-associated virus type 2 (AAV2) can inhibit tat-mediated activation of gene expression and that it can block production of infectious HIV particles in human cells. Our invention is based on a concept different from that based on trans-dominant mutant phenotypes as described recently. Herskowitz (1987, Nature 329:219- 222) discussed the usefulness of generation of mutations in genes that create a dominant negative phenotype in studying and analysing gene function. A number of groups have used this approach and for instance Friedman et al. , 1988 (Nature 335:452-454) describe a mutant herpes simplex virus trans-regulatory gene that prevented herpes virus function in cells. Trans-dominant mutants of the HIV Rev protein (Malim et al., 1989 Cell 58.205) and Tat proteins (Green et al., 1989 Cell 58.:215) have been described. We recently described a trans-dominant mutant of the AAV rep gene which has a dominant negative phenotype (Chejanovsky, N. , and Carter, B., 1990, J. Virol., in press). Friedman et al., 1988, also noted that use of the trans-dominant form of a viral gene to confer resistance to the virus may be useful as a therapeutic approach not only to herpes virus but for other viruses, and that a mutant form of tat might be used to inhibit HIV. In a review of this work Baltimore (1988, Nature 335: 395-396), described this approach as intracellular immunization. However the approach described by Herskowitz, 1987; Friedman et al., 1988; and Baltimore, 1988 is based on using a trans-dominant form of a homologous virus protein. Our invention is based on use of a trans-acting heterologous protein from a different virus to inhibit HIV growth. Our approach can be described as heterologous intracellular inhibition but not as intracellular immunization. This latter term in the concept of "immunization" recognizes that the interfering protein is homologous to the inhibited protein. Our invention is based on the rep gene of adeno- associated virus as the heterologous trans-acting gene that can be used to inhibit HIV. Adeno-associated virus, AAV, is a member of the family Parvovirdae or parvoviruses (Siegl et al., 1985, Intervirology, 23:61- 73) . There are various isolates of AAV from human and other mammalian and avian species which are distinguished by serological differences but otherwise appear very similar in function. The present invention uses a human adeno-associated virus type 2 (AAV-2 but herein after called AAV) but this work would also apply to other AAV isolates. AAV is distinguished from other parvoviruses in that it grows efficiently only in cells coinfected by a helper virus. Helper viruses include adenoviruses, herpes virus and pox viruses. In the absence of helper virus, the AAV genome is integrated into the host-cell chromosome at high efficiency but there is little or on AAV gene expression from the integrated provirus. (For general reviews of AAV see Carter, 1989 in Handbook of

SUBSTITUTESHEET Parvoviruses. CRC Press, Vol. I , pp. 155-168 and pp. 255- 282) .

AAV exhibits several inhibitory functions such as inhibition of tumor induction in hamsters by oncogenic adenoviruses (Kirschstein et al., 1968, Proc. Soc. Exp. Biol. Med. 128:670; de la Maza and Carter, 1981, J. Natl. Cancer Inst. j67:1323), decrease in tumorigenicity of cells transformed by a viral or cellular onσogene (Ostrove et al., 1981, Virology, 113:521; Katz and Carter, 1986, Cancer Research. 4^:3023) and inhibition of growth of the helper virus (Carter and Laughlin 1984, the Parvoviruses, Plenum Press, pp. 67-123) . Some of these inhibitory events apparently do not require expression of an AAV gene product. For instance, AAV inhibition of adenovirus-induced tumors in hamsters required only terminal regions of AAV DNA which contained the replication origin but no gene coding regions (de al Maza and Carter, 1981) . More recently an AAV gene, rep, has been discovered which has some trans-acting inhibitory properties.

Molecular cloning of infectious AAV genomes

(Samulski et al., 1982, Proc. Natl. Acad. Sci. USA..

29:2077; Laughlin et al., 1983, Gene. £3.:65) into bacterial plasmids allowed genetic analysis of AAV (Tratschin et al., 1984, J. Virol. 53.:611; Hermonat et al., 1984, J. Virol. 51:329) which showed that the left half of the AAV genome comprised a gene, rep, required for AAV DNA replication. The rep gene corresponded to a coding sequence (Srivastava et al., 1983, J. Virol. 4_5:555) expressed from overlapping mRNA species transcribed from two promoters p5 and p₁₉ (Laughlin et al., 1979, Proc. Natl. Acad. Sci. USA 76:5567; Carter et al., 1990, Handbook of Parvoviruses. Vol. I. CRC Press, pp. 227-254) . At least four overlapping Rep proteins expressed from the rep gene have been identified (Mendelson et al., 1986, J. Virol. 6O:823; Trempe et al., 1987, Virology 161:18). Rep78 and rep68 are coded by mRNA transcribed from the p₅ promoter and Rep52 and Rep40 are coded by MRNA transcribed from the p_1g promoter.

As well as being required for AAV DNA replication, the AAV rep gene mediates both complex positive and negative regulatory effects (reviewed in Carter et al., 1990 Handbook of Parvoviruses. CRC Press, Vol. I pp. 227- 254) . Rep is auto-regulated and shows strong negative regulation of its own synthesis (Chejanovsky and Carter, 1990, J. Virol, in press; Antoni et al., manuscript in preparation) . In the presence of helper virus, rep may activate expression of AAV genes (Tratschin et al., 1986, Mol. Cell. Biol. 6.:2883; Labow et al., 1986, J. Virol.. j5():251; Trempe and Carter 1988, J. Virol. j>2.:68). Rep may negatively regulate expression of heterologous prokaryotic genes from AAV promoters (Tratschin et al., 1986, Mol Cell. Biol. €5:2883; Trempe and Carter, 1988, J. Virol. jS2:68; Mendelson et al., 1988, Virology 166:154 and Virology 166:612) or from heterologous promoters (Labow et al., 1987. Mol. Cell. Biol. 7:1320). For expression of a heterlogous reporter gene from the AAV p₄₀ promoter rep was shown to simultaneously activate mRNA but inhibit reporter gene activity at the level of translation (Trempe and Carter, 1988, J. Virol. j52.:68).

The series of complex regulatory effects mediated by rep are not understood at the level of cellular specificity of biochemical mechanism. Thus, it is not obvious how to predict the effect of a rep gene in or upon expression of other genes. Also, the role of individual Rep proteins in these trans-regulatory effects are not clearly understood.

In the present invention we modify expression of rep to overcome the otherwise strong negative autoregulation by Rep protein on its own synthesis. We show that Rep proteins can inhibit growth of HIV and production of infectious HIV in human cells. The modification of rep expression leads to increased Rep protein and this leads to more efficient inhibition of HIV than for Rep expressed from an AAV genome background. This shows that the rep gene or the rep gene product (i.e. Rep protein) can be used to inhibit production of infectious HIV and this forms the basis of a novel approach to therapeutic intervention in treatment of HIV mediated disease.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a therapeutic approach to HIV-mediated disease by interfering with production of infectious particles in human cells.

It is another object to provide an agent to interfere with HIV growth in cells by interfering with the trans-acting gene regulation system of HIV.

It is still a further object to provide such an interfering agent by use of a trans-acting gene (or gene product) from a heterologous (non-HIV) virus. It is yet another object to develop such an interfering, trans-acting gene or gene product utilizing a gene or gene product from adeno-associated virus (AAV) , a heterologous human parvovirus.

It is still a further object to utilize the AAV rep gene or gene product to generate the heterologous, trans- acting agent to interfere with HIV growth.

It is an additional object to modify expression of the AAV rep gene or rep gene product to increase the expression level of the gene.

It is yet another object to increase expression of the AAV rep gene product by using a heterologous (non- AAV) transcription promoter. It is yet a further object to increase expression of the AAV rep gene or rep gene product by using a transcription promoter that will be induσible by gene products expressed by HIV.

It is an additional object to construct recombinant DNA molecules that express the AAV rep gene or gene product from homologous (AAV) or heterologous (non-AAV)) transcription promoters when introduced into human cells.

It is a further object to demonstrate that recombinant DNA molecules which express the AAV rep gene or rep gene product when introduced into human cells also inhibit generation and growth of HIV in human cells.

It is still another object to demonstrate that inhibition of HIV in such cells is mediated by the AAV rep gene product. Other objects and advantages will become apparent as the detailed description of the invention proceeds. BRIEF DESCRIPTION DRAWINGS These and other objects, features and many of the attendant advantages of the invention will be better understood upon a reading of the following detailed descriptions when considered in connection with the accompanying drawings and figures wherein: Figure 1 shows a schematic diagram of construction of a recombinant plasmid, pHIVrep DNA to express the AAV rep gene according to the present invention. As shown in the diagram the CAT gene was removed from pBennCAT by cleaving with the enzymes Hindlll and BamH-I. Into this site a 1968 base pair Aval fragment (obtained from pAV2 and containing nearly all the AAV rep gene) was inserted by blunt-end ligation to generate pBennHIVrep. To add the correct 3' end of the rep gene, an AAV intron and also a polyadenylation site, pBennHIVrep was cleaved with Sail and the Smaller Sail fragment was replaced with a Sail fragment from pJDT95 to yield pHIVrep. An additional plasmid, pHIV-rep was also generated by inserting the Sail fragment from pJDT95 into the opposite orientation from that in pHIVrep. pHIVrep contains a rearranged rep gene that is unable to code for functional rep proteins. The construction is explained in further detail in the text.

Figure 2 shows construction of a plasmid pHIVrepam having an amber (nonsense) mutation in the AAV rep gene. pNTC3, which contains an AAV genome having an amber mutation (am) in the rep gene, was made by converting nucleotide 1033 from C to A (Chejanovsky and Carter, 1988, Virology. 171:2391). pNTC3 was cleaved with Sstl and Hindlll to yield a 1 kb fragment spanning the region of the AAV genome containing the amber mutation. pHIVrep was cleaved with Sstl and Hindlll to yield a large fragment, a 1 kb Sstl/Hindlll fragment and a 600 bp Sstl fragment. The largest Sstl/Hindlll fragment of pHIVrep was ligated with the 1 kb Sstl/Hindlll fragment from pNTC3 to yield pHIVrepamdlS (pNT39) . This latter plasmid was cleaved with Sstl and ligated with the 600 bp Sstl fragment from pHIV to yield PHIVrepam.

Figure 3 summarizes the structure of the AAV genome and the relevant features of pHIVrep and pHIVrepam. In the upper portion the AAV2 genome is shown as a single bar with a 100 map units scale (l unit = approximately 47 nucleotides) . Stippled boxes indicate terminal repeats (replication origins) and solid circles indicate transcription promoters, (p₅, p₁₉, and p₄₀) . The poly(A) site is at map position 96. RNAs from AAV promoters are shown as heavy arrows with the introns indicated by the caret. The coding regions for the four rep proteins (Rep78, Rep68, Rep52, and Rep40) and for the viral capsids (VP1, VP2, and VP3) are shown with open boxes.

In the lower portion the relevant features of several plasmids are shown. For simplicity the plasmid DNA is represented by wavy lines at either end and the entire plasmid region is not shown. pAV2 contains an entire AAV2 genome and pJDT269 contains a deletion in the rep gene shown by the gap. p DT95 contains a deletion of a Hindi fragment (Hc2/Hc3) in the capsid gene region. pHIVLTR contains 553 nucleotides of an HIV LTR region (cross hatching) extending to nucleotide +80 relative to the RNA transcription, start site (heavy arrow) . The TAR region (+1 to +44) is shown by the open box. The vertically shaded region is about 500 base pairs of cellular DNA sequence derived from the original proviral clone pB2. pHIVrep contains the same AAV region as in pJDT95 i.e. from AAV2 nucleotide 263 to 4678 but containing the deletion of the capsid genes. AAV RNA transcription start sites are shown by the light arrows. pHIVrepam contains an amber mutation at AAV nucleotide 1033. pHIVrep has the region between two Sail sites inverted.

Figure 4 Effect of AAV rep gene on tat-mediated activation of expression of CAT from pHIVcat in human cells. Cultures of human 293 cells (10⁶ cells per culture) were transfected, using the calcium phosphate procedure, with 1 μg of pBenncat together with 10 μg of either pJDT269, pAHP, pAV2 or pHIVrep as indicated on the figure and varying amounts of a pARtat plasmid as indicated on the abscissa. The transfected cells were grown at 37°C for 48 hours then lysates were prepared and CAT activity was measured.

Figure 5 Western Analysis of Tat induction of Rep proteins in 293 cells. Plasmids pAV2 and pHIVrep were each cotransfected with increasing amounts of pHIVtat into 293 human cells. Panels A and B represent transfection of 0.1 or 1.0 μg of pHIVrep, respectively. Panels C and D represent transfection 0.1 or 1.0 μg of pAV2, respectively. In panels A, B, C and D lanes 1 to 4 represent cultures in which increasing amounts of pARtat plasmid (0, 0.01, 0.05 and 0.1 μg, respectively were cotransfected). Panel E, lanes 5 to 7, represents control transfections. Figure 6 Northern analysis of Tat induction of rep RNA transcription. Plasmids pHIVrep and pHIVrepam were each cotransfected with increasing amounts of the pARtat plasmid into cultures of 293 cells. L and H refer to transfection of 0.1 μg or 1.0 μg respectively, of pHIVrep or pHIVrepam. Panels A, B, and C refer to, respectively, 0, 0.01 or 0.10 μg, respectively, of cotransfected pARtat.

Figure 7 Induction of Rep by Tat in HeLa Cells. Cultures of 10⁶ HeLa cells were transfected by the calcium phosphate method including a glycerol-shock, with 5 μg (L) or 10 μg (H) of either pHIV rep or pAV2 or 10 μg of control carrier plasmid, pGEM4Z. pARtat was cotransfected at 2.5 μg (lanes 2, 5, 8, 11 and 14) or 5.0 μg (lanes 3, 6, 9, 12 and 15). Cells were harvested after 48 hours and the nuclear fractions were analyzed by SDS-PAGE and immunoblotting with an antibody to Rep. The Rep proteins were visualized by incubation with radiolabeled goat anti-rabbit IgG and autoradiography. Figure 8 Effect of Rep of HIV production. Cultures of 10⁶ HeLa cells were transfected by the calcium phosphate method with 10 μg of HIV proviral clone (pNL432) and the amounts of pHIVrep of pHIVrepam as indicated. All transfection mixes were adjusted to the same amount of DNA using a carrier plasmid (pUC18) . The mock transfection contained only pUClδ DNA. The medium from each culture was collected 48 hours after transfection and assayed for the HIV p24 (gag) antigen using an antigen capture assay (Dupont-NEN Research Products) . Figure 9 Western analysis of HIV inhibition by Rep. Cultures of 10⁶ HeLa cells were transfected with 10 μg of the HIV proviral clone (pNL432) and increasing amounts of pHIVrep or pHIVrepam. All transfection mixes were adjusted to the same amount of DNA using a carrier plasmid (pGEM4Z) . All transfections were performed in duplicate. Cells were harvested 48 hours after transfection and separated into nuclear and cytoplasmic components as indicated. One set of experimental samples were immunoblotted with an HIV patient serum (panel A) and the other set were immunoblotted with anti-Rep antibody (panel B) . Proteins were then visualized by reaction with ¹²⁵I-protein A (panel A) or ¹²⁵I goat anti- rabbit IgG (panel B) and autoradiography. In all panels lane 1 represents the mock transfection using only pGEM4Z; lane 2 represents transfection with pNL432 alone; lanes 3, 5, and 7 represent cotransfection of pNL432 with 10 μg, 3.0 μg and 1.0 μg of pHIVrep, respectively; lanes 4 and 6 represent cotransfection of pNL432 with 10 μg and 3.0 μg of pHIVrepA, respectively.

DETAILED DESCRIPTION OF THE INVENTION We have employed recombinant plasmids to introduce the AAV rep gene sequences into cultured human cells using the calcium phosphate DNA transfection technique and, in some cases, the lipofection technique. We assayed the effect of the rep gene on two model systems of HIV infection. In the first system using DNA transfection, we examined expression of the reporter gene CAT (chloramphenicol acetyltransferase) from an HIV-LTR sequence in the presence of a second plasmid expressing the HIV tat gene. In the second system, we transfected human cells with a plasmid DNA containing an HIV proviral clone under conditions leading to generation of infectious HIV particles. Cell Lines

All human cell lines were grown in monolayers at 37°C in 5% C0₂ in Dulbecco's minimal essential medium supplemented with fetal bovine serum (10% v/v) and the antibiotics penicillin and streptomycin. 293 cells are human embryonic kidney cells transformed by adenovirus type 5 (Graham et al., 1977, J. Gen. Virol. 36:59) obtained originally from N. Jones (Purdue University, Indiana) . HeLaJW cells are a subclone of the human cervical carcinoma cell line, HeLa, obtained from J. Rose (NIH) . SW480 cells are a human colon carcinoma cell line obtained from the American Type Culture Collection at passage number 100. Plasmids pAV2 contains an entire infectious AAV2 genome inserted into pAHP, a pBR322 based plasmid (Laughlin et al., 1983, Gene. 23.:65). pJDT269, a rep derivative of pAV2, has a deletion in the rep gene that blocks synthesis of all four rep proteins and JDT95 is a cap derivative of pAV2 (Tratschin et al., 1984, J. Virol. 51:611). pNL43 contains an infectious HIV proviral genome (Adachi et al., 1986, J. Virol. 59:284) and pARtat (pHIVtat) contains only the first exon of the Tat coding sequence expressed from the HIVLTR region of pNL43

(Gendelman et al., 1986, Proc. Natl. Acad. Sci. USA

83.:9759). pSVtat contains the first exon of the Tat coding sequence but expressed from a SV40 early region promoter (Jeang et al., 1988 J. Virol. 62.:3874). pNL432 also contains the same infectious HIV plasmid genome as pNL43 and was derived from pNL43 by deletion of flanking cellular sequences. Thus pNL432 contains only 1.5 kb of flanking cell sequence whereas pNL43 contains 9 to 12 kb of flanking sequence on either side.

Two pHIV-CAT plasmids expressing chloramphenicol acetyltransferase (cat) from an HIV-LTR promoter were used. In pBennCat, (Gendelman et al., 1986, Proc. Natl. Acad. Sci. USA 83.:9759) the HIV-LTR was obtained from an HIV provirus pB2 (Benn et al., 1985, Science 230:949) as described in Figure 1. In pUCCat the HIV-LTR-cat region of pBennCat was inserted into a pUC18 vector. pBennCat and pUCCat are functionally equivalent and collectively referred to as pHIV-cat. pHIVrep contains the AAV rep gene expressed from an HIV-LTR (Figure 1) . An American Type Culture Collection deposit under the Budapest Treaty of E^ coli containing the plasmid pHIVrep was accepted by the ATCC on June 20, 1990 and given the accession no. 68342. pHIVrepam contains an amber (nonsense) mutation in the rep gene (Figure 2) . pHIVrep contains a rearrangement in the rep gene (Figure 1) . pAHP, pUC18, pUC19, or pGEM4Z (Promega Corp., Madison, Wisconsin) were used as control plasmid DNA.

Plasmids were constructed and grown and DNA was purified using standard techniques as generally described in Sambrook et al., 1989 Molecular Cloning. Cold Spring Harbor Laboratory Press. Transfection of DNA in Cells

Monolayers of 293, SW480 or Hela cells were transfected with DNA according to the calcium phosphate precipitation procedure (Wigler et al., 1979, Proc. Natl. Acad. Sci. USA 7 :1373; Tratschin et al., 1984 Mol. Cell. Biol. 6:2884). For each experiment, cells grown in monolayer, approximately 10⁶ cells per dish or T-flask, were transfected with equivalent amounts of DNA adjusted by adding appropriate amounts of the control plasmid DNA pAHP, pUClδ or pGEM4Z. In some experiments on SW480 cells DNA was introduced into cells by lipofection using the lipofection reagent (Life Technologies Inc., Gaithersburg, Md) as described by the supplier. In calcium phosphate transfections of HeLa cells, the cells were glycerol-shocked after transfection. Assay of CAT activity

CAT activity was measured by acetylation of ¹⁴C- chloramphenicol in reactions containing l to 40 μl of cell extract as described (Garman et al., 1982, Mol. Cell. Biol. 2.:1044). All extracts were assayed in the linear range i.e. not more than 50% acetylation of the substrate. Results are expressed in arbitrary units of CAT activity where 1 unit results in acetylation of 1% of the substrate in 1 hr at 39 C. In some transfection experiments the levels of CAT activity were than expressed as a percentage of the control. Analysis of RNA

RNA was extracted from the cytoplasm of infected cells and analysed by electrophoresis in formaldehyde gels and blotting to nitrocellulose pater followed by hybridization with ³²P labeled AAV DNA and autoradiography. Analysis of Proteins

Proteins were extracted from cells and analysed by gel electrophoresis and immunoblotting (Western blotting) as described using an antibody to AAV rep protein raised against a synthetic oligopeptide (Mendelson, et al., 1986, J. Virol. .60:823) or serum from an HIV patent to detect the HIV p24 (gag) protein and its precursor p55. Analysis of HIV production

The production of infectious HIV was determined by assaying portions of cell culture medium for the p24 gag protein using the antigen capture assay kit (Dupont NEN Research Products) and for viral reverse transcriptase using a ³²p-TTP based assay (Willey et al., 1987, J. Virol. 67:139). Structure of HIV rep and related lasmids pHIVrep was constructed in order to replace the AAV p₅ promoter with the HIV transcription promoter comprising the HIV LTR and sequences extending to the Hindlll site at +80 nucleotides downstream of the HIV mRNA start site (Fig. 1,3). The pHIVrep construct contains AAV sequences beginning from nucleotide 263 immediately downstream from the p₅ TATA box. It is thus deleted for AAV p₅ promoter sequences but retains the normal start side of the AAV mRNA (p₅ mRNA) coding for rep78 and rep68 in addition to the start site of the HIV LTR promoter. As shown below only the HIV LTR start site is used in transcription from this plasmid. Because HIVrep retains the TAR region (nucleotides +1 to +44) of HIV LTR it was designed to be inducible by tat. pHIVrepam has a nonsense mutation that prevents expression of any functional rep proteins except under special conditions in the presence of a suppressor tRNA as described in Chejanovsky and Carter (1989, Virology 171:2391 and U.S. Patent Application, docket number 07/366,130) . Except for the single base charge pHIVrepam is identical to pHIVrep. pHIVLTR contains the HIV LTR but has no inserted AAV sequence. pHIVrep has a rearranged rep gene and also expresses no functional rep protein. Inhibition of tat function bv pHIVrep

In preliminary experiments we tested if rep could inhibit Tat activation of gene expression (Figure 4) . Human 293 cells were transfected with 1 μg of pHIVcat which expresses CAT from an HIV LTR and increasing amounts of pARtat which expresses Tat from an HIVLTR. As expected, in the presence of 10 μg of control plasmid DNA pAHP, there was a linear dose response for induction of CAT by Tat. When pAV2 was used instead of control plasmid there was a modest 5 fold inhibition of CAT activity. This inhibition was due to Rep because the Rep mutant, pJDT269, had no inhibitory effect. When pHIVrep was used there was a very strong inhibition of CAT expression. Not only did HIVrep abolish the induction by Tat but it also lowered the basal level of CAT expression.

In a more extensive series of experiments (summarized in Table I) we examined the effect of rep on expression of CAT from pHIVCAT in both 293 cells and SW480 cells as a function of the gene dosage of tat and rep. As shown in Table I, pAV2 showed only a modest inhibition as compared to its control pJDT269. Furthermore this modest inhibition by pAV2 was inconsistent (cf Fig. 4 and Table I for 293 cells) . In contrast, pHIVrep, consistently showed very strong inhibition and this inhibition was at least 10-fold more than the modest inhibition seen with pHIVrepam or pHIVrep. Thus the inhibitory effect of pHIVrep must be due mostly to expression of the Rep protein. The moderate inhibition by pHIVrepam or pHIVrep probably reflects competition for factors binding to the HIVLTR or TAR region because pHIVLTR showed a similar modest inhibition. Similar results were seen in both 293 cells and SW480 cells. Thus, the inhibition by pHIVrep in 293 cells is not accounted for by inhibition of the adenovirus transcriptional activator EIA present in these cells but not in SW480 cells. Similar results to those shown in Fig. 4 and Table I were obtained in other experiments when Tat was supplied from an SV40 promoter using pSVtat (B. Antoni, unpublished) or when lipofection was used rather than calcium phosphate precipitation to introduce DNA into cells (I. Miller, unpublished) . Expression of Rep proteins and RNA from pHIV Rep

The experiments above showed that pHIVrep was much more inhibitory than pAV2. To determine if this reflected a high level of Rep induced from pHIVrep by Tat we analysed protein expression in 293 cells by Western blotting with Rep antibody as shown in Fig. 5. In normal AAV infected cells Rep78 and Rep68 accumulate mainly in the cell nucleus whereas Rep52 and Rep40 are distributed in both nucleus and cytoplasm upon cell fractionation (Mendelson et al., 1986, J. Virol. 60:823..

As shown in Figure 5, pAV2 expressed only low levels of Rep proteins in nucleus or cytoplasm and there was no inductive effect of Tat. In contrast Rep78 was clearly induced by Tat at low inputs of pHIVrep (Fig. 5, panel A, tracks 1 to 4) whereas Rep52 and Rep40 were not induced. Furthermore the level of expression of Rep52 and Rep40 was about equivalent for pHIVrep or pAV2. Therefore induction by Tat was specific for Rep78 (and Rep68) in pHIVrep consistent with their expression from a chimeric HIV-rep mRNA containing the TAR sequence. In this experiment Rep68 is not well defined. The difficulty in observing Rep68 is due to several factors including a cross-reacting cell protein and a low level expression of Rep68 because of efficient splicing of mRNA. Nevertheless the experiment of Figure 5 suggests that the efficient level of inhibition of pHIVrep is due to efficient expression of rep78 or perhaps rep68. An unexpected feature is also revealed by the experiment of Figure 5. Even in the absence of Tat (lanes 1) the basal level of Rep78 expression is much higher from pHIVrep than from pAV2. This apparently reflects the fact that rep exerts a strong negative auto- regulation on expression from its homologous p₅ promoter

(Chejanovsky and Carter, 1990, J. Virol, in press) but is much less strongly auto-regulated from heterologous promoters (B. Antoni, J. Trempe, N. Chejanovsky and B.

Carter, manuscript is preparation) . Finally we note that in the experiment of Fig. 5 induction of Rep78 by Tat is readily seen at the low gene dosage of pHIVrep (panel A) but at a 10-fold higher level of pHIVrep (panel B) there is much less induction. This suggests that as the level of Rep78 is increased there is a feedback inhibition that prevents Tat action.

In other experiments using Western blotting (not shown) we confirmed that pHIVrepam and pHIVrep did not express any functional Rep protein and none could be induced by Tat. To show that induction of Rep78 protein by Tat reflected induction of the chimeric HIVrep mRNA we performed a Northern blot analysis of RNA from transfected 293 cells as shown in Figure 6. Again Tat induced the level of HIVrep mRNA at a low gene dosage of pHIVrep. At a 10-fold higher gene dosage of pHIVrep there was little additional induction. In contrast HIVrepam mRNA was induced at both high and low levels of pHIVrepam. which makes no functional Rep protein. This again is consistent with inhibition of Tat induction at high levels of Rep protein.

In Figure 6 the apparent induction of mRNA from the p₄₀ promoter of pHIVrep and not from p₄₀ of pHIVrepam reflects features of Rep regulation of AAV genes and is not directly relevant to the function of Tat. The induction of p₄₀ mRNA is due to the activating effect of Rep 78 on the p₄₀ promoter (Trempe and Carter, 1988, J. Virol. 62:68) . The chimeric HIVrep RNA from pHIVrep or pHIVrepam is the size expected if the HIV RNA start site is used rather than the AAV mRNA start site (Fig. 3) and this is confirmed in other experiments (B. Antoni, unpublished) . In further experiments similar to those of Figure 5, we demonstrated that specific induction of Rep78 and Rep68 from pHIVrep by Tat was observed in SW480 cells and Hela cells as well as in COS cells (an SV40 transformed monkey cell line) . One experiment using HeLa cells in shown in Figure 7. This experiment again shows that Rep78 and Rep68 were expressed from HIVrep in a dose- dependent fashion at a much higher level than from pAV2. Also, Rep78 and Rep68 were induced from pHIVrep by Tat. Evidence for a negative feedback also was seen in these experiments. Thus at a lower input of pHIVrep, Rep 78 and Rep 68 induction did not continue linearly with increasing dosage of pARtat. Similarly, at a higher level of pHIVrep and a high level of pARtat, Rep almost completely inhibited its own induction. These experiments showed that pHIVrep expressed higher levels of Rep78 and Rep68 than PAV2 because the negative autoregulation of rep in pAV2 was circumvented. Second, Rep78 and Rep68 could be induced by Tat and appeared to generate a negative feedback on induction by Tat. Third, pHIVrep, through the action of Rep78 (or Rep68) was highly inhibitory to Tat-mediated induction of a reporter gene CAT expressed from an HIV/LTR. This shows that the Rep78 or Rep68 was interfering with trans- activation by Tat presumable by blocking production of Tat protein. This latter point has not been directly demonstrated because we have been unsuccessful in directly observing tat protein using an anti-tat antibody in our experiments. Nevertheless, we then tested the ability of pHIVrep to inhibit production of HIV in human HeLa cells. Inhibition of Growth of HIV by Rep

Infectious HIV particles can be efficiently generated in HeLa cells transfected with an HIV proviral clone, pNL432. Accumulation of HIV into the medium from transfected HeLa cells can be measured by a viral reverse transcriptase assay (Table II) or by detection of the HIV p24 gag protein in an antigen capture immunoassay (Fig. 8) . In the cells cotransfected with pHIVRep, the level of viral reverse transcriptase was inhibited at least 50 fold whereas there was little or no inhibition (less than 2-fold) with pHIVrepam. Furthermore the inhibition was dose-dependent with respect to rep. Assay of p24 antigen in the cell medium showed a similar strong and dose- dependent inhibition by pHIVrep but not by pHIVrepam. These results are strikingly similar to the effects on CAT activity as shown in Table I. Thus release of extracellular HIV from transfected HeLa cells was almost totally inhibited by rep.

To determine that inhibition of extracellular HIV reflected inhibition of viral protein production, the experiment shown in Figure 9 was performed. Parallel cultures of HeLa cells were transfected with the HIV clones pNL432 in the presence or absence of pHIVrep or pHIVrepam. Nuclear and cytoplasmic protein extracts were prepared and analysed by Western blotting with antibody to Rep or antibody to p24 and its precursor p55 derived from HIV patient serum. As seen in Fig. 8A, the p55 and p24 proteins were readily detected in cells transfected with pNL432 and also in the presence of pHIVrepam.

However pHIVrep effectively prevented accumulation of both p55 and p24. In Fig. 8B the corresponding nuclear protein extracts show that Rep78 was expressed from pHIVrep but not from pHIVrepam. There was a clear concordance between expression of Rep78 and inhibition of p55 and p24. Thus pHIVrep inhibited intracellular production of the structural HIV proteins. In summary these results show that Rep can mediate an effective intracellular inhibition of HIV and thus Rep is a potential novel therapeutic agent against HIV. Since HIVrepam did not inhibit growth of HIV the results show that the inhibition by HIVrep was due to expression of the rep protein. This most probably reflects a function of the Rep78 protein because this is the major Rep protein expressed from HIVrep. However, Rep68 is also expressed. Furthermore Rep52 and Rep40 are also expressed at low levels from HIVrep from RNA transcripts originating from the p_t9 promoter but this promoter is not induced by Tat. It is now rendered obvious by our invention that, even if the inhibition of HIV is mediated mainly by Rep78, the other Rep proteins might also have an inhibitory function. A more general likelihood, which is an obvious extension of the present invention, is that the domain or portion of the Rep proteins responsible for inhibition of HIV could be identified by appropriate mutations or alteration by standard techniques obvious to anyone skilled in the art. Thus the recombinant construction pHIVrep is one embodiment only of the invention.

It is also now an obvious extension of the present invention that the AAV Rep protein or Rep proteins could be introduced into cells in a variety of alternative ways as a means of inhibiting HIV production. One embodiment shown here is by introduction of the rep gene as a recombinant plasmid using an HIV LTR to direct expression. For therapeutic use in a mammal the rep gene might be introduced into said mammal in different ways either by using various virus vector systems or other delivery methods, such as liposomes, or by introduction of the gene into cells which are transplanted into the mammal. One skilled in the art would recognize that a cell may be transfected with the recombinant plasmid according to the instant invention using various conventional transfer means, such as liposomes or viral vectors, e.g., vaccinia, SV-40, retroviral vectors, etc. Further, the Rep protein may be administered directly to a mammal or to said mammal's cells using a pharmaceutically acceptable carrier or a conventional means of administration known in the art, such as liposomes. It is also an obvious further development of the invention that it could inhibit productive infections of other classes of human retroviruses such as HIV2, HTLVI and HTLVII. In the present embodiment of the invention we used the rep gene of AAV serotype 2. Another obvious development of the invention is that the rep gene or Rep protein from other AAV serotypes or species could be used in an analogous way to inhibit production of HIV or related viruses.

We have illustrated one embodiment of the invention in which expression of Rep protein is inducible by use of an inducible promoter, in this case the HIV LTR which has the novel features of placing expression of Rep under direct induction by HIV gene products. It is obvious that other inducible expression systems could be used to increase the intracellular level of Rep proteins in HIV infected cells. Very recently Ventura et al., 1990 (Proc. Natl. Acad. Sci. USA 8 : 1310-1314) described a mutant adenovirus gene product that may inhibit expression from an HIV LTR transcription promoter and may have a modest inhibitory effect on production of HIV from an infectious HIV plasmid. This differs significantly from our invention in several ways. First, the gene product described by Ventura et al. was a product of a mutant Ela gene from human adenovirus, a member of the family Adenoviridae. Our invention describes use «_ of the wild- type gene product, rep, from adeno-associated virus which is a parvovirus and thus a member of the family Parvoviridae which is a completely different family with no relation to Adenoviridae. Second, the normal adenovirus Ela gene product activates HIV. The mutant Ela gene product described by Ventura et al. differs from the wild-type Ela gene by virtue of a single point mutation. As such the mutant Ela gene that may inhibit HIV could readily mutate (i.e., revert) to a wild-type Ela gene that would then activate HIV. Our invention uses a parvovirus gene, rep, with no homology to HIV and because we use the wild-type rep gene coding sequence as the inhibitory gene it cannot revert to a form that activates HIV. Third, our invention is much more efficient at inhibiting HIV as measured by production of infectious HIV from an HIV plasmid transfected into HeLa cells. As measured by extracellular reverse transcriptase activity (Table I) or p24 antigen assay (Fig. 8) our invention inhibited HIV by 59 to 78 fold (that is to less than 1.5% of the control level). In similar experiments in HeLa cells, the mutant adenovirus gene of Ventura et al. only inhibited extracellular p24 antigen at best by 3 fold (i.e. to 33% of the control). Therefore, our invention is much more effective and does not suffer from the disadvantages of reversion to wild- type as noted above for the invention of Ventura et al.

It is understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application and the scope of the appended claims.

O >

"293 cell cultures were transfected with 1 μg of pBennCAT and the indicated amounts of pARtat (lμg or 0.1 μg) and varying amounts of a competing plasmid (pJDT269, pAV2, pHIVrep, pHIVrep^", pHIVrepam or pHIVLTR). ^bThe results of CAT activity in the presence of individual competing plasmids are ex ressed as a ercenta e of the amount in the resence of the control re " lasmid

12 TABLE II INHIBITION OF HIV REVERSE TRANSCRIPTASE

ACTIVITY IN HELA CELLS BY REP

PNL432 pHIVrep pHIVrepam

MOCK 20.ug 3.0μg lO.Oug 3.0μg 10. Qua

0 1086 246 19 1670 605 0 832 462 11 1609 609

Hela cells (10⁶ cells per 25cm² T-flask) were transfected by the calcium phosphate method and glycerol-shocked. Twenty micrograms of the HIV proviral clone (pNL432) was used throughout the experiment, except in the mock transfections. The amounts of pHIVrep of pHIVrepam used are indicated in the table. All transfection mixes were adjusted to the same amount of DNA (30 μg) using a carrier plasmid (pUC19) . Supernatant from each sample was collected 48 hours after transfection and assayed for reverse transcriptase activity using the assay of (Willey et al., 1987 J. Virol .62.:139) to measure incorporation of ³²P-TTP. Aliquots (lOμl) of the reaction mix were applied to DEAE paper and the amount of radioactivity bound determined by autoradiography and subsequent liquid scintillation counting. Results are expressed as DEAE bound counts/min/10 μl reaction (equivalent to 1.67 μl of culture supernatant) . Radioactivity (62 counts/min) from mock transfected controls (pUC19) alone was subtracted. The results of duplicate transfections are shown.

Claims

23.We claim:

1. A recombinant plasmid comprising: a) a transcription promoter inducible by gene products expressed by a human lentivirus and b) a gene non-homologous to said lentivirus according to a) , said gene encoding a gene product, said gene product inhibiting the growth of said lentivirus or the production of a lentiviral protein by interfering with the trans - acting gene regulation system of said lentivirus, the expression of said non- homologous gene being increased by the transcription promoter according to a) .

2. A recombinant plasmid according to claim 1 wherein said human lentivirus is HIV.

3. A recombinant plasmid according to claim 2 wherein said human lentivirus is HIV-1.

4. A recombinant plasmid according to claim 1 wherein said non-homologous gene according to b) is from a virus family.

5. A recombinant plasmid according to claim 4 wherein said non-homologous gene is from the virus family Parvoviridae.

6. A recombinant plasmid according to claim 5 wherein said non-homologous gene is from adeno-associated virus.

7. A recombinant plasmid according to claim 6 wherein said non-homologous gene is from adeno-associated virus serotype 2.

8. A recombinant plasmid according to claim 7 wherein said non-homologous gene is the adeno-associated virus rep gene.

9. A recombinant plasmid according to claim 8 which is pHIVrep.

10. A recombinant plasmid according to claim 1 wherein said transcription promoter is the HIV long terminal repeat region (HIV-LTR) .

11. A prokaryotic cell transfected with the recombinant plasmid according to claim 1.

12. A eukaryotic cell transfected with the recombinant plasmid according to claim 1.

13. Use of the recombinant plasmid of claim 1 for the inhibition of the growth of a human lentivirus or the production of lentiviral protein in a mammal, wherein the cells of said mammal infected with said lentivirus are transfected.

14. Use according to claim 13 wherein said human lentivirus is HIV.

15. Use according to claim 14 wherein said human lentivirus is HIV-1.

16. Use of adeno-associated virus Rep protein for the inhibition of the growth of human lentivirus or the production of lentiviral protein in a mammal, wherein said Rep protein is administered to said mammal in a pharmaceutically acceptable carrier.

17. Use according to claim 16 wherein said human lentivirus is HIV.

18. Use according to claim 17 wherein said human lentivirus is HIV-1.

19. Use according to claim 16 wherein said pharmaceutically acceptable carrier is liposomes.