WO1994001551A1

WO1994001551A1 - Method of inhibiting viral replication

Info

Publication number: WO1994001551A1
Application number: PCT/US1993/006380
Authority: WO
Inventors: Julianna Lisziewcz; Daisy M. S. Sun
Original assignee: United States Of America, Represented By The Secretary, Department Of Health And Human Services
Priority date: 1992-07-02
Filing date: 1993-07-02
Publication date: 1994-01-20
Also published as: CA2139339A1; EP0649466A1; AU4666493A; AU678980B2

Abstract

The present invention relates to a method for selecting drugs suitable for use in inhibiting viral replication in vivo. The invention also relates to a method of inhibiting viral replication that involves the use of oligonucleotides complementary to specific regions of the genome of the target virus.

Description

METHOD OF INHIBITING VIRAL REPLICATION

TECHNICAL FIELD

BACKGROUND

One rationale for antiviral chemotherapy is based on the use of antisense oligonucleotides to specifically inhibit viral expression (Agra al, in Prospects for Antisense Nucleic Acid Therapy for Cancer and AIDS, E. Wickstrom, Ed. (Liss, NY, 1991), pp. 145-158; Matsukura et al in Prospects for Antisense Nucleic Acid Therapy for Cancer and AIDS, E. Wickstrom, Ed. (Liss, N.Y., 1991), pp. 159-178). It has been demonstrated that phosphodiester oligonucleotides, complementary to HIV RNA, inhibit viral replication in early infected cells (Goodchild et al Proc. Natl. Acad. Sci. (USA) 85:5507 (1988); Zamecnik et al Proc. Natl . Acad. Sci (USA) 83:4143 (1986)) but not in chronically infected cells (Agrawal et al Proc . Natl . Acad. Sci . (USA) 86:7790 (1989)), mainly because of their nuclease susceptibility (Wickstrom, J. Biochem . Biophys . Methods 13:97 (1986)). Therefore, several chemically modified nuclease resistant analogs were developed and studied for their effectiveness in inhibition of HIV replication in tissue culture (Sarin et al Proc. Natl . Acad. Sci . (USA) 85:7748 (1988); Zaia et al J. Virology 62:3914 (1988); Agrawal et al Proc. Natl . Acad. Sci . (USA) , 85:7079 (1988)). Phosphorothioate modified oligomers inhibited HIV replication in both acute infection (when virus is added to an uninfected susceptible cell line) as well as in chronically infected cells (Agrawal et al Proc. Natl . Acad. Sci . (USA) 86:7790 (1989) ; Agrawal in Advanced Drug Delivery Reviews R. J. Juliano, Eds. (Elsevier, Amsterdam, 1991); Matsukura et al Proc. Natl . Acad. Sci . (USA) 84:7706 (1987); Matsukura et al, Proc. Natl . Acad. Sci . (USA) 86:4244 (1989); Agrawal et al in Gene Regulation: Biology of Antisense RNA and DNA R. P. Erickson, J. G. Izant, Eds. (Raven Press, New York, 1992); Vickers et al Nucleic Acids Res . 19:3359 (1991); Kim et al Biochem . & Biophys . Res . Comm . 179:1614 (1991)). Interpretation of these results, however, has been hampered by the fact that even mismatched oligomers have inhibitory activity. At low concentrations, the control oligomers were less effective than the complementary oligomers, but specificity and cytotoxicity remain to be resolved. Studies on the mechanism and efficiency of antisense oligonucleotides in inhibiting viral replication can most effectively be approached in an in vitro system that mimics the process of viral infection in vivo. It is such studies that have led to the present invention. SUMMARY OF THE INVENTION

The present invention relates to a cell culture system that can be used to evaluate the in vivo efficacy of anti-viral drug treatment. In addition, the invention relates to a method of inhibiting viral replication that involves the sequential treatment of virally infected cells with antisense oligonucleotides complementary to different regions of the viral genome. Objects and advantages of the present invention will be clear to one skilled in the art from a reading of the description that follows.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGURE 1 Sequences of antisense oligonucleotide phosphorothioates and the targeted genes in the HIV-1 genome (Ratner et al Nature 313:277 (1985)). Phosphorothioate-modified oligonucleotides were synthesized using H- phosphorate chemistry on an automated synthesizer (Millipore 8700, Bredford) on 5-10 mmole scale. After the assembly of the required sequence, the CPG-bound oligonucleotide H-phosphorate was oxidized with sulphur in pyridine/triethylamine/ carbondisulfide to generate phosphorothioate linkage. The deprotection was carried out in concentrated ammonia at 40°C for 48 hours. Purification was carried out by preparative reversed phase chromatography followed by ion exchange chromatography. Finally, purified oligonucleotides were dialyzed against water and lyophilized. Oligonucleotide phosphorothioates were checked for their purity by HPLC and PAGE (Agrawal et al Proc. Natl . Acad . Sci . USA 86:7790 (1989)).

FIGURE 2 Antiviral activity of the oligonucleotide phosphorothioates. FIG. 2A.

Antiviral activity of the oligomers in short term assay. Molt cells (5xl0⁵/ml) were infected with HIV-1/IIIB. After 2 hours, the cells were washed and treated with oligomers at lμM concentration. Cells were incubated 4 days then virus expression was measured. FIG. 2B. Inhibition of HIV replication by antisense oligomers. In this short term assay, sequence specific inhibition of HIV replication was not apparent. It is one system that was used in earlier studies. Cells were split to 5x10 /ml. FIG. 2C. shows sequence specific inhibition of HIV replication using the same oligomers in the long term assay system of the invention.

FIGURE 3 Cultures set up as described in FIG. 2C were split to 5xl0⁵/ml and treated with oligonucleotides at lμM concentration every 3 or 4 days. FIG. 3A. Escape of the "SA" oligomer compared to the "Random*¹, "Mismatch" and the control (no drug treatment) . FIG. 3B. Differences in antiviral activity between the complementary oligonucleotides. FIG. 3C. Cytotoxic effect of the "C28" oligomer. Cell viability was determined by trypan blue exclusion. FIGURE 4 Sequential treatment of HIV-i infected cells with different antisense oligomers compared to repeated treatment with a single oligomer. Cells were split to 5xl0⁵ and treated with oligonucleotides at lμM concentration every 3 or 4 days. "Rotate" indicates, that at every treatment a different oligomer was added to the culture at lμM concentration. The sequence of treatment was "RRE", "rev-2", "rev-1" and "SA", sequentially.

FIGURE 5 Twentymer "SA" oligonucleotide treatment compared with 20 mer "Random" as control. At day 12, some specificity is observed but the virus escapes at the same time as the control.

FIGURE 6A Specificity of "SA" and gag (24 and 25 mer). At days 14 and 18, a sequence specific block is observed. FIG. 6B. Gag and "SA" escape, gag somewhat later than SA.

FIGURE 7 Efficacy of the antisense phosphorothioates on primary T cells. Two viruses were studied. FIG. 7A. Primary isolate of HIV-1 labeled as HIV-571 and; FIG. 7B. HIV-1/IIIB, which is a laboratory isolate.

FIGURE 8A. HIV-1 replication after long term drug treatment. After 35 days of treatment of the cells with lμM of different oligomers, an aliquot of the cells was split and carried in the absence of drug. FIG. 8B. Long term (59 day) inhibition of HIV-1 replication using oligonucleotides at lμM dose. FIG 8C. HIV-l replication at lower dose (O.lμM) of antisense treatment. Cells were treated with lμM antisense phosphorothioates for 39 days, then an aliquot of the cells was split and treated several weeks with O.lμM of the same oligomers.

"Rotate" treatment is described in FIG. 4. FIG. 8D. HIV-l replication at lower dose (O.OlμM) of antisense treatment. Cells were treated with lμM of antisense phosphorothioates 39 days, then O.lμM antisense phosphorothioates an additional 14 days (see FIG. 8C) , then an aliquot of the cells was split and treated several weeks with O.OlμM of the same oligomers. "Rotate" treatment is described in FIG. 4.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a novel culture system that simulates in vivo conditions of viral infection, particularly, retroviral infection, most particularly, HIV-l infection. The culture system can be used to evaluate the long-term efficapy of anti-viral drug treatment, for example, antisense oligonucleotide treatment. The invention further relates to a method of reducing the viral burden in an infected individual. The method involves the sequential treatment of virally infected cells with a combination of different antisense oligonucleotides. The method has the advantage that it prevents the formation of escape mutants of the target virus.

In HIV infected individuals, only a small percentage of the CD4+ cells are infected and producing the virus; the majority of the cells are uninfected and CD4+. In vitro chronically infected cells do not mimic in vivo conditions because they are CD4-, consequently, reinfection cannot occur. A better model for drug studies is an acute, low multiplicity of infection (MOI) , where only some fraction of the cell population harbors virus and other cells are uninfected and CD4+. The present invention relates, in one embodiment, to a cell culture system that mimics these in vivo conditions. A preferred embodiment of the culture system of the invention is described in detail in the Examples that follow. This system extends the treatment period over weeks rather than days and thus permits simulation of a treatment schedule that can be given to a virally infected patient. The culture system is particularly well adapted to drug screening where the target virus is characterized by a high mutation rate (for example, HIV) . In general terms, the system of the invention can be described as follows. Susceptible cells, advantageously primary cells (for example, T cells or acrophage when HIV is the target virus) are contacted with the target virus under conditions such that infection can occur. Virus particles that do not associate with the cells are removed by washing. The amount of virus used in this initial step is, advantageously, selected so as to mimic the conditions of in vivo infection (e.g., low MOI in the case of HIV) . Subsequent to infection, treatment with the test drug (for exeunple, antisense oligonucleotide, or stabilized derivative thereof, as described below) is initiated. The treatment conditions (including culture cell number and density) are monitored and appropriate test drug levels are maintained. In the case of replicating cells, splits are made (for example, every 2-4 days) so as to maintain a healthy, replicating cell population. The treatment time and time subsequent to treatment are selected so as to ensure sufficient opportunity for escape mutants to be produced. The culture system to which the invention relates can be used to test drugs for their ability to inhibit HIV replication, as well as Simian Immunodeficiency Virus (SIV) HTLV-1 or DNA viruses such as HSV, HHVG, EBV, hepatitis. The cells are selected on the basis of their susceptibility to infection by the target virus. Examples of cell/virus combinations include T cell: HIV, SIV, HTLV; B cells: EBV; neurons: HSV. Advantageously, the cells are primary cells derived from the patient to be treated. The use of patient-derived cells permits customization of the treatment regimen.

Drugs that can be tested in the system to which the invention relates include antisense oligonμcleotides complementary to specific regions of the genome of the target virus. Examples of other types of drugs that can be tested in this system include RT inhibitors (eg, AZT and DDI) , and protease inhibitors for HIV. To ensure sufficient stability for in vivo use, the oligonucleotides used in this system are chemically modified. For example, one of the chemically modified forms of oligonucleotides used is phosphorothioates (see also Agrawal et al Proc. Natl . Acad . Sci . USA 85:7079 (1988)). The oligonucleotides are also designed so as to be of an optimum length (see, for example. Figures 5 and 6, and Brief Description thereof, as compared to data presented below relating to the gag 28 mer) . Oligonucleotides approximately 28 bases long have been found to be advantageous in the case of HIV-l treatment. Such properly targeted oligonucleotides can prevent the formation of escape mutants. Factors that must be taken into account in determining optimum length include stability, selectivity, and ease of administration of the oligonucleotide.

Oligonucleotides for antisense inhibition are, advantageously, directed against mRNAs of overlapping regulatory genes. Preferred target sequences are highly conserved between different viral isolates to minimize the change for virus escape. This factor is particularly important when HIV is the target virus since there are differences between isolates even within one patient. However, less conserved sequences may also prove to be effective on an individual isolate basis if antisense oligonucleotides are custom designed against primary HIV isolates from individual patients. Targeting non-coding sequences (like "SA" in the case of HIV-l) may allow the virus to escape, even though virus replication may be reduced compared to the controls. In the case of HIV, another effective target sequence is expected to be a functional RNA, for example, RRE (see Chin J. Virol . 66:600 (1992)). Both RRE and rev-1 are effective inhibitors of HIV-l replication. However, it was found in the long term study that rev-l inhibits approximately ten fold better than RRE. Since multimerization of the Rev protein is required for Rev function, a critical amount of Rev must escape from antisense inhibition for virus production. However, if only a single RRE escapes from the antisense block. Rev might be able to recognize RRE and activate virus expression.

An oligomer directed against the gag gene can be expected inhibit HIV replication efficiently and without the generation of escape mutants. It has been previously demonstrated that sequences essential for packaging, situated around the gag initiation codon, form a stable secondary structure (Harrison et al in RNA Tumor Viruses, J. Coffin, I. Chen Eds. (Cold Spring Harbor Press, Cold Spring Harbor, 1991) , p. 235) . The gag oligomer may disrupt these structures, inhibiting viral packaging, in addition to the translation of the gag mRNA.

HIV has a high mutational rate and, therefore, drugs designed to treat HIV infection may induce the formation of escape mutants. To overcome this problem, combination chemotherapy has been suggested for treatment of HIV-infected patients. This therapy involves different drugs directed against different targets, such as reverse transcriptase (RT) inhibitors combined with protease inhibitors.

Antisense treatment of HIV-infected individuals, even when a highly conserved region is targeted, may result in formation of escape mutants. Targeting different sequences either in combination or sequential treatment schedules can be expected to result in different selection pressures on the virus with little time to develop escape mutants. The first treatments can, for example, consist of a mixture of oligomers or one targeted to a highly conserved region followed by sequential administration of alternatively-targeted oligomers. The beneficial effects of sequential treatment using antisense oligonucleotides directed against different sequences of the viral RNA are exemplified below. These results underscore the usefulness of a combination of antisense oligonucleotides for treatment of HIV-infected patients.

The oligonucleotides to which the invention relates can be administered, for example, by i.v. injection. Suitable in vivo doses can be extrapolated from in vitro results obtained using the culture system of the invention. One skilled in the art will appreciate that the amounts to be administered will vary with the drug (target sequence in the case of antisense oligonucleotides or chemical modification thereof) and the patient status (including the viral burden of the patient) . In addition, it may be advantageous to vary dose levels during the treatment regimen.

Certain aspects of the invention are described in greater detail in the non-limiting Examples that follows.

Example I

The effect of antisense oligonucleotides on HIV replication was tested in a CD4+ cell line (Molt3) infected with low MOI of HIV-1/IIIB. After 2 hours of infection, the cells were washed and treated with oligonucleotides. The cultures were split every 3 or 4 days and retreated with the oligonucleotides. Virus replication was monitored at the cellular level by syncytia formation and p24 membrane expression (Sarin et al Proc . Natl . Acad. Sci . (USA) 85:7748 (1988)), and in the supernatants by p24 antigen capture assay (DuPont) and RT assay (Sarin et al JNCI 78:663 (1987)). The initial infection is characterized by 3% positivity for p24 membrane expression after 4 days in untreated cells. Therefore, the majority of cells remain CD4+ and thus susceptible to reinfection during the treatment period.

To study the inhibition of HIV-l replication, twenty eight base long oligonucleotide phosphorothioates were chosen because it has been shown that the specificity is increased by the length of the oligomers used at equivalent concentrations (Agrawal et al in Prospects for Antisense Nucleic Acid Therapy for Cancer and AIDS E. Wickstrom, Eds. (Wiley Liss, New York (1991) , pp. 145-158) . Five different target sequences (FIG. 1) were selected in the HIV-l genome. Since tat and rev are able to trans-activate HIV gene expression and are essential for the virus replication, two oligomers, "rev-1" (Matsukura et al Proc. Natl . Acad. Sci . (USA) 86:4244 (1989)) and "rev-2", complementary to these overlapping reading frames, were evaluated. Rev regulates HIV gene RNA expression though interaction with the rev-response element (RRE) . Therefore, the "RRE" oligomer was directed against the rev-response element overlapping the known rev binding site (Daefler et al Proc. Natl . Acad. Sci . (USA) 87:4571 (1990); Malim et al Cell 60:675 (1990)). This "RRE" oligomer was chosen after screening a number of potential anti-RRE oligonucleotides. In addition to blocking the rev-RRE interaction , "RRE" could potentially interfere with the translation of the viral envelop gene. Another oligomer, called "gag" was directed against the mRNA of that structural gene. Oligonucleotide "SA" is complementary to the major splice acceptor site on the first coding exon of the tat gene (Arya et al Science 2298:69 (1985)), which is also located in the open reading frame of the non-essential vpr gene. To determine the specificity of the antisense oligomers, the biological effect has to be compared to the same size oligomer which is not complementary to any cellular or viral gene. Three control non-specific oligonucleotide were chosen: "Random" sequence was synthesized as a degenerate oligonucleotide, by coupling a mixture of four nucleotides at each stage (theoretically, it contains 4 = 7.2xl0¹⁶ sequences) to investigate the extent of sequence non-specific inhibition; "Mismatch" has the "SA" sequence with five, bases altered (Figure 1 underlined) serve to control "SA" specific inhibition; and "C28" is a phosphorothioate oligodeoxycytidine, S-(dC)28« ^τ**^e latter homo-oligomer has been shown to have a significant antiviral effect. (Matsukura et al Proc. Natl . Acad. Sci . (USA) 86:4244 (1989); Agrawal et al. Proc. Natl . Acad. Sci . USA 85:7079 (1988); Gao et al J. Biol . Chem . 265:20172 (1990)).

The amount of drug used for long term studies has been determined based on the minimal concentration which inhibited viral infection >60% in a 4 day study. Short term inhibition by different oligomers (Fig. 2A) demonstrated that all compounds inhibited HIV replication grater than 60% compared with the positive control (cells infected in the absence of drug) . There were no more than 26% differences in antiviral activity among the complementary and control oligomers in the short term assay.

Twenty five days after infection, high levels of virus replication were detected in cultures treated with the "Random" and the "Mismatch" oligomers demonstrating that the control oligomers failed to inhibit HIV-l replication (Fig. 2B and Fig. 3A) . Since these controls were at the peak of the acute phase of the invention at day 25, this time was chosen to evaluate the specificity of the inhibition by the antisense (complementary) oligonucleotide. Inhibition of HIV-l replication by all complementary oligonucleotides was greater than 99.8% (detected by a quantitative p24 ELISA assay) compared to the "Random". Less than 1% of cells treated with the sequence specific antisense oligomers expressed detectable p24 antigen on the surface and only 3% of the "SA" treated cells showed syncytia formation (Fig. 2B) . In these cultures, the number of viable cells were similar to the uninfected control suggesting that the long-term oligonucleotide phosphorothioate treatment is not toxic to cells. The acute phase of an infection is characterized by extensive syncytia formation

(>75%) , membrane antigen expression (>40%) and high virus levels in the supernatants (p24>l μg/million cells) . Cells which survive enter into a chronically infected phase characterized by the absence of CD4 receptor at the cell surface. Since CD4 is the main receptor for HIV-l infection of T cells, reinfection does not occur and production of virus declines.

At day 25, an aliquot of the cells were split without addition of the antisense oligomers. After four passage in the absence of drugs, all cultures showed acute infection as described above. FIG. 2C demonstrates that recovery of virus replication was delayed depending on the oligomers used for treatment during the first 25 days. At day 32, the "SA" treated cells showed maximal replication, which might be expected considering that "SA" does not inhibit the expression of any essential gene. At day 35, "gag" and "RRE" treated cells converted to an acute phase of infection consistent with specific inhibition at the late phase of the virus replication. The "rev-2" followed by "rev-1" were the last cells to progress into an acute phase of infection. These oligomers presumably inhibit the early phase of virus replication by interfering with the activity of regulatory genes.

The test system used for these experiments allowed time for the virus to generate escape mutants, a phenomenon not previously appreciated in short term studies. FIG. 3A shows that at 32 days after infection, a population of virus escaped from the inhibitory effect of the "SA" oligomer. At this time, 66% of this culture showed syncytia and 20% of cells expressed the p24 membrane antigen, although this was still significantly less virus compared to cultures during the acute infection (see "Random" and "Mismatch" at FIGURE 2B) . Inhibition of HIV replication in cultures treated by "rev-1", "rev-2", "gag" and "RRE" oligomers were greater than 98% (detected by p24 antigen capture assay of the supernatants) , compared to the "SA" treated one

(FIG. 3B) . The RT assay is not quantitative at the lower range. However, the RT results supported the p24. These results indicate that there was an escape from the inhibitory effect of "SA" that was not apparent early in treatment.

Expression of the Tat protein can result from alternatively spliced messages utilizing an upstream splice acceptor for the second exon (Arya et al Science 229:69 (1985). A shift in utilization of an alternative splice site or a mutation in the targeted sequences might have occurred during the "SA" treatment. Since HIV-1/IIIB contains a defective vpr gene, it is not known whether "SA" oligomer alters the function of this protein. Interestingly, "rev-1" oligomer inhibited 95% better than "rev-2" (see the quantitative results on the p24 antigen capture assay, which is supported by syncytia, RT and membrane expression) , even though they contain 22 nucleotides of overlapping sequence. This result indicates that minor shifts in the targeted sequences can cause variances in the efficacy of the oligomers. Differences in antiviral activities of the complementary oligomers were detected at day 32 with the most efficient oligomers after long-term treatment being the "rev-1", "gag", followed by "RRE" and "rev-2". These differences were maintained as long as the cultures were monitored. In the system used, high levels of viral replication were detected 11 days later in cultures treated with the control oligomers than the HIV-l infected but untreated control (FIG. 3A) . This result suggests that oligonucleotide phosphorothioates have a sequence independent antiviral effect, which not inhibits but delays virus replication. The simultaneous addition of virus and high concentrations of complementary or control oligomers (4μg/ml; 5μM) resulted in no apparent viral replication suggesting inhibition at the level of viral entry, reverse transcriptase or integration. (HIV-l infection of Molt3 cells was carried out by simultaneous addition of virus and oligonucleotide phosphorothioates (different length and sequence) . The treatments were repeated 3 times in 11 days when the cells were split, then the treatment was stopped and the cultures were carried seven more weeks in the absence of drugs. Cells infected in the absence of drug (positive control) showed high level of p24 membrane expression (one week after infection 8%, eleven days after infection 60%). During the time of the experiment (9 weeks long) neither virus replication nor any toxicity could be detected in the cultures treated by the different oligonucleotide phosphorothioates suggesting the virus was never integrated.)

A phosphorothioate oligodeoxycytidine ("C28") was previously reported to inhibit HIV-l and herpes simplex virus type 2 replication. The culture treated with the "C28" homo-oligomer produced low amounts of virus throughout the study period, similar to the culture treated with the "rev-1" oligomer. Early on, there were no apparent differences in cell growth rates when compared to cells treated with the complementary oligomers. After 25 days when the treatment was dropped, the virus production in "C28" paralleled the "rev-1". Thirty nine days after infection, the culture treated with the "C28" oligomer began to show evidence of toxicity. At this time, the same number of cells was split and treated with lμM, 0.5μM and O.lμM "C28". FIG. 3C demonstrates that after one passage, less than 5% of cells were viable under the treatment of lμM and 0.5μM "C28". However, cells treated with O.lμM "C28" recovered. After the second passage, there were no viable cells in the cultures treated with lμM and 0.5μM "C28". The O.lμM treated cells remained healthy. The persistent anti-viral effect of "C28" may not reflect a sequence specific inhibition of HIV gene expression, but may be due to a cytotoxic reaction.

Example II

Tp minimize the possibility of generating escape mutants in vivo, combination treatment using antisense oligonucleotides directed against different sequences of the viral RNA was evaluated. The sequential treatment was evaluated with "RRE", "rev-2", "rev-1" and "SA", designated as "Rotate" in FIGURE 4. During this study, "SA" treated virus escaped after 32 days and "rev-2" treated cells showed a higher level of virus replication compared to the "rev-1" or "RRE" treated cells. Sequential treatment kept the virus replication at the level of the "rev-1" treated cells which, when used alone, gave the best anti¬ viral activity between the studied oligomers. This experiment indicates, that even the drug "SA", which alone is ineffective in that escape mutants are quickly generated, can be effective in combination with other oligomers.

Example III

Primary T cells are the main target of HIV-l infection. In the present study, two different HIV- 1 isolates were evaluated for antisense inhibition. Fig. 7A shows sequence specific inhibition of HIV-l- 571 replication. HIV-1-571 is a primary virus isolate (gift from Dr. Paulo Lusso) and does not grow in established T cells lines. All three antisense oligonucleotides that were effective in the long term assay system were effective with this isolate.

Fig. 7B shows the results of a control study on primary T cells using the established laboratory isolate HIV-1/IIIB. This isolate was used for the study on the Molt3 cell line. In both cases, it was possible to show sequence specific inhibition of viral replication.

Example IV

Molt 3 cells were treated with antisense oligonucleotides as defined in Figure 8. The results shown in FIG. 8A demonstrate that even after long term treatment with the oligonucleotide (lμM) , replicating virus is present in all cultures. Virus replication, however, is suppressed by antisense inhibition at a concentration of lμM (FIG. 8B) . The results shown in FIGS. 8C and 8D demonstrate that after initial treatment at lμM, the effective dose can be reduced, depending on viral burden and the antisense target. The implications of this result on combination therapy are that the dose of, for example, "rev-1" or "gag" can be reduced without inducing viral replication. The concentration of "RRE", however, cannot be reduced.

* * *

All references cited hereinabove are hereby incorporated by reference. One skilled in the art will appreciate from a reading of this disclosure that various changes in form and detail can be made without departing from the true scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. A method of preventing the formation of escape mutants in cells infected with a retrovirus comprising contacting said cells with at least two antisense oligonucleotides complementary to different regions of the retrovirus genome, wherein said oligonucleotides are sequentially contacted with said cells under conditions such that replication of said retrovirus is inhibited whereby a first of said at least two oligonucleotides is contacted first with said cells and subsequently a second of said at least two oligonucleotides is contacted with said cells.

2. The method according to claim 1 wherein at least one of said oligonucleotides is complementary to a conserved region of the retrovirus genome.

3. The method according to claim 1 wherein at least one of said oligonucleotides is complementary to rev mRNA.

4. The method according to claim 1 wherein at least one of said oligonucleotides is complementary to gag coding sequence and packaging signal.

5. The method according to claim 1 wherein at least one of said oligonucleotides is complementary to RRE.

6. The method according to claim 1 wherein said oligonucleotide is a chemically modified.

7. The method according to claim 1 wherein said oligonucleotide is about 28 bases long.

8. The method according to claim l wherein said retrovirus is HIV.

9. An in vitro culture system for testing an anti-viral agent comprising: i) contacting a culture of cells susceptible to infection with a virus to said virus under conditions such that infection of said cells is effected to the extent that infection with said virus occurs in vivo; ii) contacting said infected cells resulting from step (i) with different concentrations of an agent to be tested; iii) culturing said infected cells for a time period that corresponds to an in vivo treatment protocol for said virus and determining which of said concentrations inhibits replication of said virus throughout said time period.

.

10. The method according to claim 9 wherein said virus is HIV and said anti-viral agent is an antisense oligonucleotide complementary to a functional region of the HIV genome.

11. The method according to claim 10 wherein said period of time of step (iii) is a period sufficient for mutants to develop.