CA2050473A1 - Inhibition of hiv using synergistic combinations of nucleoside derivatives - Google Patents

Abstract

ABSTRACT

The present invention relates to the use of synergistic combinations of nucleoside derivatives for inhibiting human immunodeficiency virus (HIV) replica-tion, thereby limiting HIV infection. In a particular embodiment, the purine nucleoside analogue dideoxyino-sine combined with the pyrimidine nucleoside analogue 2',3'-dideoxy-2',3'-didehydrothymidine (d4T) exhibit strong synergistic activity and diminished cytotoxic activity toward mammalian cells.

Description

2g35~73 WO90/11081 1 PCT/~S~0/01424 INHIBITION OF HIV USING SYNERGISTIC
COMBINATIONS OF NUCLEOSIDE DERIVATIVES

l. INTRODUCTION
The present invention relates to the use of combinations of nucleoside derivatives for inhibiting human immunodeficiency virus (HIV) replication, thereby limiting the effects of HI~l infection. The crux of the invention lies in the discovery that nucleoside derivatives used in combination have synergistic effects, such that the combined effective dose of these agents is lower than the sum of the therapeutic dosages of either drug used individuallyi increased anti-viral activity is not associated with a commensurate increase in cytotoxicity; in fact, combinations of nucleoside derivatives are less toxic to uninfected cells when compared to identical compounds administered separately.

2. BACKGROUND OP THE INVENTION

2.l. HUMAN IMMU~ODEFICIENCY VIRUS
Human immunodeficiency virus (HIV) is a human retrovirus believed to be the causative aqent of acquired immune deficiency syndrome (AIDS~ and AIDS related complex (ARC). The HIV virion or virus particle is a sphere that is roughly l000 angstrom units across. The particle is covered by a lipid bilayer membrane derived from the outer membrane of the infected host cell. Studding the viral membrane is an envelope glycoprotein which is synthesized as a precursor of 160 Kd and subsequently processed into two glycoprotein~: gp41 which spans the lipid bilayer, and gpl20 which extends beyond the lipid bilayer. The envelope covers a core made up of proteins designated p24 and pl8. The viral RMA is carried in the core, along with several copies of the enzyme, reverse transcriptase, which catalyzes the assembly of viral DNA.
The HIV genome contains three genes that encode the components o~ retrovirus particles: env (which codes 2 ~ 7 3 WO90/~1081 PCT/~S90/01~2 for the envelope proteins), gaq (which codes for the core proteins), and ~ (which codes for reverse transcriptase). These three genes are flanked by stretches of nucleotides called long terminal repeats (LTRs). The LTRs include sequences that have a role in controlling the expression of viral genes. However, unlike other retroviruses, th~ genome of HIV includes at least five additional genes, three of which have known regulatory functionsO and the expression of which is thought to have an impact on the pathogenic mechanisms exerted by the virus. The tat gene encodes a protein that functions as a potent trans-activator of HIV gene expression, and, therefore, plays an important role in the amplification of virus replication. The rev, or trs/art gene can upregulate HIV synthesis by a transacting antirepression mechanism; rev enables the integrated HIV
virus to selectively produce either regulatory proteins or virion components. In contrast, the nef, or 3'-orf, gene appears to down-regulate virus expression by producing a cytoplasmic protein which, presumably via a second messenger, inhibits transcription of the HIV genome. The vif, or sor gene is not essential for virion formation, but is critical to the efficient generation of infectious virions and influences virus transmission in vitro. The pr, or R gene encodes an immunogenic protein of unknown function.
The critical basis for the immunopathogenesis of HIV infection is believed to be the depletion of the helper/inducer subset of T lymphocytes, which express the CD4 antigen, resulting in profound immunosuppression.
Viral killing of these immune cells is thought to be a major factor contributing to the crippling effect HIV has on the immune system. The env~lope glycoprotein plays an important role in the entry of HIV into CD4 positive host cells. The gpl20 portion has been shown to bind directly 2 a ~
WO90/110~1 PCT/~S90/01~2 to the cellula~ CD4 receptor molecule, thereby producing HIV's tropism for host cells that express the CD4 receptor, e.q., T helper cells (T4 cells), macrophages, etc.
After HIV binds to the CD4 molecule, the virus is internalized and uncoated. Once internalized, the genomic RNA is transcribed into DNA by the enzyme reverse transcriptase. The proviral DNA iS then integrated into the host chromosomal DNA and the infection may assume a "dormant" or latent phase. However, once activation occurs, the proviral DNA is transcribed. Translation and post translational processing results in virus assembly and budding of mature virions from the cell surface.
When active replication of virus occurs, the host CD4+ cell is usually killed. However, the precise mechanism by which HIV exerts its cytopathic effect is unknown. A number of mechanisms for the immunopathogenesis and cytopathic effect of ~IV infection have been proposed: the accumulation of large amounts of unintegrated viral DNA in the infected cells; massive increase in permeability of the cell membrane when large amounts of virus bud off the cell surface; speculations that HIV may induce terminal differentiation of infected T4 cells, leading to a shortened life span. There is growing evidence that both the CD4 molecule and the virus envelope play a role in cytopathic effect in HIV infected cells by somehow promoting cell fusion. A prominent feature in the cytopathology of HIV infection is the formation of multinucleated syncytia formed by the fusion of as many as 500 cells which appear to be induced by the gpl20/gp41 envelope proteins. In contrast, HIV-infected macrophages may continue to produce HIV without cytopathic effects for long periods of time; it is believed that the macrophage is a major reservoir for HIV and may be responsible for transporting virus into the central 7,3 wo90/1l081 PCT/~S90/0l4t4 nervous system (Gartner et al., 1986, Science 233:215-21~).
To date, there is no cure for AIDS~ Vaccine trials are currently underway in an attempt to control the spread of the virus among the population. However, efforts at controlling the course of disease within an in~ected patient have been directed mainly towards the use of antiviral agents.

2.2. ANTI-HIV MEDICATIONS
Various therapeutic approaches are currently being explored in efforts to decrease the morbidity and mortality of HIV infection tYarchoan et al., 1938, Scientific American 259:110-119; Bartlett, 1988, J.A.M.A.
260:3051-3052; De Clercq, E., 1986, J~ Med. Chem.
29:1561-1569; Yarchoan, R. and Broder, S., 1987, N. Engl.
J. Med. 316:557-564). Knowledge of the physiology of the HIV virus has resulted in a number of rationales designed to interfere with viral replication or dissemination of infection. Intervention could potentially prevent virus binding to target calls, fusion with cell membranes, uncoating of viral nucleic acid, reverse transcription of the HIV RNA genome into DNA, transcription or translation of viral mRNA, processing Of viral proteins, assembly into mature virions, or other events related to replication or infectivity.
Several methods of preventing the binding of HIV

to cell membranes are being tested; these are directed toward inhibiting interactions between the HIV envelope glycoprotein, gpl20, and CD4 antigen on target cells, such as helper ~ lymphocytes. The binding of gpl20 to CD4 antigen has been shown to be associated not only with viral penetration of cell membranes, but to syncytia formation as well (Sodroski et al., Nature 322:470-474;
Lifson et al., 1986, Nature 323:725-728: Stevenson et al., 2 ~ 7 3 WO90/ll08l pcrl~sso/f 1988, Cell 53:483-496). Smith et al. (1987, Science 238:17~4-1707) has shown that soluble CD4 antigen can block HIV infectivity by binding to viral particles before they encoUnter CD4 molecules embedded in cell membranes.
Alternatively, anti-idiotype antibodies, directed toward anti-CD4 antibodies, have been shown to bind to HIV virus ln vitro, presumably by possessing protein configurations similar to CD4 determinants ~Dalgleish et al., 1989, UCLA
Symposia on Molecular and Cellular Biology. J~ Cell Biochem. Supp. 13B, p. 298). In addition, several large, sulfated, negatively charged molecules, including dextran sulfat~, have been shown to inhibit HIV replication and syncytia formation ln vitro (Yarchoan et al~, supra).
Anti-sense oligonucleotides have been found to t5 inhibit virally i~duced syncytium formation and expression of viral p24 protein (Agrawal et al., 1988, Proc. Natl.
Acad. Sci. U.S.A. 85:7079-7083). These synthetic nucleic acid polymers, complementary to HIV mRNA, were designed to inhibit the translation of viral mRNA into protein;
forming phosphoramidate and phosphorothioate derivatives of anti-sense oligonucleotides has produced compounds substantially resistant to andonuclease degradation.
Ribavirin, comprising a ribose moeity and a triazol ring, iS believed to act as an analogue of the nucleoside guanosine. It has activity a~ainst several RNA
viruses in vitro, primarily, it is believed, by interfering with the guanylation step required for capping of viral mRNA. Ribavirin has heen reported to sUppress the replication of HIV in cultures of human T lymphocytes 3 (Mc~ormick et al., The Lancet, Dec. 15, 1984:1367-1369).
Alternatively, post-translational processing of viral proteins can be disrupted; castanospermine, a plant glycosidase activity, has been shown to reduce synctium formation and HIV infectivity (Wall et al., 1988, Proc.
Natl. Acad. Sci. U.S.A. 85:5644-5~48).

' 2 0 ~ rl 3 woso/11081 PCl/~S90/01~24 The final steps in HIV replication involve i~s exit from the host cell and infection of new cellular targets. Alpha inter~eron has been found to suppress HIV
replication ln vitro (Ho et al., Lancet, March 16, 1985:602-604), may reduce viral buddinq, and has been shown to have direct antitumor activity against Kaposi's sarcoma, a malignancy associated with AIDS. A lipid compound, AL 721, composed of neutral glycerides, phosphatidylcholine, and phosphatidylethanolamine in a 7:2:l ratio has a demonstrated ability to extract chloresterol from cellular membranes, and appears to decrease HIV infectivity (Sarin et al., 1985, N. Engl. J.
Med. 313:1289-1290).
Several methods for controlling HIV infection are being tested which relate to inhibition of viral reverse-transcriptase. Because mammalian cells lack endogenous reverse transcriptase activity, these methods off~r potentially selective inhibition of virus replication with minimal host cell cytotoxicity. Agents which are believed to inhibit reverse transcriptase include phosphonoformate (Sandstrom et al., Lancet June 29, 1985:Ll480), suramin tMitsuya et al., 1984, Science 226:172~174), rifabutin (or ansamycin, Amand et al., Lancet, Jan. ll, I986, p. 97) and nucleoside derivatives, described infra.

2 ~ 3 WO90/l1081 PCT/~S90/0l12 2.2.1. NUCLEOSIDE DERIVATIVES
Nucleoside derivatives are modified forms of the purine (adenosine and guanosine) and pyrimidine (thymidine, uridine and cytidine) nucleosides which are the building blocks of ~NA and DNA. Reverse transcriptase and DNA polymerase will bind and incorporate nucleoside derivatives into the nascent DNA chain provided that the 5' carbon oP the derivative can bind to the 3' hydroxyl group of the previous nucleotide (i.e. by ~orming a phosphodiester linkage); subsequent nucleotides can only be added if the nucleoside derivative contains a means for linking its 3' carbon to the 5' phosphate of another nucleotide. Many of the nucleoside derivatives under t5 study as potential anti-HIV medications result in premature termination of viral DNA replication before the entire viral genome has been transcribed. These derivatives lack 3' substituents that can bind to subsequent nucleosides and result in chain termination.
AZT (3'-azido-2',3'-dideoxythymidine, azidothymidine, zidovudine) is a nucleoside derivative which has been shown to be effective in-the treatment of patients with AIDS and ARC; data indicates that AZT
increases the median survival of AIDS patients (Fischl et al., 1987, N. Engl. J. Med. 317:185-191; Creagh-Kirk, et al., 1988, J.A.M.A. 260:3009-3015). Preliminary evidence suggests that AZT may be efficacious in children with AIDS, HIV associated psoriasis (Bartlett et al., supra), AIDS associated dementia (Schmitt et al., 1988, N. Engl.
J. Med. 319:1573-1578) and HIV associated thrombocytopenia (Pottage et al., 1988, J.A.M.A. 260:3045-3048). AZT is phosphorylated by mammalian cells to form AZT triphosphate (an analogue of thymidine triphosphate) which is believed to inhibit the production of viral DNA by at least two mechanisms: competitive inhibition and chain termination WO90/11081 ~ PCT/~S90/01~2 (Yarchoan et al., supra). Competitive inhibition occurs because viral reverse transcriptase binds AZT triphosphate more tightly than native nucleotides; chain termination results from AZT incorporation because AZT lacks a 3' hydroxyl group, and therefore cannot bind to subsequent nucleotides.
Dideoxynucleosides are nucleoside derivatives which lack hydroxyl groups at both the 2' and 3' carbon residues, and therefore result in DNA chain termination.
In addition, like AZT, the 5' triphosphate products of 2', 3'-dideoxyadenosine, dideoxyguanosine, dideoxycytidine, and dideoxythymidine selectively inhibit viral reverse transcriptase and cellular ~ and ~ DNA polymerase, but do not interfere with the function of DNA polymerase , the major DNA synthetic enzyme utilized during cell division (Edenberg et al., 1978, J. Biol. Chem. 253:3273-3280;
Waqar et al., 1978, Nucl. Acids Res. 5:1933-1946, van der Vilet et al., 1981, Biochemistry 20:2628-2632; Waqar et al., 1984, J. Cell. Physiol. 121:402-408). In vitro, adenosine, guanosine, inosine, cytidine, and thymidine 2', 3' dideoxynucleosides inhibited HIV virus replication (although the thymidine derivative showed less inhibitory activity); essentially complete suppression of HIV virus was obse~rved at doses that wer lower by a factor of 10 to 20 than those needed to inhibit the proliferation or immune reactivity of T cells (Mitsuya and Broder, 1986, Proc. Natl. Acad. Sci. U.S.A. 83O1911-1915). Ahluwalia et al. performed initial studies on the cellular pharmacology of 2',3'-dideoxyinosine (1987, Biochm. Pharm. 6:3797-380). PCT publication W0 87/01284, international publication date March 12, 1987, relates to the inhibition of in vitro infecti~ity and cytopathic e~fects of HIV by purine 2', 3' dideoxynucleosides, including 2', 3' dideoxyinosine, 2', 3'-dideoxyguanosine, and 2', 3'-dideoxyadenosine. European patent application 0273277, woso/11081 PCT/~S90/01~2~
- 9 - 2~ 7.3 publication date July 6, 198~, relates to the use of 3 ~-deoxythvmidin-2'-ene (3'-deoxy-2',3'-didehydrothymidlne d4T), which lacks 2' and 3'hydroxyl groups and has a double bond between 2' and 3' carbon atoms, in treating patients infected with a retrovirus. D4T, also referred to as 2',3'-didehydro-2',3'-dideoxythymidine, or 1-(2,3-dideoxy-~-D-q ~ ero-pent-2-enofuranosyl)thymine, was found to inhibit HIV reverse transcriptase, and had anti-HIV
activity comparable to AZT (~ansuri et al., J. Med. Chem.
in press). D4T may also have less toxic effects than AZT
(Mansuri et al., supra; Ghazzouli et al., 1988, ICAAC, Abstract 1301, p. 344).
Recently, acid stable 2',3'-dideoxy-2'-fluoronucleosides which exhibit antiviral effects have t5 been described. See, European Patent Application, Publication No. 0287313 A2 which describes 2',3'-dideoxy-2'-fluoroadenosine and 2',3'-dideoxy-2'-fluoroinosine and European Pa~ent Application Publication No. 0292023 A2 which describes a fluoro pyrimidine derivative, 2'3'-dideoxy-2'-fluoro-arabinofuranosylcytosine (F-ddC).

2.2.2. ~
Because HIV depends on human cells to supply the machinery necessary for the completion of its life cycle, inhibition of viral replication is frequently associated with cytotoxic effects. AZT therapy is associated with bone marrow toxicity (Richman et al., 1987, N. Engl. J.
Med. 317:192-197) and consequently, decreased production of red blood cells (resulting in anemia), platelets (resulting in thrombocytopenia),and white blood cells (resulting in leukopenia). According to Bartlett et al.
(1988, J.A.M.A. 260:3051-3052), 40 percent of AZT-treated patients ultimately develop anemia, requiring dosage reduction or transfuslon, and leukopenia, with a decreased ~n~73 WO9Q/l108l PCT/~S90/01~2~

number of granulocytes, thereby increasing susceptibility to infe~tion; only about 60 percent of patients with ARC
or AIDS seem to be able to tolerate continued ~ZT use after one year of therapy (Pettinelli, c. B., F~inberg, J., and the AIDS Clinical Trials Group: Safety and tolerance of zidovudine in patients with AIDS and advanced ARC, abstracted; Fischl, M.A., and the AIDS Clinical Trials Groupo The safety and efficacy of two doses of zidovudine in the treatment of patients with AIDS, abstracted; both read before the Fourth International Conference on AIDS, Stockholm, June 15, 1988, and cited by Bartlett).
In order to avoid these toxic effects, AZT is being tested in combination with other antiviral agents, such as acyclovir sodium, foscarnet sodium, interferon-~, ~, or ~, interleukin 2, or ampligen. Other studies, alternating AZT with 2', 3' dideoxycytidine therapy (ddC) are also in progress (Yarchoan et al., 1988, Lancet 1:76-81). Continuous high doses of ddC (which is relatively marrow-sparing) for more than eight to twelve weeks is associated with the development of painful peripheral neuropathy; it is hoped that by alternating AZT and ddC, the toxic effects of both will be minimized. A
therapeutic regimen which is both effective and clinically tolerable by AIDS patients is actively being sought by health care researchers.

3. SUMMARY OF THE INVENTION
.
The present invention relates to the use of synergistic combinations of nucleoside derivakives for inhibiting human immunodeficiency virus (HIV) replication and thereby limiting HIV infection. According to the invention, certain nucleoside derivatives u~ed together exhibit greater anti-viral effects and less cytotoxicity at lower concentrations than either drug used separately, 2 0 ~
WO90/11081 PCT/~S90/01~2~

thus maximizing treatment of HIV infection and minimizing toxic side-e~fects.
In a particular embodiment of the present invention, the purine nucleoside analogue dideoxyinosine (ddI) and the pyrimidine nucleoside analogue 2',3'-dideoxy-2',3'-didehydrothymidine (d4T) can be used to inhibit HIV replication. It is shown, by way of example, that ddI and d4T exhibit strong synergistic anti-HIV
activity; the combination of ddI and d4T produced a greater anti-HIV effect but was associated with less cytoxicity than either compound used separately.
In another embodiment of the present invention, combinations of fluoronucleoside analogs may be used with d4T to inhibit HIV replication. For example, the purine nucleoside analog, 2',3'-dideoxy-2'-beta-fluoroinosine (F-ddI) may be used with d4T.

3.1. DEFINITIONS AND ABBREVIATIONS
The following terms as used herein whether in the singular or plural, shall have the meanings designated.
Combination Index (CI): a numerical representation of the synergistic or antagonistic effects of drug combinations wherein a CI less than one represents synergy, greater than one represents antagonism, and equal to one represents additivism. The CI value is obtained using an isobologram equation and computer simulation according to Chou and Talalay (1984, Adv.
Enz. Reg. 22:27-55 and 1987, ln ~New Avenues in Developmental Cancer Chemotherapy,~ Harrap and Connors, eds., Bristol-Myers Symposium Series, Academic Pr~ss, pp. 37-64; Hartshorn et al., 1987 Antimicrobial Agents and Chemotherapy 31:168-172).

~090/11081 PCT/~S~0/0142 A combination index may be d~termined with the equation:

( DX ~1 ( DX ~ 2 ( DX ) 1 ( D2X ) 1 where (DX)l is the dose of agent l required to produce x percent effect alone, and (D)l is the dose of agent l required to produce the same x percent effect in combination (D)2. Similarly, (DX)2 is the dose of agent 2 required to produce x percent effect tO alone, and (D)2 is the dose required to produce the same effect in combination with (D~l. As described in Hartshorn et al. (supra), the slope of dose-effect curves indicate whether the ag~nts have mutually exclusive effects (e.g. similar mode of action) or mutually nonexclusive effects (e.g.
independent mode of action).

I the agents are mutually exclusive, then ~ is 0 (i.e., CI is the sum of two terms); if the agents are mutually nonexclusive, ~ is l (i.e., CI is the sum of three terms).

Synergy: the action of two or more substances to achieve an ~ffect greater than that of either substance used individually. synergism in antiviral activity should be construed to mean greater antiviral activity;
synergism in cytotoxicity should be construed to ~
mean greater cytotoxicity.
Tissue Culture Inhibitory Dose (i.e. TCID50) the amount of virus necessary to reduce viral expression by a given percentage, i.e., 50 percent.
Tissue Culture Toxic Dose (i.e., TCTD50 the amount i.e. of drug required to reduce the number of tissue culture cells by a given percentge (i.e., 50 percent).

Wo~/11081 PCT/~S90/(~142 4. DESCRIPTION OF THE FIGURES
Figure l. Antiviral activity of equal total concentrations of d4T (broken line), ddI (dotted line) or d4T + ddI (solid line) as measured by p25yag binding, expressed as percent inhibition.

5. DETAILED DESCRIPTION OF THE INVENTION
The present invention relates to the use of synergistic combinations o~ nucleoside derivatives for inhibiting the replication of HIV, thereby limiting the 1 effects of ~IV infection. According to the invention, combinations of certain nucleoside derivatives exert stronger anti-HIV effects but are associated with less cytotoxicity than individual nucleoside derivatives at comparable total drug concentrations.
Although the applicants are under no duty or obligation to explain the mechanism by which the invention works it may be that by supplying two or more nucleoside derivatives, any one of which may substitute for naturally occurring nucleosides in viral reverse transcriptase activity, the likelihood that these derivatives will be incorporated into the viral DNA sequence (and consequently terminate reverse transcription by preventing elongation of nascent DNA) is substantially increased.
For the purpose of clarity of disclosure, and not by way of limitation, the description of the present invention will be divided into three sections: (l) nucleoside derivative combinations; (2) ln vitro assay for demonstrating the HIV-i~hibitory effect of nucleoside derivative combinations, and (3) therapeutic uses of HIV-inhibitory nucleoside derivative combinations.

2 ~ 7 3 WO90/l1081 pCT/~S90/0l42 5.1. IDENTIFICATION OF SYNERGISTIC NUCLEOSIDE
DERIVATIVE COMBINATIONS
According to the present invention, nucleoside derivatives which may be used in combination include, but are not limited to, 2',3'-dideoxyadenosine (ddA); 2',3'-dideoxyguanosine (ddG): 2',3'-dideoxyinosine (ddI);
2',3'-dideoxycytidine (ddC); 2',3'-dideoxythymidine (ddT);
2',3'-dideoxy-2',3'-didehydrothymidine ~d4T) and 3'-azido-2',3'-dideoxythymidine (AZT). Alternatively, halogenated nucleoside derivatives may be used, preferably 2',3'-dideoxy-2'-fluoronucleosides including, but not limited to 2',3'-dideoxy-2'-fluoroadenosine; 2',3'-dideoxy-2'-fluoroinosine; 2',3'-dideoxy-2'-fluorocytoslne;
and 2',3'-dideoxy-2',3'-didehydro-2'-fluoronucleosides including, but not limited to 2',3'-dideoxy-2',3'-didehydro-2'-fluorothymidine tFd4T). Preferably, the 2',3'-dideoxy-2'-fluoronucleosides of the invention are those in which the fluorine linkage is in the beta configuration, including, but not limited to, 2'3'-dideoxy-2'-beta-fluoroadenosine (F-ddA), 2',3'-dideoxy-2'-beta-fluoroinosine (F-ddI), and 2',3'-dideoxy-2'-beta-fluorocytosine (F-ddC).
The combination of 2',3'-dideoxyinosine (ddI) and 2',3'dideoxy-2',3'-did~hydrothymidine (d4T) represents a preferred embodiment of the invention. The combination of 2',3'-dideoxy-2'-beta-fluoroinosine (F-ddI) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T) also represents a preferred embodiment. For a more detailed description of the 2',3'-dideoxy-2'-fluoronucleosides and the 2',3'-dideoxy-2',3'-didehydro-2'-fluoronucleosides, see copending application Serial No. 120051, filed November 12, 1987 by Sterzycki, Mansuri and Martin which is incorporated by reference herein in its entirety.
In order to evaluate potential therapeutic efficacy of combinations of nucleoside derivatives, these ~5 2 ~
WO90/~lQ81 PCr/~S90/0l~2 combinations may be tested ~o~ anti-viral activity according to methods known in the art. For example, the ability of a nucleoside comhination to inhibit HIV
cytotoXicity, syncytia formation, reverse transcriptase activity, or generation of viral RNA or proteins may be tested ln vitro.
Combinations of nucleoside derivatives over wide concentration ranges for each may be tested for antlviral and/or cytotoxic activity, and combination indeces may be derived, according to methods outlined in Section 6.1.
infra. A preferred method for demonstrating the HIV
inhibitory effect of nucleoside derivative combinations is set forth in section 5.2 and example section 6, infra. In addition, combinations of nucleoside derivatives may be tested for anti viral activity ln vivo using an animal model system for AIDS, such as the simian immunodeficiency virus (SIV) system (Kanki et al., 1985, Science 230:951-954); however, because different species of animals (or even different cell types within one species) differ in the efficiency with which they phosphorylate these drugs, extreme caution should be used in extrapolating the experimental results obtained in one species (or in one cell type within a species) to another (Yarchoan and sroden~ 1987, N. Engl. J. Med. 316:557-564: Waqar, 1984, J. Cell. Physiol. 121:402-408; Furman et al., 1986, Proc.
Natl. Acad. Sci. U.S.A. 83:833-7; Ono et al., 1979, Biochem. and Biophys. Res. Commun. 88:1255-1262).
Therapeutic as well as cytotoxic doses for these combinations may be established, and those combinations Which show greatest synergy, i.e., which are associated with highest anti-viral activity and/or lowest cytotoxicity, may be considered for use in humans.

WO9~/11081 PCT/~S90/01~2 5 . 2 . IN VITRO ASSAY FOR DEMONSTRATING
THE HIV-INHIBITORY EFFEcT OF
NUCLEOSIDE ~E~IVATIVE COMBINATIONS
The inhibitory activity of the nucleoside derivative combinations may be tested using an in vitro assay system such as any of those described in Section 6 et seq. herein. For example, the efficacy of any nucleoside derivative combination selected may be assessed by its relative ability to inhibit (a) the formation of syncytia, and tb) the production of HIV
particles in HIV-infected cells ln vitro using a target cell line that can be infected with HIV. In general such cell lines would be of a T-cell or myelocytic/monocytic lineage or another cell type transfected with the gene for CD4. Alternatively, if the nucleoside derivatives were linked to a "targeting" molecule e.g., a monoclonal antibody, hormone, growth factor etc. the target cell line should express the appropriate cell surface antigen or receptor for the molecule conjugated to the nucleoside derivative.
In order to assess the efficacy of the nucleoside derivative combination chosen, the target cell should be grown in vitro, infected with HIV and treated with an array of doses of the nucleoside derivatives before, during or after infection (e.q., limited dilution techniques can be used). The inhibitory effect on viron production may be assessed as described, ~ , by measuring (a) production of HIV by assaying for HIV
proteins, such as the 25 gag viral core protein or reverse transcriptase, in the culture media; (b) the induction of HIV antigens in cells by immunofluorescence; or (c) the reduction in syncytia formation assessed Yisually. The ability of a nucleoside derivative combination to inhibit HIV is indicati~e of the inhibitory activity and WO90/11081 PCT/~S90/01~21 ~ 17 -inhibitory dose of the nucleoside derivative co~bination tested.

5.3. THERAPEUTIC USES OF HIV-INHIBITORY
SYNERGISTIC DERIVATIVE COMBINATIONS
The synergistic nucleoside derivative combinati~n may be used in accordance with the invention ln vivo to prevent the formation of syncytia and the production of HIV virions and, thus, inhibit the progression of HIV
infection within an exposed patient. Effective doses of tO nucleoside derivatiYes formulated in suitable pharmacological carriers may be administered by any appropriate route including but not limited to injection ~ , intravenous, intraperitoneal, intramuscular, subcutaneous, etc.), by absorption through epithelial or mucocutaneous linings (e.q., oral mucosa, rectal and vaginal epithelial linings, nasopharyngeal mucosa, intestinal mucosa, etc.); etc.
In addition, nucleoside derivatives may be mixed in any suitable pharmacological carrier, linked to a carrier or targeting molecule le.~., antibody, hormone, growth factor, etc.) and/or incorporated into liposomes, microcapsules, and controlled release preparations prior to administration in vivo.
In another embodiment, synergistic nucleoside derivative combinations may be usPd in conjunction with other treatments for HIV infection. For example, and not by way of limitation, synergistic nucleoside derivatives '~
may be used in conjunction with other antiviral compounds that inhibit reverse transcriptase activity (e.g., AZT);
with soluble CD4 or monoclonal antibodies that inhibit HIV
absorption to cells; or with other cytokines and growth factors (e.g., interferon ~, ~ or ~; tumor necrosis factor, interleukin-2, granulocytic/monocytic colony stimulating factor; colony stimulating factor-l, etc.);
~5 ~a~7,~
WO90/~1081 PCT/~S90/01~2 and with agents that are used to treat other microorganisms or viruses that opportunistically infect AIDS patients.
In another embodiment of the invention, synergistic nucleoside derivatives can be used in v _ o to test the efficacy of different drugs. In this regard~ it would be advantageous to obtain bone marrow samples from an AIDS patient and expose the marrow in vi~ro to the drug, compound or regimen being tested in order to identify those that, for example, kill the virus yet spare 1 normal cells, or those that will stimulate marrow repopulation. However, bone marrow derived from AIDS
patients exhibits very poor marrow colony formation ln vitro. This is probably due to infection of the precursor CD 34 cells with HIV. As a result, the efficacy o~
various drugs tested ln vitro is difficult or impossible to assess. The use of synergistic nucleoside derivatives to inhibit HIV in such an assay system should increase marrow colony formation ln vitro and therefore, permit the screening of various other protocols and drugs on the marrow activity ln vitro.

6. EXAMPLE: INHIBITION OF HIV INFECTION
USING DIDEOXYDIDEHYDROTHYMIDINE AND
DIDEOXYINOSINE
The experiments described below demonstrate the inhibitory effect of dideoxydidehydrothymidine (d4T) and dideoxyinosine (ddI) on HIV infection of target cells of -T-cell origin in culture in vitro.

6.l. MATERIALS AND METHODS
6.l.l. CELLS AND VIRUS
CEM-F cells were originally derived from the acute human lympyhoblastic leukemia and represent an established T lymphoblastoid line. They are available at 2 ~ 7 3 W~90/11081 PCT/~S90/01~2 the American Type Culture Collection as CCRF-CEM cells (ATCC No. CCL ll9).
The LAVBRU strain of human immune de~iciency virus (HIV) was obtained from Dr. Luc Montagnier, Institut Pasteur, Paris, France. q'he virus was adapted ~o CEM-F
cells, and stored in small aliquots in liquid nitrogen.
The titer of the virus was détermined every two to three weeks and is expressed as TCID50 (tissue culture inhibitory doses; i.e., the amount o~ virus necessary to reduce the viral expression by 50%). In all experiments described in the subsections below, a viral dose of 50 TcID50 was used.
Both the CE~-F cells and the HIV virus were grown in LAV/CEM medium. The medium consists of RPMI 1640, supplemented with 1% glutamine, lOO U/ml of Penicillin, l00 ~g/ml Streptomycin, 2 ~g/ ml Polybrene, and lO~ fetal bovine serum.

6.l.2. NUCLEOSIDE DERIVATIVES D4? AND DDI
Dideoxydidehydrothymidine (d4T, BMY27857-3/8, lot #26630-23A) and dideoxyinosine (ddI, BMY40900, lot $25879-46-1) were obtained from Pharmaceuticl Research and Development Division of Bristol-Myers Company. Both compounds were resuspended in 2-3 drops of dimethyl sulfoxide (DMSO, D-8779), Sigma Chem. Co.) and LAV/CEM
2~ media. The suspension was then sonicated (Microultrasonic disrupter, Kontes) and brought to 1 mM f inal concentration. A`ll dilutions were made from that l mM
stock In order to test the combined anti-HIV effect of th~ drugs, two different s ts of experiments were done.
In the first set, the drugs were mixed so that one was kept at constant concentration, while the other drug concentration was varied. For example, d4T was kept at either 0.0l, 0.l, l or 10 ~M, while ddI range was O.l, l, 10 to 100 ~M. In this setup the ratio o~ the drugs in wo90/11081 PCT/~S90/01~2 each mixture was different. In order to more precisely determine the degree of possible synergy between d4T and ddI, in the second set of assays the drugs were mixed at l:5 ratio starting at l0 and 50 ~M, and diluted twofold.
This way the ratio of the drugs was kept constant through all dilutions. The data was evaluated using the "Dose-Effect Analysis with Microcomputers~ software (Chou, J., and T-C. Chow, Elsevier-Biosoft Publishers, 1986). The same two protocols were used to evaluate toxicity of d4T
and ddI on uninfected cells, except that in the equal drug ratio experiment the ratio of the drugs was l:l, with a starting concentration of lO0 ~M for both drugs.
Azidothymidine (AZT, BMY27755/7, lot #2300-38) was used as a positive control in all assays.
6.l.3. DETERMINATION OF THE PROLIFERATIVE

For each assay 2 x 104 CEM-F cells/well were plated in 96 well plates. The cells were mixed with each different combination, each set up in quadruplicate. The total volume/well was 250 ~l. Plates were incubated for six days at 37C and 5% CO2. On day six, cells were labeled for four hours with l ~Ci/well of 3HTdR (New England Nuclear Corp. specific activity 5.7 Ci/mmol) harvested onto glass fiber filters and counted in the scintillation counter (Hartzman, R.J., Segall, M., Bach, M.L., and Bach, F.H., 1971, Histocompatibility matching.
VI. MiniaturiZation of the mixed leukocyte culture test: A
preliminary report. Transplantation, 11:268-273). The results were expressed as percent 3HTdR uptake of the control which consisted of cells incubated without drugs.
In addition, the results can be expressed as a tissue culture toxic dose 50 (TCTD50), which represents the amount in ~g of the drug or drugs necessary to reduce the number of cells by 50% as compared to the control.

~3'~7 3 WO ~0/ Z 1 08 I PCT/l 'S90/0 1 12 6.1. 4 . IN~IIBITION OF HIV REPLICATION
The CEM-F cells were plated in 96 well plates at 2 x 10cells/well and mixed with 50TCID50 ~ virus for 45 minutes. After 45 minutes nucleoside derivatives were added to each well and incubated as described for the proliferation assay. At the end of day six the supernatants were tested for the presence of viral antigen using an antigen capture ELISA assay described below. The assay uses two monoclonal antibodies against the viral core protein p24gag (Hu, S.H. et al., 1987, Nature 328:721-723; and Kinney Thomas, E. et al., 1988, AIDS
2:25-29).

6.1.5. ANTIGEN-CAPTURE ASSAY
15For this assay Microtiter plates (96 well plates) were coated with two monoclonal antibodies: 25-2 (ATCC
#9407) and 25-3 (ATCC #9408), each diluted at 1:2500.
These antibodies (capture reagents) are specific for p24, p40, and p55 HIV gag proteins. Horseradish peroxidase (HRP)-conjugated human IgG purifi~d from a serum of a seropositive individual was used as a signal. The absorbance (450/630 nm) is determined after the addition of substrate - tetramethyl benzidine (TMB). The OD
readings fa~l into three categories: Experimentals =
values from wells containing cells, viral inoculum and nucleoside derivative~s); and Controls = values from wells containing cells and virus (100%); and Background = values from wells containing viral inoculum alone. The background value was subtracted from all the OD values.
The antiviral effect of the nucleoside derivative(s) is expressed as the percent p24gag binding of the control.
For example, if that value is 20%, it means that the viral replication, measured indirectly through p25gag binding, is inhibited by 80%.

7 ~
WO90/11081PCT/~'S90/0~424 6.l.6. SYNCYTIA FOR~ATION
Prior to collecting supernatants for the an~igen-capture assay, all the wells were visually examined. This was done to assure that the infection took place and also to check the state of the cells. The syncytia werP easy to observe, and although they were not counted, the differences in their numbers between wells were very apparent.

6.1. 7 . ANALYSIS OF CYTOTOXICITY
The toxicities of D4T and DDI were tested against the host cell CEM-F (ATCC CCLll9~. CEM-F is a T-cell line that constitutively expresses the CD4 receptor, and therefore, is an appropriate host cell target for HIV
infection. The effect of D4T and DDI on proliferation of uninfected cells was assessed by measuring thymidine incorporation in uninfected CEM-F cells treated with D4T
and DDI.

6.2. RESULTS
?~
6.2.l. HIV INHIBITION BY A NUCLEOSIDE
DERIVATIVE COMBINATION
In vitro assays for p25gag protein revealed a strong synergistic effect of d4T in combination with ddI
for inhibiting p25gag expression, representin~ antiviral activity-greater than either compound used individually.
The results of binding assays for p25gag are shown in Table I; a comparision of a combination of d4T and ddI to either drug used at the same total concentrations is shown in Figure l.
Table I presents the results of a series of experiments used to discern the optimal concentrations and ratios of concentrations of d4T and DDI for achieving maximal anti-viral activity. A wide range of WO90/11081 P ~/ ~ 0~0~2 concentrations were tested by evaluating logarithmic increments in the csncentration of each derivative in multiple combinatio~s, using a "checkerboardn pattern of testing.
-~
Table I
Percent Inhibition of Viral Replication d4T and ddI*
concentration of ddI (~M) Concentration of D4T(~M) 0 ~ 10 l00 0 - l 3 2 86 .0l 2 25 13 38 83 .l l 26 ll 13 76 l 24 52 52 75 78 * as measured by p25gag binding When either d4T or ddI were used in conjunction with AZT, at a concentration of 1:10:50, a synergistic effect was achieved which was less than the synergism observed for the combination of d4T and ddI (see Table III
and Section 6.3).

6 . 2 . 2 . I:)IMI~JISHED CYTOTOXICITY OF A COMBINATION
OF NUCLEOSIDE DERIV~TIVES
Cytotoxicity assays measuring 3H-thymidine incorporation show that d4T and ddI used in combination are not more cytotoxic to CEM-F cells than the same total concentration of either compound used individually, despite increased toxicity to virus. The re5ults of these assays over a wide range of concentrations of d4T, ddI, or ~5 2 ~
WO90/11081 1'Cr/~'S90/0~21 ~ 24 d4T in combination with ddI in 1:1 ratio are shown in Table II.

Tabl e I I
.
5Percent Cytotoxicity of CEM Exposed to Increasing concentrations of d4T and/or ddI*
concentration of ddI (~M) Concentration of D4T (~M) 0 .1 1 10 100 _ . _ _ _ _ _ _ _ . _ _ . . _ _ _ _ . _ O ~

* as measured by 3H-thymidine uptake.

6.3. DISCUSS ON
The data presented above indicates that the nucleoside derivatives d4T and ddI exhibit a strong synergistic antiviral effect when used in combination.
For example, referring to Table I, a concentration of l~M
of d4T resulted in 24 percent inhibition of viral activity, and a concentration of 10~M of ddI was .
associated with 2 percent inhibition of viral activity;
however, l~M of d4T combined with 10~M of ddI achieved 75 percent HIV inhibition. Figure 1 contains a graph which shows the antiviral activities of d4T, ddI, and (d4T + ddI
at the same total concentration of nucleoside, and clearly shows the synergistic effects of d4T used in conjunction With ddI.
35 .

2 ~ 3 W090/110~1 PC~/~S90/0l~2 The combination index (CI) is a numerical representation of the synergis~ic or antagonistic effects of drug combinations. The CI value is obtained using an isobologram equation and computer simulation according to Chou see section 3.1). A CI value of less than one represents synergy (i.e., the whole is greater than the sum of its parts) CI greater than one represents antagonism (i.e. the wh~le is less than the sum of its parts) and CI equal to one represents additivism, (i.e., the whole is equal to the sum of its parts). Table III
shows-the CI values of ~d4T + ddI), (d4T + AZT), and (ddI
+ AZT) for antiviral effect, when AZT, d4T, and ddI
concentrations wer in the ratio of 1:10:50.

7 ~) WO90/11081 PCT/~S90/0142 .
Table III
Anti-~IV Effect CI Values at 50, 70, and 90 Percent Inhibition of Virus 50% 70% 90%
Compound D4T + DDI 0.024 0.049 0.16 D4T + AZT 0.53 0.430.30 AZT + DDI 0.77 0.740.75 Because of the synergistic relationship between d4T and ddI, lower concentrations of nucleoside can be used to achieve effective viral inhibition. As shown in Table II, increased antiviral activity of d4T and ddI
combinations is not associated with increased cytotoxicity, and therefore, selective antiviral activity has been achieved. In addition, studies to establish the TD50 (toxic dose) for d4T and ddI singly, or in combination, revealed that whereas the TD50 for both d4T
and ddI indiv _ually were graeater than lOo~M, the TD50 for the combination of D4T and DDI was 730 ~M (data not shown). The CI for cytotoxic effects of combinations of AZT, d4T, and ddI, at concentration ratios of l:lO:lO, respectively, show that at concentrations that inhibit 90 of HIV activity, the combination of d4T and ddI is markedly less than equally virocidal concentrations of either (d4T + AZT), or (ddI ~ AZT) (see Table IV).
Therefore, not only is the combination of d4T and ddI more virocidal than either d4T or ddI considered individually, (d4T + ddI) is less cytotoxic. The therapeutic window for treating HIV infection is therefore broadened by using these nucleoside derivatives in combination; therapeutic ~.5~73 WO90/11081 PCT/~S90/0112 anti-HIV effects may be achieved in patients without risking dangerous or painful complications.

Table IV
Cytotoxicity CI Values at Nucleoside Concentrations Corresponding to 50 and 90_Percent Viral_Inhibition 50% 90%
Compound D4T + DDI 2.17 6.40 D4T + AZT 2.45 l.66 AZT + DDI 2.21 2.37 . ~

-

Claims (42)

WHAT IS CLAIMED IS:

1. A method for inhibiting HIV comprising bringing HIV-infected cells in contact with a synergistic combination of nucleoside derivatives.

2. The method according to claim 1 which is performed in vitro.

3. The method according to claim 1 which is performed in vivo

4. The method according to claim 3 which is performed in a mammal.

5. The method according to claim 3 which is performed in a human.

6. The method according to claim 1, 2, 3, 4, or 5 in which at least one of the nucleoside derivatives is a 2',3'-dideoxynucleoside.

7. The method according to claim 1, 2, 3, 4, or 5 in which one of the nucleoside derivatives is 2',3'-dideoxyinosine.

8. The method according to claim 1, 2, 3, 4, or 5 in which one of the nucleoside derivatives is 2, 3'-dideoxy-2',3'-didehydrothymidine.

9. The method according to claim 1, 2, 3, 4, or 5 in which one of the nucleoside derivatives is a 2',3'-dideoxy-2'-fluoronucleoside derivative.

10. The method according to claim 9 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-fluoroinosine.

11. The method according to claim 10 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-beta-fluoroinosine (F-ddI).

12. The method according to claim 9 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-fluoroadenosine.

13. The method according to claim 12 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-beta-fluoroadenosine (F-ddA).

14. The method according to claim 9 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-fluorocytosine.

15. The method according to claim 14 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-beta-fluorocytosine (F-ddC).

16. The method according to claim 3 which is used in conjunction with other anti-HIV treatment.

17. The method according to claim 3 which is used in conjunction with one or more therapies for HIV-associated disorders, including opportunisitic infections.

18. The method according to claim 1, 2, 3, 4, or 5 in which at least one nucleoside derivative is chemically linked to a second molecule.

19. A method for inhibiting HIV comprising bringing HIV-infected cells in contact with a combina-tion of 2',3'-dideoxyinosine (ddI) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T).

20. The method according to claim 19 in which the ratio of d4T to ddI is between about 1:1 to 1:10.

21. The method according to claim 19 in which the ratio of d4T to ddI is about 1:5.

22. A method for inhibiting HIV comprising bringing HIV-infected cells in contact with a combina-tion of 2',3'-dideoxy-2'-beta-fluoroinosine (F-ddI) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T).

23. A method for inhibiting HIV comprising bringing HIV-infected cells in contact with a combina-tion of 2',3'-dideoxy-2'-beta-fluoroadenosine (F-ddA) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T).

24. A method for inhibiting HIV comprising bringing HIV-infected cells in contact with a combina-tion of 2',3'-dideoxy-2'-beta-fluorocytosine (F-ddC) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T).

25. A method for preparing a composition which inhibits HIV comprising combining synergistic nucleoside derivatives.

26. The method according to claim 25 in which at least one of the nucleoside derivatives is a 2',3'-dideoxynucleoside.

27. The method according to claim 25 in which one of the nucleoside derivatives is 2',3'-dideoxyino-sine.

28. The method according to claim 25 in which one of the nucleoside derivatives is 2',3'-dideoxy-2',3'-didehydrothymidine.

29. The method according to claim 25 in which one of the nucleoside derivatives is a 2',3'-dideoxy-2'-fluoronucleoside derivative.

30. The method according to claim 29 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-fluoroinosine.

31. The method according to claim 30 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-beta-fluoroinosine (F-ddI).

32. The method according to claim 29 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-fluoroadenosine.

33. The method according to claim 32 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-beta-fluoroadenosine (F-ddA).

34. The method according to claim 29 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-fluorocytosine.

35. The method according to claim 34 in which the fluoronucleoside derivative is 2',3'-dideoxy-2'-beta-fluorocytosine (F-ddC).

36. The method according to claim 25 in which at least one nucleoside derivative is chemically linked to a second molecule.

37. The method according to claim 25 in which the synergistic nucleoside derivatives are 2',3'-dideoxyinosine (ddI) and 2',3'-dideoxy-2',3'-didehydro-thymidine (d4T).

38. The method according to claim 37 in which the ratio of d4T to ddI is between about 1:1 to 1:10.

39. The method according to claim 37 in which the ratio of d4T to ddI is about 1:5.

40. The method according to claim 25 in which the synergistic nucleoside derivatives are 2',3'-dideoxy-2'-beta-fluoroinosine (F-ddI) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T).

41. The method according to claim 25 in which the synergistic nucleoside derivatives are 2',3'-dideoxy-2'-beta-fluoroadenosine (F-ddA) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T).

42. The method according to claim 25 in which the synergistic nucleoside derivatives are 2',3'-dideoxy-2'-beta-fluorocytosine (F-ddC) and 2',3'-dideoxy-2',3'-didehydrothymidine (d4T).