CA2227867A1 - Anti-viral guanosine-rich oligonucleotides - Google Patents

Abstract

A method and compositions for treating viral infection in vitro and in vivo using a guanosine-rich oligonucleotide. The oligonucleotides have sufficient guanosine to form a guanosine tetrad. Also provided are oligonucleotides of at least two runs of at least two guanosines. Also provided are guanosine-rich oligonucleotides and methods for treating viral infections in humans, and a method for designing guanosine-rich oligonucleotides having anti-viral activity and integrase inhibition activity.

Description

CA 02227867 l998-0l-26 ANTI-VIRAL GUANOSINE-RICH OLIGONUCLEOTIDES

The present invention is a cnntinnAtinn-in-part of U.S. Patent Application Serial No 08/535,168. The present invention also claims the benefit of the 35 U.S.C 111(b) provisional applicAtic-n~: U.S. Patent Application Serial Nos. 60/001505, 60/014,007, 60/013,688, 60/015,714, 60/015,74 and 60/016,271.
BACKGROI~ND OF THE INVENTION
Field of the Invention The present invention relates generally to the field of oligonucleotide ~h~mi~try and anti-viral pharmacotherapy. More sFecifil~Ally, the present invention relates to novel gnAnn~in~-rich oligonucleotides and their use as novel anti-viral agents.

Description of the Related Art General In Vitro Studies Previously, it was believed that "Anti~.on~e" oligonucleotides inhibit viruses by hl~r~illg with protein translation via an RNA:DNA duplex structure. More recent lese~l-,h, however, in~ Atos a variety of possible ~,tsrhA.,i~ by which oligo-nucleotides inhibit viral infections. For example, oligodeoxycytidine (poly SdC) inhibits HIV-l. Marshall et al., PNAS (1992) 89:6265-6269, ~ cll~5ed the potential m~hA.~ (co~ cLiliv~ inhibition) by which oligodeoxycytidine directly inhibits viral reverse Lla-ls~ L;~se. Poly SdC also inhibited AMV
reverse Lld-lscli~LAse and Pol I (Klenow rl_~,lllcllL) and POIY111C:1A-Se ~, ,~ and y. Previously, Matsukura et al., PNAS (1987) 84:7706-7710, used a similar phosphorothioate d~:liv~Livt: of oligo-deoxycytidine to ~i.omnn~tr~t~ inhibition of HIV-1 in culture. Marshall and Caruthers, Science (1993) 259:1564-1569, reported the use of ~lirhQ~phQrothioate oligo-nucleotides, e.g., Anti~en~e-specific, random nucleotide c~.,,,l.i,,AIinn~ and oligodeoxycytidine against HIV-l. In all cases, the mP~hAni~m of action was attributed to a direct inhibition of HIV-l reverse transcriptase. Other potential ,I"~.~h?~ mc of anti-viral action of oligonucleotides were pnstlllAtt?c~ by Boiziau et al., PNAS (1992) 89:768-772, e.g., promotion of RNAse H activity and inhibition of reverse tran~cli~L~se initiAting cDNA synt_esis. In A~ltlition, Goa et al., Molecular Pharmacology (1992) 41:223-229 reported that phn~rhorothioate oligonucleotides inhibit human DNA polymerases and RNAse H, and the adsorption or penetration of the virus into cells. Iyer et al., Nucleic Acids Research (1990) 18:2855-2859 reported that if a base was removed from an anti-sense polynucleotide forming an abasic site, the compound did not lose its activity which argues against CA 02227867 l998-0l-26 W O 97/03997 . PCTAUS96/11786 the need for the formation of an RNA:DNA ~nti~rn:~e mrrli~t.od hybrid for anti-viral activity. Stein et al. have characterized the interaction of poly SdC with the V3 loop of HIV-1 gpl20, and post ll~tr-l that the specific interaction of poly SdC with the HIV-l V3 loop may be a ~ r~ m by which an oligonucleotide could inhibit HIV-l in vivo.
S It is known that synthetic oligonucleotides may be ~lr~ignPd which are capable of binding to duplex DNA to form triplex DNA. See U.S. Patent No. 5,176,996 Hogan & Kessler issued January 5, 1993. That patent discloses a method for making synthetic gll~no~inr-rich oligonucleotides which are targeted to specific seq7lrnre~ in duplex DNA and which form collinP~r triplexes by binding to the major groove of the DNA duplex.
SpeciFIc In Vitro StudiestIn Vitro HIV Tnhihitinn With T30177 Infection with the human immllnotlPficiency virus type 1 (HIV-l) and the subsequent development of acquired immunodeficiency ~y~ le (AIDS), has become a threat to public health on a global scale. Pl~:v~llLillg further spread of this disease is a major health priority worldwide.
Although HIV-l was confirmed to be the c~us~liv~ agent of AIDS as early as 1984, few drugs and no vaccines are ~rre-;liv~ at picvcllLillg the ultimate onset of AIDS in HIV-l seropositive individuals. This is due, in large part, to the comple~iry of the c~us~liv~ agent itself, the dynamics of virus prod~lrti~n and the speed at which drug-l~si~L~ulL mutants can arise. Ho, et al., Nature 373:123-126 (1995); Wei, et al., Nature 373:117-122 (1995).
Infection of T-cells by HIV-l results in the insertion of proviral (double-stranded) DNA
into the host cell genome. C~off, S.P., Annu. Rev. Genet. 26:527-544 (1992). The i"~
process involves both the sequence-specific and seqnrnre inrltopf~n-lPnt entlonllrl~olytic and strand transfer activities of the virally encoded integrase enzyme. Katz, et al., Ann. Rev. Biochem.
63:133-173 (1994); Vink, et a., Trends in Genetics 9:433438 (1993). Once the proviral state is established, the infection may lllanir~ itself in several ways inr~ ing a latent infection in which viral replication is not measurable until the cell becomes activated or through a chronic infection in which dividing or non-dividing cells p~ lr~llly release virus in the absence of any ~;yL~dlllic effect. In ~ litinn, recent reports on the kinetics of virus production (and rlr~r~nre) indicate a dynamic process in which virtually a complete repl~rrmrnt of wild-type virus by drug-resistant virus in plasma can occur after only two to four weeks of drug therapy. Ho, et al., Nature 373:123-126 (1995); Wei, et al., Nature 373:117-122 (1995). For this reason it is of utmost importance to develop new anti-HIV-l agents which can complement, by additive or ~ylle~ ic activity, current therapies.
One relatively new a~pluacll used in the development of antiviral therapeutics for HIV-l is the use of oligonucleotides ~ ign~d as ~nti~n~e agents. Letsinger, et al., Proc. Natl. Acad. Sci.

CA 02227867 l998-0l-26 = wo 97/03997 PCT~US96/11786 USA 86:6553-6556 (1989); Lisziewicz, et al., Proc. Natl. Acad. Sci. USA 90:3860-3864 (1993);
Milligan, et al., J. Med. Chem. 36:1923-1937 (1993). While much effort is being spent on rationally ~ ign--d oligonucleotides such as ~nti~en~e agents there have also been recent findings of multiple ~lt~?rn~tive " ,~ " ,~ by which oligonucleotides can inhibit viral infections. Gao, et al., J.B.C. 264:11521-11526 (1989); Marshall, et al., Proc. Natl. Acad. Sci. USA 89:6265-6269 (1992); Ojwang, et al., J. AIDS 7:560-570 (1994); Rando, et al, J. Biol. Chem. 270:1754-1760 (1995). For example, Stein et al. (Stein, et al., Antisense Research and Development 3:19-31 (1993))have(-h~ r~ dtheinteractionofoligodeoxycytidine,c(."l;l;"i"gaphosphorothioate(PT) backbone (poly (SdC)) with the v3 loop of HIV-1 gp 120. It was ~lPt~rrnin~d that poly (SdC)28 specifically interacted with the positively charged V3 loop with a Kd of d~loxilll,~l~ly 5 x 10-7M.
Stein et al. (Antisense Research and Development 3:19-31 (1993)) then pnst~ t~d that the interaction of poly (SdC) with the HIV-1 v3 loop may be a m~h~ni~m by which poly (SdC) could inhibit HIV-1 in vivo. More recently, Wyatt et. al. (Wyatt, et al., Proc. Natl. Acad. Sci. USA
91:1356-1360 (1994)) have described the interaction of a short G-rich oligonucleotide, syntht-~i7~d with a total PT backbone, which also interacts with the v3 loop of HIV-1 gp 120. In addition, we have previously reported that oligonucleotides CO~.I;.i"i"g only deoxyguanosine (G) and thymidine (T), synth~i7~d with natural ph~sphn~ st~-r (PD) interml~lPQ~ o linkages, were capable of inhibiting HIV-1 in culture. Ojwang, et al., J. AIDS 7:560-570 (1994). The most emc~riolle lllelllbel of he G22 this dG-rich class of oligonucleotides, I100-15, was found capable of folding upon itself to form a structure stabilized by the formation of two stacked gu~no~in~-tetrads which yielded a gll~nn~in~-octet. Rando, et al, J. BioZ. Chem. 270:1754-1760 (1995). Furthermore, it was observed that the positions of the gn~no~in~ bases in the I100-15 sequence, found in both the tetrads and conn~cting loops in that structure, were extremely important to the overall anti-HIV-1 activity of the oligonucleotide. Rando, et al, J. Biol. Chem. 270:1754-1760 (1995).
Site of Activity Studies-Viral I~ d:,e Tnhihiti~n Two events which are chal~.eL~ Lic of the life cycle of le~ vhuses can be utilized for therapeutic hlLelvellLion~ One is reverse Ll~ls~ Lion, whereby the single-stranded RNA genome of the leLl~vil~ls is reverse transcribed into singled-stranded cDNA and then copied into double-stranded DNA. The next event is integration, whereby the double-stranded viral DNA gelleldLcd by reverse Lldlls~ Ldse is inserted into a chrnm~-s-)m.o Of the host cell, establishing the proviral state. Integration is catalyzed by the leLlvvildl enzyme integrase which is encoded at the 3'-end of the pol gene. Varmus, et al. MobileDNA, pp. 53-108, Am. Soc. Microbiol, Washil.~Loll, D.C.
(1989). Integrase first catalyzes the excision of the last two nucleotides from each 3'-end of the linear viral DNA, leaving the terminal coll~elv~d dinucleotide CA-3'-OH at these recessed 3' ends CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 (Fig. C-lA). This activity is referred to as the 3'-processing or dinucleotide cleavage. After transport to the nucleus as a nucleoprotein complex, Varrnus, et al. MobileDNA, pp. 53-108, Am.
Soc. Microbiol, Washington, D.C. (1989), integrase catalyzes a concerted DNA strand transfer reaction by nucleophilic attack of the two viral ends onto a host chromosome. This reaction S generates a recombination interm~ t~ res~onnhlin~ an X structure, analogous to a Holliday junction inr~rm.o~i~t~ [For recent reviews see Katz and Skalka, Katz, et al., Ann. Rev. Biochem. 63, 133-173 (1994), and Vink and Plasterk, Vink, et al., Trends Genet. 9, 433437 (1993)]. Mutation analyses of the viral integrase gene ~ l dl~ that integration is required for ~rrc~;Liv~ lcLluvildl replication and that it is a legitim~te target for the design of a~lLileLlovil~l drugs (l~n~lem~n, et al., J. Virol. 69, 2729-2736 (1995); Fngllln-l et al, J. Virol. 69, 3216-3218 (1995)).
It is known that AZT nucleotides can inhibit HIV-l integrase, ~7Jlm~l~r~ et al., Proc.
Natl. Acad. Sci. 91, 5771-5775 (1994), and that ylhsrihltion or ullsaLuldlion at the 3'-position of the deoxyribose confers potency against HIV-l illL~ldse. These results suggested that the enzyme's nucleotide binding site could serve as a potential drug target. It has been shown that the potential sr~kin~ interactions gained from the heterocyclic rings can further enhance potency against HIV-l r~ e.
Recently, olig- m~rl~oti-l~s composed of deoxygu~no~in~ and thymidine have been reported to inhibit HIV-l replication. Rando, et al., J. Biol. Chem. 270, 1754-1760 (1995); Wyatt, et al., Proc. Natl. Acad. Sci. U.S.A. 91, 1356-1360 (1994). Oligonucleotides forming intr~m~lec7ll~r G4s did not block virus adsorption but rather inhibited viral-specific transcripts. Rando, et al., J. Biol.
Chem. 270, 1754-1760 (1995); Ojwang et al. J. Aids 7:560-570 (1994) SLIu~;Lu~-Func~ion Studies It is known that G-rich nucleic acid sequences can fold, in the presence of Na+ or K+ ion, to form orderly structures stabilized by gn~n~in.o tetrads. Depending on seqnl~n~e, intramolecular folds, Rando et al. J. Biol. Chem. 270: 1754-1760, 1995), dimers (Smith, F. W., & Feigon, J.
(1992) Nature (London) 344, 410-414, S--n~lqui~t~ W. I. & Klug, A. (1989) Nature fLondon) 334, 364-366; Kang, et al. (1992) Nature ~ondon) 356, 126131; Balaguumoorthy, P. & Br~hm~h~ri, S. K. (1994) J. BioZ. Chem. 269, 21858-21869), L~Lldlllele~ (Son, D. & Gilbert, W. (1990) Nature (London) 344, 410-414; Jin, et al. (1990) Science 250, 543-546; Jin, et al. (1992) Proc. Natl.
Acad. Sci. USA 89, 8832-8836; Lu et al., (1992) Biochemistry 31, 2455-2459), and higher order associations have been detect~ Such tetrad based structures have been postulated to serve as the ~ilu.;~uldl basis for telomere filnrtion (Sen, D. & Gilbert, W. (1988) Nature (London) 334, 364-366), and have been hypothesized to play a role in ~ vildl replication (Bock et al. (1992) CA 02227867 l998-0l-26 Nature (London) 355, 564-566), and Ll~ls~ JLion regulation (Marshall et al. (1992) Proc. Nalt.
Acid. Sci. USA 8,9, 6265-6269; Wyatt et al. (1994) Proc. Natl. Acad. Sci. USA 91, 1356-60).
Recently, several groups have shown that compounds which contain tetrad-based folds may have activity as potential drug compounds. Bock and colleagues have shown that an intramolecular S fold, obtained by a SELEX procedure can bind tightly to Lhlo~ ill,so as to inhibit clotting (Bock et al. (1992) Nature (London) 355, 564-566). ~tlrlitinn~lly Wyatt et al. (Wyatt et al. (1994) Proc.
Natl. Acad. Sci. USA 91, 1356-60) has shown that a dimer-wise pairing of phosphorothioate oligomers with the seqn~nre T2G4T2 (four stranded int~rmnleclll~r tetrads) gives rise to anti-HIV
activity, by inhibition of viral adsorption to the cell surface.
The present hlvellLol~ have also obtained evidence for sequence-selective inhibition of HIV-l by simple phosrho~ st~r oligonucleotides which form G-tetrad based structures. The highest activity was obtained with a 17mer, referred to as T30177, with composition G12-T5 (Rando et al., (1994) J. Biol. Chem. 270, 1754-1760; Ojwang, J. et al. (1995) J. Aids 7, 560-570), with 2 phosphorothioate linkages (1 at each end) to block cellular exnml~le~e activity (Bishop et al.
(1996) J. Biol. Chem. 271, 5698-5703). NMR evidence was obtained (Rando et al., (1995) J. Biol.
Chem. 270, 1754-1760) to suggest that, by reference to similar oligomers (Smith, F. W., &
Feigon, J. (1992) Nature (London) 344, 410-414), T30177 forms a stable intramolecular fold which is stabilized by a pair of G-tetrads, co~ rd by three single-stranded loops and a 1-2 base long tail to either side of the fold. Those prelhllillaly studies suggested that oligomer folding was coupled to K+ ion binding (Rando et al., (1995) J. Biol. Chem. 270, 1754-1760). Additional studies have suggested that T30177 and related deliv~Liv~s are potent inhibitors of HIV-l int~gr~e7 in vitro (Ojwang et al. (1995) Antimicrob. Agent Chemotherepy 39, 2426-35).

Pharrn~ kin~ti~ Studies-Single Dose ~nti~Pn~e, triple-helix, duplex decoy, and protein-binding (aptamer) oligonucleotides have been shown to have potential as drugs for the Ll~;:.Llll~llL of a variety of human clinical disorders (Stein and Cheng, 1993; Marshall and Caruthers, 1993, Science 259: 1564-1570, Chubb and Hogan, 1992, Trends in Biotechnology 10: 132-136; Stull and Szoka, 1995, Pharm. Res. 12: 465-483. A number of oligonucleotides have undergone pre-clinical testing, and several are in human clinical trials. One finding that has aroused some concern (Black et al., 1994, Antisense Pes. Dev.
4: 299-301) is the observation that total phosphorothioate oligonucleotides cause hemodynamic changes following rapid intravenous ~fimini.~tration. Severe hy~olellsion, leukopenia, complement ,, activation, and death have been reported to occur in ~lilllaLes after rapid infilci~mc of total phosphorothioate oligonucleotides (Cornish et al., 1993, Pharmacol. Commun. 3: 239-247;
Galbraith et al., 1994, Antisense Pes. Dev. 4: 201-206). These findings have raised the question s WO 97/03997 PCT~US96/11786 of whether the cardiovascular toxicity is a property of phosphorothioate oligonucleotides, or of all oligonllr-lrotiA~-s On the basis of these findings, an FDA co"",~ ,y has recommPnA~-d that cardiovascular screening be pclrolllled for the pre-clinical safety ~c~ç~ of oligonucleotides (Black et al., 1994).
s Pharmacokinetic Studies-Repeat Dose Oligonucleotides have advanced to the stage that they are now conci~ red as potential therapeutics for the ~.cdLlllellL of a variety of human diseases, and several are plcscllLly in clinical trials. Pre-clinical studies have generally shown that doses up to a~ cly 50 mg/kg are safe, but that higher doses can cause kidney and liver damage, and death (Srinivasan and Iversen, 1995, J. Clin. Lab. Analysis 9: 129-137) Bolus intravenous ~Amini~tration has posed a particular concern since it has been shown to s--",~li"l~sc result in serious hy~lLcllsivc events in plhl~Lcs (Cornish et al., 1993; Galbraith et al., 1994; Black et al., 1995). However, because the number of oligonucleotides that have been studied has been small, it is difficult to conclude at the time of making the invention whether all oligonucleotides share similar toxiritirC In particular, given the various ways of modifying the backbone of oligonucleotides (Wu-Pong, 1994, BioPharm 7:20-33) and their ability to fold into distinct three--liml-n~cit nzll structures (Stull and Szoka, 1995, Pharm.
Res. 12:465-483), Rando et al. J. Biol. Chem. 270; 1754-1760, 1995, the safety profile of dirrclcllL
oligonucleotides may be quite distinct.
Human Clinical Trials In addition to toxicological studies, efficacy studies should be carried out foroligonucleotide drugs. In the past, the preferred method of testing drug efficacy, especially in HIV-1 infected patients, was to monitor survival of treated patients. However, recent st~tictir~l studies have shown that a good indicator of anti-HIV drug efficacy is the reAllrti~n in the """ ,h~, ~
of copies of viral genome per unit of patient serum (viral load). Mellors et al. (1996) Science 272: 1167-1170. prAllrti~nc in viral load of 90%, or more plcrcl~ly 99% are desired. However, reAnrti- n~ of viral load of lesser pclccllL2lges can be useful, çspeci~lly where the trend of the overall treatment regime is con~i~trntly d~,wllw~L-d.
* * * * * *
Thus, there is a sllbst~nti~l need for antiviral drugs with novel rh~n~i~try and with sites of activity distinct from drugs ~lescllLly used. Most highly desired would be antiviral drugs whose effcacy in humans is known.

CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96/11786 SUMMARY OF l~; INVENTION
In one embodiment of the p}esent invention, there are provided methods and compositions useful in treating pathophysiological states caused by viruses, comprising atlmini~t~ring a ph:~rm~rological dose of an oligonucleotide, the dose being snffiri~-nt to inhibit production of the virus, wherein the oligonucleotide contains a high percentage of guanosine bases. In preferred embo~imr-ntc, the oligonucleotide has a three ~limrn~i~\n~l ~,Llu~;lule and this structure is stabilized by guanosine tetrads. In a further embodiment, the oligonucleotide compositions of the invention have two or more runs of two contiguous deoxyguanosines. In certain embo-limrnt~ of the present invention, the target virus is either herpes simplex virus, human immllno~--ficiency virus, human papilloma virus, human cytomegalovirus, adenovirus, and hrp~titi~ B virus.
In still yet another embodiment of the present invention, there is provided a gn~no~in~-rich oligonucleotide having a three dimensional .7llu~;Lule, wherein the three dimensional structure is stabilized by guanosine tetrads or at least two runs of two contiguous deoxyguanosines and wherein these oligonucleotides exhibit anti-viral activity. In a further embodiment, the oligonucleotides of the present invention have partially or fully phosphorothioated intrrmlrlro.~itlP linkages (backbones) or other rhrmir~l mndifir~tif,n~. In a further c;lllbo~ the oligonucleotides of the present invention have rhrmir~lly m~-t1ifit?d or ulul;lLul~l (synthetic) bases.

BRIEF DESCRIPIION OF T~; DRAWINGS
Fijgures for Section A
Figure A-lA shows a 1973 base pair Hind III to Eco Rl sub fragment of the Friend Murine T eukrmi~ Virus (FMLV) clone 57 genome. Figure A-lB shows a 172 base pair (HindIII to StuI) rla~ --L which is an expanded portion of the 1973 base pair fr~gmPnt Within this fragment is the purine rich target to which triple helix forrning olig-mlrleotitlrs are directed. Figure A-lC
shows the entire Hind III/Eco Rl FMLV rla~lllt;llL cloned into the pT7-2 plasmid (United States Biorh.omir~l Corporation) yielding p275A. In this recombinant the Hind III site is 10 base pairs downstrearn of the T7 mRNA start site. The 5' portion of the triple helix target region is 63 base pairs duwll~Llc~t;ll of the rnRNA start and the Dde I site is 131 base pairs duwl-~Ll~l- of the mRNA
start site. Figure A-lD shows the Hind III/Eco R1 FMLV fragment was cloned into pBS
(Stratagene) yielding pBSFMLV. The Hind III site, triple helix target site and Dde I site are .e~e.;~ively 50, 103 and 171 base pairs d-~w~Llc~ll from the mRNA start site.

All tables and figures are referred tO in the text with reference tO the section in which they are first described. Thus. in the section entitled, A.
General In vitro Studies ~ there are found nine tables Al-A9 and fifreen figures Figs. Al-A15.

W O 97/03997 PCT~US96/11786 Figure A-2 shows that G-Rich ph~sph~rothioated-oligonucleotides induced reduction in HSV-2 viral titer. VERO cells infected with HSV-2 were treated with various con~ntrations of the in-lir~t~d drug. The results are plotted as percent virus yield relative to VERO cells infected with virus but not treated with drug (titer = 1). The filled square (B106-62) (SEQ. ID. N0. 5) S represents a single cul-c~ dlion point (20 ~M) for this oligonucleotide. B106-96 is the fully phosphorothio~t~cl version of B106-62 (SEQ. ID. NO. 5). B106-97 is the fully phosphorothioated version of B106-71 (SEQ. ID. NO. 6). ACV (4a and 4b) is acyclovir tested against two dirr~ ,lL
stock concentrations of HSV-2 strain HG52. In two expe.illl~llL~" after virus infection and before reapplir~t;~n of oligonucleotide (BIO-96 or BIO-97), the cells were rinsed with a pH 3 buffer in order to remove all virus not yet int.orn~li7~d (96p3 and 97p3).
Figure A-3 shows MT-2 cells infected with 0.01 m.o.i. of HIV-l and then treated with various conrP~ dLions of oligonucleotide or AZT or ddC. The data represents the number of viable cells l~ i"i"g in the culture dish, i.e., not undergoing virus induced ~;ylu~dlllic effects (CPE). In this graph, 100% is the level of CPE occurring in cultures infected with virus but not treated with any drug.
Figure A4 shows the culture media taken from NIH3T3 cells chronically infected with FMLV was mixed with various con~llLldlions of I100-51 (SEQ. ID. N0. 29) or I100-12 (SEQ.
ID. N0. 27) (fully phosphorothioate version of I100-00 (SEQ. ID. N0. 20)). The ll~ Lulc~S were then assayed for the ~le~llce of viral reverse trans~ L~ The data is presented as a percent of measurable reverse Lldllscli~ldse in culture m~--1inm not treated with oligonucleotide.
Figures A-SA, A-SB and A-SC show the radio-labelled (32p) full-length or truncated mRNA
transc}ipts were analyzed by polyacrylamide gel electrophoresis, and then 4ll~ 1 by cutting out the specific lldlls~;lipL and m~ -ring the radioaclivily in a srintill~tion counter. Figure A-SA
shows that the re~ln~ti~m in full length ~ "~ directed by the T7 and T3 promoter when I100-51 (SEQ.ID. N0. 29) (anti-parallel triple helix forming oligonucleotide; FMLV2ap) was added.
Samples in which no oligonucleotide was added were counted and used as 100% Lldlls~ lion lcL~ llce points. In all other reactions 4 x lO~M of G101-S0 (SEQ.ID. N0. 12) (4e-6) was added and where in~ t.o~l G101-50 plus I100-51 at concelllldlions ranging from 2 x 10-9 to 2 x 10~ M
(2e-9 to 2e-6). Figure A-5B shows the reduction in full length l,,",~ ,t by I100-01 (SEQ.ID.
N0. 21) (FMLV2p). T7 directed LlcLllS~.;lilJL'~ were treated as in Figure A G101-50 was added to each reaction except the control (no oligo) with or without various con~entration of I100-01 or I100-ll(SEQ.ID. N0. 26) (26% G-ctl). Figure A-SC shows the analysis of truncated (63 base pair) lldllscli~t.
Figure A-6 shows inhibition of HIV-1 induced syncytia formation four days post-inf~ction SUP Tl cells were infected with HIV IDV for four hours and then treated with various cuncell~ldLions of oligonucleotides. Four days post-infection cells were scored for ~yll~;yLiulll formation. All assays were performed in quadruplicate and the average values used to plot this graph. The legend to the right of the graph in~lir~tes the symbol used for each oligonucleotide tested.
Figure A-7 shows cnntinn~od suppression of HIV-l p24 production seven days post removal ~ of oligonucleotide. Four days post-infection with HIV-lDV, the media from infected cells treated with oligonucleotides (2.5 ~LM) was removed and replaced with fresh media without oligonucleotide. The presence of viral p24 antigen was then assayed 7 and 1 l-days post infection.
All samples were assayed in quadruplicate and the average values used to plot this graph. I100-07 (SEQ. ID. NO. ~: I100-15 (SEQ. ID. NO. ~; I100-18 (SEQ. ID. N0. _); I-10021 (SEQ. ID.
N0. _). The legend to the right of the graph in~lir~tes the symbol used for each oligonucleotide tested.
Figure A-8 shows a Dixon Plot of random oligomlrleoti~lç 1232 (SEQ. ID. NO. 41) obr~inPd from kinetic analysis of inhibition of HIV-RT with respect to dNTP. The inhibition C~J~ lll Kj was dPt~rmin~-d by ~imlllf~neoucly varying dNTP (without dATP) conr~ ions at the same time as inhibitor (oligon--rl~otifle 1232). The Kj tl~. ",i"~ n was p~lr~lllled at 0.125 mM, 0.25 mM and 0.5 mM dNTP conrçntr~tion~ with c~ n~t~nt Primer-Template c~"-r~ lion of 0.2 pM. HIV-RT was used at 1 unit in each reaction. The reported values are the result of simlllt~neous intlrp.onfl.ont duplicates ~l~Lrl l l li l~i-l inn~
Figure A-9A reveals PBMCs derived from HIV-1 positive patients were mixed with HIV-1 negative PBMCs in culture medium ~;ollL~illillg drug I100-15 (SEQ. ID. NO. 33). On day 7 the cocultures were washed and lc~u~ellded in fresh medium cnnt~ining drug. The p24 levels in me linm collected on day 7 (before medium change) and day 10 were assayed for p24 Figure A-9B HIV-1 negative PBMCs from two dirr~lcll~ donors were infected with HIV-lDV and then inr~lb~trd in t_e presence of drug for 10 days at which time the culture me~illm was assayed for the presence of p24 antigen.
Figures A-lOA and A-lOB show inhibition of binding of V3 loop specific Mabs to HIV-1 gpl20 by phosrhrJrothioate L;IJlllilillillg oligonucleotides. M~trhe-1 seqnrnre oligonucleotides with either pho~pho-liPst-or (PD) or phosphorothioate (PT) backbones were assayed for their ability to inhibit the interaction of V3 loop specific Mabs with the gpl20 m~lecule: SEQ. ID. NOS. 31 (1173) and 32 (1174); SEQ ID. NOS. 24 (I100-07) and 39 (I100-21); or SEQ. ID. NOS. 42 (1229) and 43 (1230). To do this, immobilized gpl20 was preinruhated with oligonucleotides before washing and the addition of Mab NEA 9284 (panel A) or Mab NEA 9301 (panel B).
Figure A-11 shows a srh~m~tir diagram of the HIV-1 genome not drawn to scale.
~.

Figure A-12 shows analysis of DNA (PCR) and RNA (RT-PCR) extracted from SUP T1 cells three days post-infection with HIV-1. (Left Panel). PCR analysis of HIV-1 infected drug treated SUP T1 cell DNA used 0.1 ~cg of total extracted DNA for each reaction. In this ~ hlle either AZT, at 0.3 ~M which is 10 fold over the IC50 value (lane 1) or I100-15 (SEQ. ID. NO.
33) at 5.0 (lane 2) or 0.3 ,uM (lane 3) were added to SUP T1 cells at the same time as HIV-1.
Lanes 4 (AZT), 5 (5.0 ,uM I100-15 (SEQ. ID. NO. 33)) and 6(0.3 ,uM I100-15) are the results of DNA samples obtained from cells in which drug was added 8 hours post-infectinn Lanes 8 to 10 contain 10, 100 or 1000 ng of DNA extracted from HIV-1 infected control SUP T1 cells. The band corresponding to 220 bp is the predicted size of the internal ,~-actin control and the 200 bp fragment is the predicted size for the amplified portion of the HIV-1 genome. The right panel contains RT-PCR analysis of extracted RNA (1 ~g/reaction) obtained from cells treated in an nrir~l fashion as those described in lanes 1-6 of the left panel. Lanes 7 and 8 are control HIV-1 infected cell mRNA and lanes 9 and 10 are the results obtained using l-"il,r~ d untreated SUP T1 cell mRNA.
Figure A-13 shows the results of three oligonucleotides (10-sM) inrllb~trd with increasing eonr~ntrations (0,7.5,15,30,60 and 120 mM) of KCI (lanes 1-6 for I100-15 (SEQ. ID. NO. 33), 7-12 for I100-18 (SEQ. ID. NO. 36) and 13-18 for Z106-50). The nucleotide markers are poly dT.
Figures A-14A and A-14B show a line model and space filling model for I100-15 (SEQ.
ID. NO. 33). A line model (A-14A) of I100-15 folded into an intramolecular tetrad of the Oxytricha class is depicted. The 5'-end of the mol~cJlle is in the bottom left hand side. The bases (Gs) are stacked on top of each other with the 4 bases in each plane stabilized through their hydrogen bonding with each other and their interaction with the K+ ion complex in the center of the tetrad. The space filling model (A-14B) also has the 5' bases in the lower left hand corner.
The lighter colored atoms are part of the G-tetrad and the darker shaded atoms are part of the loop structures. The K+ ion is buried within the tetrad.
Figure A-15 displays a one ~lim~n.~ion~l NMR analysis of a KCl titration and thermal melting p~ for I100-15 (SEQ. ID. NO. 33).

Figures for Section B
Figure B-1. Dose 1~7~JO11~7iV~ profile for T30177, AZT and ddC. CEM-SS cells were infected with HIV-lRF(0.01 MOV and treated with various c~ " ,r~"l, illions of each drug for six days at which time the degree of HIV-1-induced ~,yll~:yLiulll f~ ti~n (~;y~opalllic effect, cpe) was addressed. The results shown are the ~t~tdg~ of three or more ~ ,;"~"~ with the standard deviations in-lir~tPfl CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 Figure B-2. Effect of T30177 on HIV-l replication in primary m--acrophages. Primary macrophages were obtained from PBMC L~ "~ innC and infected with HIV-lDV for 24 hours in the presence of the inrli~trd amount of drug. Seven days post-infPcti- n the inrrare~ r levels of p24 were qn~ntit~t~d using the Coulter p24 antigen capture ELISA kit. The results shown are the S averages of three or more exp~:lhllcllL~,.
Figure B-3. Effect of time of drug addition on the inhibition profile of T30177, AZT, and DS5000. MT4 cells, infected with HIV-lIIIB at a MOI of 1, were treated at various times during (time 0) or post-virus-infection with the test compounds at a conr-ontr~tion 100-fold higher than their l~ e~;iiv~ IC50 values. Viral p24 levels in the culture mrrlillm were monitored 29 hour post-infection. The results shown are the averages of three or more exp~lhl~llL~,.
Figure B4. HeLa-CD4-,6'-g~l~rt~ si~ e cell assays. (A) HeLa-CD4-,6'-g~l~rtQ~ e cells were inrnh~ted in medium c.",l~il.illg drug for one hour before virus was added to the culture me~linm One hour after the addition of virus the cells were washed e:~lcll~,iv~ly to remove unbound virus and extr~relllll~r test material. Forty-eight hours post-infection the cells were fixed and stained with X-gal. Blue mnltimlrlP~r cells were then counted under an inverted microscope (5). (B) HeLa-CD4-,15-g~l~rtQ~ e cells were inrnh~tel1 for 1 hour in the presence of test compound at which time an equal number of HL2/3 cells were added to each well. Cells were inrllb~ted for 48 hours at which time they were fixed, stained with X-gal and counted under an inverted microscope.
Figure B-5. Long term ~,uL*~lc~,.,ion of HIV-llIIB after Ll~:~Llll~llL of infected cell cultures with T30177. (A) MT4 cells were infected with 0.01 MOI of HIV-lIIIB and then cultured for 4 days int eh ~lesc~llce of T30177, AZT, DS5000, JM2763 or JM3100 using a conrentration of drug equivalent to 100-fold over the le,~e~;~iv~: IC50 value. After 4 days the cells were washed extensively and further inr~lh~trd in drug free mr~ lm The level of viral p24 antigen in the culture mr~lillm was monitored at various times after removal of drug from the infected cell cultures. The values given are the averages of three or more experiments.
Figure B-6. Single cycle analysis of viral DNA. CEM-SS cells infected with HIV-lSKI at an MOI of 1, were treated with T30177, UC38, CSB or ddC at the intlir~te-l time post-viral infecti~n Time 0 in-lir~tes the Llc;~L~ llL of cell cultures with drug during virus infecticn After 12 hours the DNA was extracted from the infected cells and used as a template for PCR. The conrçntration of drug used in each assay is equivalent to 10 to 100-fold over their respective IC250 values.
Figure B-7. Analysis of replicated viral DNA. CEM-SS cells were infected with HIV-lSKI
at an MOI of 1 and then treated with T30177. Fightrçn to 20 hour post-infection the low molecular weight Hirt DNA was analyzed using PCR primers which would amplify mitc!rh~ ri~

CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 DNA (A), early viral synth~-~i7t~d cDNA (C), viral gag cDNA(D) and viral 2-LTR circles (B). The drug c~ nrtontrations used were 0.0, 0.01, 0.1, 1 and 10 ~M corresponding to lanes 1 to 5 respectively. The unlabeled lane in each panel contains molecular size marker control DNA.

S Figures for Section C
Figure C-1. ~hibition of HIV-l integrase 3'-~locessillg and strand transfer and HIV-lRF
cytopathicity by gn~no~in~ quartets. (A) St~h--m~ti~ diagram showing 3'-processing (3'P, which liberates a GT dinucleotide) and strand transfer (S.T., which results in the insertion of one 3'-processed oligonucleotide into another target DNA), with 5'-end labeled (asterisk), blunt-ended oligonucleotide. (B) Left panel, Con~t?ntr~til n-response obtained from a typical experiment. The DNA substrate (21mer), 3'-processing product (19mer), and strand transfer products (STP) are shown. Lane 1, DNA along; lane 2, with hlLe~ Se; lanes 3-6, with integrase in the presence of the inllir~ttod concentrations of T30177. Right panel, graph derived from qn~ntit~tion (see Materials and Methods) of the dose response in the left panel showing inhibition of h.Le~ ,e-catalyzed 3'-processing (open squares) and strand transfer (filled squares). (C) Structures o g..~nnsin~ quartets oligonucleotides. (D) IC50 values for several a4 oligonucleotides against both activities of HIV integrase and HIV-1RF in cell culture. Insertions into the parent compound T30177 are shown by an jt~ i7~cl and nn~l~rlinPrl nucleotide while mnt~tionc are ~ ipn~t~d by a lower case nucleotide. The guanosines involved in the quartets are shaded and the loops are /1.ocign~t~-d by the corresponding ".1l"1,~ (see panel C, left).
Figure C-2. ~hibition of strand transfer and 3'-plOCeS:,illg activities of HIV-1 iuL~;l~lse by the guanosine quartet T30177. (A) Left, sc-l.rl,l~lic diagraln depicting the strand transfer assay using the precleaved oligonucleotide (19mer substrate). Right Phosphorimager picture showing inhibition of strand transfer with T30177. The DNA substrate (19mer) and strand transfer products (STP) are shown. Lane 1, DNA alone; lane 2, plus integrase; lanes 3-6, plus integrase in the presence of the intli~t~-rl c~-l,rellll~Lions of T30177. (B) Left, s~-hrlll~lic diagram depicting the 3'-processing assay using the oligonucleotide labeled at the 3'-end with 32P-col.lyce~.in (*A) (22mer substrate). Right, pho~pho,hna~,e~ picture showing inhibition of HIV-1 integrase-catalyzed 3'-processing with T30177. Lane 1, DNA alone; lane 2, with int~-gr~e; lanes 3-6, in the presence of the in~ t~rl c~nt~ntr~tinn~ of T30177.
Figure C-3. Inhibition of the DNA binding activity of HIV-l illLe~ se by g--~no~in quartets. DNA binding was l..easulèd after UV cros~linkin~ of reactions in which integrase was prein~lh~tf~d for 30 minutes at 30~C with the gl~no~in~ quartet prior to addition of the DNA
substrate. (A) Phosphorimager picture showing dirr~.ellLial inhibition of DNA binding with T30177 and T30659. Lane 1, DNA alone (20 nM); lanes 2, 8, and 14, with integrase (200 nM);

W097/03997 PCTrUS96/11786 lanes 3-7, in the plr~r~ e of the in-lir~t~d cf~nrentrations of T30177; lanes 9-13, in the presence of the in-lir~ttod concentrations of T30659. The mitig~tionc of proteins of known m- lecnl~r weight are shown to the right of the gel. (B) Graph derived from ~lu~lLiLd~ion of the does response in (A) showing inhibition of integrase binding by T30177 (open squares) but not by T30659 (filled S squares).
Figure C4. Dirrtl~lllial activities of T30177 on wild-type and deletion mutants of HIV
integrase. (A) S(~h~m~tit~. diagram showing the three domains of HIV-l int~gr~ce (B) Inhibition of wild-type IN1-288 (open squares), INl-212 (closed squares), and IN50-2l2 (open triangles) in the ~licinr~gration assay. (C) Binding of HIV-1 integrase wild-type (INl-288) and deletion mutants at a final concentration of 1 ~M to 32P-end labeled gn~noCin~ quartet T30177 at a final con~ entr~til~n of 250 nM. The mobility of plulrillS of known mol~cnl~r weight (in KDa) are shown to the right of each figure. Lane 1, T30177 alone; lanes 8-9, binding to wild-type, full-length HIV-l integrase (IN1-288) in the presence of the indicted metal; lanes 2-3, binding to IN1-212 in the presence of the in~ t~-d metal; lanes 6-7, binding to IN50-2l2 in the presence of the in~ tt-A metal are lanes 4-5, binding to IN50-288 Figure C-5. DNA binding activity of the zinc finger domain of HIV-1 int~gr~ce Binding of IN1-55 to T30177 or the viral DNA substrate (see Fig. lA, 21mer). Lanes 1, DNA alone (50 nM); lanes 2, IN1-55 (2 ~M) with no metal; lanes 3, INI-55 with m~ng~nPse (7.5 mM); lanes 4, INI-55 with m~"e~il"" (7.5 mM); lanes 5, INI-55 with l~ Psr (7.5 mM) and zinc (4.2 mM); lanes 6, IN1-55 with m~gnt~cillm (7.5 mM) and zinc (4.2 nM); lanes 7-10, INl-55 in the presence of the in-lir~t.-d CullCrlllldLiOn of zinc alone.
Figure C-6. Increased binding to and inhibition by guanosine quartets in m~gn~cillm versus m~ng~n~se. (A) Phosphorimager picture showing DNA binding of wild-type integrase to radiolabeled T30177. Lane 1, DNA alone (27 nM); lanes 2-5; binding of integrase (200 nM) in m~ng~n~se buffer to the in~ t.od co~ "l".lion of T30177; lanes 6-9, binding of integrase (200 nM) in m~gn~cillm buffer to the in~ t--cl c-)"~e"l ~ ,.1 ion of T30177. The migrations of proteins of known m~l.ocul~r weight are shown to the right of the gel. (B) Structures of T30177 and two analogs in which the int~-rmlcl~otidic linkages have been changed. (C) graph derived from qn~ntit~tion (see Materials and Methods) of the inhibition of i"l~",ce-catalyzed 3'-processing in the presence of T30177 and analogs in either m~gn~cillm or m~ng~n~se. Inhibition by T30177 (triangles), T30175 (squares), and T30038 (circles is shown either cl.lll;li~ g m~n~cinm (filled symbols) or m~ng~n~se (open symbols). (D) Table showing IC50 values for 3'-processing for the gn~nC).cin~o quartets in buffer colll;~il,;"g mz~ng~n~se and m~gn~cillm and the ratio of these values.
Figure C-7. Competition of binding to either US viral oligonucleotide (see Fig. C-lA, e 35 21mer) (A) or gn~nocin~o quartet T30177. (B) Lanes 1, DAN alone; lanes 2, with wild-type, full-length HIV-1 integrase. Lanes 3-6 in panel (A), with integrase in the presence of the inflir~tr~
c~ r~~ ion~ of T30177 added after a 5 minute preinrnh~tion with the US viral DNAoligonucleotide. Lanes 3-6 in panel (B), with integrase in the plcse,lce of the in-lir~t~-d co~ ,dLions of viral U5 DNA oligonucleotide added after a 5 minute preinrllh~ti-~n with the gn~nr ~in~ quartet T30177.
Figure C-8 Inhibition of the related lc~ vi~al integrases. (A) Inhibition of 3'-processing and strand transfer catalyzed by HIV-1 (lanes 2-8), HIV-2 (lanes 9-15), FIV (lanes 16-22), and SIV
(lanes 23-29) integrases in t_e presence of T30177. Lane 1, DNA alone; lanes 2, 8, 9, 15, 16, 22, 23, and 29, with integrase; lanes 3-7, 10-14, 17-21, and 24-28, with inL~,dse in the presence of the inflir~tr~l crnrrnfr~ti~n~ of T30177. (B) C~raph derived from ~ "~ inn (see Materials and Methods) of the dose responses in (A) showing inhibition of HIV-1 (open rect~nglrs), HIV-2 (filled rect~n~lrs), FIV (open rri~ngl~s), or SIV (filled triangles) int~gr~ce-catalyzed 3'-processing.
Figures C-9, C-10, C-13 and C-14. Three-~limrn~ n~l drawings of certain gn~nn~inr tetrad forming oligonucleotides referred to in Tables C-1 and C-2.
Figure C-11. Percentage inhibition of 3' processing by certain oligonucleotides in Table C-1.
Figure C-12. Tnhihifi~n of syncytium f~rm~tion by certain oligonucleotides in Table C-1.

Figures for Section D
Figure D-1. Structure Models. A. The seqllrnre and a structure model for oligonucleotides used in this study p lcscllLcd All four oligomers have been modified so as to include a single phosphorothioate linkage at the 5' and 3' trrmimls Proposed sites of G-quartet formation have been i~lentifi~d by dotted lines. The C~lllilluiLy of the phosphodiester backbone is i~l-ontifiP~1 by solid lines.
B. A two step kinetic model for ion induced folding of oligomers in this study. It is proposed that binding a first K+ or Rb+ ion equivalent, marked as a (+), occurs within the central G-octet, which has been i~lentifir~l by dotted lines. This first step is relatively fast, and is a~cori~tPd with higher d~l,alc"L ion binding affinity. It is also associated with ft~rm~tir,n of nn~t~rlrrcl loop ~lr,m~in~, and the resultant net loss of UV hypoch,u,,,i~,,,, as colll~ed to the initial random coil state. The second step in the process involves as many as two ~-lrlifinn~l K+ or Rb+ ion equivalents, (+), at the junction between the core octet and flanking loop regions. This second step requires .~ignifir~nt ordering of the flanking loop domains, and is therefor associated with an increase of base st~rking interaction, and a generally high activation energy.
Figure D-2. Thermal Stability of Oligomer Folding. Thermal dclldLuldLion of oligomers has been lllea~uled as a fimrtion of ion type, ion c~nrentration and strand cul~cc~L~diion. Data have been obtained at 240 nm, in 20 mM Li3PO4, pH 7, as the supporting buffer. Tm values were CA 02227867 l998-0l-26 W O 97/03997 PCTrUS96/11786 calculated from the first dclivaLivc of a plot of absorbance vs. Lclll~cldLulc, but similar values were obtained by using the midpoint of the overall absorbance change. A. Tm values for T30695 (curve a), T30177 (curve b), T30376 (curve c), and T30677 (curve d) obtained as a function of added 12 KCl concentration. B. The Tm Of T30695 obtained as a function of KCl, RbCI, NaCI or CsCl concentration. C. The strand concentration deplon-len~e of Tm has been measured at 1 mM of added KCl .
Figure D-3. Oligomer Folding Monitored by Circular dichroism (CD). CD data have been obtained at 25~C in 20 rnM Li3PO4 as a function of added ion cc,nccll~ld~iom Data have been ~lcscllLcd as molar ellipticity in units of dmole bases. A. The CD spectrum of T30695 in the presence of 0 mM (curve a), 0.05 mM (curve b), or 10 mM (curve e) of added KCl. B. The change in ellipticity at 264 nm, relative to that measured in the absence of added ion is ~lcscllLcd as a function of added KCI concentration for T30695 (curve a), T30177 (curve b) and T30676 (curve c). The overall midpoint of the measured KCI induced transition has been plotted for each oligomer: 0.02 mM, 0.15 mM and 0.27 mM, respectively C. T30695 has been treated with hlclea~illg c- nr.onfration of several dirrclcllL cations. The change in ellipticity at 264 nm was then measured as described in part B as a function of added KC1 (curve a), RbCI (curve b) or NaCl (curve e).
Figure D-4. The Kin.ofi~c of Ion Induced Folding. Ion was added to oligomers at time zero in the standard 20 mM Li3PO4 assay buffer. Data have been plescllLcd as absorbance (A) vs. time after addition of metal ion. A. Kinetics for T30177 were measured at three added KCl con~entrations: 0.2 mM (ewe a); 1.0 rnM (curve b); and 10 mM (curve e). B. Kinetics for T30695 were measured at three added RbCl concentrations: 1.0 mM (curve a), 5.0 mM (curve b) and 10 mM (curve e). For both, the data has been fit to a sum of two expon~ nti~l~ i.e. A(~) =
Alexp(-~/Tl) + A2exp(-~/T2).
Figures for Section E
Figure E-l. Mean arterial ~lcs~,ulc of cynomolgus monkeys pAor to, during and following intravenous ~.l . "i " i~ ion of ARl77 0ver ten minutes. Blood plcs~ule was contimlou~ly lllolliL~,lcd via an indwelling femoral artery catheter. The values are the mean _ s.d. of three monkeys at each dose.
Figure E-2. Neutrophil levels in blood of cynomolgus monkeys prior to, during and following intravenous ~rlmini~tr~tinn of AR177 over ten rninutes. Neutrophil levels were rlf~t~r~,nin~orl pre-dose (-10 rninutes), and at 10, 20, 40, 60,120 and 1440 minutes following the initi~til~n of the ten-minute infil~ion of AR177 into cynomolgus monkeys. The values are the mean i s.d. of three monkeys at each dose.

Figure E-3. aPTT versus time profile following a ten-minute infil~ion of AR177 to cynomolgus monkeys. aPl~ was rlPterrnin~d before and at various time after intravenous infusion of AR177 as described in the Methods section. aPTT levels returned to baseline by 24 hours in all groups. Certain aPTT values in monkeys at the 20 and 50 mg/kg dose time points, denoted by c S asterisks, e~eee~led the upper limit of the assay.
Figure E-4. Complement factor Bb cullcellLld~ion versus time profile following a ten -minute infusion of AR177 to cynomolgus monkeys. Bb was tletermined before and at various times after intravenous infusion of AR177 as described in the Methods section. Bb levels returned to baseline by 24 hours in all groups.
Figure E-S. CHS0 levels in blood of cynomolgus monkeys prior to, during and following hl~ldvenous ~lmini~tration of AR177 over ten minutes. CHS0 levels were determined pre-dose (-10 minutes), and at 10, 20, 40, 60,120 and 1440 minutes following the initi~tion of the ten-minute infusion of AR177 into cynomolgus monkeys. The values are the mean of two monkeys in the saline and 50 mg/kg groups, and three monkeys in the 20 mg/kg group. Data for the third monkey in the saline and 50 mg/kg groups, and for all of the 5 mg/kg group was not available.
Figure E-6. Plasma CmaX ~f AR177 in cynomolgus monkeys ~hll;ll~ d AR177 as a ten-minute intravenous infil~ion The plasma concellLld~ion of AR177 was ~ tern~ined by anion-exchange HPLC as ~lesrrihe-1 in the Methods section.
Figure E-7. AR177 plasma conf~Pntrati-~n versus time profiles following a ten-minute intravenous infilcion to cynomolgus monkeys . The plasma cv, ~ l d~ion of AR177 was ~ ferrninPd by anion-exchange HPLC as ~ sl~rihed in the Mefho ic section. The plasma AR177 concentration at 24 hours for the 5, 20 and 50 mg/kg groups were < 0.020 g/mL for the S and 20 mg/kg groups, and 0.24 + 0.42 ~/mL for the 50 mg/kg group.
Figure E-8. The relationship between plasma AR177 and aPTT in cynomolgus monkeysfollowing a ten-minute intravenous infusion of 5 mg AR177/kg. The plasma con~~~,lldLion of AR177 was determintod by anion-exchange HPLC as described in the Met~ section. The baseline aPTT level (at 10 minutes prior to dosing) was 32.1 _ 4.4 seconds (mean _ s.d.).Figure E-9. The rçl~tit~nchir between plasma AR177 and aPTT in cynomolgus monkeys following a ten-minute intravenous infil~ n of 20 mg AR177/kg. The plasma conrentration of AR177 was ~1etçrmin~d by anion-exchange HPLC as ~lesrrihed in the Methods section. The baseline aPTT level (at 10 minutes prior to dosing) was 41.6 + 6.7 seconds (mean + s.d.). The aPTT
value in monkeys at the 10 minute time point, denoted by an asterisk, e~ree~ d the upper limit of the assay.
Figure E-10. The rel~ti-)n~hir between plasma AR177 and aPTT in cynomolgus monkeys following a ten-minute intravenous infilsion of 50 mg AR177/kg. The plasma concentration of WO 97/03997 PCTrUS96/11786 AR177 was determined by anion-exchange HPLC as described in the Methods section. The baseline aPU level (at 10 minutes prior to dosing) was 33.2 + 4.8 seconds (mean + s.d.). Certain aPTT
- values in monkeys at the 10 to 120 time points, denoted by asterisks, e~l~eede-l the upper limit of the assay.
s Figures for Section F
Figure F-l. AR177 plasma c~-"~e~ ion after bolus IV dose 1 or 12 versus dose amount in Cynomolgus monkeys. Cynomolgus monkeys were given intravenous doses of 2.5, 10 or 40 mg/kg/day every other day for a total of 12 doses. Blood was obtained 5, 30 and 240 minutes following doses 1 and 12. The conr~ dLion of AR177 in the plasma of every monkey was determined by anion-exchange HPLC as dtosrrihed in the Methods section. There were six monkeys in the 10 and 40 mg/kg groups, and eight monkeys in the 40 mg/kg group. There was a linear relationship between each dose and the plasma Co~ liO~ that was achieved at each of the sampling times.
Figure F-2. AR177 plasma conrentration versus time profile following a bolus IV injection (dose 12) to Cynomolgus monkeys. Cynomolgus monkeys were given intravenous doses of 2.5, 10 or 40 mg/kg/day every other day for a total of 12 doses. This figure shows the Collc~nlldLion of AR177 in the plasma 5, 30 and 240 minutes following dose 12. The c~ l;on of AR177 in the plasma was detern in~d in every monkey by anion-exchange HPLC as described in the Method~
section. There were six monkeys in the 2.5 and 10 mg/kg groups, and eight monkeys in the 40 mg/kg group. There were no dlJpdlC~llL dirr~lence between the disd~e~ ce of AR177 from the plasma following the 1st (Figure F-3) and 12th doses.
Figure F-3. The relationship between the plasma AR177 co,~ dlion and aPTT in Cynomolgus monkeys following a bolus IV injection of 2.5 mg AR177/kg. Cynomolgus monkeys were given intravenous doses of 2.5 mg/kg/day every other day for a total of 12 doses. This figure shows the plasma AR177 c-~n(~nrr~tif~n versus aPTT levels 5, 30 and 240 minutes following doses 1 and 12. The concentration of AR177 in the plasma was ~ t~rrnin~l in every monkey by anion-exchange HPLC as described in the Methods section. There were six monkeys in the 2.5 mg/kg group. The baseline aPTT levels just prior to (pre-dose) doses 1 and 12 were 24.1 + 3.4 seconds and 22.1 + 2.2. There was no change in the aPTT levels at any of the time points after the 1st or 12th doses of AR177 at 2.5 mg/kg.
Figure F-4. The relationship between the plasma AR177 con~entr~tinn and aPTT in cyn~ m~lgllc monkeys following a bolus IV injection of 10 mg AR177/kg. Cynomolgus monkeys were given intravenous doses of 10 mg/kg/day every other day for a total of 12 doses. This figure shows the plasma AR177 c~ nr~ntration versus aPTT levels S, 30 and 240 minutes following doses CA 02227867 l998-0l-26 WO 97103997 PCT~US96/11786 1 and 12. The concentration of AR177 in the plasma was determined in every monkey by anion-exchange HPLC as described in the Methods section. There were six monkeys in the 10 mg/kg group. The baseline aPTT levels just prior to (pre-dose) doses 1 and 12 were 23.3 + 1.8 seconds and 21.6 + 2.2. There was a close correlation between the aPTT] levels after the 1st or 12th doses of AR177 at 10 mg/kg and the aPTT levels.
Figure F-S. The relz~til~nchir between the plasma AR177 concentration and aPTT in cynomolgus monkeys following a bolus IV injection of 40 mg AR177/kg. Cynomolgus monkeys were given intravenous doses of 10 mg/kg/day every other day for a total of 12 doses. This figure shows the plasma AR177 cu,lr~ ,dlion versus aPTT levels S, 30 and 240 minutes following doses 1 and 12. The ccncc~lL.dLion of AR177 in the plasma was A~tt-rmined in every monkey by anion-exchange HPLC as described in the Methods section. There were eight monkeys in th~e 40 mg/kg group. The baseline aPTT levels just prior to (pre-dose) doses 1 and 12 were 24.8 + 3.3 seconds and 22.5 + 2.5. Certain aPTT] values in monkeys at the 20 and S0 mg/kg dose time points at five minutes following doses 1 or 12, denoted by asterisks, exreeA~d the upper limit of the assay. There was a close correlation between the aPTT levels after the 1st or 12th doses of AR177 at 40 mg/kg and the aPTT levels.

Figures for Section G
Figure G-1. AR177 pharmacokinetics following a single IV dose of 0.75 mg/kg to humans. Four HIV-positive human patients were ~A " ,; ~ lcd AR177 at 0.75 mg/kg as a two-hour intravenous (IV) infilsion. Blood samples were collected in EDTA-coated tubes at various time points during and following the IV infilcil~n Plasma was obtained following low speed centriguation of the blood. The con~çntration of AR177 in the plasma was det~rmin~d using a validated anion-exchange HPLC method.
Figure G-2. AR177 phs1""~okin~tirs following a single IV dose of l.S mg/kg to hllm~nc Four HIV-positive human patients were ~A~ lrl~d AR177 at 1.5 mg/kg as a two-hourintravenous (IV) infilcinn Blood samples were coll~ctod in EDTA-coated tubes at various time points during and following the IV infilcion Plasma was obtained following low speed centriguation of the blood. The c~ ,alion of AR177 in the plasma was det~rmin~A using a validated anion-exchange HPLC method.
Figure G-3. AR177 phdlll,acokin~tirc following a single IV dose o 3.0 mg/kg to h~ nc Two HIV-positive human patients were aA",i"i~irl~d AR177 at 3.0 mg/kg as a two-hour hlLldve,lous (IV) infilcinn Blood samples were coll~ctl-A in EDTA-coated tubes at various time points during and following the IV infilci~n Plasma was obtained following low speed WO 97/03997 PCTrUS96/11786 centriguation of the blood. The col~c~ dLion of AR177 in the plasma was determined using a validated anion-exchange HPLC method Figure G-4. AR177 ph~rm~okin.oti~ following a single IV dose of 0.75, 1.5 or 3.0 mg/kg to humans. Ten HIV-positive human patients were administered AR177 at 0.75, 1.5 or 3.0 mg/kg as a two-hour intravenous (IV) infiusion. Blood samples were collected in EDTa-coated tubes at ~ various time points during and following the IV infil~ion Plasma was obtained following low speed centriguation of the blood. The c~-n~el,l, ~Lion of AR177 in the plasma was ~l~ot~rmin~-d using a validated anion-exchange HPLC method.
Figure G-5. AR177 Tl/2 and C.~IAX following single doses to humans. HIV-positive human patients were ~lmini~t~red AR177 at 0.75, 1.5 or 3.0 mg/kg as a two-hour intravenous infiusion.
The c- n~-ontration of AR177 was fll-tt?rmint-d in the plasma using a validated anion-exchange HPLC
method. The CMAX (m~im~l plasma conc~ Lion of AR177) and plasma Tl/Z (half-life of AR177 in plasma) were cl~t~rminf d using PKAnalyst software (Micro Math, Salt Lake City, UT).
Figure G-6. AR177 flt?~r~n~ following single doses to hurnans. HIV-positive human patients were ~.h~ .ed AR177 at 0.75, 1.5 or 3.0 mg/kg as a two-hour intravenous infilcion The con~entration of AR177 was dPt~rmin~d in the plasma using a validated anion-exchange HPLC
method. The plasma clearance was ~l~lr~ using PKAnalyst software (Micro Math, Salt Lake C~ity, Ulj.

DETAILED DESCRIPTION OF THE INVENTION
INDEX TO DETAILED DESCRIPTION OF THE INVENTION
Definitions .. ............................... ............................ 19 A. General In Vitro Studies ........................................... 21 B. Specific In Vitro Studies and In Vitro HIV Inhibition Using T30177....... ..... 47 C. Site of Activity Studies-Viral Integrase Inhibition ..................... 71 D. Structure-Function Studies ......................................... 88 E. Single Dose Ph~""~okin~ti~ Studies .................................... 99 F. Repeat Dose ph~rTn~r(~kin~tic Studies ................................... 112 G. Human Clinical Trials............................................... 127 Definitions The following terms as defined will be used in the description of the invention:OLIGONUCLEOTIDE. The term "oligonucleotide" as used herein is defined as a molecule co~ ised of two or more deoxyribonucleotides or ribonucleotides, preferably more than ten. Its exact size will depend on many factors inr~ ing the specificity and anti-viral activity of the -CA 02227867 l998-0l-26 oligonucleotide for various viruses. In addition, bases can refer to ululdLuldl (synthetic) bases used in place of an A, C, T or G.
BASE. In referring to "bases" herein, the term includes both the deoxyribonucleic acids and ribonucleic acids. The following abbreviations are used. "A" refers to adenine as well as to its deoxyribose derivative, "T" refers to thymine, "U" refers to uridine, "G" refers to guanine as well as its deoxyribose delivdLiv~, "C" refers to cytosine as well as its deoxyribose d~livdLive. A
person having oldillaly skill would readily recognize that these bases may be mr-lifie~l or delivdLiG~:d to u~Lhlli~ the m--thn~lc of the present invention. In addition, bases can refer to ulllldLuldl (synthetic) bases used in place of an A, C, T, or G.
INHIBITION. The term "inhibition" of viral replication is meant to include partial and total inhibition of viral replication as well as de~ ;ases in the rate of viral r~rlie~tinn The inhibitory dose or "therapeutic dose" of the colll~uullds in the present invention may be determined by ~csecSing the effects of the oligonucleotide on viral replication in tissue culture or viral growth in an animal. The amount of oligonucleotide a~iminict~ored in a therapeutic dose is dependent upon the age, weight, kind of COll~;ullellL llcdLlll~:llL and nature of the viral con-litinn being treated.
PHARMACOLOGICAL DOSE. The term "pharmacological dose" as used herein refers to the dose of an oligonucleotide which causes a pl~ rological effect when given to an animal or human. The rh~rm~ological dose introduced into the animal or human to be treated, will provide a 5nffiriPnt quantity of oligonucleotide to provide a specific effect, e.g., (1) inhibition of viral protein or el.~yllles, (2) inhibition of viral-specific replication, (3) plcv~lllillg the target site from functioning or (4) ~l~m~ging the duplex DNA at the specific site or (5) ablating the DNA at the site or (6) inhibiting the L dlls~ Lion/tr~ncl~tinn of the gene under the regulation of the site being bound or (7) internal inhibition of transcription or translation of the gene cont~ining the seqnenr~.
One skilled in the art will readily l~cog~ r that the dose will be dependent upon a variety of p~ ."~ , in-~hlfling the age, sex, height and weight of the human or animal to be treated, the organism or gene location which is to be ~tt~rk~(l and the location of the target seqnen~e within the Ulg~ lll. Given any set of p~r~m~t~rs~ one skilled in the art will be able to readily ~l~L~ the d~plU~ Lt~ dose.
PATHOPHYSIOLOGICAL STATE. The term "pathophysiological state" as used herein refers to any abnûrmal, lln~iesir~hle or life-thre~tening condition caused directly or indirectly by a virus.
GTO. The term "GTOs" means an olignm~ oti~l~ in which there is a high ~e.~;~llLdge of deoxygu~nocin~, or u~nt~inc two or more segm~?ntc (runs) of two or more deoxygll~nncin~ residues per se~ nL.
GUANOSINE TETRAD. As used herein, the term "gu~n~sinP tetrads" refers to the structure that is formed of eight hydrogen bonds by coordination of the four o6 atoms of guanine with alkali WO 97/03997 PCT~US96/11786 cations believed to bind to the center of a quadruplex, and by strong st~r~ing intl-r~rrion~. Of particular interest to the I100-15 class of GTO is the structure of the telomere sequenre repeat T4G4, first ~l~tect.-rl in Oxytricha. The oxytricha repeat has been studied in oligonucleotides by NMR and by crystallographic methods. See Smith et al., Nature, 1992, 356: 164-68, and Kang et al., Nature, 1992 356:126-31. As predicted from llUlll~lVUS previous physical and biorhPmi~l studies, both the NMR and crystallographic studies suggest that folding is m~ t~d by square planar Hoogsteen H-bonding among G-residues, with overall ~nt~ r~llel orientation of the four strand equivalents cu~ isillg the tetrad fold. As f~7~pect.o-1 the crystallography has shown that the structure is selectively stabilized by tight binding of a small monovalent cation to the o6 oxygen of guanosine.

The following examples are offered by way of illustration and are not int~-ntl~d to limit the invention in any manner.

A General In Vitro Studies The present invention provides m~thofl~ and compositions for treating a pathophysiological state caused by a virus, comprising the step of ~(l",i"i~l~.i"g a ph~rm~rol~gical dose of an oligonucleotide, the dose being sufficient to inhibit the replication of the virus, wherein the oligonucleotide contains snffiri~ont contiguous gn~n- sin~s so that a ~ ui~ .~ tetrad (inter- or intra-molecular) can form, and the three tlim--n~ n~l structure of the oligonucleotide is stabilized by gll~no~in~ tetrads formed at ~ld~e~;ic locations. Generally, this method of treating a virus-induced pathophysiological state may be useful against any virus. More pler~ldbly, the methods of the present invention may be useful in treating pathophysiological states caused by viruses such as herpes simplex virus, human papilloma virus, Epstein Barr virus, human immnnodl~fit~i~nf y virus, adelluvilus, lc~ hdLvl,~ syncytial virus, hPp~titi~ B virus, human cytom~-g~l-)virus and HTLV I and II.
Generally, the oligonucleotides of the present invention contain a percentage of gll~no~in~
bases high enough to ensure anti-viral efficacy. The ~ ~ is hll~ulkulL in forming tetrads which stabilize the three ~lim~on~ic)n~l structure of the oligonucleotides. Thus, the oligonucleotides of the present invention may have any percentage of gn~n~in~ bases which will allow for tetrad formation provided that the oligonucleotide exhibits anti-viral activity. Preferably, the oligonucleotides of the present invention contain two or more segm,onr~ of two or more gn~no~in-o bases, and an overall high ptlcel~Ldge of G in order to enable the oligonucleotide to form at least one qu~no~in~ tetrad.

CA 02227867 l998-0l-26 Generally, the oligonucleotides of the present invention may be capped at either the 3 ' or the 5' L~ ,i""~ with a mr~flifier. Preferably, the m~lifier is selected from the group con~icting of polyamine or similar compounds that confer a net positive charge to the end of the molecule, poly-L-lysine or other similar compounds that enhance uptake of the oligonucleotide, cholesterol or similar lipophilic compounds that enhance uptake of the oligonucleotide and propanolamine or similar amine groups that enhance stability of the m-)lec lle. t The phosphofliester linkage of the oligonucleotides of the present invention may be modified to hll~LOv~ the stability or increase the anti-viral activity. For example, a phosphodiester linkage of the oligonucleotide may be m~ ~lified to a phosphorothioate linkage. Other such modifications to the oligonucleotide backbone will be obvious to those having ~,l.lhlaly skill in this art.
The present invention also provides specific metho-lc of treating viral states. For example, the present invention provides a method of treating a pathophysiological state caused by a virus (in preferred embodiments, as specific virus such as, herpes simplex virus, human papilloma virus, Epstein Barr virus, human immnnndeficiency virus, adenovirus, lC;~ Oly syncytial virus, hrp~titic B virus, human cytomegalovirus and HTLV I and II), culllL,lisillg the step of ~lmini~tering a ph~rm~rological dose of an oligonucleotide, the dose being sufficient to inhibit the replication of the virus, wherein the three ~limen~ n~l ~Llu~;Lul~ of the oligonucleotide is st~hili7~-~l by the formation of gn~nn~in-o tetrads.
This invention ~ clt~ses a novel anti-viral technology. The total number of antiviral mPch~ni~m~ by which oligonucleotides, and ~sreci~lly G-rich oligonncleoti~hs~ work is not completely known, although the hlv~llLol~ have at least narrowed the sites of action as to certain oligonucleotide drugs as will be seen below. However, in the dirr~ virus culture systems listed above, G-rich oligonucleotides were able to ~ignifir~ntly reduce virus production in each. More importantly, actual human clinical studies have ~1ernonctrated the efficacy of the drug in reducing viral replicons in AIDS patients. The present invention is also drawn to oligonucleotides that have three ~limPn~inn~l structures stabilized by the formation of gn~nl~ine tetrads.
The present invention ~le",-",~ s poly and/or oligonucleotides inhibit growth of HIV-1, HSVl, HSV2, FMLV and HCMV and other viruses if the mn]ecllle cont~in~ a high ~cl-;~llL~ge of ribo- or deoxyribogu~nl-cine. The rest of the molecule is composed of thymine, cytosine, xanthosine or adenine nucleotides (ribo- or deoxyribo-), their d~liv~lives, or other natural or synthetic bases. The 5' and 3' termini of the oligonucleotide can have any ~tt~rhm.ont which may enhance stability, uptake into cells (and cell nuclei) or anti-viral activity. The backbone which connects the nucleotides can be the standard phnsrhc~liester linkage or any mn~ifir~ti~m of this linkage which may improve stability of the molecule or anti-viral activity of the m~lecllle (such as a phl~sphQrothioate linkage).

CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 Structural forrnulas for lc~ L~Liv~ G-rich oligonucleotides ~ sed in the instantinvention are listed below in Table A-1.

TABLE A-1.
S
SEQ ID NO 5(B106-62) 5'-~ LggLg~,L~ lLg~L~,~L~,~LLLggggg~Lyggg-3' SEQ ID NO 6(B106-71) 5'-gL~;~ lL~,~Lg~Lg~ L~ L~ g~,LLLggg~Lgggggg-3' SEQ ID NO 21(I100-01) 5'-tg~,Lgg~L~L~,Lggg~ g~,L~,LL~ ggg~lL~LL~gLgg~,L~,Lgg-3' SEQ ID NO 24(I100-07) 5 '-~ Lg~,Lgg~,L~ggL~,g~L~,Lgg~,Lg~L~,~LL~ Lgg~ Lgg~Ly,~L~-3 ' SEQ ID NO 28(I100-50) 5'-g~Lg~Lggg~,L~,LL~LLgggg~,LL~,-3' SEQ ID NO 29(I100-51) 5'-g~;lg~Lgggy,Lg~LL~LL~,ggg~ lL~LLgg~Ly,L~,Lgg~;Lg~L-3' SEQ ID NO 26(I100-11) 5'-g~L.~ L~,L~;~Lgaca.;Lyc~l~g,,l. cg;.l~ c~gtcgatg-3' SEQ ID NO 12(G101-50) 5'-~Lgg~L~LLL~,4 Lg~ LL~L~g~LLLl-3' SEQ ID NO 13(G105-50) 5'-~ggggggg~:4 Igggggggg~L4 L~yLgg-3' SEQ ID NO 14(G106-50) 5'-g lgg~Lgg~ LL~gggg~Lgg~ Lgggg-3' SEQ ID NO 15(G109-50) 5'-L~,gg~,LL4 g~;Lggggg~ LLgg~Lg~LL~-3' SEQ ID NO 16(G110-50) 5'-~Lg~L~L~Llg~,L~ L~l~-3' SEQ ID NO 17(G113-50) 5 '-g~,Lgggggg~LLg~ L~L~lLL~-3 ' SEQ ID NO 1(A100-00) 5'-tgg~,Lg~g~L~gg~,L~ggggg~,L~,Lggg~L~,Lgg~4-3' SEQ ID NO 2(A100-50) 3'-4g~Ly,~g~,Lggg~ gggg7~L~L~gg~Ly,L~gg~L~,-5' SEQ ID NO 4(A101-00) s'-g~g~g~ &~ gg~g~g~ gg~ gg-3 SEQ ID NO 18(HIV26ap) 5'-~ gggg~g~Lggg~,lggg~L~g~L-3' SEQ ID NO 19(HIV26ctl) 5'-gg~Lgg~,L~g~L~gg4;g~4;gyL~g-3' SEQ ID NO 9(B107-51) 5'-ggLggg~ L~,~Lg~Lg~,LLg~,ggggggggg~L-3' SEQ ID NO 10(B133-55) 5'-g~Lg~l~ggggg~lggggggg-3' SEQ ID NO ll(B133-55) 5'-gg~Lggg~,Lg~;Lgg~,L~gggg-3' SEQ ID NO 20(I100-00) 5'-gLL~,gg~,~ LL~LL~;~Lggg~Lg~L~g-3' SEQ ID NO 27(I100-12,PT) 5'-~ lL~ggg~,LL~,LL~,~L~gggL~,L~g-3' SEQ ID NO 22(I100-05) 5'-l~,~,Lg~L~ 4 L~gggggL~LLgggg~Ll~,LLg~L~gg~L~gLgg-cHoL
SEQ ID NO 23(I100-06) 5'-~L~,~,Lg~,Lgg~,L~,g~L~;~Lgg~,Lg~,L~,L4 Lgg~,L~ggLg.~Lg-CHOL
SEQ ID NO 25(I100-08) 5'-~ LLgggg~ Lggg~ 7gL~,g-cHoL
SEQ ID NO 3 5'-gg~,L~~,ggL~,g~L~g~Lgg-3 SEQ ID NO 30 5'-gg~,L~,~,LL~g~,L~~,~,LLgg-3~
SEQ ID NO 31(1173) 5'-gg~,L~g~,Lgg~,Lgg~ L~g-3' SEQ ID NO 32(1174,PT) 5'-gg~,Lgg~,L~g~L~g~ L~g-3' SEQ ID NO 33(I100-15) 5'-~,L ~L~g~ Lgg~,Lgg~L-3' SEQ ID NO 34(I100-16) 5'-.~L~,Lgg~,Lgg~,L~,g~LggLgggLggl-3' SEQ ID NO 35(I100-17) 5'-~ Lgy,4), ~,Lgg~LgggL~,~,L~"~,L~,L~,~ LL~,L~,ggL-3' SEQ ID NO 36(I100-18) 5'-LL~Lgg~Lgg~Lg~l~-3' SEQ ID NO 371I100-19) 5'-Lg~Lgg ,L~,~,L~,~ LL~Lgg~,L~,g~,Lg~,Lg-3' SEQ ID NO 38(I100-20) 5 '-~ L~ g~,L~,g~,LggLgg~Lg~,Lg~,LL~L~g~,L~,g~,LggLg-3 ' SEQ ID NO 39(I100-21,PT) 5'-~,Lg~ L~,g~Lgg~ Lgg~L~,L~,g~,L~,~ L~,gLL~,Lgg~,L~,g~,L~,~,Lg-3' SEQ ID NO 40(1231) 5'-gdL~;c~L~,Lc~,Lgacac.-3' SEQ ID NO 41(1232,PT) 5'-~ gL .~gtg~ r-3' SEQ ID NO 42(1229) 5 '~cCCcCccccccccCCC-3 ' SEQ ID NO 43(1230,PT) 5'-CCCCCCCCCCCCCCCCCC-3' SEQ ID NO 44(1198) S'-Lk~aLLL~gg?~r-c~;LI~g~ lg~ Lg5~ (ggCC~k;~LLLLdC-3 SEQ ID NO 45(1200) 5'_gt~ g~rggcca-3' SEQ ID NO 46(I100-25) S'-~L~,~,Lgg~,L~ g~ L~ggg-3 ' SEQ ID NO 47(I100-26) 5'-~,Lg~,Lgg~Lgg~,Lg~,g-3' SEQ ID NO 48(I100-35) S'-Lg~,Lgg~,Lgggtgggt-3' W O 97/03997 PCT~US96/11786 SEQ ID NO 49(I100-Z7) 5'-~L~,L~,g~ Lgg~L-3' SEQ ID NO 50(I100-28) 5'-~L~,L~,ggL-3' SEQ ID NO 51(I100-30) 5'-~ Lgg~L~ggtgggt-3' SEQ ID NO 52(I100-29) 5'-~Lgg~,Lgg~L-3' S

In viral yield re~lllrti~n assays, Vero cells (4 x 104 cells/tissue culture well) were inrllhat--d with oligonucleotide(s) for 14 hours before the oligoml~ leotit1~ was removed and virus (HSV-2 strain HG52) was added to the cells at a multiplicity of infection (m.o.i.) of 0.1 to 1.0 (4 x 10~ to 10 4 x 104 PFU). The infection was allowed to proceed for 10 minutes after which the cells are washed and fresh media, cQ~ i "g the same oligonucleotide was added for an ~irliti~n~l 14 hours.
Then, the cells were subjected to a freeze/thaw lysis after which the released virus was titered.

H[V-1 CULTURE ASSAY
The SUP T1 T lymrhom~ cell line was infected with HIV-1 strain DV at a mnltiplicity of inf~cti~m (m.o.i.) of 0.1 for one hour at 37-C. After the inf~ctit~n, free virus was washed off and the newly infected cells were plated (5 x 104 cells) in quadruplicate in 96 well plates that had been prepared with various ~lihlt;Qnc of oligonucleotide. The final cul .r~"l, alion of drug varied between 0.1 and 20 uM. After 3 days of in~llb~ti~n at 37 C, the plates were scored for the presence of 20 mnltimlcl~tt~-l giant cells (syncytia).
In assays ~ cignlod to inhibit syncytia formation, a number of oligonucleotides exhibited anti-HIV-1 activity The oligonucleotides and their IC50 are listed in Table A-2. I100-05 is the same as I100-01 with a cholesterol group attached to the 3' end via a triglycyl-linker. I100-08 is the same as I100-00 with a cholesterol group att~h~od to the 3' end via a triglycyl-linker. I100-07 25 was ~ecign~(l as a se~ln~n~e isomer to I100-01 and I100-06 is the cholesterol deliv~Liv~ of I100-07.
A100-00 is the same sequence in the opposite ori~nt~tion to HIB38p (A100-50). I100-07, crigin~lly ~1~cign~d as a control for I100-01 to be used in anti-FMLV ~clh~ lL~, was the most efficacious oligonucleotide tested against HIV-l.
In other ~ eli~ llL~, the HIV-l strain LAV was used to infect MT-2 cells at an m.o.i of 30 0.01. After 7 days, these cells were scored for ~,y~u~Lhic effects (CPE). In anti-HIV-1 assays in which MT-2 cells were infected at an m.o.i. of 0.01, several G-Rich oligonucleotides were able to inhibit viral-induced ~;yL~)aLlliC effects with t rrc~ iv~ dose 50's (ICSOs) in the 0.5-1.0 uM range (Figure A-3). The oligonucleotides shown in Figure A-3 were effective in the 0.5 to 1.0 uM
range, in~ ing A100-00 (HIV38p) and A100-50 (HIV38ap), A101-00 (HIV38ctl), HIV-26ctl.
35 The oligo-nucleotide HIV-26ap exhibited less efficacy in this assay with an IC50 in the 5 to 10 uM

CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 range. In Figure A-3, TE ic~lcS~llL~ buffer alone, i.e., no drug, while AZT and ddC are control drugs.

S IC50 for oligon~rl~ot~ c in an anti-HIV-1 ~7yll ~Lid formation assay.
G-Rich oli~onucleotide IC50 I100-00 3.75 ~M
I100-01 4.50 ~M
10I100-05 3.25 ~M
I100-08 3.25 ~M
I100-06 0.70 ~4M
I100-07 0.25 ~M
A100-00 3.25 ~M
Fl~ILV CULTURE ASSAY
Friend Murine T Pnkrmi~ Virus (FMLV) was grown in a chronically infected murine fibroblast cell line (pLRB215) or was prop~g~tt-d in an acute assay system by infection of NIH3T3 cells. When the chronically infected cell line was used, pLRB215 cells were split (1 x 105) into 24 well culture dishes and i~ d 16 to 20 hours at 37~C. The media was then removed and replaced with media culll~illillg various conrentr~tion~ of oligonucleotide. After l, 3 or 5 days, culture media was assayed for the pl~sence of the viral reverse Lldlls~ ldse enzyme.
In acute assays, NIH3T3 cells were split (1 x 104) into 96 well dishes and allowed to inr~lh~tr for 16-20 hours. After inrllh~tion, culture media was removed and cul,r.~ ,dl~d virus stock (10 ul) was added to each well in 100 ul of completed media cn,.li.;"i,.g 2 ug/rnl polybrene.
The virus infection was allowed to proceed for 18 hours at which time the virus co"l;~i"i"g media was removed and complete media co,~ ~illillg various c~ rrll(~ ions of oligonucleotide was added.
After 4 to 7 days, the culture media was assayed for the presence of viral reverse llalls~ Ldse.

HCMV CULTURE ASSAY
Human cytomegalovirus was cultured in the human diploid lung fibroblast cell line MRC-5.
These cells were split and placed into 24 well culture dishes and preinrnh~t~d for 24 hours with various cullcc:llLldLions of oligonucleotide (0.5 to 20 uM) in complete media. The oligonucleotide was then washed off and virus was added to the cells (~ xilll;11~1y 0.1 m.o.i.) for 2 hours at 37~C The virus was then lc~ v~d and complete media c~-"l~i"i"g the same co~-r~"l~dLion of oligonucleotide was added. Cells were then placed at 37~C for 10-12 days at which time virus in ~ the culture media was titered using a standard agar overlay procedure.

In this assay system, a 2 kb fragment (HindIII to EcoRl) of the FMLV virus (clone 57) was molecularly cloned between the HindIII/EcoRl sites 10 bp dowll~Llca ll of the bacterial T7 promoter (p275A) or 50 bp dowll~Llcalll of the bacterial T3 promoter (pBSFMLV2). A sr.hPm~ti~
S l~l~S~ lion of these two recombinant plasmids can be seen in Figure A-1. Isolated recullll,hl~llL
DNA was then digested with DdeI. Oligonucleotides were then inrnb~3tf~d with the digested DNA
and the mixture was subjected to in-vitro 1~ ,Lion using either the T7 or T3 b~rtPri~l enzymes.

REVERSE TRANSCRIPTASE ASSAY
In this assay, reverse Lldns~ L~lsc (either MMLV or FMLV from pLRB215 culture media) was in-~ub~tPd with various cnn~Pnrr~ti--n~ of oligonucleotide and then assayed using the enzyme linked oligonucleotide sorbent assay (ELOSA), the ELOSA kit which is c~lJlllelci~lly available from New Fngl~n(l Nuclear.

EUKARYOTIC IN VITRO TRANSCRIPTION
In this assay, a lccollllJill~lL plasmid cont~inin~ the HSV-1 IE175 plUlllO~Cl fused to the bacterial chlor~mrhPnirol accLylLl,..,.~r~ e gene (CAT) was linearized and used as a template for run off Ll~ls~ ,Lion studies. C~,ll"ll~.cially available HeLa cell nuclear extracts or prepared nuclear extracts of HSV-2 infected VERO cell were used.
..
I~I~IllON OF HSV-2 A(:llVllY
The oligonucleotide B106-62 was ~lrigin~lly ~IeS;gnP(1 to form a triple helix structure with a portion of the promoter region of the major immP~i~tP. early protein of HSV-2 (IE175). The phosphorothioate derivative of two oligonucleotides were 5ynthP~i7~cl and tested for anti-viral activity against HSV-2. Figure A-2 shows that the B106-62 oligonucleotide at 20 ~M was able to reduce viral titers by a~ xi",~l~ly 20% whereas the phosphorothioate version (B106-96) reduced virus by 50% in the submicromolar C~ alion range. The control oligonucleotide (B106-97), the rhl~sphorothioate backbone dc~iv~lLivc of B106-71, was also able to inhibit virus at the same levels as B106-96. Even when an c,~Lel.. ivc washing procedure at a pH of 3.0 was employed to remove excess virus not intPrn~li7ec1 during the infPcSi~n, in~nb~ti~m with both B106-96 and B106-97 was able to ~i~nifir~ntly reduce virus yield. Thus, the il~vc.lL~ . con~lllflP~l that the ",Prl~ ", of anti-viral activity was not merely a blocking of the adsorption of HSV-2 virions to cells.

WO 97/03997 PCTrUS96/11786 Figure A-2 also shows the results of acyclovir in the same molar range as the oligonucleotides. Acyclovir was tested against two dirr~iellL stocks of HSV-2 strain HG52, as illustrated in Figures A4a and A-4b.
~.
S OLIGONUCLEOTIDE SYNTHESIS
All oligonucleotides used in these examples were synthe~i7ed on a DNA synthr~i7rr (Applied Biosystems, Inc., model 380B or 394), using standard phosphoramidite methods. All oligonucleotides were syntheci7~od with an arnino mn-lified 3'-trrrnin~l, which resulted in the covalent ;.~ 1""~.,l of a propanolamine group to the 3'-hyd~ yl group or resulted in a rholesterol moiety att~rh~cl to the 3'-terminal via a triglycyl-linker. Oligonucleotides used in this example were capped at their 3'-terminal with either a propanolamine or a cht-l~sterol moiety to reduce degradation by cellular exonnrl.o~ce5. Phosphorothioate collli1;llillg oligonucleotides were prepared using the snlfilri7ing agent TETD or be~nr~--ge reagent. The 3 '-rhol~tt-rol modified oligonucleotides were prepared and purified as described by Vu et al. (in Second International Symposium on Nucleic Acids C7zemistry, Sapporo, Japan, 1993).

STABILITY AND TOXICITY
Gn~n-~in-o-rich oligonucleotides with either fiull length phospho-li.oster (PD) or fiull length phosphorothioate (PT) backbones were stable in the culture media for 4 days, while oligonucleotides consisting of a more random composition of nucleotides were rapidly degraded.
This in-lir~tr~ that the 3'-modified G-rich oligonucleotides with PD backbones were stable against both endonnr-le~e and e~-mnrle~e digestion over a defined four day incubation in culture. The concentration of oligonucleotide needed to reduce cell proliferation by 50% (TC50) of selected compounds, based on the dye metabolism assay was a~pL-.xi".;~ ly 40 to 50 ,~bM for oligonucleotides with PD backbones and 15 to 40 ~M for those compounds c~ i"i"g a PT
backbone. The TC50 for selected oligonucleotides are plescllLed in Table A-3. Stability and toxicity tests were replaced as described below CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 "o n .,, E-- A t~ ~

o ~, o ~ _ _ _ ~ ~ _ 3 o '¢ o ~ O
e ~ .~.?

a ~ ? ~ ? ~- ~ ~ ~ 3 .o .c ~ o ~ -~~ _ 3 ~ ~

~ ~ C ~ ~ C ~ C ~ _ ~ X
- s ~ s s ~D ~D ~ ~ ~ 00 00 ~ ~O ~~ C ~~ ~
~_ ~ o & --04 ~.
-- S
8 ~ ~ ~ ~ 8 o d d d d d d 8 ~ d --1 ~ ~ ~ ~ -- ~ ~ -- --~ ~ ~

O ~ _ ~ ~ ~ ~ ~: ~ ~ --I ~ ~ ~ o WO 97/03997 PCT~US96/11786 A. Cytoto~ ity and Stability Assays.
The cytotoxicity of selected oligonucleotides was assayed using the CellTiter 96~ Aqueous Non-Radioactivity Cell Proliferation Assay (Promega). This is a colormetric method for deL~ll.ulillg the number of viable cells in prolifer~tinn or chemosensitive assays using a solution if MTS. Dehydrogenase el,,ylues found in metabolically active cells convert MTS into a r ~
product. The SUP Tl cells used in the cytotoxicity assays were in log phase growth at the time of the assay. Cytotoxicity profiles for GTOs with PD backbones such as I100-15 (SEQ. ID. NO.
33) had TC50s (50% cytotoxic conrentration) in the range of 30 to 50 ~M while GTOs with PT
backbones such as I100-15 had TC50s in the 10 to 30 ~M range. The TCso for AZT in this assay format was ~pro~cill~Llely 10 ~M .
Blockage of the hydroxyl terminus of oligonucleotides has been shown by many investigators to greatly reduce degradation by cellular exonn~ ces. Therefore, all oligonucleotides used in these studies were modified at their 3'- end with either a propanolamine group or a cholesterol group. Por stability studies, 10 ,uM of GTOs were inrub~t~cl in MEM (GIBCO) supplemented with 10% FBS. Aliquots were taken after 10 min, 1 day, 2 days, 3 days and 4 days.
The aliquots at each time point were immPfli~fely extr~t-o~ twice with 50:50 phenol-chloroform solution and then plGci~iL~Led by the ~ liti~n of ethanol. The recovered oligonucleotides were 5'-end-iabeied using 'L y-~2PjATP ~und poly~ ucleûtide kir.ase. T he irlt.~grit~ of the o!igonur!eotides was then analyzed on a 20% polyacrylamide gel with 7 M urea. The results in~lir~t~d that a portion of each GTO with a PD backbone was present in the culture m~ lm for three to four days while oligonucleotides composed of a more random assulLIuell~ of all four nucleotides were rapidly degraded. In ~rlrliti~n, positions within PD GTOs where there existed two or more contiguous pyrimi~iin~?~ were more susceptible to en-lQm-(~ ce digestion than regions c-~nt~ining purines or ~ltern~ring purines and pyrimi~lin-os.
IN~IIBITIO~ OF ~V-1 PRODUCTION IN CULTURE ASSAYS
B. Long Term Su~les:~;u~ of Acute HIV-1 Infection~c in SUP T1 cells. The anti-HIV-l activity of a series of gn~nncin~lthymidine oligonucleotides (GTOs), with PD ba.,hlJolle~, c~ i,-;,-g dirfelellL sequences motifs was tested. As seen in Table A-2, one of the seq~n~e motifs tested (oligonucleotide I100-07) was 10 fold more active at inhibiting HIV-l induced syncytium formation than the other motifs tested (e.g. I100-00 shown in Table A-l). I100-07 and its delivalives (length and ~h~mir~l modifi~tions) were further tested for their ability to inhibit virus in a dose-dependent fashion by lllea~uleulellL of ~yu~;yLiulll f~rm~tion and viral p24 production.
Briefly, HIV-1DV was used to infect the SUP Tl lymphobl~ct--id cell line at an m.o.i. of 0.1 TCID50 for one hour at 37~C prior to w~lliug and leSu~ell~iOn in iucleaSillg concentrations of GTOs. The cells (2 x 104 cells/well) were inoculated in triplicate in 200 ul of RPMI 1640 c~ 10% fetal calf serum. Four days late}, the number of syncytia per well or the level of p24 in the m~ m was cletermin~od The results of these assays are pl~s~ ed in Table A-4. which results in~ t~rl that GTOs with simple PD linkages were capable of inhibiting HIV-l syncytia formation and p24 production in culture.

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 o o o o E- ~ ~ A ~

V~ ~0 0~ 0 oO ~ O Z ~~ ~
= ~~ ~V -- ~ V

~ ~ ' ~ ~~ ~ ~ ~ ~
-. _20 O

28 ~~ 2 C
._ 20 , C 2 ~, 20 2 2 O ' ' ~ . .:
V
= ~.5 ~ _ a ~ a a c~ a a a a ~

=~ 5 5 5 5 5 ~ 5 5 5 5 - ' ~ -- ~ ~~ C~ , , , , , , , ~, E-- E~ --CA 02227867 l998-0l-26 In order to r~etermine the effect of backbone mo-lifie~tion on GTO anti-viral activity, the PD backbone in two oligonucleotides seq l~nr.~c motifs was replaced with a PT backbone. The phosphorothioate containing oligonucleotides (I100-12 and I100-21) where then tested for their ability to inhibit HIV-1 induced syncytium fnrm~ti~n and production of HIV-1 p24 in the SUP T1 acute assay system (Table A4). The results from these assays in~lie~ted that the presence of sulfur molecules in the oligonucleotide backbone greatly enh~nred the anti-viral activity of I100-00 (I100-12) but had little if any effect on I100-07 (I100-21) (Table A4).
It was ~PL a.e.lL from the studies that the anti-viral activity of I100-07 was I l l~ e~ when steps were taken to reduce the length of the mrlecnle to 17 by deleting se~",~ ; from the 3'-end o (I100-15, -16, -17) but not by deletions from the S'-end (I100-18, -19, -20). To further ~1etermin~-to optimal size of the PD oligonucleotide needed for m~xim~l anti-HIV-1 activity, the I100-15 size variants listed in Table A-S were syntheci7t-c~ and assayed for antiviral activity.

WO 97/03997 pcTrus96lll786 TABLE A-5.
Tnhihition ofEnrV-lTn~n~e~ Syncy~a Using Size v~ ~ ofllO0-1~.
oligo Seqllen~e IC50 Syn.(uM) ~, I100-15* 5' ~Lg~L~,ggL~ Lgg~,L -3' 0.16 I100-25 5' ~L~,gLgg~,Lgg~,L~ ~gg -3' 0.25 I100-26* 5' ~,LggL~ggLg~,~,L~,gg -3' 0.12 I100-35 5' lg~,Lgg~,L~,g~Lgg~L -3' 1.75 I100-27 5' ~ Lg~,L~,g~,L~ ,L 3~ 4 50 I100-28 5' ~ Lg~;Lgg~,L -3' 4.50 o I100-30 5' gtg~ ~,L~gglgg~L -3' 4.50 I100-29 5' ~LgggL~,g~,L -3' > 10.00 AZT 0.02 * At 5 uM these compounds ~u~lessed virus at least 7 days post-removal of drug. All other compounds at 5 uM were the same as AZT 7 days after removal of drug.

The duration of the viral ~u~p~ sion was assayed by ~h~nging the m~linm in HIV-1infected cultures containing 2.5 uM of various oligonucleotides to complete media without added oligonucleotide on day 4 post-viral infection The pro-lllrtion of viral p24 antigen was then assayed on day 7 and day 11 post-infection The results of this experiment in-lic~f~d that the shorter variants of I100-07 (I100-15 and I100-16) as well as the PT version of this mnl~cule (I100-21), were capable of totally ~u~ ssillg HIV-1 p24 production for at least 7 days after removal of the drug from the culture medium (Table A-6). This substantial level of prolonged inhibition was >99% for I100-15, I100-16 and I100-21 when compared to the p24 antigen levels obtained for untreated HIV- 1 infected cells (Table A-6) . The 4, . ;.,, l i I ;1 l i on of p24 production relative to untreated HIV-1 infected SUP T1 cells for all oligonucleotides tested is plesellL~:d in Table A-6. The pl~sellce of sulfur molecules in the backbone of oligonucleotide I100-07 (I100-21) had a more marked effect on the red~rtion of virus seven days after removal of compound from the culture m~flinm than was observed at the four day post-infection assay point (Table A-5).

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 TABLE A-6.
Dcle~lioll of EIIV-1 p24 Antig~n in the Culture Media of GTO-Treated SUP T1 Cells.
Percent p24a s Oligonucleotide (2.5uM) Day 4b Day 7 Day 11 Control SUP T1 cells 100.0 % 100.0 % 100.0%
I100-07 6.0 % 15.9 % 8.6 %
I100-21 (PT)d 0.0 % 0.0 % 0.0 %
I100-15 0.0 % 0.0 % 0.0 %
o I100-16 0.0 % 0.0 % 0.0 %
I100-18 144.5 % 9.7 % 5.3 %
I100-19 208.0 % 21.8 % 15.0 %
1100-12 (PT) 0.0 % 0.0 % 0.0 %
a Level of detectable p24 in culture medium relative to control (infected but untreated SUP Tl cells after subtraction of bacl~l.ulld values.
b Day 4 post-inf~cfinn culture m.o-linm was replaced with fresh medium without oligonucleotide.
c SUP T1 cells infected with HIV-1 but not treated with oligonucleotides or AZT were used as positive control cells in this c~ h~l~.. L.
d 1100-21 and 1100-12 contain phosphorot'nioate backbone linkages (PT).

In control ~r~ ;",~ the culture medium from HIV-1 infected SUP T1 cells treated with AZT (4 ~M) was also replaced on day 4 post-infection with drug free media. In these expc..i~
two days after removal of AZT from the culture medium the presence of ~7yll~;yLiulll was observed in the HIV-1 infected cell cultures and by day 4 all cells were visibly infected with HIV-1.
To det~rmin~ whether the prolonged ~ul-~lc~,ion of HIV-1 was due to toxicity of the oligonucleotides, SUP T1 cells were counted for all treated samples 7 days after removal of the oligonucleotides from the infected cell cultures. The results inrlir~t~d that for cells treated with 2.5 ,uM of drug there was no difference in the number of cells when compared with control cultures (llllillrrc~rd~ untreated) of SUP T1 cells.

C. Tnhihifi~n of H~ ~s~,i.,.. in patient derived p~ .dl blood m~.o....lr~r cells (PB M Cs). I100-15 was ~e~ed for activity in PBMC cultures derived from AIDS patients.
Briefly, PHA activated ulliur~;L~:d PBMC's were added to 4PBMC's derived from patients with HIV infecti~n in the presence of varying cnn~entr~rions of oligonucleotide. Anti-HIV activity was ~sec.~ed by analyzing ::iu~r~ ;, collected every three days from these mixed cultures, for the -CA 02227867 l998-0l-26 W O 97/03997 PCTrUS96/11786 presence of HIV p24. The PHA activated PBMC's were g}own in the presence of 10 units/rnl of IL-l and medium was exchanged every three days for a period of three weeks. HIV p24 antigen production was assayed in drug-treated as compared to untreated control specim~'nc. It should be noted that the results in these ex~ (Figure A-9) observed for AZT were obtained when AZT was used at 12 uM which is roughly 300 fold greater than the ICso for this compound.

D. In-Vitro inhihi~ n of ~V-1 reverse transcriptase (RT). The ability of oligonucleotides to inhibit HIV-l RT in vitro has been well docllmPntP-l Marshall et al. PNAS 1992 89:6265-6269 have described a co-n~c~LiLive interaction at the active site as the ."P~h;~ ... by which mono- or diphosphorothioate co--l~i--i"g oligonucleotides inhibit HIV-l RT independent of whether the molecule tested was ~nti~PnSe, a random seq ~en~ ç or poly SdC.
In order to ~lPtPrminP whether I100-15 or its parent mnlPclllP, I100-07 (or the PT version I100-21), was interacting with HIV-l RT, the activity of this enzyme was assayed in the plGs~l-ce of various cullc~llLldlions of oligonucleotides. A kinetic analysis of the resultant enzyme inhibition was con-i~lrtpd to (içt~rminP the ",Pi-h ."i~m of inhibition. The GTOs ~ea.~:d to be inhibiting the RNA dependent DNA polylll~lase activity of the RT enzyme by cullll-c~Liliv~ inhihition at the active site of the enzyme.
The Kj value for all of the oligonucleotides tested is presented in Table A-7. The data indicate that for all oligonucleotides tested the presence of the sulfur group in the backbone greatly çnh~n~e-l the interaction between the oligonucleotides and the enzymes. The median inhibitory dose (ID50) for these oligonucleotides were also ~ tPd (Table A-7). The ID50 results are based on the ability of these compounds to inhibit 10 nM of HIV RT.
Short oligonucleotides (18 mers) with PD or PT backbones were assayed to ~letPrminp whether the nature of the nucleotide sequence contributed to inhibition of HIV-l RT in this assay system. Comparison of the effects of the PD versions of a GTO (1173 or 1100-15), poly dC
(1229) or a random nucleotide sç~ln~nre (1231) suggested that at this length none of the seq~lp-n~e motifs inhihit~cl RT (Table A-7). Other 18 mer PD GTO sequence motifs tested yielded similar results. Enzyme inhibition l~uniLolcd by both Kl and ID50 was observed for the PT versions of these same 18 mer oligonucleotides (Table A-7). The degree of enh~ "~ of observed enzyme inhibition for all oligon~ lPoti~lps tested when the sulfur group was present in the backbone, was between one to three orders of m~gnhll~lP (Table A-7).

WO 97/03997 PCT~US96/11786 TAIBLE A-7.
In V~ro Tnhihifion of ErrV-l RT by PD and PT OL~ eo~c.
Oligonucleotides Length Linkageb Ki (f~M) IDS0 (~LM) I 100-00 26 PD 0.37 5.0 I 100-12 26 PT 0.005 0.015 I 100-07 45 PD 0.137 2.5 I 100-21 45 PT 0.001 0.004 I 100-15 17 PD >5.0 >5.0 1173 18 PD > 5.0 ~ 5.0 1174 18 PT 0.015 0.0154 1229 (poly dC) 18 PD >5.0 >5.0 1230 (poly dC) 18 PT 0.044 0.033 1231 (GATC) 18 PD >5.0 >5.0 1232 (GATC) 18 PT 0.56 0.045 a Each pair of oligonucleotides contain the same sequenre and differ only in the nature of their backbone linkage. Oligonucleotides 1229 and 1230 were poly dC while the 1231 and 1232 oligonucleotides were a random sequence of all four bases (GATC).
b The barkhone mor~ifir~ti~nc are denoted as PD for phosphodiester and PT
for phosphorothioate.

2~
The results from this set of e~ h~ "l~ demnnctrated that I100-15 is minim~lly inhibitory to the RNA ~leppn~lent DNA polylll~,~dse activity of HIV-l RT The data also in~lir~ted that chemically modifying GTOs, poly dC or a random seqn~nre oligonucleotide greatly enh~nre(l the in vitro inhibitory activity of the m~lecllle Therefore, rhrmir~lly mo-lified oligonucleotides such as the antiviral G-rich m~lecllle ~lesrrihe by Wyatt et al. [112] has, by nature, a dirr~l~llL set of characteristics from oligonucleotides with natural PD backbones.

E. T-.hil.;l;.,~ ofthe interaction of HIV-1 gp120 with cellular CD4. The outer envelope gly~;opl~ ill gpl20 of HIV-1 mr~ t.os viral ~tt~rhm,-nt to the cell surface glyc~lolt~ CD4 in the CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96/11786 initial phase of HIV-l infection. The effects of both PD and PT modified oligonucleotides on this interaction were ex~Amint-d using a gpl20 capture ELISA kit.
The concentration of the gpl20 used in these studies (125 ng/ml) was ~lf-t~ d to be within the linear range of the ~1ett?ctiQn assay. The ability of oligonucleotides to inhibit gpl20/CD4 interactions by binding to gpl20 was ~lett?rTninPd by preinr~bAtion of the test compounds with soluble gpl20 before addition to the imrnobilized CD4. The results of this t~ illlCll~ (Table A-8) are ~l~s~ d as the crnr~ntration of oligonucleotide needed to reduce by 50% CD4 bound gpl20 (ID50 [gpl20]). The l~ local ~ elhl,~:"L was then performed to measure the ability of the oligonucleotides to inhibit these interactions by binding to immobilized CD4. In this set of o exp~lhl~ s I100-00, I100-07 and the PT versions of these two oligonucleotides were capable of preventing the interaction of gpl20 with immobilized CD4 (ID50 [CD4], Table A-8). For both sequences tested, the PT version of the oligonucleotide had ID50 values which were 50 to 100 fold lower than that of the PD version.
A fixed length (18 mer) set of oligonucleotides with either PD or PT backbones were assayed to cletrrrninr whether the nature of the nucleotide seql-~nre contributed to inhibition of gpl20/CD4 interactions. As was observed for the inhibition of HIV-l RT, the PD versions of these molecules had little or no measurable effects on the binding of gpl20 with CD4. However, the PT versions of these oligonucleotides did yield measurable inhibitory activity. The 18 mer GTO(1174)illLt,,upLedgpl20/CD4interactionsata~l..xil,,AI~ly lOfoldlowerconcentrationsthan poly (SdC)l8 (1230) while the random seqUpnre 18 mer (1232) had no measurable activity (Table A-7).

W O 97/03997 PCT~US96/11786 TAiBLE A-8.
In Vitro T.~ of IIIV-l gpl20 Interaction with CD4 by PD alld Pr Oli~o-...- lPo~ c.
Oligonucleotide Linkagea ID50 tgpl20](,uM) ID50[CD4](,uM) 5I 100-00 PD 3.50 18 I 100-12 PT 0.08 0.475 I 100-07 PD 0.80 4.25 I 100-21 PT 0.07 0.048 1173 PD > 100 > 100 1174 PT 0.09 0.45 1229 (poly dC) PD > 100 > 100 1230 (poly dC) PT 1.00 3.25 1231 (GATC) PD >100 >50 1232 (GATC) PT > 10 > 10 ~ l~ach pair of oligonucleotides contain the same seqll~nre and di~er only in the nature of their backbone linkage.
b The backbone ml llifir~tion~ are denoted as PD for ph~pho~ t~r and PT
for phosphorothioate.

F. Oligo~ otide ~t~.deLions with the v3 loop of HIV-1 gpl20. It had been reported previously that poly SdC oligonucleotides were able to bind to the third variable loop domain of HIV-l gpl20 (v3 loop). The degree of int~r~.~tion was reported to be dependent on the length of the oligonucleotide studied, with a rapid decrease in binding affinity observed for compounds shorter than 18 nucleotides.
It was noted that the ~ t,-cti~n antibody used to monitor inhihitif~n of gpl20/CD4 interactions in the capture gpl20 ELISA KIT (HRP-c~-GP120) as described above (Table A-8) recognized an epitope in the gpl20 v3 loop (~ ,c~ s i~r~" ~ m)~ For this reason, control expclh-lc .l~ were pclrul.l-ed to d~- . ",i"~ whether the observed inhibition of gpl20/CD4 int~r~rtion~ was due in part, or in whole, to i..Lc.rclc -ce with the HRP-c~-gpl20 ~let~cti~n antibody.
The results in~ ted that I100-07 and I173 (PD backbones) did not inhibit the detection of im~nobilized gpl20. However, the PT oligonucleotides tested (I100-21 and 1174) were able to slightly inhibit the ~i~tection of gpl20 at oligonucleotide co~ Lions above ~ ~M. This level WO 97/03997 PCT~US96/11786 of inhibition was too small to account for the ID50 [gpl20] values presented for I100-21 and 1174 in Table A-8.
Further analysis of oligonucleotide intrr~rtions with the v3 loop was c~n~lnrted using a v3 loop specific murine Mab, NEA-9284 (Figure A-10). PT oligonucleotides were able to inhibit binding of NEA-9284 to gpl20. The presence of bound gpl20 specific Mab was determined using a HRP-labeled goat-(x-mouse antibody. The results of these ~ elhll~llL~ in~ ted that PT
oligonucleotides were able to inhibit binding of NEA -9284 to gpl20. The ID50 for the most active oligonucleotide (1100-21) was a~lu~hllal~ly 4 to 7 ~M. This concentration is approximately 10 to 30 fold higher than the IC50 for this oligonucleotide against HIV-1 in culture (Table A-8). The o PD oligonucleotides tested did not inhibit the binding of any Mab to gpl20. Therefore, it was determined to be unlikely that this was the mrch:lni~m by which the PD GTOs such as I100-07 (and hence I100-15) were inhibiting HIV-1.

G. Analysis of ~IIV-1 RNA and DNA in single cycle assays. Total RNA and DNA wereextracted from SUP T1 cells 36 hours after infection with 0.1 m.o.i of HIV-lDV. In this assay, the infected cells were treated with I100-15 or AZT at various time points before, during or after infection. Harvesting of the infected cells at 36 hr post-infection allowed for the analysis of appr-~xim~t-?ly one round of viral replication. A schematic diagram of the positions of the PCR
primers used in the DNA and RNA analysis is shown in Figure A-11.
Total extr~rted DNA was analyzed using a PCR primer set which would amplify a 200 bp portion of the viral genome ~p~nning the repeat element (R) into the gag gene. The primer set detected full-length or nearly completely synth~-~i7rd viral DNA. This is the last region of the minus strand of viral DNA that is synthe~i7t-d Thus, for DNA to be detected by this primer set, two template-~wiL~;hillg events have occurred and contiguous 5'LTR to gag sequences must be present on either the minus or plus strand of DNA.
In the same reaction mixture, a PCR primer set which would amplify a 220 bp region of the human ,5-actin gene was used. The results intlir~ted that in cells treated with AZT there was a marked decrease in viral DNA synthesis when the drug was added up to 4 hrs post-infection (data in Figure A-12 shows zero hour and 8 hour time of ~ lition studies). The effects of I100-15 on the early rounds of viral DNA synthesis was minim~l The results of this c~-y~ L in~lir~trd that I100-15 did not inhibit virus entry into the cells because of the ~letect~ble levels of viral DNA even in samples treated with I100-15 at the same time as virus infection (zero hour a~l~liti-~n). Furthermore, it suggested that I100-15 had a dirr~ lL
~,erll~"ism of action co~ raled to AZT.
-WO 97/03997 PCT~US96/11786 ~lrlition~l e~ L~ using alternative PCR primers suggested that there may be alterations in the viral DNA synthesis caused by I100-15. The observed amplification products, when primers clustered in the U3 region of the virus were used, yielded a banding pattern which was not predicted and obviously dirr~l~ll- from the infected cell control (untreated) and the AZT
s treated infected cell samples.
RNA ç~fr~rtrd from HIV-l infected cells was analyzed by RT-PCR. In this assay, the r ~"~ e primer of the PCR primer pairs was used with MMLV RT and extracted mRNA tosynthesize cDNA strand. The resultant cDNA was then used as a temrl~tr in PCR re~rtil~n~ Two RNA primer sets were used to analyze nn~Flirecl (primers rl and r2) and spliced (primers rl and o r3) HIV-l tr~n.ccripts. Predicted sizes of the ~mrlifird products were 101 bp and 214 bp for the unspliced and spliced species lc~e~;Lively. The same ,~-action primers used for the analysis of the DNA samples were used as controls in this ~ lhllt:llL.
The results obtained using primer pair rl and r3 are depicted in Figure A-12. The results of this ~ lhll~lll clearly intlir~trd that a reduced level of HIV-l specific LldllS~ t was observed in samples treated with I100-15 in the samples treated with drug at the same time as virus infection (zero hours). It was also clear that while samples treated with AZT had reduced levels of viral cDNA, viral mRNA was still being produced. The same decrease in HIV-l specific Llduls~ t was observed in viral infected cells treated with I100-15 when the rl and r2 primer pair was used (data not shown).
H. Slru~Lulal analysis of I100-15 and I100-26. I100-07, and its derivative products inrlllflin~E~
I100-15 and I100-26, are composed entirely of deoxygll~nocin~ (G) and deo~yLhylllidine (T). These G-rich oligonucleotides were purified using anion exchange reverse phase HPLC. Using this procedure the oligonucleotide is purified in the ~lc:s~llce of sodium ions. Monovalent cations are 2s known to encourage self-~soei~t.od ~Llu~;Lules for G-rich m~ cnl~5, all of which involve formation of G-tetrads. The G-tetrad formation involves the formation of eight hydrogen bonds by coor-lin~tit n of the four o6 atoms of guanine with alkali cations believed to bind to the center of a quadruplex, and by strong stacking interactions. The oligonucleotides purified using anion exchange clll~ ugraphy then have an opportunity to form inter- or intra-molecular tetrads. The tetrad structure can be strengthened by replacing the sodium ion with potas~iulll.

I. Non~ ..~g gel analysis. I100-15 (17 mer, Table A-5) was analyzed using no~ n~tnring polyacrylamide gel electrophoresis. In this ~ lhll~UL, trace c(",r~.,~ inn~ of r~-liolz~he~
oligonucleotide (10-7M) was inrub~trd with iu~ d~hlg co~r~ irm~ of cold oligonucleotide (up to 10 5M) before gel analysis in the presence of monovalent cation. Under the gel cnn-litinn~ used, CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96/11786 I100-15 migrated as a unique band faster than a random coiled (denatured) 17 mer oligonucleotide would and it was shown to do so in a co,-~e"l,dLion in~ pen-l.onf fashion (data not shown). This was in contrast to I100-18 (16 mer, 10 fold less active than I100-15) which appeared to migrate as multiple species in a C~ CellLldliOn dependent fashion under the same gel conditions (data not shown). The same phen~ m.on~ was observed when 10-5 oligonucleotide (total cold and radiolabeled oligonucleotide) was inrllh~t~d with hlcleh~illg c~ nrenfration of KCI (Figure A-13). I100-15 migrated as a unique species at all col~e~ Lions of KCl while I100-18 and Z106-50 (ggttg~gggLlggg) migrated as multiple species.
The results from these assay suggested that I100-15 folds into an intramolecular structure o while other G-rich oligonucleotides such as I100-18 and Z106-50 aggregate into higher order int~rm~lçclll~r structures. It was noted that the total phosphorothioate oligonucleotide G-rich compound described by Wyatt et al., P.N.A.S. 1994 91:1356-66, with the sequence T2G4T2, was claimed to fold into an int~rml~]eclll~r tetrad. Therefore, I100-15 (PD backbone) is structurally and rh.omi~lly dirrelellL from the oligonucleotide reported (ISIS PT oligonucleotide).
J. Tetrad Sl~u~ . Principally due to its role in telomere form~tion, the structure of four stranded nucleic acid tetrads has been well studied. Most euk~lyuLes possess a repe~ting G-rich sequenre of the form (T/C)nGm where n= 1-4 and m= 1-8. Of particular interest to the study of the I100-15 class of GTO was the ~LlueLule of the telomere seqn~onre repeat T2G4, first ~l~tectod in Oxytricha. The Oxytricha repeat has been studied in oligonucleotides by NMR, Smith et al., Nature 1992, 356: 164-68, and by crystallographic m.otho~lc, Kang et al. Nature, 1992, 356: 126-31.
As had been predicted from llulllel~u~ previous physical and bioehemi~l studies, both the NMR
and crystallographic studies suggested that folding is m~ t--d by square planar Hoogsteen H-bonding among G residues, with overall ~ntir~r~llel orient~ti~n of the four strand equivalents 2s COlll~liSillg the tetrad fold. As expected, the crystallography has shown that the structure is selectively stabilized by tight binding of a small monovalent cation to the o6 oxygen of guanosine.
But ~ul~ ingly, both NMR and crystallography confirm that the folded structure possess ~ltern~ting syn/anti glycosidic bond angles (as opposed to all anti for most duplex structures).
Feigon and colleagues have used NMR and modelling to deduce the structure of a 28 base-long oligonucleotide (G4T4G4T4G4T4G4,0xy 3.5) which is capable of forming a well-defined all-~ntip~r~llel inrr~m~ c~ r tetrad, Smith et al., Nature 1992, 356:164-68. The present hlvellL~
postulated that if the GTO I100-15 were to fold to form a stable intramolecular tetrad, its NMR
properties would be expected to be similar to those of the Oxy 3.5 mnlecllle.
In the folded state, the salient NMR char~rt~ri~ti~s of the intramolecular Oxy 3.5 tetrad 3s were as follows:

WO 97/03997 PCT~US96/11786 1. Narrow linewidths, indicative of ~ ll,rl ~ormation only.
2. Induction of well-defined gll~nocin~ Nl Hoogsteen imino resonances in the 11.2 to 11.7 ppm range. The rh~.mi~l exchange rate of these protons is very slow, reflective of the high positive COOp~ldLivi~y of tetrad folding and dissociation.

3. Spectral simplicity, indicative of a single pred--min~nt folded structure, rather than an equilibrium among dirrt lellL folded structures.

4 Tnrr~h~;e H8-Cl' and hlL~ll.asc H7-C2" NOE colllle.:LiviLy which ~lPm~n~lc a pattern of alternating syn-anti glycosidic bond angle throughout the "tetrad stem"
of the folded structure.

K. One ~lim~nC;~n~l NMR analysis. Displayed in Figure A-15 is a line model for I100-15, folded to form an intramolecular tetrad of the Oxytricha class. From a physical perspective, the possibilitythat an intramolecular tetrad .7Llu~;Lulc might form in high KCI or NaCI is not .7ul~lk7illg.
What was surprising was the fact that this model proposed a stem region Cu~ ,illg a single G-lS octet and hlL~lvellillg loop regions which were only two bases long.
In order to test the general feasibility of this model, a detailed 3D m~l.ocnl~r model for a I100-15 was constructed. In so doing, the hlv~nLuls ~,.,.".orl that the 8 G's comprising the octet core of the structure formed a standard square planar octet, and that glycosidic angles were as in the crystal and NMR structures of the antiparallel Oxytricha tetrads, Smith et al., Nature 1992, 356:164-68, and Kang et al., Nature 1992, 356:126-31. Additionally, a single K+ ion was introduced into the center of the G-octet, with octahedral coordination to GO6. Initially, 2 base loop structures were created so as to connect elements of the octet without disruption. Subsequent to this initial po~t~ ti~m, the structure was subjected to m~ h~ni~ with full electrostatics, employing (~h~rmm p,fl,1lllt~f~ in Sybyl.
2s After refin~-m~nt it was observed that cooldilldL~s of the octet core were not ~ignifi~ntly altered and that backbone p~ within the loop domains were within acceptable energetic limits.
First, the structure was very C'J"'1'~''l nearly spherical, with the three loop regions and the 5' "GT tail" comprising the surface of the tetrad core. Based upon this structure, it appeared likely that interaction with cellular macromolecules would be heavily cl-)min~t~-d by the structures of these surface loops. In that regard, the hlv~llLol~ believe that it may be illd~JplU~lidLe to think of such interactions as "tetrad binding." The in~hlcion of G-tetrads in such a structure may not be important as a recognition element per se, but instead provides a latticework upon which an orderly loop array is po~itinnP-l CA 02227867 l998-0l-26 Further, although the loop regions did not appear to be under mrch~nic~1 stress, they were short enough so that they possç~.ced very high configurational freedom. Because of those severe length constraints, it was found that all feasible loop models display a distinct "rabbit ears"
structure, wherein the two base planes of the loop region are nnxt~rkt--l and point outward from the center of the octet core. Such rigid, IlllX~ k~-t1, single strand loop character was very distinctive as c~ aled to other known folded nucleic acid structures. Therefore, varying the seq~llonre or rh-~mir~l structure of these loops, one at a time, was n~cecx~l y to d~t~-rmint- if bonding interactions between these loops and cellula} macromolecules are hll~olL~lL to the observed anti-HIV activity.
The structures described above pnxse~xed a single G-octet core, which was known to be the .llilli.. l"lll structure required for mlc.lr~ticn of tetrad formation. Therefore, when paired with the observed short loop size, the intramolecular tetrad structure proposed for I100-15 is best described as meta-stable, relative to other more robust tetrads which have been described in the li~ dLul~.
An increase of the core from 2 to 3 stacked tetrads, or an increase in the length of flexibility of one or more loops would be expected to increase the thermodynamic and/or kinetic stability of this structure ~i~"iric~ ly. Thus, the observed anti-HIV activity can be h~ uved by sequence m~l~lific~ric-n which rnh~nres the stability of the underlying tetrad latticework.
Finally, it was observed that I100-15 and homologues display profound rç~ixt~nre to cellular mlrl~xes. One hlL~l~xlillg aspect of the proposed structure was that, even in the loop dom~in.c, phosphodiester linkages are generally buried from interaction with large solutes, such as a nnrle~e. The structure analysis proposed defined local phosphodiester backbone xLl u~;lulc: at low resolution. When paired with explicit biorht-mir~l analysis of phosphodiester cleavage rate, it is possible to define sites for selective introduction of backbone mnriifir~tion in I100-15 homologies, for the purpose of extending the biological half life in vivo.
The gel electrophoresis data described above suggested that I100-15 spends very little time 2s as a random coil at 25~C, under native salt con-lition~. Although the gel data rules out interm- lec~ r associations, the data do not cull~Ll~ the oligomer to any particular folded monomeric structure. Oligonucleotide folding in I100-15 has been studied employing a cu---l~i--aLion of high resolution NM~ and methods.
Stable formation of a discrete octet core, mr~ trcl by tight binding of a single monovalent ion is crucial to the model described above. Given that G-N1 imino protons give rise to sharp, charSlrt~ri~tic lH NMR signals in such a structure, focus has been on the puL~s~iu--- ion df p~n~1rnre and te...~el~Lu.c deprn~lrnre of I100-15 folding, as ~sec~ed by IH NMR at 500mHz.
For these mea~ul~l..c..L~, I100-15 was synth~i7~d at 15uM scale employing fast deblocking "Expedite" rhrmi~try on a Milligen synthe~i7~r. Subsequent to purifir~ticn by ~ n~tllring anion-3s exchange chio~Lography in base (lOmM LiOH, 0.2 to 0.7M NaC1), oligomer purity was CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 co"ri""rd by denAtnring gel electrophoresis (7M urea, 65~C). For NMR, the oligomer was desalted and transferred into 201nM LiC1 adjusted to pH 6.0, which ,.,i..i..-i~s folding to form tetrads. Oligonucleotide strand cc,llc~llLldLion was held constant at 2.7 mM. NMR was measured in H20, employing a Redfield pulse sequt-nre to saturate the water resonance, as ~lrsrribed s previously, Dittrich et al., Biochemist~y, 1994 33:41114120.
In Figure A-15 a KCI tit~r~tinn is displayed. At 300~K, in the absence of added K+, imino proton signals cannot be resolved in the 10-12 ppm region. Snhse(l~lent to addition of KCI, snbstAnti~l nallowi"g of imino signals was obtained, c~ at an added KCl conc~llLldLion of 3mM, which is very close to one added K+ equivalent per octet. Above 4 mM, it can be seen that o at least two classes of imino resonance can be detected in the 10-12 ppm range with roughly equal hlL~llsiLy: a broad envelope from 10-11 ppm, upon which several sharp resonances are ~,ul-elilll~osed in the 11-11.5 ppm region.
By analogy with rhrmir~l shifts of other G tetrad structures, the hlvt:llLc,l~, LcllLdLivt:ly ascribed the sharp imino signals to the 8 Hoogsteen H bonds of the core octet. The broad envelope was ascribed to the G and T imino r~osonAnrçc c~ . iblllrd by the loop and 5' trrmin~l ~inm~inc C~ r~ll with pllblichrd tetrad NMR data, a broad envelop of signal was ~irtectec7 at 9 ppm, which most likely results from nm-cllAlly slow exchange of guAnnsinf~ N2 protons engaged in Hoogsteen pairing.
In order to better distinguish the two classes of iminolH signal and, ~ litinnAlly, to investigate the gross stability rh;~ r,i~lirs of the folded 1100-15 structure, thermal melting analysis, at 2.7 mM in strands, 6mM KC1, 20mM LiCI, pH 6.0 over the range from 300~K to 345~K was pt:lr~lllled.
SnkstAntiAI line nallowillg of "Hoogsteen" imino proton signals was seen at 310~K, which appears to be arculll~ lir~ by broA~irning of the poorly resolved imino envelope at 10.7 ppm. This caused a n~llowillg of the plateau above 310~K, giving rise to 7-8 well-resolved imino protons at 320~K. By reference to the NMR behavior of the O~tricia tetrad and other tetrad structures, the formation of 7 to 8 narrow, well-resolved imino lc;sol~,res at elevated ~rmrp~rAtllres strongly suggested that in the ~lestillce of one bound K+ ion per octet equivalent, 1100-15 folded into a discreet tetrad structure, stabilized by the 8 Hoogsteen H-bonds of the ~le~ullled octet.
In the range from 330 to 340~K, the imino proton spectrum undergoes an abrupt transition, which is likely to be l~l~7~llLdLiv~: of CO~t;ldLiv~: unfolding of the octet. Stability of this kind, accu",l.;1"ird by a~dleuLly high thermal coopc~ldLiviLy is very striking indeed, and is generally indicative of a single, well-defined folded oligonucleotide structure.
The origin of the shallow ~ CldLUlC~ pPn~ nre of the spectral ~aldulleLt:l~" leading to çnhAnred lH resolution at 320~K, remains to be llrtc-rmin~ It is likely to have resulted from CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 weak intermolecular association which occur in t'ne rnillimolar strand concentration range. This i.,~el~lei~Lion is born out by preliminary analysis of spectral parameters as a function of strand concentration (not shown). Independent of hl~ Lion, the data suggested that high quality NMR
data may be obtained for exchangeable and non-exchangeable I100-15 (SEQ. ID. NO. 33) protons s at 35~C, 20 mM, LiCl, 6mM KCl and 2rnM in strand equivalents.

lON OF HCMV A(: l lVl l Y
Several dirr~ oligonucleotides reduced HCMV titers in tissue culture. Each of the oligonucleotides cnnt~inl d a dirr~lcllL p~lcellL~ge of gn~nosin~ residues and a Llirrt;l~ll, number of total nucleotides in the polymer. The results of this assay are depicted in Table A-9. All oligonucleotides were capable of reducing viral titer in culture in~ tling G101-50 which contained only 53% G residues (16 out of 30 total nucleotides). In Table A-9, the length and percent gn~no~in~ nucleotides is in~ tt-~l for each oligonucleotide tested.

Ol;g~ -' lf of ~ Tnhihifinn of HCMV Activib Viral Yield in plaque fo.~ g units (PFU) oligonucleotide (%G) 20 Oligo. G101-50 (53%) G105-50 (80%) G106-50 (78%) G109-50 (65%) G113-50 (64%) Conc. 30mer 31mer 27mer 29mer 24mer None 4.5x103 PFU 4.5x103 PFU 4.5x103 PFU 4.5x103 PFU 4.5x103 PFU
20.0 ~LM 0 4.5xlOI PFU 2.5xlOI PFU 8.0xlOI PFU 3.5xlOI PFU
25 10.0 ~LM 2.5xlOI PFU 1.8xlOZPFU 4.0xlOI PFU 4.5xlOi PFU 4.0xlOI PFU
1.0 yM 7.0x102 PFU 1.9xlOZ PFU 6.0xlOI PFU 1.5x102 PFU 5.0x102 PFU
0.5 ILM 8.0x102 PFU 2.7x102PFU 1.3x102 PFU 3.0x102 PFU 5.4x102- PFU

W O 97/03997 PCT~US96/11786 In NIH3T3 cells chronically infected with FMLV, oligonucleotides (Fig. A-l) were capable of inhibiting virus production. However, oligonucleotide controls in this experiment were capable of inhibiting virus production in culture.
IN VITRO ENZYMATIC ASSAYS
Culture media cont~ining FMLV reverse Lldlls~ Ldse (RT) was mixed with various conrt-ntrations of I100-51 or I100-12, (the phosrhrrlir5t~r backbone of I100-51 having been ml ~lifird to a phosphorothioate backbone). Reverse Llzlns.,li~Ldst: was lllea~uled as described in the section entitled "Reverse T.dl-s~ L~se Assay" above. Figure A-4 shows that both oligonucleotides were capable of inhihiring the RT enzyme. Inhibitory concentrations for 50% reduction in RT
o activity was between 0.5 to 1 ~M for I100-51 and less than 0.5 uM for I100-12.
The I100-51 (FMLV2ap), ~tteml~tto(l full length transcription directed by either the T7 or T3 poly~ ldses (Figure A-5a). As can be seen in Figure A-l, full length ~ldllscli~L~ directed by the T7 promoter would be 131 bases long while full length ~ directed by the T3 p~ ol~l would be 171 bases long (position of the Dde I site relative to the mRNA start site). The se~lenre isomer of I100-51 (I100-01 = FMLV2p), de~ignPd parallel to the target strand was also capable of 5ignifir~ntly inhibiting Lla~lscli~lion from the T7 plOlllOLc~l (Figure A-Sb). However, only the anti-parallel triple helix forming oligonucleotide FMLV2ap inhibited via ~ r~ ion of Ll~ls~;.i,L.Lion as can be seen in the build up of a truncated Lldlls~ in the reaction mix (Figure A-SC). The truncated LldllS~ t analyzed in Figure A-SC was a~ hlldLely 63 bases long and m~trhrd the predicted size rld~lllc~ when p275A was used as a trTnrl~t.o (T7 promoter). G101-50 (53% G) inhibited T7, but not T3 directed, transcription by a mrrh~ni~m other than ~trrml~tion (Figure A-SA) since no Llu,lcdLed Ll~ulscliLL~ were observed when this oligonucleotide was used alone.
I100-11 (26% G) increased the level of specific LldllS~ L~ directed by the T7 ~lollloL~l (Figure A-4b).
In experiments ~lr~ignPd to monitor inhibition of tr~n~r-rirtinn initi~ti~n of the HSV-l IE175 promoter, using oligonucleotides, both specific and control G-Rich oligonucleotides were capable of inhibiting eukaryotic l.,..-.c~-~ il-Lion when a HeLa cell extract system was used. The oligonucleotides used were B133-54; B133-55 and B107-51 as specific inhibitors via potential triple helix mPrh~ni~m of action and G101-50 and I100-11 as the low G-content control oligonucleotides.
The experiments described above clearly flrllll~ ldled the anti-viral activity in tissue culture assays for several G-Rich oligonucleotides against HSV-2, HIV-1, HCMV and FMLV. In iti~n, G-Rich oligonucleotides specifically inhibited the bacterial RNA polymerase ell~yllles T7 and T3, the FMLV and HIV-l reverse Lld,ls~ Ldse enzyme and eukaryotic RNA polymerase.

B. Specific In Vitro Studies and In Vitro HIV Inhibition Usin~ T30177 As was ciPnn-~nxrrated by the inventors in the studies initially c~-n-illcted as described below, T30177 is an oligonucleotide composed of only deoxygn~n-)~inl? and thymidine, it is 17 nucleotides in length is the same sequence as I10075 (SEQ. ID. NO. 33), and it contains single phosphorothioate internucleoside linkages at its 5' and 3' ends for stability. This oligonucleotide does not share xignifir~nt primary sequence homology with, or possess any complr~ y (~ntix--n~e) seqll~n~e motifs to the HIV-1 genome. As shown below, T30177 inhibited replication of multiple laboratory strains of HIV-l in human T-cells lines, peripheral blood lymphocytes, and macrophages. T30177 was also shown to be capable of inhibiting multiple clinical isolates of HIV-1 and plCV~llLillg the ~;yLoL~aLllic effect of HIV-l in primary CD4+ T-lymphocytes. In assays using human peripheral blood Iymphocytes there was no observable toxicity associated with T30177 at the highest cullcelllldLion tested (100 ,uM), while the median inhibitory COllCcllLlaLiOn (IC50) was d~otermin~ to be in the 0.1 to 1.0 ,uM range for the clinical isolates tested, resulting in a high therapeutic index for this drug. In temporal studies, the kinetics of ~ litit)n of T30177 to infected cell cultures inrlir~t~cl that like the known viral adsorption blocking agents dextran sulfate and Chicago sky blue, T30177 needed to be added to cells during, or very soon after, viral infection.
However, analysis of nucleic acids e~ d 12 hr-post infection from cells treated with T30177, at the time of virus infection, ~st~hlixh~od the plesellce of Illlilllrgl~lrd viral cDNA, inrhltling circular proviral DNA, in the treated cells. In vitro analysis of viral enzymes revealed that T30177 was a potent inhibitor of HIV-1 integrase reducing ~l~yllldlic activity by 50% at con~entrations in the range of 0.01 to 0.10 ~M. T30177 was also able to inhibit viral reverse Lldlls~ Ldse activity, however, the 50% inhibitory value obtained was in the range of 1-10 ~M depending upon the template used in the el~yllldLic assay. No observable inhibition of viral protease was detected at the highest cu-lcel~lldLion of T30177 used (10 ~M). In ~ elill~lll~ in which T30177 was removed from infected cell cultures 4 days post-HIV-1 infection, total ~ul~les~ion of virus production was observed for more than 27 days. Polymerase chain reaction analysis of DNA extr~rt~tl from cells treated in this fashion was unable to detect the presence of viral DNA 11 days after removal of drug from the infected cell cultures. The ability of T30177 to in_ibit both labc,ldLuly and clinical isolates of HIV-l and the expe~lilll~llL;ll data suggested to the hlv~llLol~ that T30177 leL~ llL~d a novel class of ill~ se inhibitors, in~lir~ting that this compound was a viable c~nfli~l~t~ against ev~ln~ti--n as a therapeutic agent for HIV-1 in humans.
In the present study the inventors disclose the ~ ix,,, by which a variant of I100-15 (T30177) was able to inhibit multiple HIV-1 laboratory strains in acute and long-term ~u~ ssion assays. The data in~lir~t~d that T30177 is a potent and selective inhibitor of HIV-1 via at least two s 3s m-orh~nixmx One m~ch~nixm involves illL~lrc~lillg with CD4- and gpl20-m~ ted cell fusion O 97/03997 PCT~US96/11786 events However, T30177 is 100-fold less errcc~ive in inhibiting gp 120-induced cell fusion events than it is at inhibiting an early event in the viral life cycle, suggesting a specific point of interdiction distinct from that of blocking virus/cell interactions. The data also clearly showed that T30177 is a potent inhibitor of the HIV-1 integrase enzyme in vitro and that by blocking these events in the viral life cycle T30177 is able to ~iU~)~)lCSS virus production for prolonged periods after an initial short Llc~lLll~cllL regimen with the drug.

Materials Used iu In Vitro H~V Tnhih;~ion Studies OLgol.u.leotides. The deo~y~ nosin~-rich and other oligodeoxynucleotides used in this study were synfh~i7~ purified, and characterized as previously reported. Ojwang, et al., J. AIDS
7:560-570 (1994); Rando, et al, J. Biol. Chem. 270:1754-1760 (1995). The seqll~nre and phosphorothioate (PT) pattern of the oligonucleotides used in antiviral assays is shown in Table B-7.
Materials. Zidovudine (3'-azido-3'-deoxythymidine, AZT) and the nucleoside analogs 2',3'-dideoxyionsine (ddI) and 2',3'-dideoxycytidine (ddC) were obtained from the AIDS Research and Reference Reagents Program, National Institute of Allergy and Infectious Diseases. Dextran sulfate (DS5000) was purchased from Sigma, and the bicy~;l~ll deliv~Live~ JM2763 and JM3100 (De Clereq, et al., Antimicrob. Agents Chemother. 38:668-674 (1994)) were obtained from Johnson Matthey (West~htost~r, Pellusylvdllia). Chicago sky blue (CSB) was obtained from the Drug Synthesis and ~h~mi~try Branch, National Cancer Institute.
Cyl~ x;-~y Analysis. The cytotoxicity of T30177 was assayed as described above. The con~ i( n of drug n~c.o~ry to give one-quarter (TC25), one-half (TC50) or 95 % (TCg5) of the lll~illlUlll inhibition of growth response was then dcLellllhled. The degree of cell proliferation was ~i~tt?rmin~od according to the I~ 1111r~ l's instructions.
2s In other e~ the effect of T30177 On the viability of primarY human PBMCs, PBLs and macrophages was tl~otermin~-d using the trypan blue dye exclusion technique. Griffiths, B., IRL
Press, p. 48 (1992), or by ,~ g the degree of [3H]thyrnidine or [3]1eucine uptake in these cells (McGrath, M.S., et al.Proc. Natl. Acad. Sci. USA 86:2844-2848 (1989)).

Antiviral assays HlV-l i. rc~iol. assays using cell lines. Laboratory strains of HIV-1, HIV-2, simian immnnndehciency virus (SIV), or the low passage isolate HIV 1DV (Ojwang, et al., J. AIDS 7:560-570 (1994)), were used to infect established cell lines using the in~ tt~rl multiplicity of infection (MOI) of virus, for one hour at 37~C prior to washing and lc~u::~,cllsion in m.o~ lm cont~ining increasing cn~r.-l.,linn~ of drug. The infected cells (2 x 104 cells/well) were inoculated in -WO 97/03997 PCT~US96/11786 triplicate in 200 ,ul of complete medium which contains RPMI 1640 (Life Technologies) supplem~-nt~d with 10% FBS, penicillin (50 U/mL), sll~Lullly~;in (50 ~Lg/mL) and L-glrlt~min~, (2 mM). Four to 6 days post-infecti-~n, drug treated and control wells were analyzed for HIV-1 induced cytopathic effects, for the presence of viral reverse transcriptase (RT) or viral p24 antigen in the culture m.o~ m Buckheit, et al., AIDS Research and Hurnan Retroviruses 7:295-302 (1991); Ojwang, et al., J. AIDS 7:560-570 (1994); Rando, et al, J. Biol. Chem. 270:1754-1760 (1995). CyLv~dLhic effects (CPE) were lllolliLol~d by either direct counting of HIV-1 in-lnt~t~cl syncytium forrnation or by staining cells with the tetrazolium dye XT or MTT. Buckheit, et al., AIDS Research and Human Retroviruses 7:295-302 (1991). The AZT resistant strain of HIV-1 (ADP/141) was kindly provided by Dr. Brendan Larder and the AIDS Directed Programme Reagent Project, Medical Research Council, F.ngl~n~1 HIV-l infe~inn of PBMCs. Peripheral blood monomlrle~r cells (PBMCs) were isolated from blood of HIV-1 negative and h~patiti~ B virus (HBV) negative (healthy) donors by Ficoll/Hypaque density gradient centrifilg~tion, cultured as described by Gartner and Popovic (Gartner et al., In Tel~hniqll~s in HIV Research, p. 59-63 (1990)), then a.;livdLed with phytohr., .r,~ h ,l i "; " (2 ~g/rnL) and cultured in RPMI 1640 medium supplemented with 15% fetal bovine serum (FBS) and human reculllI3illallL int,orlellkin 2 (IL-2, 30 units/rnL). After 3 days PBMCs (2 x 105 cells/well) were infected with various isolates of HIV-1 at a multiplicity of infection (MOI) of 0.01. After 2 hours at 37~C cells were washed and treated with various cullc~ lalions of T30177 or AZT, as described by Buckheit and SW~I~L1O111, id. (1991). The medium was changed on day 3 or 4 post-infection and fresh drug was added at these times. Seven days after infecti~n HIV-1 replication was analyzed using the Coulter p24 antigen-capture assay. Assays were performed in triplicate. Data was obtained by ~e~;~lu~hotometric analysis at 40 nm using a Molecular Devices Vmax plate reader.

~V-1 infection of PBLs. Human peripheral blood lymphocytes (PBLs) were isolated from blood drawn from HIV-1 and HBV seronegative donors. PBLs were isolated by Ficoll-Hypaque density gradient centrifugation. The PBLs were suspended in culture m!~-linm (RPMI 1640 medium supplemented with 2 mM L-glllt~min~, 20% FBS and 50 ~Lg/mL gent~mi~in) and the cells counted using the trypan blue exclusion t~hniqll~ After adjustment of cell density to 1 x 107 cells per mL
with culture m~--iillm, the suspension was placed in a T-75 culture flask and inrnb~t~d flat at 37~C
in a hnmi~lifi~ tmf~ph.ore of 5 % CO2 for 2 hours. The non-adherent cell population was der~nt~d into a sterile disposable flask. PhytnhPm~gglulillill (PHA-P) was added to the PBL sll~pl-n~ n at a cullcellLldlion of 2 ~g/mL and the PBl plc;~dlion was then further in~llb~t--d at 37~C for 48 W O 97/03997 PCTrUS96/11786 hours. At this time an aliquot of the culture was used for virus i.lr~liviLy studies. PBLs (5 x 105 cells/well) were infected with HIV-1 isolates at an MOI of 0.2. This level of infection yielded a s~ticf~rtQry virus control RT activity value result at day 7 post-infection (Buckheit, et al., id.
(1991)). Two hours post-infrctinn, the cells were separated from the virus by centriguation, washed twice with culture m.o-linm, and suspended in culture mr-linm cnnt~ining IL-2 at a co~r~ ,dLion 30 units/mL and at a cell density 2 x 105 PHA-P-stimnl~trd PBL cells/0.1 mL of culture mr~ m Seven day post-infectinn, HIV-1 replication was analyzed using either the RT or p24 assay systems. Data was obtained in the p24 assays by ~ecllu~llotometric analysis at 450 nm using a Molecular Devices Vmax plate reader.

Tnhihifi~n of acute infection of ~ human macrophages. Human macrophage cultures were established as described by Crow et al. Crowe, et al., AIDS Research a7u~ Human Retroviruses 3(2):135-145 (1987). Briefly, PBMC's isolated from HIV-1 and HBV seronegative donors was allowed to adhere to glass at 37~C for two hours in calcium and m~gnP~illm free PBS
rs (pH 7.4). The non-adherent cells were aspirated and the adherent cells were washed three times with cold PBS. The adherent macrophages were scraped free from the plate, counted, and inncnl~t~l into 96 well plates at a ~;ol.c~ .dLion of 105 cells/well in RPMI 1640 mr~lillm suppl~ d with 10% hurnan serum. The llla~l~hages were cultivated in RPMI 1640 with 10%
human serum. After inrnh~tion overnight at 37~C the ll.a~ hages were infected with HIV-lDV
at a multiplicity of infectinn of 0.1 for 24 hours at 37~C in the plesellce of the in-lir~tt-d amount of drug. Unabsorbed virus was then washed off and the cells were further inrllbatrrl for 7 days at 37~C in complete mr-lillm suppl--m~-ntrd with the in~lir~t--d amount of drug. On day 7 post-infection the adherent macrophages were washed extensively with PBS and lysed with detergent.
Cytoplasmic HIV p24 levels were then ~ d and percent inhibition were c~lc~ t~d and compared to control infected but untreated cells.

Long term ~U~ iUIl studies. Long term ~u~le~sion assays were pelr~ ed in MT-4 cells infected with HIV-lMB (MOI of 0.01) using drug concentrations lc~l~S~ lLi--g 1, 10 or 100-fold over the median IC50 value for each compound. Four days post-inf~octinn, cells were washed twice with phosphate-buffered saline (PBS) and resuspended in complete medium without drug (day 0). Viral breakthrough was ~-lu--iLol~d at several time points by llledsul~lllc:llL of viral p24 antigen production in the culture mr-lillm or the presence of intr~relllll~r viral DNA as described previously, (Rando, et al, J. Biol. Chem. 270: 1754-1760 (1995)).

W 097/03997 PCT~US96/11786 Other viral assays. Re~ dLoly syncytial virus (RSV strain A2), and inflnPn7~ A (FLUA strain H3N2) virus assays were p~lrulllled as ~lesrrihed by Wyde et al. (Wyde, et al., Drug. Dev. Res.
28:467472 (1993)) while Herpes Simplex viruses types 1 and 2 (HSV-l, HSV-2) plaque re~l< tion assays were ~lrull-led as previously described. Lewis, et al, Antimicrob. Agents Chemother.
38:2889-2895 (1994) Vesicular stom~titi~ virus (VSV), Vaccinia virus, Sindbis virus, Coxsackie virus B4, Polio virus-l, and Semliki forest virus assays were p~.rul.--ed as described by De Clercq.
De Clereq, E., Antimicrob. Agents Chemother. 28:84-89 (1985). The alc:n,lvilidae assays (Junin and Tacaribe viruses) were pclrolllled as flPs~rihed by Andrei and De Clercq. Andrei, et al., Antiviral Res. 14:287-299 (1990). Punta Toro virus (ATCC VR-559) and Yellow fever virus (vaccine strain 17D) assays were p~lrulllled using Vero cells.

Flow ~IV~U~11;C analysis of HIV-l infected ly~~ les. Seven days post-HIV-l infection of PBMCs, the infected cell culture medium was analyzed for HIV-1 production using the p24 antigen-capture assay. In ~A~1ition~ cells from both the drug treated and control wells were analyzed s for CD4 and CD8 antigens by cytofluululll~Lly. Briefly, cells were washed and treated with fluolocl.lu.lle-labeled monoclonal antibodies to CD4 or CD8 (Becton Di~l~ ;"CU~). The cells were washed again and fixed with 2% pdl~rullllaldehyde before analysis. Cri~m~n, et al., ~low C~ytometry and Sorting, p. 229-230 (1990) and Crowe et al., AIDS Res. Hum. Retroviruses 3:135-145 (1987).

Single cycle analysis of HIV-1 cDNA. CEM-SS cells (2 x 106 cells/well) in 0.5 mL of c~-mpl~t~
medium were infected with HIV-ls~CI at a MOI of 1.0 for 45 minutes on ice at which time complete culture medium (10 mL) was added to the cells. The infected cells were then pelleted (1000 RPM
for 10 min. at 4~C), washed twice and aliquoted into a 24-well flat bottom plate (2 x 105 cells/well). The inllir~t~cl amount of drug was added to the infected cell cultures at various times during or post-i--r~ ,- The cells were harvested 12 hours post-infoctirn at which time cell pellets were lysed in 100 ~I polymerase chain reaction (PCR) Iysis buffer (50 mM KCI, 10 mM Tris-HCl (pH8.3), 2 5 mM MgCI2, 0.1 mg/mL gelatin, 0.45% Nonidet P40, 0.45% Tween 20 and 75 ,ug/mL PloLt:i-lase K) at 50~C for one hour followed by 95~C for 10 mimlt~ The Iysate was stored at -20~C until use.
PCR analysis of viral cDNA was pt:lrulllled using 10 ~L of total cell Iysate in a 100 ~L
reaction buffer as previously described (Rando, et al, J. Biol. Chem. 270: 1754-1760 (1995)). The primers used were 5'-ATAATCCACCTATCCCAG TAGGAGAAAT-3' and S'-TTTGGTCCTTGTCTTATGTCCAGAATCG-3' which will amplify a 115 bp segment of the HIV-l genome. The cycle conditions used were 95~C for 10 minutes to denature the DNA, CA 02227867 l998-0l-26 followed by 30 cycles of 95~C for 75 seconds, 60~C for 75 seconds, and a final rYtrn~ n step at 60~C for 10 minutes. Thirty ,uL of the amplifir-~tinn reaction were rnixed with 10 ul Of ~_32p_ l~helP~1intrrn~1probe(5'-ATCCTGGGATTAAATAAAATAGTAAGAATGTATAGCCCTAC-3'), placed at 95~C for 7.5 minutes and then ~nn--~lrc~ at 55~C for 15 minutes. The resultant ~ludu were sep~trd by electrophoresis on a 10% polyacrylamide gel.

Analysis of viral r~p~ CEM-SS cells (2 x 107) were infected with HIV-lSKI (MOI of I) for 45 minutes at 37~C with gentle mixing. Following virus ,I~ hl l lrl ll the cells were gently pelleted, washed twice and resll~p(?n~l~fl in complete tissue culture mPf~illm The cells were then divided into aliquots, treated with various cnnrr~.l.,.linn~ of drug and placed in T75 culture flasks. The cells were inrub~trd at 37~C for 18-20 hours and then harvested by centriguation. To extract nucleic acids for analysis of HIV-1 intrgr~ti~n low- and high-mnlrcul~r weight DNA were prepared from HIV-1 infected cells (untreated or treated with hl~LI,a~illg en~lr~ lion~ of drug) according to the protocol originally described by Hirt (Hirt, B.J., J. Mol. Biol. 26:365-369 (1967)) and modified by Gowda et al. Gowda, et al., J. Immunol. 142:773-780 (1989).
DNA (300 ng), obtained from the low-mnlecl~l~r weight Hirt fractions, was used as the template in PCR analysis undergoing a 30 cycle amplifir,~tinn reaction using the contlitir~n~
described by Strink~erer et al. (Strink~crrer, et al., J. Virol. 69:814-824 (1995)). PCR primer sets int~hl~ d control primers for the ~mplifir~tinn of mitorhnnArial DNA (sense, 5'-GAATGTCTGCACAGCCACTTT-3'; ~ r~e7 S'-ATAGAAAGGCTAGGACCAAAC-3';
amplified product, 427 bp); primers for the detection of early viral lldl.sc;liL,lion events (M667 and AASS primers as described by Zack et al. (Zack. et al., CeU 61:213-222 (1990)), amplified product, 142 bp); primers for the rletrcticn of the viral gag gene (sense, S'-AGTGGGGGGACATCAAGCAGCCATGCAAAT-3'; antisense, 5'-l-l-lGGTCCTTGTCTTATGTCCAGAATG-3', amplified product 300 bp); and primers for theectinn of circularproviral DNA (sense, S'-CCTTTTAGTCAGTGTGGAAAATCTCTAGCA-3';
;r~ e, 5~-CAG TGGGTTCCCTAGTTAGC-3', amplified product,536 bp). PCR products were.s~L,~"1~rd by agarose gel electrophoresis and vi~ i7rd by ethi~lillm bromide staining.

30 Reverse l~ ~se enzyme i.. l.il.;l-~-- assays. Purified lecol.. l,i.. a.. t RT (HIV-lBHIo) was obtained from the Ul-iv~ y of ~l~h~rn~ Center for AIDS research. The erl7yme assays utilized three dirr~lcll- template:primer systems, primed liboso-.ldl RNA, gapped duplex DNA, and poly(rA)p(dT)I2 l8 to evaluate the inhibition of HIV-1 RT as described by White et al. (White, et al., Antiviral Res. 16:257-266 (1991), and Parker et al. (Parker, et al., J. Biol. Chem. 266:1754-3s 1762 (1991)).

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 ase enzyme assays. Purified Lccoll~'L~ill~lL HfV-1 inrrgr~ce enzyme (wild-type) was a generous gift from Dr. R. Craigie, Laboratory of ~olPcul~r Biology, National Institute of Diabetes and Digestive and Kidney Diseases. The enzyme (0.25 ~M) was preinrllb~t~d in reaction buffer at 30~C for 30 minutes All 3'-piuces~.i,lg and strand-transfer reactions were p~.ru-.l~d as described previously by Fresen et al. (Fresen, et al., Proc. Natl. Acad. Sci. USA 90:2399-2403 (1993)) and ~ mfl~r et al. (~7llm-1~r, et al., Proc. Natl. Acad. Sci. USA 91 :5771-5775 (1994)).
Enzyme reactions were qll~nrh~-d by the addition of Maxa~;n-Gilbert loading dye, and an aliquot was electrophoresed on a flt-n~t~lring 20% polyacrylamide gel. Gels were then dried and subjected to autoradiography using Kodax XAR-2 film or exposed in a M~lecnl~r Dyll~l~ics Ph~.crhoTm~ger cassette.

~-~te~se assays. HIV-1 protease enzyme (R~rh-3m) was diluted to 166 ug/mL in 50 mM NaOAc, S mM DTT, 2 mM EDTA, and 10% glycerol (pH 5.0) and stored as 10 ul aliquots at -20~C.
HIV protease substrate I (Molecular Probes) was diluted to a working cu"r~ ir.n of 0.32 rs nmol/~L. Enzyme (20 ~LL), substr~t~ (20 ,uL) and drLtg (20 ~L) were added to each well of a i~loliL~L plate. Positive and negative controls were evaluated in parallel. Fluolcacc:llce was d on a Lab~.y~ llls Fluoroskan II using 355 nm for rYrit~tion and 460 nm e~";~ ", wavelengths at 37~C at time zero and at 30 minute intervals for 2 hours.

HeLa-CD4-,B-g~ ce cell assays. Two different assays using genetically engineered HeLa cells were pe.~----ed as described previously. Rllrkh~it et al., AIDS Research and Hurnan Re~roviruses 10:1497-1506 (1994). These assays utilized the HeLa-CD4-LTR-~'-g~l~rt~.ci~l~ce cell line (Kimpton, et al., J. Virol. 66:2232-2239 (1992)), which employ a tat protein-induced ~.d,-sa~iv~Lion of the ~'-g~l~rtoci(1~ce gene driven by the HIV-1 long terrninal repeat (LTR). One assay involved infecting the HeLa-CD4-LTR-,B-galactosidase cells with HIV-1 while the second assay monitored the expression of ~-g~l~rtoci-l~ce after inrnh~ti~n of the HeLa-CD4-LTR-,6'-gal~ ce cell with HL2/3 cells. Rllrkh.oit et al., AIDS Research and Hutnan Retroviruses 10:1497-1506 (1994); Ciminale, et al., AIDS Research and Hun.~an Retroviruses 6:1281-1287 (1990). The HL2/3 cells express both the HIV-l envelope ,lycu~lol~in and tat gene product so that co-cultivation of these cells with the HeLa-CD4-LTR-,~-g~i~rtosi-l~ce cells would allow for CD4- and gpl20--mf ~ t~d cell fusion. The extent of cell fusion can t'nen be .ll~,lliLul~d by the degree of tat transactivation of LTR-driven ,6'-g~l~rtr.ci-i~ce expression. Buckheit, et al., AIDS
Research and Hurnan Retroviruses 10:1497-1506 (1994); Ciminale, et al., AIDS Research and Hurnan Retroviruses 6:1281-1287 (1990).
3s W O 97/03997 PCTnUS96/11786 Results of the In Vibro HIV T..l.;l.;l;.... Studies As described above, the anti-HIV-1 activity, in cell culture assays of the olig~,lucle~,Lide (I100-15) composed entirely of G and T was çst~hli~hPd by the illV~ Ol::~. See also, Ojwang, et al., J. AIDS 7:560-570 (1994); Rando, et al, J. Biol. Chem. 270: 1754-1760 (1995). I100-15 was s found to inhibit HIV-lDV in SUP Tl cells with a median h.1libiLuly cQI~r~ nn (IC50) of 0.125 ~M. I100-15 was 5ynth.o~i7rd with an nnmn~lifird (natural) PD intrrmlr1Orcirlr linkage and a prop~no1~minr group attached to the 3'~ to increase the stability of the olignnllr1rQti~lr T30177, a ml~rlifiPf1 variant of I100-15, has the same seqllrnre as I100-15 but contains an hydlokyl moiety at its 3'-l~ ~..;...~ and a single PT ;..~ rleoci~ linkage at both the 5'- and 3'-ends.

Cy~ Y;- :~y Assays. The ~;ylo~ y of T30177 was ~ L~ ...;--.od using several dirr~ cell lines and primary human cells as described above. The TC2s, TCso and TC95 values obtained are shown in Table B-l . The cytotoxicity profile obtained for log phase growing cells was variable depending upon the cell line used, while the slower growing PBMCs, PBLs, and ll~aclo~ ages all tolpr~trd S the compound at cu .rc ~ ;nns çYree~1in~ 100 ~M as lllolliLol~d using the trypan blue ~3xrhl~inn~
[3H]thymidine uptake, or [3H]leucine uptake tprhniqllrs Table B-1. CYIULO~iCilY of T30177 in Pst~hli~hrd cell lines and prirnary cells.

CYTOTOXICITY (~LM)a Cell Type TC2s TCso TC9s Cell Linesb CEM-SS 50 8 + 3.2 92.0 + 3.0 > 100 MT4 34 + 4.0 70 + 7.1 > 100 CEMx174 10 + 2.5 50 + 5.2 > 100 2s MT2 27 + 3.5 61.2 i 5.5 > 100 AA5 45.66 + 2.0 94.2 ~ 3.1 > 100 U937 > 100 > 100 > 100 Vero > 100 > 100 > 100 NIH3T3 > 100 ~ 100 > 100 Primary human PBLs > 100 > 100 > 100 cellsC
PBMC > 100 > 100 > 100 Maclu~hagc:s > 100 > 100 > 100 ~- lC25, lCso, and lC95 values are t e c~ linn~ o 1iOl77 required o inhibit 25%, 3s 50% and 95% of growth (cell lines) or cell survival (primary human cells).

CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 b The cytotoxicity of T30177 in human cell lines was ~1~e~ led using log phase growing cells.
c The cytotoxicity of T30177 in primary human cells was ~l~terrnined using trypan blue exel~ n terhni~ o or by Ille~ g the uptake of [3H]thymidine or [3H]leucine on slow growing primary cells.
Tnhihitif n of Viral R~ . in Cell Lines. CEM-SS cells were infected with HIV-lR~at an MOI of 0.01 and treated with T30177, AZT or ddC for six days. In this assay system T30177 inhihited HIV-lRp replication in a dose-dependent manner with an ICso value of 0.075 ,uM
while the control drugs, AZT and ddC, had IC50 values of 0.007 and 0.057 ~M respectively (Figure B-l). T30177 was then assayed against ~ltliti~n~l strains of HIV-l in a variety of dirf~
cell lines. The results from these assays showed that the degree of inhibition observed for each strain of HIV-l analyzed was greatly inflnen~ed by the cell line used (Table B-2). In addition, as observed for DS5000, T30177 was inhibitory for the AZT-l~ L~lL strain of HIV-l tested (ADP/141) which has four mutations in its RT gene (67N, 70R, 215F and 219Q).

Table B-2. Inhibitory effects of T30177, AZT, and DS500 on viral replication.

IC50(~M)a Virus Cell Lile T30177 AZT DS5000b s HIV-1 strainsC
SKI CEM-SS 0.025 + 0.006 0.022 + 0.0001 MT2 0.06 + 0.001 0.66 + 0.005 RF CEM-SS 0.075 i 0.007 + 0.0002 0.0002 MT2 0.270 + 0.04 0.03 + 0.005 MT4 0.037 + 0.03 - 0.018 + 0.02 DV SUP T1 0.06 + 0.004 0.03 + 0.005 IIIB CEM-SS 2.83 + 0.17 0.002 + 0.0003 MT2 1.94 + 0.12 0.01 + 0.004 MT4 0.15 + 0.02 - 0.034 i 0.016 SUP Tl 0.6 + 0.06 0.03 + 0.006 AA5 < 0.32 < 0.003 ADP/141 MT4 0.27 + 0.05 - 0.032 ~t 0.008 HIV-2/SIV strains HIV-2Ron MT4 27.5 + 11.6 - 0.082 + 0.088 HIV-2,~o MT4 5.98 + 1.05 - 0.084 + 0.086 SIVMAC25~ MT4 1.5 + 1.2 - 0.548 ~t 0.48 The IC50 value is the c-",~ Qn of drug required to inhibit virus production by 50%.
The results plese~ d are the averages of three or more ~ lhll~llL~.
25 b For DS5000 the ~M units are an ~.o~hllalion based upon the average mol~c~ r weight (5000) of the m~t~ri~l used in these studies.
c The MOI used for all HIV-l, HIV-2 and SIV strains tested was 0.01.

T30177 was also tested for its ability to inhibit labu-~luly strains of HIV-2 and SIV. The results (Table B-2) from these assays indicate that T30177 is more active against the strains of HIV-1 and SIV tested than against the two strains of HIV-2 tested (ROD and EHO). In ~Mitil~n~
T30177 was found to be inactive against a variety of enveloped and nonenveloped viruses tested 35 (Table B-3) with IC50 values found to be grater than the highest concellll~Lion of drug tested (20û
,ug/mL or 37 ~M). This is in contrast to DS5000 which was found to be a potent inhihiror of all of the enveloped viruses tested except Vaccinia and Sernliki forest viruses (Table B-3).

W 097/03997 PCT~US96/11786 Table B-3. Inhibition of viral replication in cell lines treated with T30177 or DS5000.

IC50(~g/IIlL)a T30177 DS5000 T30177 DS5000b Envelope Viruses:
,. 5 HSV-l (KOS) >200 2 400 >400 HSV-2 (G) >200 2 400 >400 HSV-l TK (B2006) >200 2 400 >400 HSV-l TK (VMW1837) >200 2 400 >400 Sindbis virus ~200 10 2200 >400 o Sernliki forest virus >200 >400 2200 >400 Vesicular Stomatitis virus > 200 20 400 > 400 Vaccinia virus > 200 > 400 400 > 400 Punta Toro virus >200 10.9 >200 >400 Yellow Fever virus >200 26 >200 >200 RSV (A2) >200 4 >400 >200 Tnflu~n7~ A (H3N2) > 125 >200 >400 Junin virus > 50 13 > 50 > 200 Tacaribe > 50 13.5 > 50 > 200 Non Enveloped Viruses:
Coxsackie virus (B4) >200 >400 2400 >400 Polio virus-l >200 >400 2400 >400 Conc~,lLld~ion o~ drug required to reduce virus-induced cy op~rh--geni~ ity by 50% (lC50).
The assay results are p~e~ ed in ~g/rnL units. For T30177 5.4 ,~bglmL is equal to 1 ~M
and for DS5000 5 ,ug/mL is ~ Ox;~ rly equal to 1 ~M.
b The ",i"i"""" c(-~lrt~ lion required to cause microscopically detectable ~lt~r~tion~ is norrnal cell morphology (MCC). The results plc;s~llL~d are the averages of the three or more ~ llL~.
Tnhibit;~~ of ~V-1 R~rlir~tinn in r~ h~.al Blood Cells. The primary targets of HIV-1 infection in vivo are CD4+ T lymphocytes and macrophages. Therefore in the following set of expelilllc:llL~ the hlvt;llLol~ tested the efflcacy of T30177 on HIV-l replication in PBMCs, PBLs and macrophages.
Activated PBMCs were infected with laboratory strains of HIV-l and cultured in the presence of T30177, AZT or ddl. Tl~Lll~llL of infected PBMCs with T30177 inhibited the replication of the all four HIV-l isolates tested with IC50 values ranging from 0.12 to 1.35 ~M

WO 97/03997 PCT~US96/11786 (Table B4). In this assay AZT was more efficacious against all HIV-l isolates tested, on a molar scale, than T30177 while at the same time T30177 was more potent than ddI against the two HIV-l strains tested. It is also illL~ Lhlg to note that HIV-llIIB was more susceptible to T30177 in assays pelru.llled using PBMCs than in assays using T-cell lines (Tables B-2 and B-4).

Table B-4. HIV-l replication in primary hurnan cells treated with T30177, ddI or AZT

ICso~Ma Virus Strains (Cells) HIV-l Isolate T30177 ddI AZT
Laboratory Isolatesb IIIB 0.12 ~ 0.006 0.74 + 0.05 0.003 ~t 0.0002 (PBMCs) JR~sF 0.28 + 0.04 2.0 i 0-5 0.0025 + 0.001 RF 0.75 + 0.13 NDc 0.272 ~t 0.003 MN 1.35 + 0.10 ND 0.053 + 0.001 Clinical Isolatesb WEJO(SI) 0.30 + 0.01 2.18 + 0.026 0.017 + 0.0001(PBLs) BAKI(SI) 0.23 + 0.005 2.61 + 0.003 0.020 + 0.006 WOME(SI) 0.71 ~ 0.002 0.41 + 0.008 0.025 + 0.0003 ROJO(SI) 3.9 + 0.02 0.87 + 0.001 0.052 + 0.0004 JOGA(NSI) 0.33 + 0.004 ND > 1.0 BLCH(NSI) 3.08 + 0.006 ND 0.022 ~t 0.0008 VIHU(NSI) 1.3 + 0.02 1.21 + 0.009 0.036 + 0.0007 S. E. Asia 0.58 + 0.003 ND 0.06 ~ 0.005 2s N.Amer. #1 0.25 + 0.003 ND 0.01 + 0.004 N.Amer. #2 2.92 + 0.005 ND 1.65 + 0.007 942716 0.86 + 0.006 ND 0.002 i 0.003 942751 0.38 + 0.003 2.2 + 0.02 0.028 ~ 0.0025 Concentration of ~rug required to i hibit viral product on by 50% (ICso) was ~il'.t~ nnin~od using the Coulter p24 antigen capture or RT assays.
b Antiviral assays were performed using laboldLoly strains of HIV-l in peripheral blood mon()m~rlr~r cells (PBMCs) or using syncytium in-l--ring (SI) or non-:;y~l.;yLiulll in~inring (NSI) clinical isolates of HIV-l in PBLs.
3s c The value was not deterrnined (ND).

~ i =
The thel~u~ic potential of any anti-HIV drug is dependent upon its ability to inhibit clinical isolates of the virus obtained from dirr~ t geogr~rhit~l locations. Therefore, the hlve~ evaluated the ability of T30177 to inhibit the infection of PBLs using a variety of clinical isolates of HIV-1 which were both Syll;y~iu~l in~ cing (SI) and non ~yll~;~yliulll incln~ing (NSI) strains of HIV-1. In ~litit~n, the isolates used in this study had their origins in dirr~lcllL
" geographic regions. After infection with HIV-1 the PBLs were cultured in the presence of T30177, AZT or ddI for seven days. T30177 inhihit~cl the viral replication of all the HIV-1 isolates tested with IC50 values ranging from 0.23 to 3.08 ~M (Table B-4). In the same assay, AZT and ddI had IC50 values ranging from 0.01 to 1.65 ~M and 0.41 to 2.61 ~LM, respectively. It is important to note that T30177 was active against both NSI and SI isolates and was very active against he JOGA
isolate which was obtained from a pediatric patient. The JOGA isolate was also observed to be relatively resistant to AZT treatment (Table B-4).
Another major target cell of HIV-1 infecti~n is the macrophage. Fully ~Iirre.~l.Li~Led macrophages were infected with HIV-lDV and treated with T30i77 or AZT. T30177 ~ignifi~ntly inhibited HIV-1 replication in macrophages (Figure B-2). However, due to the long exposure of cells to virus (24 hours), T30177 and AZT worked best when ~lmini~tered at con-entrations above the IC50 values obtained for these drugs in assays performed in established cell lines.

Variations in Viral MOI. To investigate the effect of variations in the MOI on the anti-HIV-1 activity of T30177, CEM-SS or MT4 cells were infected with various MOIs of HIV-1RF or HIV-1IIIB (Table B-5). Unlike AZT, T30177 was much less sensitive to changes in the viral MOI. For example in these assays when the MOI of HIV-lRF was changed from 0.01 to 1.28, T30177 only exhibited a 14-fold increase in its IC50 value while at the same time the IC50 value for AZT
increased over 1000-fold (Table B-S).

-CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 Table B-5.
Effect of changes in viral multiplicity of infection (MOI) on the anti-HIV-l activity of T30177 and AZT.
IC50/IC90 in 231lMa HIV-lbMultiplicity ofT30177 AZT
Isolate/Cell Infection RF(CEM-SS)0.01 0.20/0.50+0.01/0.030.01/0.19+0.001/0.001 0.02 0.41/1.50+0.03/0.040.02/0.47+0.009/0.046 0.04 0.60/1.56+0.01/0.020.07/0.86i0.005/0.03 0.08 0.70/1.56 ~0.01/0.080.50/1.0+0.01/0.005 0.16 0.87/1.6~t0.01/0.030.6/> 10.0~t0.05 0.32 1.25/4 7+0.15/0.27 8.5/> 10.0 0.64 2.64/4.75~t0.05/0.16 > 10.0/> 10.0 1.28 2.81/4.77+0.04/0.06 > 10.0/> 10.0 IIIB(MT4)0.02 3.1/6.6iO.23/0.8 0.037/0.22+0.003 0.01 2.7/9.2+0.03/0.250.01/0.03~t0.002 0.3 3.38/7+0.15/0.50.15/3.3 ~t0.01/0-05 1 6.8/26+0.53/5.1 0.42/3412+0.1/10 a The concellLldLion of drug needed to limit virus production by 50 (IC50) and 90 ((IC90) percent as measured in the cpe assay b The strain of HIV-1 and cell line used for each assay is intlir~tef~

Effect of T30177 on CD4 and CD8 T-cell Snhs~tc One of the principal immunological markers correlated with plo~lc~ion to AIDS is the decline in T lymphocytes which express the CD4 cell ~ "~i"i"g marker (CD4). The change in CD4+ T-lymphocytes is usually lll~uiLol~d by noting changes in the ratio of CD4+ to CD8+ lymphocytes in the blood. To ll.oterTnin.o the effect of T30177 Llc:dLlllc:lll on the CD4/CD8 ratio, CD4 and CD8 antigen expression was analyzed on the surface of cultured PBMCs seven days post-infection with either laboratory strains or clinical isolates of HIV-1. In these e~pr,;",~ dL~ llL with either AZT or T30177 increased the number of CD4+ T-cells in the cell culture, relative to untreated infected cultures (Table B-6). The observed increase in CD4+ cells was ~ep.on-l.ont on the drug con~ dLion used and was inversely correlated with the level of virus production (Figures B-1 and Tables B-2 and B4). These results suggest that the blockage of HIV-1 replication p~r~llel~ the ~u~ples~ion of the ~;ylop~Lllic effects of the virus in primary human lymphocytes.

CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 oo a~ ~~ ~~~~ oooo oooo oooo 0OOO
V

~ 0~ ~~ ~~

V V ~ ~ ~

~a~ ~ ¢ ¢ ¢ O O O ~o,, ,,~ 0 0 _ O
V~ ~ ~ aO ~ aO ~ O ~ ' ~ aa z_OO ~ ~0 z~OO ~0 z_OO
._ a Z~ 0_~ 0O,O,,O o_u, 0OOO 0_~
-CQ 8 ~ ~ z ~ ¢ ¢
~ ~, Z ~
~ a .oo C~

o _ c~ D O ~ ~ ~ O ~ ~ ~ O c~ ~ ~ O ~ ~ ~ -- C~
E- a C~ ~ ~ ~ ~ ~ ~
¢ ~
O oo ¢ ~ ¢ o O ~ ~ ~ ¢ ~¢ ¢ ~ ~ ~ E- ~ ¢ ¢ ¢
V 0~ O Q ~ ~ X ~ Z ~ Z - ~ ~
a Z Z O _ ~ O O 0. 0O 0 _ ~" _ O O O O _ ,,, -- C ~ Z ~y CA 02227867 l998-0l-26 -O O ~
o o o o ."
cr~ c~ co co O

~t ~ ~ C~ D

O o o O

_ o o ~, ._ O _ ~ O ~

r&~

.~
-O O ~

.~ O
~a~ ~ O
~'I ~ ~ ~ D

O ~ ~ 5~ ~ O

-L C
-- O o 0 CC C~
O _ ~ O ~ C~

S --WO 97/03997 PCT~US96/11786 In vitro some HIV-l isolates infect CD4+ lymphocytes, shed infectious virus into the culture medium but do not cause destruction of the infected cells. Garry, R.F., AIDS 3:683-694 (1989). This may explain results obtained when the hlvc:llLul~ used the North American isolate number 1 ~N. Amer. #1, Table B-5).
When this virus was used to infect PBMC's, in the absence of drug, a CD4/CD8 ratio of 0.35 was observed 7 days post-infection. At the same time analysis of the culture medium from cells infected with this isolate revealed the presence of viral p24 antigen ~Table B-4) which 5nggpctpd that a productive viral infection had occurred.

Time of drug addition studies. T30177, DS5000 or AZT was added to MT4 cells infected with HIV-llllB
~MOI of 1) at various times post-infPctinn Test c.,llll,uullds were added at a ~-n~ .r,.l~ .lir~n 100-foldhigher than the ~lrl~ d IC50 value for each drug in the standard assay pelrulllled using MT-4 cells and the IIIB
strain of HIV-l ~Table B-2). Viral p24 antigen levels were IllulliLuled 29 hour post-infecti~n The results of this assay indicate that pû~Ll,o~ ~hlg the addition of T30177 for one hour was enough to ~ lly reduce the inhibitory effects of this compound in a fashion similar to that of DS5000 and clearly different from AZT
which lost its ~lole~live capacity when added to the cell culture medium 3 or 4 hours post-infection (Figure B-3). A similar result was obtained when cn~ g T30177 with CSB, a known inhibitûr of both virus binding to cells and fusion related events (Clanton, et al., J. Aids 5:771-781 ~1992)), in that the antiviral activity of both T30177 and CSB was greatly reduced if added to infected cell cultures one hour post-virus infection ~data not shown).
HeLa-CD4-,~ cell studies. To dirr~ the effects of T30177 on early events in the viral life cycle, through intPgr~tinn and ~ SP~l '' ' 11 production of the tat gene product, from the inhibition of HIV-1 gp 120-mP~ tPd cell fusion two ~ protocols were employed. The first protocol IllolliLul~,d the effects of the drug on the ability of HIV- 1RF to infect and/or replicate within HeLa-CD4-LTR-,B-galactosidase cells and was performed as described in Methods. In this ~ I drug interdiction at any step in the viral life cycle through the production of the tat gene product would cause a decrease in C~ iVll of the ~-g~ tngifl~c~P gene, the l ~ ion of which is regulated by the HIV-l LTR. The results show that T30177 is a potent inhibitor of ,~-g~l~rtncifl~ce production in this assay with an IC50 value of 0.009 ~uM, while the IC50 value obtained for CSB in the same PYpPrim~nt was 0.26 ~cM ~Figure B4A). In control ~
T30177 had no observable direct effect on B-g~ tnci~l~ce en~yme activity at c~."t~ innc up to 10 ~LM
~data not shown).
The second protocol used was a virus-free assay designed to monitor CD4- and gpl20-m~i~tPcl cell fusion events. In this assay T30177 was able to interfere with the fusion process ~Figure B-4B). However, the observed IC50 value (l ,~LM) was ~luxilllaLely 100-fold higher than that needed to interfere with ,B-gzll~ctc~citl~ce production in the virus infection assay (Pigure B-4A). In the same assay system the IC50 value observed for CSB h~ ed ~p~v~ill~L~ly 3-fold to 0.8 ~LM over the cn~ ~"~ ion needed to interrupt ,B-g;~l~rtoci,l~ce production in the virus infection assay (Figure B4).
,.

WO 97/03997 PCT~US96/11786 The thrce--limPnci~ structure of an oli~l~nllr1~oti-l~ with the sequence of T30177 is stabilized by the form~tinn of an intr~m~llec~ r G-octet, (Rando, et al, J. Biol. Chem. 270: 1754-1760 (1995)). Previously the hlv~llLu-~ have reported how the repl~rf~m~nt of one of the Gs involved with tetrad formation with a deo~y-~A~ (A) reduced the anti-HIV-l activity of the resultant molecule. Rando, et al, J. Biol. Chem.
270:1754-1760 (1995). To ~lPtt-rminP the effects of intr~m~-lec~ r tetrad formation in T30177 on the observed inhihiti~n of ,B-g~l~rt~)~iA~e production in the two assays p-~ -L~d in Figure B~, the hlv~llLul:, tested T30526, an olig~ rl~ootif~p in which a dA has been sllbstitllted for a dG at a position that would interrupt the form~tion of one of the two tetrads involved in the G-octet. T30526 has the sarne partial PT
patterns as T30177 (Table B-7). T30526 has the same partial PT pattern as T30177 (Table B-7). T30526 was found to be a~ O~illldL~ly 100-fold less potent that T30177 in inhibiting HIV-lRF production in culture assays (Table B-7), 10-15-fold less potent at inhi~iting virus-infected cell ~B-g~l~rt~ciA~c~ production (Figure B-4A) and did not inhibit cell fusion at the highest cu"rr~l,, l ion of drug tested (20 ~LM, Figure B-4B).

-~W ~ ~ ~ ~ ~ A ~~ ~A O ~A ' ' 'o O

~ O ~ o o ~n ~ O o O o A ~ A ~ ~

~? _..._ ~ o o ~o o o o ~o s c -- ~ A A A O o O ~
~ C

s ~ O ~ A A ' ~ ' O O - é

:-- 00 ,_ O ~ O ~, ~ ~ ~ O O O O o S

C S
m ~ i E-- u c ~ ~ ~ 8 c~ o o c ~ ~ ~
c ~, ) ~ c ~ ~ ;~ 5 rc ~ ~ ~ c - e ~ 9 9 e . . _ ._ _ c u, ~ O

rc ~s ~ c ~ u~ 1-- 00 'D c.~ O C:~ -- oo ~ _ _ O ~C O
o o o o o o o o o o o o fi ~ ~ v V ~
~-- E-- E-- ~-- V E-- E-- E-- ~ E-- V ~ ~ '~ V

~ : .

W 097/03997 PCT~US96/11786 c Olignmlrlrotil1.oc were 5ynthrci7rd with either total phngrhnrli~ctf~r (PD) backbone, total phosrhnrothioate (PT) backbone, or partial phn~ lluLLioate (pPT) backbone, in which the 5'- and 3~-prnllltim~te int~rn~lrl.ongi~e linkages were phosphorothioate.
S d For DS5000 the ~M units are an d~pl.~ inn based upon the average mnlrclll~r weight (5000) of the material used in these studies.
Long term ~U~ iOn of HIV-l. In separate ~ r~ lr ~1~, HIV~ lB infected MT4 cells were treated with T30177, AZT, DS5000, or the bicyclam cu.ll~uul.ds JM2763 or JM3100 for four days using drug COllCellLldLiOnS equivalent to 1, 10 or 100-fold over their lt:~})C~;LiVC IC50 values (Table B-7). The IC50 values used for JM2763 and JM3100 were from previously reported results, (De Clereq, et al., Antimicrob. Agents Chemother. 38:668-674 (1994)). After four days in culture the cells were washed and then further cultured in complete medium without drug. The cells were ~onilu~cd daily for the ayl,e~d.~cc~ of viral-induced ~y~CyLiulll formation and every second or third day for viral p24 antigen in the culture medium. In cells treated with T30177, at 100-fold over the IC50 value (d~ylu~hlli~Lely 10 ~M), ~ul)ylc~ion of virus P24 production was observed for at 1st 27 days after removal of drug from the infected cell culture (Figure B-5).
FU1L11~1111U1~7 there was no d~t~ct~hle viral cDNA (by PCR analysis) in cells ~ min-~dup to 11 days after the removal of T30177 from the infected cell culture (data not shown). Cells treated in the same fashion with AZT, DSS000, JM2763, or JM31000 had ...~,aiu~dl)le levels of viral p24 antigen in the culture medium within 3 days after removal of the drug (Figure B-5). The degree of continned ~u~ ioll was cnntingPnt upon the c~- ,r~ linn of T30177 used in the assay and the duration of the drug h~,aLlllc:llL regiInen (data not shown). The cu..cellLldLion and duration of ~ I regimen data are c~ r~ I with those previously reported for I100-15, (Rando, et al, J. Biol. Chem. 270:1754-1760 (1995)).
To ~ t~nin~ if exposure of cells to T30177 protects them for 5~hsequ~nt infection with HIV-l, cultures of HIV-l infected MT4 cells treated for 4 days with T30177 (100-fold over the IC50 value) were washed and then .~;-.rc. ~ed with HIV-llllB before ~c~ ;nn in fresh culture medium without drug. In these assays there was no protection of cells from the second round of viral infection (data not shown).

Single cycle analysis of viral cDNA. Total DNA from HIV-lSKI infected CEM-SS cells was isolated 12 hour post-infection and analyzed for the presence of viral cDNA as ~ crrihed in Methods. In this ~
viral cDNA was detected in cells treated with 1 or 10 ~LM T30177 ~d~lu~-illldL~ly 10- to 100-fold over the IC50 value) even when the drug was added to the cell culture at the time of virus infection (Figure B-6). This is in contrast to the results obtained when the adsorption blocking drug CSB (10 ~LM), the mlrl-oo~ RT
irlhibitor ddC (10 ~LM), or the nnnmlrl~oQi~ip RT inhibitor UC38 (1 ,uM) were used as control drugs. UC38 is an analog of ~ l.ii"~ b"~ P Bader, et al., Proc. Nat~. Aca~. Sci. U.S.A. 88:6740-6744 (1991);
McMahon, et al., Proc. Natl. Acad. Sci. USA (1995). As expected there was no detectable viral DNA in cells treated during, or very soon after, virus infection with any of the three control drugs when used at cu c~..l...linns 10- to 100-fold over their reported IC50 values (Table B-7, Figure B-6).

Analysis of rerlir~f~d viral DNA. The h~vcllLul~ have pr-eviously reported on the presence of viral cDNA
in T30177 treated SUP T1 cells 36 hour post infection with a lower MOI of HIV-lDV. Rando, et al, J. Biol.
Chem. 270:1754-1760 (1995). As described above, viral cDNA was also detected in T30177 treated cells 12 hour post-infection with a high MOI of virus (Figure B-6). To tlt~tf~rrninf~ the extent of viral r~plir~ti~n S within these cells PCR primers were used which would dirrc ~lLi~Lc between the different stages of viral replication through the production of circular proviral DNA (2-LTR circles). The results of these Ci'.~)ClilllCll~::i indicate that viral replir~tion has occurred in the T30177 t}eated cells up to an inl-1n-1in~ the production of 2-LTR circles (Figure B-7).

Tnhihiti~n of viral enzymes. Oli~l mlrl~-otides with PT backbones have been reported to be much more potent inhibitors of HIV-1 reverse tr~nsrrirt~e (RT) than the same m-lleclllPs with PD backbones. Ojwang, et al., J. AIDS 7:560-570 (1994). T30177 was able to inhibit HIV-1 RT however, the ~-O~ .l ion needed to inhibitthe enzyme by 50% was above 5 ,ILM when gapped duplex DNA or RNA:DNA tf mpl~t~os were used (Table B-8). It is hl~elcsli~g to note that when the primed ribosomal RNA template was used the IC50 value for T30177 was in the 1 ~LM range (Table B-8).
Table B-8.
k~hihiti~n ~~f l~~~.,1~;................................. ,. I HIV=1 Reverse T~

IC50(~LM) Template T30177 AZT 5'-~
poly(rA) p(dT)12-18 11.0 0.59 4.2 0.6 gapped duplex DNA 8.0 0.47 10.0 0.40 ribosomal RNA 1.2 0.019 0.36 0.008 a The Crlllr~ Lion required to inhibit enzyme activity by 50% (ICso) is given for duplicate c~ lellL~ in ~M units.

T30177 was also tested for its ability to inhibit HIV-1 protease and integrase cl,~yllle~.
When co"~ n~ of T30177 up to 10 ~M were used in protease inhibition assays no effect on CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 the viral enzyme was observed (data not shown). However, when assayed for its effect on HIV-l lase, T30177 was able to reduce both the 3'-processing and strand transfer activities of the integrase enzyme with IC50 values of 0.092 and 0.046 ~M, respectively (Table B-7).
To determine if the seqllenre, three ~limrn~ n~l structure, rhrnnir~l composition of the S backbone or a collll.illdLion of these p~ contributed to the observed anti-hlL~ dse activity of T30177, the hlvellLol~ ~yllllle~ l and tested for enzyme inhibitory activity the olignmlrlroti-le~
shown in Table B-7. T30038, T30175, and T30526 are variations of T30177. T30340, T30341 and T30659 are variations of the Llll~,llll.ill-binding aptamer seqnr-nre reported by Bock et al. Bock, et al., Nat~re 355:564-566 (1992). Both the dG-rich s~quenre of the anti-HIV-1 oligonucleotide T30177 and the Lhl.3lll1Jill binding aptamer have been shown to fold upon themselves to form structures stabilized by intramolecular G-octets. Rando, et al, J. Biol. Chem. 270:1754-1760 (1995).; Srhlllt7r, et al., J. MoZ. Biol. 235:1532-1547 (1994); Wang, et al., Biochem. 32:1899-1904 (1993). Oligonucleotides T30531, T30658 and T30662 are variations of the ;."li~ e compound GEM91 reported to be a potent inhibitor of HIV-l . Agrawal et al., Antisense Research and Development 2:261-266 ~1992).
The IC50 values for each of these oligonucleotides tested in the hl~egldse assay are shown in Table B-7. The results of this experiment indicate that any of the seqnrnre motifs tested were potent inhibitors of the HIV-1 integrase enzyme when the oligonucleotides were ~y~ e~ d with a PT backbone. When the number of PT linkages in the backbone was reduced to one linkage at each end of the molecule (pPT) the lhlollllJill binding aptamer (T30559) and the anti~en~e sequence (T30662) no longer displayed anti-integrase activity while the level of inhihiti-)n observed using T30177 was relatively the same as that observed using the total PT version of this molecule (T30038). For compounds with total PD backbones only the total PD version of T30177 sequence motif was able to inhibit viral i~lr~ ee with IC50 values of 170 and 125 nM for the 3'procec~ing and strand transfer enzyme activities, respectively. T30526, the tetrad-disrupted mutant version of T30177, was still able to inhibit viral hlL~gldse protein in this assay, albeit at a conrentration 2-to 3-fold higher than that observed using T30177.

C~ o-.~ of the In Vitro HIV Tnhihi1inn Studies The illVt~ e~p~ntlPd upon the earlier obs~lvdLions of their initial studies (see also, Ojwang, et al., J. AIDS 7:560-570 (1994); Rando, et al, J. Biol. Chem. 270:1754-1760 (1995)) CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 on the anti-HIV-l activity of dG-rich oligonucleotides by ~ n~ dLillg the efficacy of T30177 against multiple laboratory strains and clinical isolates of HIV-l. Using the cytotoxicity (Table B-l) and the efficacy data (Tables B-2 and B4), it was found T30177 to have a wide range of thPr~pelltir indices (TIs) depending upon the viral strain and cell line used in a given assay. For example, when T30177 was used to inhibit HIV-ls~ in CEM-SS cells a TI of 3680 was obtained.
However, when measuring the effect on HIV-lRF in MT2 cells, the TI for T30177 was only 226.
The va}iability in efficacy of T30177 in PBMCs and PBLs, which depended upon theclinical isolate tested, was very similar to the variation in activity observed for the nucleoside analogs AZT and ddI. It is hlL~ Lillg to note that an ~lv~h-l~L~ly 20 fold variation in the IC50 value was observed for T30177 when used to inhibit HIV-llIIB in CEM-SS cells (2.8 ~M) versus PBMSc (0.12 ,uM) (Tables B-2 and B-3). An ç~pl~n~tion for this observation rnay be that when viruses are ~,u~ag~d contimlously in homogeneous cell lines the "adapt" to those cells and begin to display phenotypes dirr~ l from low passage clinical isolates. Therefore, results obtained using clinical isolates to infect heterogeneous populations of primary cells (PBMCs or PBLs) may be more predictive of in vivo efficacy than data generated using labu~aLol~ strains of HIV-l in est~hli~hPd cell lines. It is unlikely that HIV-lIIIB is a ieSi~L~lL strain of HIV-l since T30177 was more ~rr~ ve against this virus in PBMCs than in cell lines. However, given the well do~ " ".~. ~Ir.d ability of HIV-l to mutate and thus develop rçci~t~nre to known therapies, efforts are underway to determine if resistant mutants can arise after ~ aL~.e--L of HIV-l infected cells with T30177.
In m~cll~ni~m of action studies it was found that T30177 displayed some antiviral which in-ljr~te~l a mPrhzlni~m of action similar to the known blockers of virus adsorption or virus mPrli~ted cell fusion such as dextran sulfate and CSB (Figures B-3 and B~). Like CSB and DS5000, T30177 needed to be added to cells at the time of or soon after virus infection. However, T30177 is 100-fold less effective in inhihiting gp 120-induced cell fusion events than it is at inhibiting early events in the viral life cycle, suggesting a specific point of interdiction with virus distinct at least from that of CSB. In ~ hion~ the antiviral profile of T30177 also displayed other characteristics which distinguished T30177 from DS5000 and CSB. ~or example, while DS5000 is active against a wide range of enveloped viruses, T30177 appears to be a more selective inhibitor of l~Ll~vil~ses with ~ .x;.. ll efficacy displayed when used to inhibit strains of HIV-l (Tables B-2 and B-3).
Exp~li,l,el.L~l results ~,esG"Led in Figure B-6 show that unlike control drugs CSB, AZT, and UC38, when T30177 was added to cell cultures during virus infPctir,n it was unable to CA 02227867 l998-0l-26 completely block viral infectitm even when used at conrPntrations 100-fold ove} the IC50 value ( ~
10 ~M). Furthermore, analysis of viral DNA fiPnn~tn~trated that viral replicative intermP~ tPs inrhl-ling circular proviral DNA were present in infected cells treated with T30177 (Figures B-6 and B-7). This data, coupled with the ability of T30177 to completely ~U~plCS5 virus outbreak (Figure B-5), and possibly clear virus from infected cell cultures, after removal of drug from infected cells (a profile not observed for AZT, DS5000, JM2763 or JM3100), suggests that a second mPrhA~ of action, distinct from inhibition of virus binding or inhibition of cell fusion events, is at work One possible alternative mPrh~ni~m is that T30177 illLclrclcs with the viral illLc~ldLion process. A cullllJilldLion of activities inr11l-1ing inhibition of virus ;~ r]""~."l or intrrn~ tion, virus-mP~ tpd cell fusion events and viral hlLe~ldLion could explain the loss of virus from infected cell cultures. This typothesis is supported by the observation that T30177 is a potent inhihit~r of HIV-1 intPgr~e function in vitro (Table B-7) and by the observed arcllmlll~til~n of circularized proviral DNA in the low-m- lecnl~r weight Hirt DNA fractions (Figure B-7).
It is clear that highly charged molecules such as DS5000 and oligonucleotides with total PT backbones are PxrellPnt i"hilli~r,)~ of the illLc~lase enzyme in vitro. However, since T30177, of all the pPT molecules tested Ill~ ;.i"Pd its level of enzyme inhihhory activity (Table B-7), it is unlikely that the mPrh~ni~m of inhibition is totally based upon a polyanion effect as seen for compounds such as DS5000 and suramin. Carteau, et al., Arch. Biochem. Biophys. 305:606-610 (1993). It is unclear at this time whether the G-octet structure, with the two base long dG loops, found in T30177 is of L,aldllwullL importance for inhibition of viral integrase since the G-octet seqnPnre found in T30659 did not inhibit hlLe~ ldse activity while T30526 (tetrad disrupting mutant) was able to inhibit enzyme activity albeit at a reduced level.
While the time of drug addition studies would suggest illLcl Çclcnce with virus intPrn~li7~tion as a key mPrh~ni~m of action for T30177 it is also clear that readily ~l~Ptect~hlP viral nucleic acids do enter the cells. It is quite possible that T30177 inhibits HIV-1 via several ~lirrclen~ mPch~nicm~
of action. Another possibility is that T30177 is carried into the cell along with the infecting virus or is slow to ~rcnm~ tP within cells (Bishop et al. 1996 J. Biol. Chem. 271:56988-5703) hence the need to add drug during virus infection. Fxl.cli",.~ signpd to address these possibilities are underway.
The recently reported cll~lgcllce rate of drug-lc~i~L~ulL virus to current approved therapies for HIV-1 (T"~ of a~pi.~xi",~ ly 2 days) suggests that single drug therapy for this virus cannot succeed (Ho, et al., Nature 373:123-126 (1995); Wei, et al., Nature 373:117-122 (1995), and CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 therefore, a likely Lled~ llL regimen for any new drug r~n~ t~o would be in collll)hla~ion with one or more other drugs which have differing antiviral m-rh~ni~m~ of action. Further experimentation might ~let~rmin~o that the actual m~h~ni~m of action for T30177 may not be via either inhibition of virus binding/intern~li7~rion or inhibition of viral integration, however, it is S unlikely that this oligonucleotide is acting via the same mP~h~ni~m as drugs ~;ullc;llLly in use for HIV-1. In additional studies the Applicants have d.-t~rmin.od that T30177 is stable in serum and within cells, with a half-life measured in days ~Bishop, et al. J. Biol. Crzem. 1996 271:5698-5703).
This hlrolllldLion taken together with the ability of T30177 to suppress HIV-1 for over four weeks after an initial LleaLll~llL regimen, in culture, makes this class of compounds an dLLld ;~iv~ ç~nrli~1~t~
for development of oligonucleotide-based therapeutic agents for HIV-1.

C. Site of Activity Studies-Viral Inte~rase Inhibition Next the illVt;lltUl:!i undertook studies to d~llloll~tldLe the potent inhibition of HIV-1 integrase by oligonucleotides Cu~ i..g intramolecular gn~no~in~ quartets or octets al~bl~;viaLt:d (G4s) and to provide better underst~n~ing of the structure-activity results from a series of these structures and the site of molecular interactions with HIV-1 hlL~ ldse. The relevance of these findings with respect to HIV-1 integrase binding to its DNA substrate and to dilll~liGdLion of the l~ vilal genome was also reviewed.

Materials Used in Site of Activity Studies ~aldlion of oli~o~ f~ f ~ lldl~s and ;,.l.;l,;l.~.~ The following HPLC purified oligonucleotides were ~ulcllased from Mi(1l~n~ Certified Reagent Company (Midland, TX):
AE117, 5'-ACTGCTAGAGATl-rTCCACAC-3'; AE118, 5'-GTGTGGAAAATCTCTAGCAGT-3';
AE 1 5 7, 5 ' -GAAAGCGACCGCGC C-3 '; AE 1 46, 5 ' -G G A C G C C A T A G C C C C G G C G C G G T C G C T T T C - 3 '; A E 1 5 6, 5 ' -GTGTGGAAAATCTCTAGCAGGGGCTATGGCGTCC-3 '; AE 1 1 8S, 5 ' -GTGTGGAAAATCTCTAGCA-3'; RM22M, 5'-TACTGCTAGAGATTTTCCACAC-3'. The AE117, AE118, and the first 19 nucleotides of AE156, correspond to the U5 end of the HIV-1 long termin~l repeat (LTR).
To analyze the extents of 3'-processing and strand transfer using 5'-end labeled substrates, AE118 was 5'-end labeled using T4 polynucleotide kinase (Gibco BRL) and y-[32Pl-ATP (Dupont-NEN). The kinase was heat-inactivated and AE117 was added to the same final cnn~.ontr~tion WO 97/03997 PCT~US96/11786 The mixture was heated at 95~C, allowed to cool slowly to room L~ aLule, and run on a G-25 Sephadex quick spin colu nn (Boehringer IVI~nnh~im) to separate ,7nn~7led double-stranded oligonucleotide from unincorporated label.
To analyze the extent of strand transfer using the "precleaved" substrate, AE1 18S was 5'-S end labeled, ,7nn.?~le-7. to AE117, and colurnn purified as above.
To analyze the choice of nucleophile for the 3'-~rocessillg reaction, AE118 was 3'-end labeled using Ix-[32-~-cordycepin triphosph,7te (Dupont-NEN) and terminal L~ r~Ldse (Boehringer Manheim). Fngl~m~7n, etal, Cell67, 1211-1221(1991); Vink, etal., NucleicAcids~es. 19, 6691-6698 (1991). The Ll~~ ase was heat-illa~;Liv~Led and RM22M was added to the same final concentration. The nixture was heated at 95~C, allowed to cool slowly to room L~ dLulc:, and run on a G-25 spin colurnn as before.
To c7etermin~ the extent of 30mer target strand generation during ~7.i.cint~gration, Chow, et al., Science 255, 723-726 (1992), AE157 was S'-end labeled, ,7nnlos71~d to AE156, AE146, and AE117, ,mnt~z7l~r7 and colurnn purified as above.
Oligonucleotides composed of deoxygnS7nr~7cin.o and thyrnidine were synth~ci7~r7, purified, and inrn~h,7ted with poLds~,iulll ion to generate the G4s. The gn~7nocin~ quartet (G4) fornung structures were then purified as previously ~7es~ riher7 Rando, et al., J. Biol. C~em. 270, 1754-1760 (1995).

Integrase ~l~)t~:~ and assays. Purified l~c~ wild-type HIV-1 integrase, deletion mutants IN1-2l2, INso-2ss, INso-2l2, gl7chm,7n et al., Proc. Natl. Acad. Sci. U.S.A. 90, 3428-3432 (1993), and INI-55 and site-directed mutants INFI8sKlc28os and INFI85 VC280S~12Nn~16N were generous gifts of Drs. T. Jenkins and R. Craigie, LaboldL~ly of Molecular Biology, NIDDK, NIH, Reth~s~k7 MD. Dr. Craigie also provided the expression system for the wild-type HIV-1 int~gr,7ce A
plasmid ~nrorling the HIV-2 hlL~,Idse was generously provided by Dr. R.H.A. Plasterk (Netherl,7n~7c Cancer Tnctir77tec). Purified rec~,l.l7 hl~lL wild-type FIV and SIV i~ es were generous gifts of Drs. S. Chow (UCLA) and R. Craigie (NIDDK), respectively.
Integrase was prein~-77bS7ted at a fina7, cu-~r~ n of 200 (for H7V-1 and HIV-2) or 600 nM (for FIV and SIV) with inhibitor in reaction buffer (50 mM Nacl, 1 mM HEPES, pH 7.5, 50 ,uM dithiothreitol, 10% glycerol (wt/vol), 7.5 mM MnCI2 or MgCL2 (when specified), 0.1 mg/mL
bovine serum albumin, 20 mM 2-lllel~ .t--et7r7,7n~-1, 10% dilll~L7llyl sulfoxide, and 25 mM MOPS, pH 7.2) at 30~C for 30 minutes. When m~7gn~cillm was used as t7he divalent metal ion, CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 polyethylene glycol was added at a final conrentration of 5% to increase activity as previously described (Engelman & Craigie, 1995). Preinrllhs-tion for 30 minutes of the enzyme with inhibitor was pelro~l--ed to o~Lh~ increases the inhibitory activity in the 3'-processing reaction (Fesen et al., 1994). Then, 30 nM of the 5'-end 32P-labeled linear oligonucleotide substrate was added, and S inrnh~tion was cr~ntinll~d for an additional 1 hr. The final reaction volume was 16 ~uL.
Di~h~ .dlion re~rtionC, Chow, et al., Science 255, 723-726 (1992), were p~lrulllled as above with a Y oligonucleotide (i.e., the branched substrate in which the U5 end was "i..~.aL~d"
into target DNA) was used.

Ele~llu~horesis and 4~ n Reactions were quenched by the addition of an equal volume (18 ~L) of loading dye (98% deionized f..",.~",i-l.o, 10 mM EDTA, 0.025% xylene cyanol, 0.025%
bromophenol blue). An aliquot (5 ~L) was electrophoresed on a den~tllring 20% polyacrylamide gel (0.09 M Tris-borate pH 8.3, 2rnM EDTA, 20% acrylamide, 8M urea). Gels were dried, exposed in a Molecular Dynanucs Ph~sph~rim~ger cassette, and analyzed using a Molecnl~r Dynamics phosph~rim~ger (Sunnyvale, CA). Percent inhibition was r~ nl~tecl using the following eqn~tion 100 X[1 - (D - C)/(N - C)], where C, N, and D are the fractions of 21mer substrate collv~.L~:d to l9mer (3'-processing product) or strand transfer products for DNA alone, DNA plus integrase, and integrase plus drug, l~e~;lively. IC50 was determined by plotting the drug c~ l ion versus percent inhibition and del~"lli"illp the concentration which produced 50% inhibition.

W cros~linking CA~ i. The method used has been described by Fnglem~n et al.
Engelman, et al., J. Virol. 68, 5911-5917 (1994). Briefly, inrrgr~e (at the in~1ir~trl1 concentration) was inrllh~tpd with substrate in reaction buffer as above for 5 minutes at 30~C. Reactions were then irr~ ted with a UV tr~n~ oi (254 nm wavelength) from 3 cm above (2.4 mW/cm2) at room L~ cildLule for 10 minutes. An equal volume (16 ~L) of 2X SDS-PAGE buffer (100 mM
Tris, pH 6.8, 4% 2-mercaptoethz~n~-l, 4% SDS, 0.2% bromophenol blue, 20% glycerol) was added to each reaction. Twenty ~L aliquots were heated at 95~C for 3 minutes prior to loading on a 12%
or 18% SDS-polyacrylamide gel. The gel was run at 120 V for 1.5 hours, dried, and exposed in ~ a Phn~phorlm~ger cassette. For inhibition of DNA binding expe~ .,L~ (Fig. C-3), integr~e (200 nM) was preinrllb~ted with the gu~no~ine quartet (at the in-lir~tl-cl cvllce:l.Lldlion) for 30 minutes CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 at 30~C prior to the snbseqllent addition of the radiolabeled viral DNA substrate (20 nM). For the ComretitiQn expe.i.,.ellL~ (Fig. C-7), integrase (200 nM) was preinrllh~t~d with either the radiolabeled viral DNA substrate (20 nM) or T30177 (20 nM) for S minutes at 30~C prior to the liti~n of Culll~eliLOl DNA at the infli~t~c~ cullCellL~dLiOn.

Results of the Site of Activi~ Studies G~ quartet oliL~ o~ inhibit ~IV-1 IL~ 1dS~:. The inhibition of HIV-1 integrase by a series of oligonucleotides which can form G4s is shown in Figure C-1. Oligonucleotides T30177 and T30659 (Ojwang, et al., Antimicrob. Agents Chemother. 39, 2426-2435 (1995)) fold upon themselves into structures stabilized by two G4s stacked upon each other to fo~n a gl~nnsinlo octet (Rando, et al., J. Biol. Chem. 270, 1754-1760 (1995); Schultze, et al., J. Mo~. BioZ. 235, 1532-1547 (1994)). I-Lc-c~Li--gly, T30177 is active against HIV-1 in cell culture and against purified HIV-1 i..l~.ase in vitro (Ojwang, et al., Antimicrob. Agents Chemother. 39, 2426-2435 (1995)) while T30659 is not. For example, inhihitinn of both the 3~ uce~ g and strand transfer activities of HIV-1 inrtogr~ce (Fig. C-lA) by T30177 was observed in the n~n~m~ r range (see Fig.
C-lB).
In order to asce.L~il- why T30177 was rrre~;~ive and T30659 was not, the hlvt:lllul~ made a series of compounds to iu~;lr~ lly change one cu.-l~uulld into the other. The ~Llu-;lules of these cull.~ou-~ds are shown in panels C and D of Figure C-1. The differences between T30177 and T30659 (i.e., the presence of ~ lition~l bases at both ends, dirre~e~L sequ~n~es in all three loops, and ç~ct.on~ion of loop 2) ,I.~..irr~l themselves in dramatic increases in the IC50 values (Fig.
C-lD). To distinguish the contributions of each of these changes, the hlvr--Lu~ first added the same 5'- and 3'-nucleotides to T30659 as are present on T30177, yielding T30674 (Fig. C-lC).
These changes did not confer potency (Fig. C-lD). Then it was undertaken to change either loop 1 to obtain T30675 (Fig. C-lC) or the three bases in loop 2 into those found in T30177, yielding compound T30677 (Fig. C-lC) Neither change by itself cu"rr~ed potency (Fig. C-lD).
However, when the change was accomplished in two of the loops to resemble T3û177, yielding T30676 or T30678 (Fig. C-lC), the h~vt:lllol~ were able to ~i~nifir~ntly hll~luve the activity over that of T30659. Lllele~Lillgly, a two- to three-fold decledse in potency was also observed when a second quartet was unable to form, yielding T30526 (Fig. C-lD). These data suggest not only that the octet structure is critical but also that the loops are h~o~ulL for int~r~tinn with HIV-1 integrase.

CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96/11786 - The activities of the oligonucleotides in the cellular assays did not strictly correlate with the in vitro anti-integrase activity (Fig. C-lD). The correlation is complicated by the .lirr~;lell~ial stabilities and susceptibilities to nuclease digestion of the oligonucleotides in vivo (Joshua O.
Ojwang and Robert F. Rando, unpublished).
In Figure C-1, G4 oligonucleotides were initially tested in a dual assay which ,l,ea~ul~s both 3'-processing and strand transfer. Craigie, et al., Cell 62, 829-837 (1990); Katz, et al., Cell 63, 87-95 (1990). A strand transfer assay using "preprocessed" (3'-recessed) 5nhstr~te (19mer in Fig. C-2a, left panel) was also performed to tle5~rmin~ whether the strand transfer reaction was truly being inhibited or whether the inhibition of the 3'-processing reaction caused the decrease in the subsequent strand transfer products. TnhihitiQn of strand transfer using this s~bstr~t~ was observed in the same co~lc~lllldLion range (Fig. C-2A, right panel) as that seen with the blunt-ended, duplex oligonucleotide substrate (Fig. C-lA, top). Therefore, G4 oligonucleotides inhibit both steps of the integrase reactions: 3'-proces~hlg and strand transfer.
Inhibition of 3'-ploce~ g was confirm~d using DNA suhstr~tes labeled at the 3'-end, (F.nglem~n, et al, Cell 67, 1211-1221 (1991), Virlk, et al., Nudeic Acids ~es. 19, 6691-6698 (1991)) (Fig. C-2B, left panel), showed that all of the G4s tested inhibited glycerolysis, hydrolysis, and circular nucleotide formation to the same extent (Fig. C-2B, right panel). Thus, G4 oligonucleotides exert a global inhibition of the three nucleophiles in the 3'-proce~i"~ reaction (glycerol, water, or the hydio~-yl group of the viral DNA termiml~) Having ~lemr,n~trated that the catalytic activities of illL~lds~ could be inhibited by G4 oligonucleotides, the hlv~llLo~ next ex~min~d whether DNA binding was also affected. They performed UV cro~linking of int~gr~ce-DNA reactions to address this question. Cro~linkin~ of substrate DNA to integrase followed by electrophoresis results in a product having a molecular weight of a~ o~il,laLely 39 kDa (F.nglem~n et al., 1994, yo~hin~g~ et al., 1994). As seen in Figs.
C-3A and C-3B, binding of HIV-l integrase to radiolabeled U5 DNA ~ub~lldk; was inhibited by preinrnb~ti~ n of the enzyme with T30177 in the same concentration range as its IC50 value for strand transfer (lanes 3-7). In contrast, preinr -b~tion of the enzyme with T30659, which was poorly active in the 3'-processing/strand transfer assay (Fig. C-lD), resulted in only modest inhibition of DNA binding even at a T30659 cu~ ion of 500 nM (Fig. C-3A, lanes 9-13).
o,~ll~e of the ~V~ ,.a3e zinc finger region for gll~nosine quartet oligonnrl~oti(l~
il~t~.a~Lions. Integrase can catalyze in vztro an a~a,~lli reversal of the DNA strand transfer W O 97/03997 PCT~US96/11786 }eaction called tlixintrgration. Chow, et al., Science 255, 723-726 (1992). In contrast to the 3'-procçs~ing and strand transfer rç~rti-mX, disintegration requires neither the N-terminal zinc-finger region nor the C-t~rmin~l DNS-binding domain of hlL~ dse. Ru~hm~n, et al., Proc. NatZ. Acad.
Sci. U.S.A. 90, 3428-3432 (1993). For this reason, the HIV-l integrase catalytic core domain, In50~2l2 (Fig. C4A), can be use din the intramolecular ~ , d~ion assay and for testing the site of action of inhibitors ~7nm~lPr~ et al., Proc. Natl. Acad. Sci. 91, 5771-5775 (1994);
M~7llm~t?r, et al., AIDS Res. Hum. Retrov. 11, 115-125 (1995).
In the ~lixint~gration assay, only the Inl~288 and INl-2l2 proteins (Fig. C4B) were inhibited by T30177 (with IC50s of 270 and 600 nM, respectively) while neither IN50-2l2 (Fig. C4B) nor IN50-288 (data not shown) showed more than 30% inhibition at a 3 ~M conrçntration of T30177.
The conr~ntrati~-n of T30177 required for inhibition of di~hl~ldLion was higher than that required for inhibition of either 3'-~ ces~hlg or strand transfer These results are col~ lll with those observed with other mol~cnl~s (Fesen et al., 1994, Ma7nm-1~r et al, 1994). This obs~l~/dLion suggests that the active site of HIV-l i, ~ x~ may tolerate drug-induced protein or DNA distortion during the ~ ;, d~ion reaction, C~ ~xi~ . l with the relative t~ ranre of i ., l~ ;".xe to mutagenesis of either substrate features (Chow & Brown, 1994) or protein structural ~lomainx (Bushman et al., 1993) in this reaction. This is the first example of an HIV-l hlL~ lase inhihitor requiring the enzyme zinc-finger region for inhibitory activity. These results suggest that the zinc-finger may assist in stabilizing binding to T30177.
This hypothesis was inv--sfig~trd further by ",.. "il."i"g binding of wild-type, full length integrase (INl-288) and of the deletion mutants to radiolabeled T30177. The conrentration of T30177 required for DNA-protein complex f~rm~ti~n was the same as that required for complex form~ti-n using the viral U5 DNA substrate (i.e., in the 20 nM range). UV cro~linkin~ assays, Engelman, et al., J. Virol. 68, 5911-5917 (1994), showed that INl-288 formed a DNA-protein complex of the e~rectt-d m~ lecnl~r weight in the absence or plcsc~llce of added mang~nrse (Fig.
C4C, lanes 8 and 9). The INl-2l2 protein, which has previously been shown to bind to linear viral DNA only at high conrrntrations (d~Lo~illldL~ly 2.56 ,uM) and only in the presence of divalent metal ion, (Fngelm~n, et al., J. Virol. 68, 5911-5917 (1994)), was able to crosslink to T30177 with the same efficiency as wild-type illlr~la~e in the absence or presence of added m~ng~n-ose (lanes 2 and 3). The IN50-288 protein, which contains a nonspecific DNA-binding domain, was also able to crosslink to T30177 with the same effficiency as wild-type hlLe~ldse in the absence or pies~llce of added m~ng~n~s~ (lanes 4 and 5), conxixt~nt with its ability to bind to viral U5 DNA (Engelman CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96/11786 et al., 1994). The extent of crocclinking was cignifi~ntly tlimini~h( d in the case of t'ne core mutant INs0-2l2 cu~ aled to INl-2l2 in the absence or presence of m~ng~n~se (compare lanes 2 and 3 with 6 and 7, faster migrating complex). The higher molecular weight species in lane 6, having t'ne ~rect~d mnlec~ r weight of a dimer, has been reproducibly observed, but its density has not been confirml-cl These data support the notion that the N-t~rmimlc of HIV~ .se assist in the formation or stabilization of an HIV-1 i..lt~ e-T30177 complex, perhaps by binding the oligonucleotide .

DNA-binding activities of the HIV~ er~se zinc finger ~ln~in To further analyze the binding of the N-terminal zinc finger region to T30177 and Culll~ldle these results to the viral U5 substrate, UV crosclinking was pc~r~ ed with an Inl~55 deletion mutant (Fig. C4A) cont~ining only this dornain. As seen in Figure C-5, this mutant could not bind either the T30177 oligonucleotide or the viral DNA s~lhctr~t~- when only m~ng~n.-se or .. ~ ,i.. was present left and right panels, lanes 3 and 4). However, the IN1-55 protein could bind to both DNAs int he presence of zinc and either m~ng~n.ose or m~gn~cillm (left and right panels, lanes 5 and 6).
Signifi~ntly, the Inl~55 protein was able to bind to the T30177 G4 oligonucleotide, but not the viral DNA substrate, in the ~ s~-~ce of zinc alone (left and right panels, lanes 9 and 10). These results are in accord with the known zinc-binding ability of this domain. Rnchm~n, et al., Proc. Natl.
Acad. Sci. U.S.A. 90, 3428-3432 (1993), Burke, et al., J. BioZ. C*em. 267, 9639-9644 (1992).
But they also suggest that the N-terminal domain of inrl~gr~ce has DNA binding capabilities on its own. Finally, these ~LJe~ emlmctr~t~ cu~ a~ble binding of the HIV-l i..l~;"~e zinc finger domain to an oligonucleotide c~ illillg in G4s than to a double-stranded, linear, viral DNA
oligonucleotide when both m~ng~n~se (or m~gn~cillm) and zinc are present but more efficient binding to the G4 oligonucleotide under non physiological c~ n-liti- ns (zinc alone). The inventors also found that the nucleocapsid protein of HIV-1, a nucleic acid ~nnP~ling protein which contains two CCHC zinc fingers and which is eccenti~l for di..l~.iG~ion of the l~:Lluvilal RNA genome, Ts ~ ih~chi, et al., J. Virol. 68, 5863-5870 (1994), was able to bind efficiently to T30177 (data not shown). The ability of zinc to confer DNA binding ability on the IN1-55 protein was e~min~-1 by repl~em~nt of this ion with other transition metals. C. ~ l ll with ~,~ecl~uscu~ic data (Burke et al., 1992), only zinc was able to induce detectable DNA binding to the G4 oligonucleotide (data not shown).

W 097/03997 PCTrUS96/11786 Increased potency of ~,,.u.,,~ quartets in m,~ cinm. In contrast to INI-55, the extent of crocclinking (and pl~,ulllably binding) of wild-type hlLe~;ldse to radiolabeled guanosine quartet was increased in the presence of m,gnP~ m relative to m"ng~nPsç at several c~ r~ dLions of the guanosine quarter (Fig. C-6A). This observation led us to examine whether the inhibitory activity of T30177 and analogs could also be ~"l~ re~l by buffer cn"l;;,~i,.g m,l~nPcinm In order to address this ~lesti~m, the hlv~llLol~, tested three versions of T30177 as shown in Fig. C-6B.
T30175 has the same base sequçnre as T30177 but is composed entirely of phosphorothiodiester inrPrn-l~leotidic linkages. The inhibition of 3'-processing catalyzed by HIV-1 integrase by these gll,mo~inP quartets is shown in Fig. C-6C. Both T30175 and T30177 showed four to five-fold increases in potency when m"gnP~illm was used as the divalent metal instead of mAng,lnese. In contrast, T30038 showed no ~ignifir~nt increase in potency when m,l~,.P~ ll" was used as the ion (Fig. C-6D). These data are in accord with the hlcledsed stability CUll:~L~l~, for mAgnecillm-nucleotide complexes when oxygen replaces sulfur (Pecoraro et al., 1984). The opposite is true for m~ng;mps~p. Therefore, the greater inhibitory potency of T30177 in buffer c~-"l;-i"i-~g ms~gnPSillm versus m,mg~nPse may reflect a lc;~uh~ llL for mAgnPsillm ion coor lin,ltil-n along the phospho~liPstPr backbone of T30177 in order to confer hlllibiLoly activity and ~ illlulll intPr"rtion of T30177 with HIV-l intPgr~ce. This coordination can occur with more stability when either T30177 or T30175 are assayed in buffer cv~ llillg m,gnPcillm rather than m,mg~nPse and is li rc~ in a greater potency against 3'-processing.
DNA competition t~ ~ents. The relative affinity for the G4 oligonucleotide was probed by dLLclll~Lillg to compete off the i.,l~"~~ bound to r,~-liol,lhPlP~I HIV-l viral U5 DNA with hlL;leasi.~g c.---r~.,l.dLions of unlabeled T30177 (Fig. C-7A). The cullvel-,e t:~uelilll~llL, where binding of integrase to r~ hçle~l G4 oligonucleotide was carried out prior to t'ne ,1~1~1iti~n of increasing c~-llre~.l.dLions of nnl:~hPl.ocl HIV-1 viral U5 DNA, was also p~:lr.Jllll~d (Figure C-7B). In each case, a band having the d~pdlCll~ mobility of an integrase-DNA complex was evident. In Fig. C-7A, the viral DNA-integrase complex has a mnlPcJll,lr weight of 38,500 while in Fig. C-7B, the T30177-hlL~,ldse complex has a molecular weight of 37,000. Neither complex could not be culll~led off by either c~ ilul DNA even at c~-nrç-ntrations where the colll~eLiLol was in 500-fold excess (Fig. C-7A, lane 6). Similar results were seen when the In1~212 and IN50-2l2 proteins were used in cnmpetition ~ (data not shown). Therefore, the stability of the G4 oligonucleotide DNA-integrase complex is colll~dble to that of the viral DNA-integrase complex 7 PCTrUS96/11786 is comparable to that of the viral DN~-integrase complex. Ellison, et al, Proc. Natl. Acad. Sci.
U.S.A. 91, 7316-7320 (1994); Vink, et al., Nuc. Acids Res. 22, 4103-4110 (1994).
Tnhihition of related It:~liv~dl i~ s~s. T30177 was tested for inhibition of the related leLlovildl integrases from HIV-2 (van Gent et al., 1992), simian immunodeficiency virus (SIV) and feline immunodeficiency virus (FIV) (Vink et al., 1994b). As seen in Figs. C-8A and C-8B, T30177 inhibited 3'-processing catalyzed by HIV-l integrase in the expected co,~ .d~ion range (Fig. C-8A, lanes 2-8; IC50 = 55 nM). Inhibition of HIV-2 integrase (using HIV-l DNA) was also observed in the same range (lanes 9-15; IC50 = 90 nM). However, FIV integrase was inhibited at three-fold higher co~r~~ dLions of T30177 (lanes 16-22; IC50 = 175 nM), and SIV integrase was inhibited at seven-fold higher c.~l-r~ ions to T30177 (lanes 23-29; IC50 = 420 nM). Therefore, the T30177 G4 oligonucleotide displayed some selectivity among the lentiviral integrases.

CQn.~ n~ Regarding the Site of Activity Studies The present study fl~m~ i1les for the first time the binding of DNA gn~n~in~ quartet structures to HIV-l integrase, and that some oligonucleotides recently shown to exhibit antiviral activity are potent HIV-l integrase inhibitors.

G--q..~ Quartet OL~ oti~lpc are Novel and Potent ~nhihit-)rs of HIV-1 integrase.
Oligonucleotides composed of deoxygu~n~inf and thymidine and forming gu~no.~in~-tetrads (G4) structures have previously been shown in inhibit HIV replication. Rando, et al., J. Biol. Chem.
270, 1754-1760 (1995); Wyatt, et al., Proc. Natl. Acad. Sci. U.S.A. 91, 1356-1360 (1994). Two "~ h~ m~ have been invoked. First, some oligonucleotides have been shown to bind to the V3 loop of the envelope protein gpl20 and subsequently inhibit virus adsorption and cell fusion.
Wyatt, et al., Proc. Natl. Acad. Sci. U.S.A. 91, 1356-1360 (1994). Secondly, oligonucleotides such as those described in th~e present study also inhibit viral-specific L.alls~ ~, Rando, et al., J.
Biol. Chem. 270, 1754-1760 (1995), plc~u~llably by inhil~iting viral hlL~ldlion. Ojwang, et al., Antimicrob. Agents Chemother. 39, 2426-2435 (1995). The present finding that inhihhi(ln of the HIV-lRF strain in cell culture parallels that of purified integrase in vitro in the series of G4 oligonucleotides tested (Fig. C-lD) further ~lf m~ L~s the possibility that HIV-l integrase can be targeted by some G4 oligonucleotides.

WO 97/03997 PCTnJS96/11786 G4 oligonucleotides differ from previously published HIV~ lL~.ds~ inhibitors in several ways. (Table C-1) First, they are among the most potent inhibitors to date with IC50's in the nanomolar range. Their potency range is comparable to flavone, Fesen, et al., Biochem.
Pharmacol. 48, 595-608 (1994), and tyrophostin d~livaLives, M~7llm~1~r, et al, Biochemistry 34, in press (1995), which, however, generally fail to show antiviral activity. Secondly, the zinc finger domain of HIV-1 integrase contributes to the inhibition by G4 oligonucleotides, as truncation mutant e~yllles lacking this domain are resistant to the G4 oligonucleotides. This property is unique, as all the other inhibitors to date are active against the HIV int~gr~e catalytic core domain.
(Table C-2) M~7nmrlt~r, et al., Proc. Natl. Acad. Sci. 91, 5771-5775 (1994); ~7llm~ r, et al., Mol. Pham. sllbmitt~-l (1995); Fesen, et al., Biochem. Pharmacol. 48, 595-608 (1994); M~7llm-l~r, et al, Biochemistry 34, in press (1995). Finally, G4 oligonucleotides form stable enzyme complexes that carmot be displaced by excess viral DNA oligon..~leoti~

CA 02227867 l998-0l-26 W097/03997 PCT~US96/11786 _~
C~S ~
C ~ ~ o .

3a ~ 6 ~ O O O ~0 ~ ~¢ a 8 o _ * ~ G
> ~ ~ P.- ~ ~ ~ C C 0~ ~ ~ ~ G ~ C C ~ ~ C ~ C , C

~ o ~ ~ o o o o o o o o ~ '~

W O 97/03997 PCT~US96/11786 ~: _ s .CC ~CC ~ C~
c ~ .', o ~ E c D ~
_ ' ~) o ~ ~
E :' ~ 50 E E C _ _ E C E E E ~ E E ~C E E - , o~ s o ~, t-- l-- r- ~ ~-- 1~ ~ t~ ~ l~ ~~ ~ ~ ~n o o c ~ ~ '--~ ~ o ~ E
-- C C C~ X
C, ~ C" ~ ~ C~ ,c ~ ", o ~ ,o ~
o ~ ~L~ S C O ~ D -3 co oc ~ E ~ a ~ 3 ~ ~ G S C
~ O O ~ .~ ~~ ~ o o O O
~ ~ ~ - - - -~ ~ ~ ~ ~ ' - - -E ~ ~ ~' E 5 C~ O O o o o o ~,~ ~ ~ s s C~ Cu~ ~ ca C~ C
~ ~ Z s -- , _ x ~ CD C S ,.~ e~
-' c ~ ~ ~ ~ ~ ~ ~ ~ E-- c ~, o ~ ~ c ~ & c c ~ & c c ~ ~ ~ 3 D ~ ~ ~ ~ - ~ Z
<~ ~ z U~ @ _ ~ 5 -- _ ~ ~ ' --' 5 C--C _ E

- CO =~ C- C Y = Cc = ~ ~ O ~
~ ~ a ~ cococo - cc E 3~ ~ ~ CC C 5 CO C CO CO CO ~ ~ O; ~ ~ C,~ -- _ ¢
.: . ~ _ ..... ~,, ., ~CO CO CO C'' CO CO C 'Ce~ -- ~-- CY
c~ ; C C co ~o co ~ E s3 ~

c D C o o ~ cc 'C D ~o ~-- ~ J
'~"~'3 Cc=' C Cc~ -- ~D
C C C,C ~, _ o S ~ S C
o o ~o o o o o o o o oo o o o -- ~ 3 ~
~ ~ ¢'--~E-- C cuc~~ o ~_ ~ - CA 02227867 1998-01-26 PC~/US 96~1 I 786 IPEAAJS ~ T ~

ZZZOZ-ZZZZZZZZO ZOZOZ~ ZZozZZZoZ ZZozZoZ ZZZZZZZ~
HHHHHHHHHHHH HHH HHHHHHH HHHHH HHHHHHH
~ ~ ~ ~ o~ ~ ol ol ~ ~ ~ ~ ~ ~ ~ a ~ ol ~ ~ cY ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ol ~
__________~_, ~__ _______ _____ _______ S~ ~ h ~ h 5~ h 5~ h ~ h S~ h ~
-e E 8z E 8e~ ee e e eeeeeee eee8e eeeeeee ~o Ln ~ ~ ~ .
O . ~. ~ O ~
H

O t- ' , H
v ~ O r o ~ ~ ~ rv, ~ ~ ~
h ~ a) VV v V I I I h IIHI aS IIII
_r ~-~'~t~l~t- t~7 0 h h h h ~ V ~ u) u) ul u~
. o ~ ~ ~~ ~ ~ '~ Ul ~r~; ~i r~; r~w 1l 1l 1l 1l ~1l 1l 1l 1l 1l 1l O ~O ~I E~ E~ E~ E~ E~ E~ E~ ~ ~ h -~1 -~1 -~1 -~1 E~ E~ E~ O E~ E~HE-I m m HHHH
O _ ~ . - r - - - -: - - ~:::: ~ r ~ ~ ~ ~
U~ ., ~ ., ., ., -. -. -. -. -. -, -, oo s~s V U s U s s S S~ S S S s s s' 's 's ' ' '' ' s s 's 's -S S U U ~ U U~D~ ~ EL E~, E, E E E E E E~ El E E E, E' E, E E E, . El El El El ~ ~
Ql L~ ~ L~ L~ L~ L~ L~ L~ L~ ~ Ll. ~ ~ ~ S~ ~ ~ ~) ~ ~ ~ ~ ~ ~

111111111111 111 1111111111111~11111 111111111111 111 ~111111111111111111 t-- ~ o ~1 a~ ~ ~ In ~D ~ a~ alo ~ In ~ ~ n o ,~ o ~ ~ ~n o ~ ~

o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o AM~ ~

CA 02227867 1998-01-26 PCT~S 96/1~7~' ~PE.WS ~ 4 o~ ~97, _ _ _ _ _ _ _ ~ _ ~ ~ ~ ~ ~ Ca 3 ~
t- CC~ co ~ ~ ~ O co ~o C 3 ~ ~a c~ ~C = ~3 o o o z o n ~ cl Q ~ a c~ O c H H H H H H H H H ~ ~ ~ ca ~ ~ ~ ~ ~~
C" ~~' ~,D C~ C ~ &
~3 cn cn ~ u~ ~ cnu~ cn O ~ _ ~d -- r~
____ ___ __ , c& - ~ C c~
S~ S~ S~ S~ h S~ S~h S~ C ca ~ 'a ~ ~, tn e e 8 e 8 8 8 8 8 3, O co~ c~ v~ ~na cCa o ~
~~ t~ t~ t~ ~ ~ 7 0 ~ tV ~ ~~_ tv cC C~ 8 ~
t~ N~tn ~~ .c ~va ~c 8 3 ~~ rv- ca c o & c ~~ c x ~ ~ tn ; C ~ c~ c~ O 9 ,~ ~ ~v v to 3 t~ O .v .v ~ cO ,~ 8 ca r'~ S 8 c ~ ~ ~ 9 ~n t~ ~ ~ &

E~ ~ ~ _ CC ~ O ~ ~ C5 t ~ -oo $v ~ $ m 9 o 9 ~~ C5 ~t~ ~ ~
~t ~ 3 8 ~ CC5 tV O C ~_ r_~ t~ ~V
~ ~' '~ ~ I S~ I ~ tV ~ ~ ~ C ~ C5 ~ ~ m n Il 1l 11 1l t~ A ~V O C~O C ~V O C ~ m ~ ~
r~ vc~ ; ; ~ ~n cc5 C~ ~V ~V C5 8& .3n t-n E--5 u. u. u U ~ ~ C5 0 3 -C& ZZ 9 ~V V 3 ~
St~ S~ t; ti3U. . t30 t~ L ~v & CV ~_ tV L -- c, o '' -~-Z ~ ~Z u7 ~V C5~C 9 ~~rn tV a ~ C ~
_0 &.r~7 cn rn C ~V ~V ~ . ~_ _ _ _ __ _ _ _ _C5 ~V C & C7 ~V C5 0 ~ ~V ~V c r~7 u~7 r~l r~7t~l r~)r~7r~7 r7 ~ Cn ~ ~ 9 ~ ~ ~ S c5 ~ ~
7 ~ C5 9 ~rnV ~ ~V ~ ctn C 3 ~

.c E cO 9 C 93 ~ ~
1 7 1 ~ r ~cn 3 ~-- ,_ ,v -- o ,v ~ ~ 7 ~ 7 7 ~~~~V C S C CC ~rn Oc r- S
_ 7 7 7 7 ~ ~ ~ C5C '~ '~ rn ~ O~V C~5 ~ rO
r, 9 rn 7 ~ t'~)c 5 .~--C ~ i ~c~~ 9 ~ ~i n ~ ~ cO~ cr5 u7 ul ul u7 ul u7 u7 u7ul ~ cr5 C~7 ~ C $ 1_ ~ Cr5 r C ~ Oc 3 L ~' -- 9 0 C~ r~ C ~7 s~ ~V ~~ L ,V
r53 C _C c ~ ~_ C 5 E-- ~ ~v r7 ~ Ul ~D Co ~ t~ 3 ~5 ~ ~ ~ 5 0 ~ _cn7 9 Cr5 -O 7~V Ct--~ ~ U7 ~ ~0 ~ C~S~ O ~V ~rn ~~ 3 o tV C _C cn ~~7 r~7 r~7 t-~ r-7t~l t~r~~7 ' ~C ~CV r rn C~. ~C ~ 3 c 5 ~ ~
F~ ~ ~ ~ . ~ ~ ' 3Q V C . ~ ~ C ~~ C~ O'X C
~ cv ~, _c _ _ C _~ V ~ ~n V ~V ;~

AM~N~ 9~

CA 02227867 1998-01-26 ~CT/U~ 9 6 / 1 I ?8 h IPE~Js 14oc~ 7.
CLl ~ o E-- ~
~D
C

o C

C

., ~Cl, o~,~
-O d-. _~~, _ _ J ~ ~ V~
~ ~ ~V
3 ~~ ~
~AV P_-- 5.1 C ~ô -~
~'~ _ A
,C 'OU' _ C C' ~ ,C

,C~_ C
~ 3 o ~D ~-- 3 c~
O .. ~ o O
O D ~ ~
c ~ ~ ~ ~ E

~.~ *, o ~ o ~
~n C ~11 ~ c o_ o _ o ' AMENDED SH~ET

, CA 02227867 1998-01-26 P ~ ~ S 9 6 / ~ 8 lPE~US 14 O~T'9 z -C~

-- ZOz ZOZZOZOZOZZZOZOZO ZZZOZZ~ZOZOZZZZ ZoZZZZZZZZOZZ
HH HHHHHHHHHH HHHHHHHHHHH HHHHHHHHHHH
o ~~ '0~'0~ '0~'~ w'~'~' ~'a ~'~~'~'~'~~ ~'~~'~' w'~~'~~'~'~'~'w'~'~'~'~' ~ __ __________ ___________ ___________ ~ ~ ~
~t ~ In o o o o o o o u~
Ll7 o o o o ~ ~ o ~ o o~ o o - ~~ ~A~ ~0 ~0 0 ~ ~~LnO~0~0d'~ ~1Ul~o~ Ln~Do,looooooo --~OO ~OOOOOOOOO OOOOOOOOOO OOOOOOOOOOO
O ~ ~
-~ 5 c ~ O O ~ O O ~O ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ O O O O o O o ~ ~ o o ~-- L~ ,~ . . o ~ cd oo~ooooooooo oooooooooo ooooooooooo ~ v~

V 1 1 1 1 1 1 , , , , ~ , ~ , , 1 1 1 1 1 1 1 1 1 E 1 _ O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O
Et E~ E~ E~ E~ E~ E~ ~ E~ E~ E~ E~

~ENDEfl 9~

' CA 02227867 1998-01-26 PCT/US s6rI~7~
lPE~JS 14 ~ '97 L ~ V~ C
o o ~ ~ .

_ _ _ ~ _ _ _ _ ~ ~
z z z z o o o o o 5 o-3 H H H H H H H H H C
~lot~ ~, ~,~,~, ~,~l ~ c.5 o ~ C~ C~
In C c,~
A O ~ 5 ~

o o o oul o ul u) o 5 C ~-o o ~ ~~ ~ ~ o ,~
o o o oA O O O O C.~
C~ ~ C.~ 00 ~ ~ o o o ~ o o ~ c 5 In O O L~ D
~ ,1 o ~1 ~ C
o o o oo o o o ~ ~ ~

~ C O
rn ~ .
__ _ _. ,~ . _ r~l r~ ~ ~ C
t C C~
~ ~ n ~ 3~ ~ .C

_ _ r~ ~ 7 1 ~ ,C ~ ~ ~n 6S ~ 5c r~l ~ Ul ~D t-- ' r~ 1 ~ O ,j~
J ~r . _ C . _ ,~

O O O O O -I o ~-- o C~
r r rrl r r ~'~ r _C S ~ ~

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 integrase and on the structure and seql~Pn~e of the G4s. These findings also suggest that novel AIDS therapies could be based upon G4s as inhibitors of HIV-1 integrase.

D. Structure-Function Studies As shown previously, the illV~ Ul~ obtained evidence for inhibition of HIV-l infection by LledLlllcllL with phr~rhodiester oligonucleotides co,-li~;"i"g only G and T bases. ~l~liti~n~l studies noted above suggested to the inventors that such oligomers were potent inhibitors of HIV-1 integrase, in vitro. The highest activity was obtained using the 17mer, referred to as T30177, with composition Gl2-T5. NMR evidence suggested to the illVell~Ul~ that T30177 forms an intr~mnlPclll~r fold which is stabilized by a pair of G-tetrads, connected by three single 5tr~n-1Pd loops, with a 1-2 base tail to ether side of the fold.
Thus, the hlv~llLol~ undertook studies to ~IPtPrminP sequence dPpçn~lPnre of theintr~m-llec~ r folding mPrh~nicm, in a set of four closely related 16-17 base oligonucleotide homologues, with seql~Pnrçs in the range G10-12-T4-7. The original T30177 colll~uulld was in~hl(l~P-l along with three delivdliv~s which were decignPd so as to alter the structure of loop domains, while keeping the pair of G-tetrads intact. Based on thermal .1~. ,,.l ~ " ,~ n CD and kinetic analysis, the hlvellL~ were able to show that a single base alteration within the loop or tail ~lom~inc can produce a very large change in folding stability. The K+ ion ~leppn(lpn~e of these data suggested a pl~lhllillaly model wherein the loop and tail domains interact to form stable metal ion-binding sites. A 16mer deliv~Liv~ (T30695) was ~lP~ignPd within the context of that model, with the intent of enhancing the interaction between K+ and the 5' tPrmiml~ of the oligomer. The illV~ Ol:j showed that T30695 folding is indeed more stable than other members of the group and is highly specific for K+, as ~ ssed from the ion de~Pn-lPn~e of therrnal d~lldluldlion, CD spectra and UV ~IPtPC-tpd folding kinetics.
To assess the rel~tinn~hir between biological activity and formation of the ion-selective oligomer fold, the hlv~llLol~ colll~d tertiary ~llU-;Lul~ stability at three K+ concentrations with the capacity of the folded oligomers to inhibit the HIV-1 hlL~ldse enzyme in vitro, or HIV-1 inft~ction in cell culture. The stability and activity data are found to be highly correlated, as a filn~tion of sequence alteration, suggesting that formation of the stable intr~m-~leclll~r fold may be a prerequisite for both integrase inhibition and anti-HIV-1 activity. Although the structure of the folded state has not yet been co. rllllled at high resolution, the data presented here suggested that the structure of the T30695 complex with K+ ion may be of ph~ ignifir~n~e and could serve as the basis for additional hll~l~,v~llltllL of the observed HIV-1 activity.

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 Materials and Methods for the Sll u~ e/Function ,S~--l;r~
Oligon-u~l~oti~lp Synthesis. All oligonucleotides used in this study were synthpci7pcl on an Applied Biosystems Inc. DNA synthPsi7Pr, model 380B or 394, using standard phosphoramidite rhPmictry, or fast deblocking Expedite chPmictry on a Milligen synthesizer. All oligomers poccPsced 2 S pho5rhorothioate linkages (one on each tPrminllc) which were introduced by the H-phosphonate method. Oligonucleotides were purified by ~leL~dLive anion exchange HPLC, on Q-Sepharose.
Chain purity was co"l~, ~II~d by analytical Q-sepharose ~,hlol.~dLography, and by d~llaLulillg electrophoresis of 32P labeled oligomers on a 20% polyaerylatnide (19:1), 7M urea gel matrix (Rando et al., (1994) J. Biol. Chem. 270, 1754-1760, 17). In all inct~nres, greater than 90% of the purified oligomer was determined to be full length. Oligomer folding was l-lolliL-~lcd by native gel electrophoresis on 15% acrylatnide (19:1) rnatrix in TBE. Folded, 32P labeled samples were loaded subsequent to ~nnP~Iing in 20 mM Li3P04, pH 7.0, 10 rnM KCl at 7 uM in strands, as described below.

AnnP~lin~. Prior to UV, CD or kinetic analysis, oligonucleotides were ~nnP~lPd at 20 mM Li3P04, pH 7 at 3-15 uM in strands. Samples were heated to 90~C for 5 min and then inrllb~tPd for 1 hour at 37~C. Metal ion could be added as the chloride either before or after the 37~C incubation, with no ~,,ea~,uldble dirr~.~,nce in final state, as ~ccPsced by UV, CD or gel analysis. As accPcced by native gel electropho}esis (not shown), this ~nnr~ling method was found to produce a single product with mobility con.cictrnt with a folded "~ OI"~- over the strand collct:llLldLion range from 3-15 uM, at all ion conrPntrations described.

Ultraviolet Spectroscopy. UV ll.easulc~ llL~, were obtained on a HP 8452A diode array spe~;Llolllc:l~l, using a HP 89090A L~ ldLule regulator. Except where noted, thermal d~llaLul~Lion profiles were obtained at a rate of 1.25~C/min over the range from 20~C-80~C, on samples at 20 mM Li3P04, pH 7, at 7 uM in strands. Absorbance was monitored at 240 nm, which was determinf~d to be the point of m~im~l temperature induced change. For melting analysis, metal ion was added to the desired co~ "l,dLion, followed by a one hour pre-incubation at 37~C, to ensure competP ~nn~ling, Folding kinetics were obtained by manual a~l~lition of metal ion at t=O, followed by absorption ll.ea~ulclll~llL at 264 nm. Mixing dead time was determined to be lO sec.
Kinrtir,c were llwlliLolcd over tne range from 10 sec to 15 min at 25~C.

Circular I)~ oi;. 1. CD spectra were obtained at 25~C in 20 mM Li3P04, pH 7, at 15 uM in strands, on a Jasco J-500A spectrop~ L~l. Metal ion was added to the desired concentration, followed by one hour of pre-inrnb~ti--n at 37~C. Each ~e~;Llulll in the text ~e~l~s~llL~ 5 averaged CA 02227867 l998-0l-26 scans. To conforrn to tra~liti-n~l standards, data are ~l~sellL~d in mola} ellipti~ity (deg-crn2-dmolE-1) as measured in base, rather than strand equivalents Antiviral assay. The RF laboldLoly strain of HIV- 1 was used to infect established cell lines for S one hour at 37~C prior to washing and resuspension in medium c~-"l;~i"i,.~ hlc;lc:a~illg c-J,~rr~-l-dLions of drug. Four to six days post-infection, drug treated and control wells were analyzed for HIV- 1 induced ~;yLuL~aLllic effects, for the pl~s~llce of viral reverse ~l~ulscli~lase (RF) or viral p24 antigen in the culture mPtlinm as previously described by Ojwang et al. (Bishop et al., (1996) J. Biol. Chem. 271, 5698-5703). Purified l~colll~hlallL HIV-l illlc:gl~se enzyme (wild-type) was a generous gift from Dr. Craigie, Laboratory of M- l~cul~r Biology, NIDDK. All 3'-L~ioces~,illg and strand-Lldll~,r~l~ were performed as described previously by Fresen et al. (Fresen et al. (1993) Proc. Natl. Acad. Sci. VSA 90, 2399-2403) and M~ rn~r et al. (Mzl7llm-1~r et al. (1994) Proc.
Natl. Acad. Sci. USA 91, 5771-75).

Results of SL~u.,luuc;/F~ liuu Studies The structure of the oligonucleotides in this study are ~lc;~,~uled in Figure D-lA. For the purposes of clarity, they have been l~lc;s~uL~d in the context of a particular folding model which places eight of the gn~n~l~in~s as a central octet and the rPm"inrlPr of the oligomer in either a loop region, or as part of a 1-2 base long tail region at the 5' or 3' L~:llliulus. Previous electrophoresis and ID NMR data (Rando et al., (1994) J. Biol. Chem. 270, 1754-1760) have strongly suggested that T30177 folds so as to form an intr"mt l~cnl~r G-tetrad based ~7Llu~Lule which is stabilized by a single central G-octet. Therefore, for T30177, the simple model ~l~sc~llL~d in Figure D-lA is qn~t~ly Sllh,~ r(l by structural data. The validity of a similar structural model for the other ,..~. .h~., of the series is legitim"tely ~nm~d based upon their seqnPnre similarity, to be tested in terms of the data plesell~d below.

Thennal Dt:udLuldLAon Analysis. Based upon previous NMR data, and the general lil~ldLuie, the hlv~ Lol~, poshll~t~ that folding of T30177 should be strongly ~lepPn~lPnt upon K+ binding. To quantify this, they measured the therrnal stability of T30177, as a function of added KCl co"~ . dLion. Coupled equilibrium theory predicts that, in the instance that K+ binding stabilizes fi~rm~tion of an intr~m~,leclll~r fold, measured TM values should increase linearly with the Log of the KCl concentration. Such data are shown in Figure D-2A, line b. It is seen that in the presence of 20 mM Li3PO4, measured Tm values for T30177 increase from 38~C to 65~C in the range from 0.1 to 10 mM of added KCl. This very large increase in Tm below 10 mM of added KCl, in the presence of 20 mM of Li3PO4 as buffer, argues strongly that the effect of K+ binding is not a simple ionic strength effect.
The inventors have noticed that the measured Tm values for T30177 are consi~t-ontly higher, by 1030~C, than has been seen for other small intramolecular folds (Smith, F. W., & Feigon, J.
(1992) Nature (London) 344, 410-414; Schultze, et al., J. (1994) J Mol. Biol. 235, 1532-1547).
Since T30177 differs from these other homologues only in terms of the proposed loop dom~in~, the h~velllol~ have 5ynth~ d homologues of T30177 where the central G-octet remains constant, but where the loop dom~in~ to either side have been motlifi~d by addition or rep~ rn~ont of a single base. In the context of the simple folding model (Figure D-lA) the T30676 homologue is j~l.ontir~l to T30177, but has been mo~lifi~d so as to add an additional G into the topmost loop of the structure. As seen in Pigure D-2A, line c, this one base ~ lition produces a 20~C decrease in Tm over the entire range of K+ ion tested. Similarly, the T30677 homologue was prepared (Figure D-lA), which is ;tlt-nti~l to T30177, but has been m~l-lifitod so as to convert a pair of Gs in the bottommost loop domain. As seen in Figure D-2A, line d, this two base loop substitlltio~ produces a 30~C decrease in Tm over the range of K+ ion tested.
In the context of these sub~L~lLial stability changes, the illV~llLol~ sought to confirm that the general ",~ "i~m of folding had not been altered by base 5~lhsfitlltion Thus, Tm analysis was l~lJeaL~;d at 1 mM KCl as a filn~ti~n of strand concentration in the range from 3 to 10 ~M (Figure D-2C). As seen, a measurable strand c~Jl l~ lion depen-l~nt e could not be ~ t.oct~d over this three fold range of v~ri~tion, for any of the d~liv~lives, thus verifying that the folding equilibrium remains intr~ml-lecnl~r throughout. This was c.~ od by native gel electrophoresis, which contiml~-d to display a single folded oligomer state (not shown), similarly, it was observed that the thermal dirr~it llce s~e~;Llulll for all three homologues was very similar (not shown).

Oligomer Design I~ uvt~ . Based upon the nml~ lly high thermal stability of T30177, relative to intramolecular folds in the lil~ldLule, and upon the 2040~C decrease in Tm observed as a fun~tio~ of what should have been a simple loop modification (Figure D-lA), the hlvt:llLvl~
con~h~ d that interactions within the loop dom~in~ may contribute to stability. Specifically, it is proposed that K+ ions may engage in stable binding to the loop ~l~m~in~ of T30177. Simple docking c~lclll~tions, p~lrolllled with BIOSYM software (not shown) suggested that the TGTG loop configuration at the lower face of these folded homologues could, in cooperation with the p~
G-tetrad, give rise to tight K+ coordination which is similar to that seen when K+ (or Na+) ion coordinates between G-tetrads (Bishop et al., (1996) J. Biol. Chem. 271, 5698-5703) In the context of that proposal, the hlv~llLul~ noticed that, if the pennltim~t~ T were removed from the 5' t~ lhlus W O 97/03997 PCT~US96/11786 of T30177, the distribution of nucleotide bases in the uppermost face of the fold would be similar to that of the lower face, but with one less intrrnllrl,ootidyl phosrh~tr linkage.
Those cr,n~ rations served as the basis for the design of the 16mer oligonucleotide, T30695 (Figure D-lA). As seen in Figure D-2A, line a, even though T30695 is one base shorter than the S T30177 homologue, it was found to melt at ~lo~hlldl~1y lO~C higher ~ ,c~ , over the entire K+ range tested. As for the other homologues, Tm values for T30695 were found to be strand co~r~-l-a~ion independent, confining the general ~imil~rity of the folding process (Figure D-2C).
For T30695, the K+ ion ~ pen~ltonre of thermal stability was very striking. In the presence of 20 mM of Li3PO4 as buffer, measured Tm values increase from 40~C to 65~C over the added KCl range from 50 EM to 1 mM. Again, this ion deprn~ nre argues that the observed stabili7~tic-n is likely to result from site-specific ion binding, rather than simple ion-screening effects.
In order to explore the selectivity of ion binding by T30695, Tm values have been measured for alkaline metal ions with differing radius: Na+ (0.99A), K+ (1.38A), Rb+ (1.49A), and Cs+ (1.69A). As seen in Figure D-2B, ~ig.,iri-~." K+ ion selectivity is ~i~tectrrl Although Rb+ is very similar to K+ in general rh~mi~l properties, and differs by only +O.llA in ion radius, it is seen that the Rb+ complex with T30695 melts at d~ xi",;1l~1y 20-30~C lower temperature over the entire cuncelllldlion range studied. Na+ ion and Cs+ ion, which differ from K+ in ion radius by -0 37A and +0.29A, respectively, are seen to be even more destabilizing.
Similar ion binding selectivity were obtained by this method for the T30177 homologue (not shown).

Circular Di~ . In order to explore the nature of these ion binding effects, the illV~
monitored the folding of T30695 by circular dichroism (CD) mrtho-lc. It is known that G-quartet based folding, both intra and int~rm~ cnl~r, gives rise to large induced ellipticity values (Balagurumoorthy, P. & Br~hm~rh~ri, S. K. (1994) J. Biol. Chem. 269, 21858-21869; Jin, et al.
(1992) Proc. Natl. Acad. Sci. USA 89, 8832-8836; Lu et al. (1992) Biochemistry 31, 2455-2459;
Gray et al. (1992) Methods in Enzymology 211, 389406). Stable tetrad folds are chal~.;Lt;li~t:d by n~mron~elv~Liv~ spectra, with maxirna at 264 nm (~ lxlO+5 deg-cm2/dmol) and 210 rim (~Sx10+4 deg-cm2/drnol) and a minima at 240 rim (4x1044 deg-crn2/dmol).
In Figure D-3A, the i~lvellLol~ l.. o.liLo.~d the CD s~e~;Llul-- of T30695 at 0, 0.05 and lOrnM
KCl. As seen, at the highest added KCl cu"r~ ion, the induced CD ~ecLlulll is very sirnilar to that predicted for an orderly G-tetrad based fold. IllLclc:,Lillgly, the ~e.;L.u,ll obtained in the absence of added KCl is indicative of ~ignifir~nt folding in the absence of added K+ ion, at 25~C
in the supporting 20 mM Li3PO4 buffer. Tm data for T30695 in Figure D-2A suggests an extrapolated Tm near to 20~C in the limit of very low added K+ ion. That extrapolated value is c-n~i~rent with the structure evidenced at 0 mM KC1 in Figure D-3A.
Detailed K+ titration data of that kind have been presented in Figure D-3B, for T30695, , T30177, and T30676. As seen, all three oligomers displayed a generally similar increase in elliptic as a function of added K+ ion c~ ~~ ,l . dlion, which is conci~tent with the hypothesis that they fold ire a fashion similar to the simple model of Figure D-lA. However, the ion col-e~ill,dlion depen~ n~e of the folding process is 4~ ively dirr~.ellL for the three. As would have been predicted from the Tm data of Figure D-2A, it was found that the coupling between K+ ion binding and folding is stronger for T30695 (transition midpoint near to 0.02 mM), than is the case for T30177 (0.15 mM) or T30676 (0.27 mM). The T30677 oligomer, which was the least stable of the oligomers tested by Tm analysis, showed very little elliptirity change over the 0- 100 mM KC1 range, and therefore has not been ~lesellled in Figure D-3B.
Closer inspection of the data in Figures D-3A and D-3B suggested that the CD titration for T30695 is biphasic, with a first step cc-mrlete-l by 0.1 rnM, and a second step which is complete in the 1-2 rnM range. In order to confirm that the K+ induced folding process involves two steps, the inventors have ~elr~ d CD titrations with dirrelGllL alkaline metal ions (Figure D-3C). The hlv~ ul~ Rb+ induced folding of T30695 is associated with an overall ellipticity increase which is very similar to that induced by K+. This argues that the Rb+ and K+ complexes are folded in a similar fashion. However, as expected from the Tm dirr~lcnces seen in Figure D-2B, it is observed that Rb+ ion induced folding is qll~ntit~tively dirrel~llL, occ lrring only at relatively high added ion collc~ ld~ion (0.5 rnM midpoint as culll~Jdled to 0.02 rnM for K+). This confirmc that Rb+ is a much poorer effector of the folding process. A very similar K+ vs. Rb+ dirreiellLial was seen for T30177 (not shown), which suggests that the two oligomers display similar overall ion binding selectivity.
The biphasic character of the T30695 folding process now easily detected upon aMition of Rb+
(Figure D-3C). The m~gnihl-1e of the CD change associated with the first and second ion induced steps are similar for both K+ and Rb+, c.",li"";"g that the folding process has not been ~ignifi~ntly altered in a qualitative fashion by Rb+. For col~ on, it was observed that folding of T30695 as a filn~ti~n of Na+ ion binding is not biphasic, and is associated with a total ellipticity increase which is no larger than that of the first transition seen in the presence of K+ and Rb+ ion.
One hlLel~lt;ldLion of this dirrel~llce is that Na+ ion is capable of driving the first, but not the second step in the folding process. This proposal will be ~ cllcced below.

Folding KinPti~c. In order to investigate the two step folding process in more detail, the hlvellLo have measured the kinetics of oligomer folding, for T30695 and T30177 at 25~C in the standard W O 97/03997 PCT~US96/11786 20 rnM Li3PO4 buffer. Data were obtained by manual ~ tion of K+ or Rb+ ion at time zero, followed by mea~ul~llc:llL of UV absorbance change at 264 nm, in the 0 to 300 second time range In Figure D4A, K+ ion has been added to T30177 at 0.2 uM (curve a), 1.0 rnM (curve b) or 10 uM (curve c). In Figure D-4B, Rb+ ion has been added to T30695 at 1 uM (curve a), 5 uM (curve S b) or 10 uM (curve c). These three values are a~pl~illldL~ly those required to obtain the midpoint, endpoint and ten times the endpoint of the K+ induced (Figure D-3B, curve b) or Rb+ induced folding process (Figure D-3C). Although not shown, the kinetic data described below were found to be nucleic acid co,)r~ n in~ penrlPnt over the range from 3-10 uM in strands, cu"r" .";.,g that the folding process is intramolecular.
As seen in Figure D4A, upon addition of K+ to T30177 to 0.2 mM (curve a), a single slow kinetic process is (lpt~pctpd with a time con~f~nt near 18 sec. IllL~le~Lillgly, this component is hyp~lclll-,l.lic, inrlir.~ting a net loss of base stacking interaction during this first step of the folding process. Upon addition of snffiri~nt K+ to drive the folding transition to completion (1 uM, curve b), a second kinetic component is detected ( ~ = lS sec, t2 =lx104 sec). The second component is hy~oclllulllic, indicative of a net increase in base stacking, and is very slow. Upon an a~ ition~l increase of K+ ion to 10 ~M (curve c), the first kinetic colll~oll~--L becomes nearly too fast to be detected in the current ~ dldLUS, while the time cull~L~ulL for the second step has decreased to about 50 sec. Very similar kin~tirs~ but ~ rly 20-fold slower, have been obtained upon ~ ition of Rb+ ion to T30177 (not shown).
In Figure D4B, the hlvc~llLul~ have p~lrc,ll,led a similar folding analysis on T30695, but with Rb+ instead of K+ ion. This was done because, for T30695, the kinetics of K+ induced folding were too fast to be detected in the simple optical d~palaLu~ employed. As seen in Figure D4B, upon addition of Rb+ to lmM (curve a), a single slow kinetic process is ~lrt~cte i, similar to that obtained at low K+ ion c.Jllct:llLldLion with T30695 (Taul=48 sec). Again, this component is hyp~l~llloll ic, infiir~ting a net loss of base stacking intrr~rtiQn Upon ~ 1ition of snffiri~nt Rb+ to drive the T30695 folding transition to completion (5 uM, curve b), a second kinetic component is ClPtPCt~l Again, the second component is hypochromic, indicative of a net increase in base stacking. Upon ~lrlitil~n~l increase of Rb+ ion cu''r~~ f;on to 10 ~LM (curve c), the first kinetic component becomes nearly too fast to be ~lPtPctP~l while the time CUll~i for the second step has decleased from about 60 sec (curve b) to about 16 sec.
Although these initial kinetic data are not s~ffiriPnt to solve for rate cu ~ , the abs,.ll,;~lce-detected kinetic data for both T30177 and T30695 are con~i~tpnt with the equilibrium binding data obtained by CD (Figure D-3, B and C). Both trrhniqlles suggest that for K+ and Rb+, the ion induced oligomer folding process is aphasic. Kinetic data obtained with Na+ ion (not shown), suggest that only the first, lly~el~ L. --lic transition is obtained at any c~",c~"~ldLion in the 0-200 mM range. That observation is also generally c~n~ ell~ with Na+
titration data (Figure D-3C). A structural model is proposed below to rationalize those observations.
-S A Rel~tiu~lli~ Between SLIu~lule and Fnn~inn The inventors' interest in T30177 and its ~ d~;liv~Lives has arisen because this class of oligonucleotide is a potent inhibitor of HIV infection in culture (Rando et al., (1994) J. Biol. Chem. 270, 1754-1760; Ojwang et al. (1994) J. AIDS
7, 560-570; Bishop et al., (1996) J. Biol. Chem. 271, 5698-5703; Ojwang et al. (1995) Antimicrob. Agent Chemotherepy 39, 2426-35), and in vitro, has been shown herein to be the most potent inhibitor of HIV-l integrase to have been illentifit?cl thus far (see also Ojwang et al.
(1995) Antimicrob. Agent Chemotherepy 39, 2426-35). In Table D-l, there is provided a catalog of the melting l~ dluLes of the closely related set of derivatives used in this study, as an index of their stability as an imr~m~lec~ r tetrad-based fold. Stability has been plesellL~d at three dirr~cl~L added K+ ion concentrations, ~p~nning a range of Tm values which differ by 50~C. This was done to ensure that stability-activity correlations would not be limited to any particular K+ ion co~r~ lion~
Three kinds of activity data have been pl~s~lllc:d. In~,dse inhibition by these oligonucleotides has been monitored for both the 3' excmnrlea~e and strand transfer activities of the purified HIV-l illlr~ e (Ojwang et al. (1995) Antimicrob. Agent Chemotherepy 39, 2426-35). Data are l"esellL~d in Table D-l as the IC50, in nM of added oligonucleotide.
Antiviral activity has been obtained as described herein and elsewhere (Ojwang et al. (1995) Antimicrob. Agent Chemotherepy 39, 2426-35), and is plc:sc~ d as the IC50, in nM, of added oligonucleotide .
Inspection of Table D-l suggests that, relative to any added K+ ion cnnr~ntration, there is a correlation between thermal stability of the folded state and the capacity to inhibit the exnmlrle~se or strand transfer activity of purified HIV-l h~L~,ldse. A qualitative correlation is also obtained when colll~ g thermal stability with lllea~u,t;d anti-HIV activity in cell culture.
A rel~ti- n~hip between thermal stability and function can only be me~nins~fill for folded structures which are very sirnilar. However, given the sequence ~imil~rity among these four homologues in Table D-1, and the ~imil~rity of their ion-induced folding process, the correlations are likely to be m~ningfill CQ~ O ~ Re~ the St~ e/Function Studies Data were obtained suggesting that the anti-HIV oligonucleotide drug T30177 and its homologue T3005, fold via intramolecular G-tetrad formation, to yield a structure which is W 097/03997 PCTnUS96/11786 stabilized by K+ ion binding. It is well known from the li~ aLulc~ that alkaline metal ions can stabilize G-tetrad formation (Williamson, J R (1994) AnnuZ Rev. Biophys. Biornal. Struct. 27, 703-730). What distinguishes the behavior of these two oligomers is the nml~n~lly high stability of the folded state (Figu}e D-2A), the high selectivity shown for K+ ion (Figures D-2B and D-3C) and the possibility that K+ coordination may be strongly coupled to loop structure within the oligonucleotide fold (Figures D-2A and D-3B). Co~ "( with the idea that ion binding ,, may occur with G-tetrads and with loops, the i~lvelll~ have observed that the folding of T30695 and T30177 appears to occur as a two step process, as detected by eqnilihrillm (Figure D-3) and kinetic methods (Figure D-4).
In order to relate these various observations, the hlve~lLol~ have found it useful to propose a simple, two step folding model (Figure D-lB). They suggest that the first, higher affinity ion binding step occurs by coor-lin~tinn of metal ion with the central-most pair of G-tetrads, thereby gell~ldLillg a core octet which is similar to that seen in related intr~mnleclll~r folds (Willi~m~on, et al. (1989) Cell 59, 871-880; Panyutin, et al. (1990) Proc. Natl. Acad. Sci. USA 87, 867-870; Smith, F. W., & Feigon, J. (1992) Nature (London) 344, 410414; S-'hlllt7~, et al.
(1994) J. Mol. Biol. 235, 1532-1547). It is proposed that, by analogy to those other, better understood G-tetrad based structures, this first ion binding step has rather modest selectivity among the alkaline metal ions (Williamson, J. R. (1994) Annul Rev. Biophys. Biornal. Struct.
27, 703-730). The inventors propose that the second step in the folding process involves binding of ~ itinn~l ion equivalents to the loop regions of the structure. It is also proposed that this second process, which occurs at higher added ion co~ . dLion (Figure D-3) and which is ~sori~t~od with the slow kinetic step of Figure D4, is coupled to a rearrangement of the loop ~lom~in.~ to yield two ~lrliti~n~l sites for metal ion coor lin~tion In preliminary mnrl~-ling studies (not shown), the hlv~llL,l~ have confirm~ that orderly structures of the proposed kind can be obtained in which cdllJollyl oxygens from T and G base plains are ~lga~ d in the loops so as to complement the end of the G-octet, resulting in oçt~h~flr~l coor~lin~tinn of one K+ equivalent at each of the two junctions between loop and core octet dom~ins. It is proposed that this capacity for ~tlrlition~l K+ ion binding is the origin for the l~ le stability of T30695, the corollary being that other homologues described in this work are less stable because they have lost one or the other of the proposed K+
coortlin~tion sites. A second corollary of the model is that the high ion selectivity seen for these oligomer folds is dolll;l.~l~d by the structural re4uhclll~llL~ for ion binding to the loops, rather than from ion binding within the core octet. Prelilllillaly NMR data (Ding & Hogan, unpublished data) suggests that the ~ ition~l binding step involves 2 equivalents of K+, yielding 3 K+ equivalents per oligomer fold, at saturation.

W O 97/03997 PCT~US96/11786 cOI,rll Ill~ion of this model awaits detailed structure analysis. However, the data at hand (Table D-1) suggest that formation of the ion-selective oligomer fold described herein may be a n,-cess~ry pre-condition for anti-integrase and the overall anti-HIV activity of these compounds. As such, lcr l.r l l l~l~l of the present folding model could prove useful as the basis for ph~ l h~ ov~ llL.
"

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 o ~
-- ~ ~ U, ~ ~
:C A

C E- +i -H +1 ~

~ o ~ ~ ~

~D

~ â K
a ~ ~

O
~E ~ ~C' ~ ~E
", '_ -~E ~ = C

~ ~ s O C~
C

_ ~ J o ;~ ~ J L' J ~ ~ 5 o 3 E~ c E-- ~

W O 97/03997 PCTrUS96/11786 Pharmacokinetic Studies E. Single-dose hemodynamic toxicity and pharm~cokin~ti~ of a partial rh~ sFh-n oll.ioate anti-HIV oligo .- 1F~ (AR177~ following iull~velluus infusion to ~ O1~1C
monkeys As part of the pre-clinical ~se~",~"l of AR177, a toxicity study of AR177 (T30177) was cond~lctfd with the objective of est~hli~hing the dose-l~pol1se relationship between intravenous infusion of AR177 and hemodynamic paldlllcL~l~ in cynomolgus monkeys. Intravenous infusion is the proposed route of ~dmini~tration of AR177 to humans. The present study was conducted using the short term infilcic-n protocol recommfndf d by the Food and Drug ~dmini~tration (Black et al., 1994), with mea~,ul~ nl of central blood pl~5~,ule, serum ~hf~mictry, hematology, coagulation factors, complement factors, and plasma AR177 concentrations.

Materials and Methods for Pre-Clinical TOX--Q~-OY Screens Materials. AR177 was synthf~ci7rd at Aronex on a Milligen 8800 oligonucleotide synthf~i7f~r, and made into a stock solution at 25 mg/mL in sterile phosphate-buffered saline. AR177 has a mnk~c~ r weight of 5793 daltons, and is a fully neutrali_ed sodium salt. The structure of AR177 was cha~ f'd by phosrhnrus and proton NMR, sequencing, base composition, laser Resorption mass ~L,e~;LI. ll~lly, anion exchange HPLC and polyacrylamide gel electrophoresis. The AR177 was ~yl~Ahl~Lt:ly 94% pure accoldillg to HPLC and electrophoretic analysis. All analyses are c~n~i~tf~nt with the proposed structure.
For HPLC analysis of plasma AR177, tris was obtained from Fisher, NaBr and NaCI were obtained from Sigma, and mf th~nnl was ~ulcl~ased from J. T Baker. The Gen-Pak Fax anion-eAcll~lge HPLC column (4.6 x 100 mm; cat. no. #15490) was purchased from Waters.
Dosing. Twelve exp~, ;",~.,I;.lly naive cynomolgus monkeys were ~esignf~d to four groups of three animals each. Prior to dosing, each animal was lightly sedated with a c~ hlalion of krL;.",i,~f' (1 O mg / kg) and ~ 7Pp~m (0.5 mg / kg), and a catheter was introduced into the femoral artery for lcculllillg central arterial ~lGS~,ul~. Monkeys were given a single intravenous infil~inn of 5, 20, or 50 mg AR177/kg or saline over ten minutes through a cephalic vein catheter using a Harvard infil~inn pump. Arterial blood samples were drawn at -10, +10, +20, +40, +60 and +120 minutes relative to the initi~tion of infusion into EDTA-c~ illillg tubes for hematology, complement factors, coagulation assay, serum rhf~mi~try, and plasma AR177 .l~L~, .,.i"~linn At 24 hours post-infiusion, blood was drawn via the femoral vein into EDTA-cnlll;~illillg tubes for these same p~udllleL~l~,. The c~ rr~ dlion of AR177 in dosing solutions was c~ ~" rh 11If d post experiment by absorbance at 280 nrn on a spectrophuLo~ L~l. For the ~lft~-- ~-,i";11inn of AR177 by HPLC, the WO 97/03997 PCT~US96/11786 plasma fraction was obtained by low speed centrifilg~tion of blood, and stored at -20 ' C until used.
Electrocardiograms (ECGs), central pLC~S:~iUl~::, and heart rate were recorded continllously for 120 minutes following the initi~tit~n of dosing. Table E-1 ~ulllll~ui~cs the study design. The animals were observed twice daily for ph~rm~eotoxic signs and general health beginnin~ two days before S dosing and for seven days following dosing. The monkeys were not necropsied at the end of the study.

Serum ~ y parameters. The following were drL~ rl sodium, potassium, chloride, carbon dioxide, total bilirubin, direct bilirubin, indirect bilirubin alkaline rho.srh~t~ce, lactate dehydrogenase, aspartate aminotr~ncfer~ce, alanine aminoLl~l~rclase~ gamma-gluL~ullyl~ cfer~ce, c~l~inm, phosphorus, glucose, urea nitrogen, c~ i"i"~o~ uric acid total protein, albunun, globulin, rhole5terol and triglycerides. The samples were analyzed at Sierra Nevada Laboratories (Reno, NV).

~f~ tnlogy and co~gnl~hon paraIneters. The following were rieterminet1 red blood cell count and morphology, total and dirr~ ial white blood cells, hemoglobin, h~m~tor.rit, pLoLlllullll~i time, fibrinogen, mean cell hem~ gk~bin, mean col~us.;ular volume, mean col~us~;ular hemoglobin cu~,r~"l,dLion, platelet count, and activated partial thromboplastin time. Hematology p~
were ll~t~rminPfl at Sierra Nevada Labn,,,l~ os (Reno, NV).
C , ~ t factors. The complement split product Bb and total hemolytic complement CHS0 were ~letermine~l The choice of mP~cnring the Bb split product, as opposed to other complement factors, was based on a pllbli.ch~ study showing the involvement of the alLcllldLivc pdLhw~y in complement activation induced by oligonucleotides (Galbraith et al, 1994). Complement ~1~L~I ",i";~lionc were p~lrJlllled in the laboldL~.ly of Dr. Patricia Giclas at the Complement Laboratory, National Jewish Center for Tmmnnnlogy (Denver, CO).

~LC analysis of AR177 plasma co ~.Liulls. AR177 was assayed in the plasma using an anion-exchange HPLC method on a Waters HPLC system with a 626 pump, 996 photodiode array detector, 717 ;~ o~",l-ler and ~illenninm system software controlled by an NEC Image 466 Co~ uLcl. Buffer A consisted of 0.1 M Tris base, 20% m~th~nnl, pH 12, and Buffer B c- ncicted of 0.1 M T}is base, 1.0 M NaBr, pH 12. The anion-exchange column (Gen-Pak Fax column) was equilibrated at 80% buffer A/20% buffer B for 30 minutes before each HPLC run. Fifty microliters of 0.2 ~ filtered, neat plasma were analyzed per run. The elusion c~n-liticnc were: a) five-minute isoaatic run at 80% A/20% B. during which the majority of the plasma proteins eluted, b) ~A 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 25-minute linear gradient to 30% A/70% B during which AR177 elutes, c) five-minute Socratic run at 30% A/70% B. d) one-minute linear gr~ nt to 100% B. e) t-wo- minute run at 100% B for column clean-up, and fl two-minute linear gradient to 70% A/30% B for the step in the HPLC
clean-up. The high pH (12) of the elusion buffers was n.ocess~ry to ~ o(~i~t~ AR177 from tissue con~tihl~nt~, which bind AR177 tightly around physiological pH. AR177 is completely stable at pH
12. This method can clearly distinguish between the full length AR177 and n-1, n-2, etc. species, which are potential metabolic products. The ultraviolet detection wavelength was 260 nm. The flow rate was 0.5 mL/minute in all steps. Column clean up between runs was p~lrulll,ed by a 500 pL
bolus injection of 0.1 M Tris base, 2 M NaCl, pH 10.5, followed by: a) ten-minute linear gr~ nt to 60% A/40% B. b) one-minute linear gr~ nt to 100% B. c) a three-minute isocratic run at 100% B and d) one-minute linear gradient to 80% A/20% B.
A standard curve was gen~rat~d by spiking AR177 into cynomolgus monkey plasma inorder to achieve cu,.~ .dLions of 0.04 to 128 ~/mL. The plasma standards and unspiked plasma (control) were run on the anion-exchange HPLC column using e above con~liti( n~. The Waters Mill~nninm software was used to ~ilotçrminp the area under the peak for each AR177 standard at 260 nm. The HPLC peak area versus AR177 ~;ol~-e~,l,,.liQn was plotted using Cricket Graph III
1.5.1 sOrLW2~ There were one to three HPLC replicate runs per AR177 standard. The limit of qu~ntit~tion was 25 ng/rnL (50 pL injection), whereas the limit of ~t~ctit n was 5 ng/mL (50 pL
injection). The overall correlation coefficient of the fitted lines on the standard curve plots was greater than 0.999 on two dirr~lcllL standard curves used in this study. The standard curve was linear over an ~ hl~L~ 3,200-fold range. The variability of the replicates was 1-2% at all con~entr~tions. There was one HPLC run per monkey plasma sarnple. This method was v~lirl~t.-~l Further details about the method will be published elsewhere (Wallace et al., snhmitt~od).

Pharrn~.~okin~tir ~a~ 7. The volume of distribution (Vd)was c~lrnl~tt-d by dividing the total dose ~tlmini~tçred by the con~çntr~tic-n at the end of the infil~ n (Rowland and Tozer, 1995). The CMAX, (~ ll conrentration) was taken from the plasma cu,u~"l.i-~ion at the cnn~ ion of the ten-minute intravenous infilci~n Results/Clinical Ol~:,t:. ~'dLiUllS and Hemo~lylld- ~c Pdl d~ l S
Aside from an ~ntiro~gulant effect ~i~os~rihed below, there were no in~ ti~n~ of ~ignifi~nt toxicity. No clearly Ll~zlLll~llL-related changes in blood pressure (Figure E-1), heart rate (data not shown) or electrocar~liogr~phi~ activity (data not shown) were observed, no animals died following AR177 infilsion One high-dose (50 mg/kg) monkey exhibited a rise in arterial ~lGS~,ulc during the infusion followed by a decline to ~lu~ill~L~ly 20-30 mm Hg below the pre-infusion blood ~lC~i~iUl~. These changes are qualitatively similar to, but less pronounced, than those seen in monkeys given total pho~phnrothioate oligonucleotides (Galbraith et al., 1994). Although suggestive of a treatment effect, the alterations in blood plC~S~ul~ in the subject animal could not be clearly distinguished from normal flllr*l~ti~n~ that occurred in other animals, in~-hl-ling one control S monkey. The only Llc:dllllGll~ related clinical sign was emesis during the infusion, which occurred in two of the three animals in the 50 mg/kg group and all of the monkeys that received the 20 mg/kg dose.

Serum Chemistry There were no changes in any of the serum rh~mi~try pdl~ i that could be attributed to AR177.
T~Pms~tQIogy There were no changes in hematology values attributable to AR177. Neutrophil counts were increased to a sirnilar extent in all groups, in~ ing the saline control group, probably as a result of the stress associated with the exp~lhl~llLdl procedure (Figure E-2). The chald~ lic ncuLl~llia and rebound neutrophilia that has been reported with other oligonucleotides did not occur in the AR177-treated monkeys, which is CO,)~ with the relatively small changes in complement Bb split product (Figure E4) and CH50 (Figure E-5) levels.

Co~ .n parameters The most salient effect of AR177 observed in this study was a pronounced, albeit transient, dose-dependent, reversible prolongation of aPTT in the 20 and 50 mg/kg groups, which reflected inhibition of the inrrin~ic coagulation paLllwdy. There was at least a four-fold prl-long~ti-)n of aPTT
in the 20 and 50 mg/kg dose groups at the c~ n~hl~inn of the infusion of AR177. Det~rmin~ti~ n of the upper aPTT value was limited by the range of the assay. (See Figure E-3). This change was reversible in both dose groups. The aP~T was hl~:lcdsed beyond the upper limit of the assay in the 50 mg/kg group for all or most of the two-hour lllollilc lhlg period, but had returned to normal by the following day. R~elin~o aPTT values were reest~hli~h~d by two hours after L~l",il~;~lion of dosing with the 20 mg/kg dose. In the S mg/kg group, only a small and tr~n~ nt rise in aPPT was observed, and there was no change in pLuLlllullll~ time (PT). Similar changes have been observed with other oligonucleotides, and are believed to be, at least in part, attributable to direct and reversible binding of the oligonucleotide to Lhlulllbill (Henry et al., 1994, Pharn~nee~ti~ Res. 11:
S353, 1994). PT was affected to a much lesser extent than aPI~ in the 20 and 50 mg AR177/kg groups (data not shown), in~ ting little or no effect on the ~x~ ' p~Lhw~y.

C~ A ~ ' ' activation Plasma levels of the complement split product Bb, a marker for activation of the ~ " .~ I i ve p~Lllw~y, were increased 60-85% over baseline in the 5 mg/kg group, approximately 2-fold over baseline in the 20 mg/kg group, and appr-~im~trly 2- to 4-fold in the 50 mg/kg group at the end of infusion. (See Figure E-4). The elevation in Bb persisted through the duration of the 2-hour monitoring period, but the values had returned to normal by the following day. These increases in Bb were small in m~gnihl-lr. There were also small and tr~n~i~nt decreases in the CHS0 levels (Figure E-S) in the 20 and 50 mg/kg doses, but there was no dose-CH50 level relationship. In co" ri " ";,l ion of this minimz~l change in CH50, AR177 had no effect on complement CH50 at doses up to 236 ~/mL when it was tested in vitro in human or cynomolgus monkey plasma. (See below).
Thus, a large increase in complement activation, and resulting characteristic n~uLluAucllia and rebound neutrophilia, that has been reported with other oligonucleotides (~lhr~ith et al., 1994) did not occur in the AR177-treated monkeys (Figure E-2).

Plasma AR177 ~UAI _~AILA dliOAI
Plasma co~r~ntrations of AR177 were m~cim~l at the end of the infilci~n and derline~l Llleleanc:i with an a,u,ulv~illlaL~ initial half-life of 20-30 minutes (Figure E-6). Another more comrlet~ study in cynomolgus monkeys has shown the terminal half-life to be ~u~ur~ xilll~l~ly 24 hours (See below). These half-lives are much longer than that reported by Lee et al. ((1995) Pharmaceut. Res. 12: 1943-1947) in cynomolgus monkeys for GS-522, a 15mer oligonucleotide that has a tetrad structure similar to AR177. No metabolites of AR177 could be observed in the plasma at any time point or dose. This contrasts with the results of Lee et al. (1995), who found ~i~"i riri1 .,1 amounts of shorter species of ~S522 in monkey plasma following intravenous infil~ n The results with AR177 suggest that AR177 does not undergo metabolism. There was a direct r~l~ti-n~hir between the AR177 plasma Cmax and the dose that was ~ ed as a ten-minute intravenous infusion to the monkeys (Figure E-5). Plasma Cmaxs of 83.2 +/- 7.2, 397.8 +/- 30.8, and 804.7 +/- 226.3 ~g/mL were achieved for the 5, 20, or 50 mg/kg doses, respectively, at the end of the infusion (+10 minute time point) (Table E-2; Figure E-7).
The initial volume of distribution (Vd) of the three doses ranged from 200-248 mL (mean + s.d.) (Table E-2) at the conrlll~ion of the intravenous infilsion The mean body weight of the monkeys in the AR177 dose groups was 3.67 kg. ~sllming that plasma volume is 4% of body weight (Davies and Morris, 1993, Pha)7n~7ce~Ji~ Res. 10: 1093-1095), the plasma volume would be 147 mL. Thus, the initial Vd was slightly greater than the plasma volume.
In general, there was a direct relationship between the plasma conc~llLldLion of AR177 and aPTT for the 5 (Figure E-8) and 20 (Figure E-9) mg/kg doses. For the 50 mg/kg dose (Figure E-W 097/03997 PCT~US96/11786 10), the aPTT values were off-scale during the two-hour sampling period so it was not possible to drterrninP the rel~tirln~hir between the plasma concentration and aPPT. There was a no effect plasma AR177 conrentration versus aPPT of d~ xill.,.lrly 60-100 ug AR177/mL, above which there was prolongation of aPTT. Doubling of aPTT was ol)sel ved at plasma AR177 concentrations of d~lo~illldl~ly 100-250 ug AR177/mL. Tripling of aPTT was observed at plasma AR177 conrentr~tions of d~lo~dllldLt:ly 250-300 fig AR177/mL, after which no correlation was possible because the aPTT values were beyond the aPTT assay range. The disd~edldllce of AR177 from plasma was roughly correlated with the return of the aPTT to b~ inr, which is collsi~L~llL with direct and reversible binding of the olignm-rleoti-ie to one or more clotting factors. By contr~t there did not appear to be a correlation between the plasma concentration of AR177 and complement split product Bb (data not shown).
In ~ ition to the in vivo study in cynomolgus monkeys, an in vitro study was performed which invçsti~trd the effect of AR177 on the coagulation- cascade and complement activity in cynomolgus monkey and human plasma (coagulation) and serum (complement), respectively.
AR177 caused a two-fold increase in aPTT at a conrentration between 30 and 59 f~g/mL of human plasma, whereas the compound caused a two-fold increase in aPTT at a c~ ..,r~.,l . d~ion between 118 and 236 ~Lg/mL of cynomolgus monkey plasma in vitro. AR177 had no effect on Lhl.~llll~ill time in human plasma, but caused d~lo~illldL~ly a 2.5-fold increase in Lhlollll~ill time in cynomolgus monkey plasma at 236 ~g/mL. AR177 had no effect on either fibrinogen or complement CH50 at doses up to 236 ug/mL in human or cynomolgus monkey plasma. AR177 caused a 30% increase in pl-,Lhlol.ll~ill time in human plasma and a~lo~illldL~:ly a 15% increase in ~ LI-lullll,hl time in cynomolgus monkey plasma at 236 ~Lg/mL.

D;~ n Using an i~lrntir~l dosing regimen to that used in previous ~ lhllc~llL~ that resulted in profound hemodynamic effects, the present study showed that AR177 was very safe. Although limited c~nrhlcionc can be drawn from the present study because only one partial ph~phorothioate (AR177) was ~x~minr-l, it is possible that the lack of the cardiovascular toxicity is due to the limited number of phosrh- rothioate linkages (two) in AR177. It is also spernl~trd that the lack of toxicity could be due to the three-dimensional (i.e. tetrad) shape of AR177 (Rando et al., 1995, J.
Biol. ~hem. 270: 1754-1760). In c- --l i- ~";.lion of the lack of toxicity of AR177 found in the present study, AR177 does not cause toxicity when it is ~-lmini~t-red as a bolus intravenous injection to cynomolgus monkeys every other day at 40 mg/kg for a total of 12 doses (see below).
There was minim~l activation of the complement system following ~lmini~tr~tiQn of AR177. Small increases (24 fold) in plasma Bb levels occurred at plasma AR177 c~,llc~llLldLions W O 97/03997 PCT~US96/11786 as high as 750 ~L/mL after a dose of 50 mg/kg given as a ten-minute hlLl~ve~lluu:, infusion. Minimal ~, changes (-25 %) in the CHX levels occurred at plasma AR177 ~ ;. .l lc as high as 750 ~L~g/mL
after a dose of 50 mg/kg given as a ten-minute hlLldvclluus infusion. The c~--~' activation that was seen with AR177 at these high doses did not even result in hy~ulcll~iull. By contrast, G-Alhr~ith et al. (1994) have reported that GEM91, a 25-mer phh~ .luLllioate oli~ oti~
caused an 80% decrease in complement CH50, a 700% ir.crease in the level of comr-lPTnPnt C5a, and death in two out of four monkeys following the hlLI~vclluu~ infusion of 20 mg/kg over ten minutes. The Illr-~llA..;~.., by which olig- n--rlPoti~Ps activate the complement system is UllhlIUWII.
However, this phPnr~mPnnn bears l~ A~lr~ to a similar phPl~...l.~u.lu in human patients during dialysis in which contact between blood and dialyser ~ h~ A ~ l~ induces cul ~ activation and profound ncuLlu~cnia (Heierli et al., 1988, Nephrol. Dial. Transplanr 3: 773-783; Jacobs et al., 1989, Nephron 52: 119-124). The present work indicates that although AR177 induces some minimal c~ activation, this is not ~ d into the h~lllodyll~ullic toxicity that has been seen with other oligl.llll~ Iroti~lP5 AR177 ~h.. ;ll;~ll Al;(~n }esulted in the dose-~ 1 hll,ihili.. ll of the intrinsic co~A~ll~tir,n pathway, reflected by prol on ~ti- n of aPTT. The effect was maximal at the end of infusion and was reversed in parallel with cl A ~ P of AR177 from plasma. The inhihitir,n of coagulation was cignifir~nt at the highest dose level, but marginal and not cull~id~ .d clinically ~i~,.liri. ~.ll at the 5 mg/kg dose level. An All~;l o~gul~nt effect has been reported to be a class effect of oligomlrlPQti-iPc (Henry et al., 1994). The Al l;~oA~ Ant results with AR177 thus agree with results that have been seen with other oligom~rlPQti~iPc~ although AR177 is 40-fold less potent than the Lhlulllhill-binding aptamer oligom~rlPoti~lP that has been reported by Griffin et al. (1993).
In col-rl~ .,l AR177 does not cause mortality, cardiovascular toxicity, or AltPr-Atinnc in clinical ~h~ Lly in cynomolgus monkeys rcceivil~g doses up to 50 mg/kg as a ten-minute hlLl~lvcllous infusion. However, there was a lcvcl~ible prol-ng~tir,n of co-gnl~ti~n time at doses of 20 and 50 mg/kg. Taken together, the data suggest that AR177 does not have the helllo-lyllalllic toxicities that are ~Accoci~tPd with total pho~,huluLllioate oli~ c and can be Alhll;ll;~ ,d safely as an hlLIav~:lluu~ infusion over ten minutes.

TABLE E-l. Monkey Dosing Tnfc~rrn-Atif)n ., ~~ Dose Monkey Dose levelDose Conc. volume weight(kg) Group Treatment (mg/kg) (mg/mL) (mL/kg) Male Female mean + s.d.

W 097/03997 PCT~US96/11786 Placebo 0 0 4.0 1 2 3 67 i 0.38 2 AR177 5 1.25 4.0 2 14.12 i 0.83 3 AR17720 5.0 4.0 2 13.67 + 0.29 4 AR17750 4.0 4.0 1 23.23 i 0 76 s Table E-1 - Monkey dosing : - - r,J ~ - Cynomolgus monkeys were given ten-minute, illLIaV~;lWU5 infusions of 5. 20 or 50 mg AR177tkg at a volume of 4 mL/kg.

TABLE E-2 Plasma AR177, Vd, aPTT and cr~mrlrnnlont Bb values at CMAX

Dose AR177 plasma Vd (mL) aPTT Bb (mg/kg) (~g/mL) (seconds) (~Lg/mL) 83.2 i 7.2247.6 i 21 4 45.9 i 11.4 0.59 i 0-07 397.8 i 30.8184.5 i 14.3> 166 8 i 23.8 1 48 i 0 44 804 7 + 226 3200.7 i 56 4> 170.0 i 34 61.28 i 0.46 Table E-Z - Plasma AR177 Cm,x, aPTT and .-1 , ' Bb levels. The AR177 plasma CMAX, aPTT, and Bb values are the means i standard deviations of data at the + 10 minute tune point (end of the infusion). The baseline (10 minutes prior to dosing) aPTT levels were 32.1 i 4.4.
41.6, 6.7 and 33.2 i 4.8 seconds for the 5, 20, and 50 mg/kg doses, I~,~y~;Liv~ly (mean i s.d.).
The baseline (10 minutes prior to dosing) Bb levels were 0 44 i 0.14, 0.78 i 0.46 and 0 49 i 0.21 ~L/mL for Me 5, 20, and 50 mg/kg doses. Volume of ~;ictrihlltinn (Vd) = dose/plasma CMAX, whe}e the dose is the total mg of AR177.

W O 97/03997 PCTrUS96/11786 .
~ ~ o o ~
,. -- ~ -- -- ~ _ -- '-- o o o o ~ ~ ~ ~ ~ ~ o ~ ~
~, ~ ~ ~ ~ o o o ~ ._ ~ c ~ , O O 00 _ O
-- +l +l ' ~ _ +~ +l o, ~ ~ o ~
~ J ~~ . ~ ~ ~t ~ O O O
C~ ~
o ~ ~ - 2 ~ ~ ~ ~ ~ o o t-- ~ ~ ~ ~ ~ ~ a~ ~ ~

~ ~ 2 -~ E c E ~ E C C ~ E

cq .~

V '-- -- o _ ~C~ ~ C~ ~ ~ ~ ~ o ~Q ~ or~ o ~ ~ o X G 5~ a~ ~~ cr~ c~
._ C ~ O ~ o~, ~ O
~ ~ ~ ~ o ~. ~ ~ o ~ 2 ~ ~. ~
"' ~ ''~~~r ~ ~ -- _ c ~~ t-- ~
_ 3 ~D ~ ~ _ O ---- ~ _ ~ ~D _ _ C ~> '~ '' '' c~ U~ V ~ ~L~ V~
c ~ E E ' E E ~ -~ E E

_ CA 02227867 l998-0l-26 --o o -~o O

,-o o -a O

O
~D
O ,_ :' -V~
O
U~ C

-~ .

C~
a C~ .

CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96/11786 o~ ~oo ~ ~ o ~' ~ ~ ~ ~
~o U~ o ~ _ ~ -- o o -- o -- ----+l +l +l +l +l +l+l +l +~ +l +l +l +l oo +l +
~ ~ ~ _ U~ ~ ~ _-- o o ~~ _ .. .
t-- ~ ~ ~ ~ O O o:~ ~ O, o ~ ~ -- o _ -- o o o o ~ ~ ~
o +l +l +l +l +l +l +l +l +l +l +l +l +l +l +l +l _ ~, _ ~ U-~ ~ _ o o ~ ~

o ~ O -- -- ~ o o O O ~t 'r E +l +, +, +l +l +l +l +l +l +l +l +l +l +l +l ~
~ ~ ~ oo _ ~ ~ ~ X ~ ~ ~ ~ ~ ~ C-- o v E E
E E o ~ ~ x E
o ~ - ~ - - ~ o - - E -c~ = ~ ~ _ ~ ~ ~ o-- ---- ~ o o O O ~ u, ~ ~ O
~ E ~+1 +1 +1 +1 _ +1 ~ +1 +1 +1 +1 ~ +l +l +l ~
O ~ 00 u ) ~ 00 ~O00 00 :~ O O O O c~

.- _ _ o ~ _ o o ~ o o ~ ~
+l +l +l ~ +l +l +l +l +l +l +l +l O +
Oc~ ~" o, O~ ~ ~ ~ O~ ~

3= ~ ~ o - _ - o o o o o o ~ o o +l+l +l +l +l +l+l +l +~ +l+l +l +l +l ~ ~
~ ~ ~ ~ ~ o o o o c~l ~ ~ ~ ~ O ~ ~ ~ ~ ~ ~ ~ ~ ~t ~
~o +l+l +l+l +l +l +l +l+l +l +l +l +l +l +l +l C ~ D V V ~ a V v ~L~ ~4 v v c~
, o - ~ E 5 5 ~ E ~ 5 ~ ~ 5 ~ E ~ ~
~ ~ ~o ~ ~ cO~ ~o ~ ~ ~0 u~ ~, ~O~ 0~

WO 97/03997 PCT~US96/11786 . .

s ..

C~
_ .o e C~

' W097/03997 PCT~US96/11786 o o o _ -- ~o ~ ~ -- ~ ~
~ +l +l +l+l +l +l +l +l +l +l +l +l o ~_ ~ O,t~ -- ~ a~ O ~ ~ ~ +, E

o o o o ~ oo _ ~ _ _ .,, +l +l+l+l +l +l +l +l +l +l +l +l o o ~ '~ ~ ~ -- o U~

~. o o o o o _ o o o ~ ~ _ ~ 2 +l +l+l +l +l +l + ~ ~ _ ~, o~ ~
,~, _ ~_ ~ -- ,_ _ V
~ E
~ -- o o oo g-- o o _ -- ~ _ _ ~ ~
v~+~ +~+~ +~ +l +l +l +l + o C o ~ C
~ r -- +l +l+l +l +l +l +l +l +l +l +l +l o - ~ ~ ~ o ~ .

-- o o o ~
o ~ ~ ~ +l +l +l+l +l +l +l +l ec o - 'o x o~~ ~t o.
o o o o -- o o o ~ c~

~ ~ c U--~~~ ~ _ ~ ~~ ~ _ tr~ ~~ ~ _, F. Repeat-dose toxicity and pharmacokinetics of a partial l~hn~y~ o!l~;o~te anti-HlV
oligo--~ lt~ (AR177) following bolus ~ ~ldV-:llUL~7 ~ alion to cynomol~le monkeys AR177 is a 17-mer partial phosphorothioate oligonucleotide with the sequence 5'GTGGTGGGTGGGTGGGT-3', with sulfurs at the f~rmin~l intermleleo~ e linkages at the 3' and S' ends. It is a potent inhibitor of HIV integr~ce and HIV production in vitro (Rando et al., 1995;
Ojwang et al., 1995), and has a long tissue half-life in rodents (unpnhli~h~-d data). AR177 does not have an ~nti~n~e- or triplex-based m~h~ni~m of action. A previous study has shown that AR177 does not cause the char~-teri~tif hypotension or n~ulluy~llia of other oligonucleotides (Cornish et al., 1993; Galbraith et al., 1994) following a ten-minute intravenous infusion, at doses up to 50 mg/kg (Wallace et al., sllhmitted, 1996). As part of the pre-clinical ~c~~~"..-.,l of AR177, an intravenous toxicity study of AR177 was c- nr~ t~cl in cynomolgus monkeys with the objective of establishing the clinical and histopathological changes that occur following repeated doses.
Materials arld M~tho~le for Repeat Dose Studies Materials. For HPLC analysis of plasma AR177 conrPntr~tir~n~ tris was obtained from Fisher, NaBr and NaCl were obtained from Sigma, and methanol was yulcllased from J. T. Baker. The Gen-Pak Fax anion-exchange HPLC column (4.6 x 100 mm, cat. no. #15490) was ~u~hased from Waters.
AR177 was synth~ci7Pd at Biosearch, a division of PerSeptive Biosystems, on a Milligen 8800 oligonucleotide synthe~i7~r, and vialed at 25 mg/mL in pho~haL~urr~lc~d saline. AR177 has a mf~lec~ r weight of 5793, and is a fully neutrali_ed sodium salt. The structure of AR177 was ~llald~ li;Gcd by phosph(lrus and proton NMR, sequencing, base composition, laser Resorption mass spe~;Ll~ Lly, anion exchange HPLC and polyacrylamide gel electrophoresis. All analyses were con~i~tent with the proposed ~Llu~;Lulc. The AR177 was d~lu~hlldL~ly 94% pure according to HPLC and electrophoresis analysis.
The monkeys used in this study were laboldlc,ly bred (C.V. Primates, Tn~lon.oei~ or Yunnan National Laboratory, China) and were exp~ lly naive prior to the study. The age of the monkeys was 3 to 61/2.
Dosing. AR177 was ~t1rnini~tered intravenously over 1-2 minutes into nn~e-l~tt-d monkeys every other day for 23 days (12 doses) by injection into the femoral vein. (See Table E-l). The monkeys were not sedated, but were restrained during dosing. The highest dose level (40 mg/kg/injection) was selected based on observations in a previous single-dose study of pronounced ~nriro~ulant activity of AR177 at a dose of 50 mg/kg infused over 10 minutes (Wallace et al., submitted). A
comparable or greater degree of anticoagulation was expect~d to occur with fast (1-2 minute) infusion of 40 mg/kg, and was confirmed by the results of this study. The dosing schedule (every other day) was chosen in order to avoid e~ces~iv~ a~cumnl~tion of the test material, which, based on pharmacokinetic data obtained in rats (Wallace et al., s~-bmittlod), would be expected to occur with daily ~tlmini~tration.
The monkeys were observed twice daily for general health, changes in appetite and clinical signs of adverse events. Body weights were ~l~ea~u~d within a few days prior to the first dose (Day 1) and aLlL lo~illlaLely weekly thereafter. Ele~ ocdl.liographic (ECG) .~coldi,lgs were obtained from all animals prior to the study and on Day 22, and from lec.,vely animals on Day 35. Blood samples were collected for evaluation of serum chemistry, hematology and coagulation ~Jdl~llll~Lel:~ from all animals prior to the initiation of the study, on the first day of dosing (Dose l; Day 1), and on the last day of dosing (Dose 12; Day 23). The sample collection on Days 1 and 23 was timed relative to dose ~ n~ d~ion in order to chald~;L~ ; possible acute effects on hematology pdldlll~:~el~i. An additional clinical pathology evaluation was cnn~lllrted for all animals on Day 24, as well as for lcc~v~ly animals on Day 37. Blood was collPcted from all animals at S minutes, 30 minutes and 4 hours post-dosing on Days 1 and 23 for analysis of the plasma AR177 c~ c~l.d~ n On Day 25 (two days after the last dose), three males and three females from each group were hllm~nely ~ h~ d and necropsied, while the l~ i"i"~ two animals each in the high-dose and control groups were contim~d on study for an ~ 1iti~n~1 two-week tlcdLl~lcllL-free "lec~vt:ly"
period, and were e~lfh~ni7Pd on Day 38. Complete gross ne~ ies were p~.r~,.l.,ed on all animals at their 5rhp~ rd L~l"~;";~ n Urine was collected from each animal during necropsy by bladder puncture and ~ d for routine urinalysis. Weights of 13 major organs were recorded, and numerous tissues were collectr.l preserved and processed for histology.

Serum .l.~ pard~ . Serum rhemi~try was determined pre-study, on day 24 (one day after the 12th dose), and on day 37 in the recovery monkeys. The following were r1rterminr~l sodium, pot~illm, chloride, carbon dioxide, total bilirubin, direct bilirubin, indirect bilirubin alkaline phQsph~t~ce, lactate dehydlogc..ase, aspartate aminoLl2~ Ç~,dse, alanine aminotransferase, g~mm~gl~ ., Iyl I ~ .3"~Ç~,~dse, c~lrinm, phnsphorus, glucose, urea nitrogen, ~ Lillille7 uric acid total protein, albumin, globulin, rholesterol and triglycerides. Serum chemistry was det~rmin~od at Sierra Nevada Laboratories (Reno, NV).

~ms~tology and ro ~ ;on parameters. ~em~tology and coagulation p~ were determined 9-11 days prior to the start of the study, just prior to ~flmini~t.oring doses 1 (day 1) and 12 (day 23), five minutes after dosing (coagulation only), 30 minutes and 4 hours following dosing, one day after the 12th dose (day 24), and in recovery monkeys at sacrifice (day 37). The following W O 97/03997 PCT~US96/11786 were iPt~rminp~7 red blood cell count and morphology, total and dirr~lcllLial white blood cells, hemoglobin, h~ms7tocrit, ~loh7llolllbill time, fibrinogen, mean cell hemoglobin, mean corpuscular volume, mean corpuscular hemoglobin c-~l -r~ inn, platelet count, activate partial thromboplastin time, and D-dimer. ~Pms7tr~logy was ir-t~rminPd at Sierra Nevada Laboratories (Reno, NV).

AR177 plasma ~LC analysis. Blood was taken for plasma analysis of AR177 just prior to, and at 5, 30 and 240 minutes following a~7minictrStion of doses 1 and 12. The plasma fraction was obtained by low speed centrifilgS7ti-7n of blood, and stored at -20 C until analyzed for the AR177 col-rr~ 1l l dLion. Plasma AR177 con~çntrations were assayed using an anion-exchange HPLC method on a Waters HPLC system with a 626 pump, 996 photodiode array detector, 717 Snlt~sSlmr71er and Mill~nnillm system software controlled by an NEC Image 466 computer. Buffer A was 0.1 M Tris base, 20% m~thsln~l, pH 12, and Buffer B was 0.1 M Tris base, 1.0 M NaBr, pH 12.anion-exchange column tGen-Pak Fax column) was equilibrated at 80 % buffer A/20 % buffer B for 30 minutes before each HPLC run. Fifty microliters of plasma were analyzed per run. The elusion cnn-7itionc were: a) five-rninute isocratic run at 80% A/20% B. during which the majority of the plasma proteins eluted, b) 30-minute linear gradient to 30% A/70% B during which the AR177 eluted, c) five-minute icorrs7ti~ run at 30% A/70% B. d) one minute linear gradient to 100% B. e) two- minute run at 100% B for initial column cleanup, and fl two-rninute linear gradient to 70%
A/30% B for the initial step in the clean-up method for HPLC column clean-up. The high pH (12) of the elusion buffers was n~c~cssl~y to iiccoCis7t~ AR177 from tissue conctitn~-ntc, which bind AR177 tightly around physiological pH. AR177 is completely stable at pH 12. This method can clearly distinguish between the full length AR177 and n-1, n-2, etc. species, which are potential metabolic products. The W ~7r--tçcfi~-n wavelength was 260 nm. The flow rate was 0.5 rnL/minute in all steps. All runs were performed at room l~"p~ e. Column clean up between runs was ~tlru~ ed by a 500 ~L bolus injection of 0.1 M Tris base, 2 M NaCI, pH 10.5, followed by: a) ten-minute linear gradient to 60% A/40% B. b) one-minute linear gradient to 100% B. c) a three-minute isocratic run at 100% B and d) one-minute linear gradient to 80% A/20% B.
AR177 was spiked into cynnm- l~nc monkey plasma in order to achieve cu~ ~~ ;onc of 0.0635 to 125 ,ug/mL for the standard curve. The plasma st~ndards and unspiked plasma (control) were run on the anion-exchange HPLC column using the above c~m-.7itif7nc. The Waters Millrnninm software was used to ~7~ ",i"~ the area under the peak for each AR177 ~L;llldald at 260 nrn. The HPLC area versus AR177 conrçntration was plotted using Cricket Graph m 1.5.1 sOrLw~c~ There were two HPLC replicate runs per AR177 standard. The areas which l~ s~ d the lowest c~ lalion were at least two times the bac~luulld area at 260 mn. The overall correlation co~ nt of the fitted lines on the standard curve plots was greater than 0.999. There was a linear CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96tll786 concentration versus A260 relationship over a ~ llll 6,500fold range. The variability of the replicates was 1-2%. This method was v~ tt~
.

Necropsy and _istopathology. A complete necropsy was conducted on all monkeys, and included ex~min~tion of the external surface of body (body orifices; dosing site; cranial, nasal, paranasal, thoracic, ~h(l~min~l and pelvic cavities), and the external surface of the brain and spinal cord. The organ weights of the adrenals, epididymies, liver, pituitary, spleen, thyroids, ~dLhyluids7 brain, heart, lungs, prostate, testes, uterus, cervix, kidney, ovaries, seminal vesicles, and thymus were recorded.
A histopathological ~c~e~m-~nt was made of 46 hematoxylin and eosin-stained tissues by a ~ hldl~y pathologist. These in(~ d tissues from the cardiovascular, digestive, respiratory, ulogelliL~l, lymphoid/hematopoietic, skin/musculo~k~let~l and nervous systems, and all major organs.
Results Clinical obst~ lions. No animals died during the course of the study, and there were no effects on body weight. The only LledLI~ ll-related clinical sign was an in-i~lPn-~e of discoloration around the eyes in three high-dose ~nim~l~, which occurred on only one occasion (Day 16 or 18) for two of the animals, and on four consecl~tive days (Days 18-21) in the third animal. The latter monkey also had swelling around the eyes on Day 18. The reaction was transient and was limited to the high-dose group.

ECG, clinical .~ .y, urinalysis and hPm~tology. No abnorrn~liti~s in the ECG recordings were noted, and there were no Ll~dLll~nL-related changes in serum chemistry or urinalysis pdldlll~ i. The only changes in hematology parameters considered possibly LlcdLIll~ L-related were an acute and transient increase in lymphocytes in the high-dose group, and an acute decrease in eosinophils which was seen in all groups, but appeared to be more pronounced in the AR177-treated groups. Both of these changes were observed shortly following dosing on Days 1 and 23 (i.e., those days when clinical pathology was evaluated at several time points post-dosing), but were largely absent on Day 24 (one day after the last dose). The values generally rçm~inPd within the normal range and were not conside}ed indicative of ~ignific~nt toxicity.

N~lv~y and Histo~alllology. No clearly tl~dLIll~llL-related histopathologic changes were seen in any organs or tissues, and no effects on organ weight were evident. Eosinophilic material was seen in a few tubules in the m~clnll~ry area of the kidneys of three monkeys in the high dose group on day 25, but was not seen in the controls or in the recovery animals. Although this may be Ll~aL~ L-related, eosinophilic material can s~ L; " ,~s be observed in the renal tubules of healthy, untreated monkeys.

Plasma AR177 co~ lion. Figure F-1 shows that there was no difference between the AR177 plasma concentrations that were achieved after the first and twelfth (last) doses at either 2.5, 10 or 40 mg/kg. The plasma c~ -"r~ . dLion versus time profile of AR177 is shown in Figure F-2. At the earliest s~mpling time point (five minutes after initi~ti~n of dosing), m~im~l plasma levels of 35.79 + 5.99, 135.43 _ 16.19 and 416.54 + 54.55 ,u/mL were achieved for dose #1 at 2.5, 10 and 40 mg/kg (Figure F-2). At the earliest sampling time point (five minutes after initi~tic-n of dosing), m~im~l plasma levels of 33.98 + 9.98, 113.71 ~t 26.55 and 386.39 + 70.29 ,u/mL were achieved for dose ~12 at 2.5, 10 and 40 mg/kg. The decay kinetics of the 2.5 mg/kg dose appeared to be dirrelell~ than the decay kinetics of the 10 and 40 mg/kg doses after either dose 1 (Figure F-2), although no definite conclusions can be drawn because of the limited number of time points.
No metabolites (i.e. n-1, n-2, etc.) could be observed in the plasma at any time or any dose. This c.,"ri""~ results in rats showing no metabolism of AR177 (Wallace et al., snbmitted)~

C.o~ inn ~ n~ i. Dose-(lep-on~lrnt ~ntiro~gulant activity was ",,",i r~ d at the 10 (Figure F-4) and 40 (Figure F-5) mg/kg doses, whereas there was no ~ntiro~gulant activity following the 2.5 mg/kg dose (Figure F-3). This activity was evident from the prol- ng~ti~-n of activated partial thromboplastin time (aPTT), which reflects a primary effect on the intrinsic coagulation pathway.
Following both the 1st and 12th doses, mean aPTT in the 10 mg/kg group was increased to a~ xi~ lrly twice the pre-dose value by 5 minutes post-dosing, but had returned to baseline levels by four hours. Following both the 1st and 12th doses, mean aPTT in the 40 mg/kg group e~reerlPd the upper limit of the assay five minutes after dosing. By 30 minutes post-dosing, aPTT
values in the 40 mg/kg group had ~lrrlinrd to ~L.~xi".i1lrly 2 to 4-fold above the pre-dose level.
By four hours, the aPTT had returned to the pre-dose levels in all but one monkey.
The rel~tinn~hir between the AR177 plasma concc~llLldLion and aPTT is also shown in Figures F-3, F-4, and F-S for doses 2.5, 10, and 40 mg/kg, respectively. There was a no effect plasma AR177 conrentr~tion versus aPTT of a~lo~ l~L~ly 60-100 ~g AR177/mL, above which there was prol~ ng;~ti- n of aPTT. Doubling of aPTT was observed at plasma AR177 col-r~ Lions of approximately 100-220 ~g AR177/mL. Tripling of aPTT was observed at plasma AR177 concentrations of ~ o~ ly 220-300 ~g AR177/mL, after which no correlation was possible because the aPTT values were off-scale. There was no change in aPTT after the first or twelfth doses of 2.5 mg/kg (Figure F-3), since the AR177 plasma cnnrrntr~tion did not reach the threshold of ~ hll~LLt;ly 60-lOO ,ug/mL. There was a m~im~l two-fold increase in aPTT after the first CA 02227867 l998-0l-26 WO 97/03997 PCTrUS96/11786 or twelfth 10 mg/kg doses (Figure F4). The elimin~ti- n kinetics of AR177 and the return of aPTT
to baseline levels were sirnila{ after the first or twelfth doses.
Di~ ;o These results indicate that AR177, ~hlli~-ixlfLed as bolus intravenous injections up to 40 mg/kg every other day for 12 doses, did not cause mortality, hixtop~thological or cardiovascular events that have been d~scrihed for other oligonucleotides (Galbraith et al., 1994; Slhliv~sall and Iversen, 1995). The only xignifir~nt change that was observed was a prolongation of aPTT, which was reversible. To our knowledge, this is the first oligonucleotide that has not been observed to cause liver and kidney toxicity following intravenous a lmini~tration.
The structure of AR177 may collLlil)uL~ to its lack of general toxicity. AR177 contains only two phosphorothioate bonds at the 3 ' and 5' termini. These phosphorothioate bonds were .lf ~ignf d to help prevent endonnrl~xe-induced cleavage of AR177. We xpeclll~t~ that the small number of sulfurs may have reduced the pLu~en~iLy to bind to proteins, a phrnomenon that has been observed for full phosphorothioates, which has been specnl~ted to cause toxicity (Slhliv~s~l and Iversen, l99S). AR177's three-~imfncion~l shape may also contribute to its lack of toxicity. AR177 has been shown to form a structure in which hydlog~ll bonds form between deoxyg~ n-~xinP residues to create a "(~-tetrad" (Rando et al., l99S). This tetrad structure imparts a c~ L shape which makes it l~ Li~lL to ~lfgra~l~tinn (Bishop et al., 1996) and may make it relatively non-toxic by ",i";",i,i,~p reactive sites. The rçci~t~nre to degradation has been noted in single and repeat dose rh~3rm~rokinf tirs studies in rodents (Wallace et al., s~hmitted)~ and in a more complete rh~rm~rokinrtir study in cynomolgus monkeys which showed a terminal plasma half-life of greater than 24 hours (data not shown).
The results of the AR177 plasma analysis ~ifm~n~trated that there was no dirr~ience between the AR177 plasma co,-r~ t)ns that were achieved after the first or twelfth (last) doses of 2.5, 10 or 40 mg/kg. These results can be hlhl~lchd to mean that AR177 does not induce metabolic ~ yllles that would, if they were in-lllcerl, reduce the c-nrf ntration of AR177 by increasing its metabolism. This has the important implication that repeat doses of AR177, at least when given every other day for 23 days, will not result in rh~rm~rokinetic tolrr~nre The results of the AR177 plasma analysis dr~ llxll~L~d that there was a close rel~fionxhir between the AR177 plasma cullcellLl~Lion and aPTT. There was a no effect plasma AR177 concentration versus aPTT of appro~im~tçly 60-100 ,ug AR177/mL, above which there was prolongation of aPTT. The ability to prolong coagulation has been noted to be a feature of other oligonucleotides (Bock et al., 1992; Henry et al., 1994). An oligonucleotide composed of deul~y~ u~ nxin~os and thymidines has been dçsrrihe~l that binds to Lhlollll~ill (Paborsky et al., 1993), d~lllt.ll~l.aLes sequence dependent inhibition of coagulation in vitro (Bock et al. 1992), has a _ CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 G-tetrad structure (Wang et al., 1993), and is active as a short acting anticoagulant in vivo (Griffin et al., 1993; DeAnda et al., 1994) The structure of t'ne oligonucleotide, (GGTTGGTGTGGTTGG) bears some resPrnhl~nre to AR177 (GTGGTGGGTGGGTGGGT). Both oligonucleotides formG-tetrad structures. A comparison of the ~ntiro~gulant properties of these oligonucleotides in~lir.~tPS t that the oligonucleotide is approximately 10-100 times more potent than AR177. An ~ ",i"~lion of the anti-HIV properties of the oligonucleotide showed that it had little or no anti-HIV activity (unpublished data). Thus, although both oligonucleotides are composed of deo~y~ in~s and thymidine, and form G-tetrads, they have distinct biological properties.
In conr~ ion~ mini~fration of up to 40 mg/kg of AR177 to cynomolgus monkeys by bolus hlLldvellous injection every other day for 23 days was well toler~tP~l No mortality or clinical signs of ~ignific~nt toxicity occurred. The most salient ~ltPr~ti~n in clinical pathology ~
was the prolongation of aPTT in the 10 and 40 mg/kg groups, which reflects inhibition of the intrinsic coagulation p~lhw~y. The approximate (loubling of aPTT observed in the middle-dose group (10 mg/kg) is considered to be marginally clinically ~ignifir~nt following bolus intravenous injection. The severe inhibition of coagulation in the 40 mg/kg group may not be dose limiting since aPTT values had returned to baseline levels fours hours following dosing. It is probable that prolongation of aPTT at these doses could be cil.;ulllv~ d by ~.l."i"i~. i .g AR177 as a slow infusion over the course of several hours in order to stay below the threshold for antir-o~gulation7 which was established to be 60-100 ~g/mL of AR177. The absence of clinical pathology abnorm~liriPs or tissue histop~th- logy at even the highest dose (40 mg/kg) after repeated intravenous ~-lmini~tration suggests that there is little potential for cnmlll~tive toxicity with T31077 with any type of ~rlmini~tration.
***

CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 ~ ~ E-. ~ ~ :~
~ _ 3 o ~, ~ C
o E ~ 3 -", c r ~ ~ o O O ~ A ~ ~
3 E ~ ~ ~ ~ ~ - '~
E
~ 3 - c W c~ ~ _ ~ 2 _ 'C ~L, ~ g ~ ~o C

~: a o ~ o ~ c O E o r- ~ ~3 9 ~, ~ c~,o O ~ _ ~ o ~ _ ~1 E
~ ¢ ~: ~ r,~ ~ c O

o -- ~ ~ ~ _E ~

~n o u~

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 V~
~ V

o ~ cn o _ ~ --.

'-- F
O
a ~ +1 5 -~~
~ ~" ._ _ i_ --O o ---:~+l * * +l ~ '' a ~ ~ ~
~i ,., C4 .~

O O O O .
+l +l+l +l 5 a ~ ~ ~ ~ 3 ~ : 3 o o o o C~5 --;~+l +l+l +l c~ a-- ~ ~ ~

~,0 ~0 +l +l +l +l ~
a ~ e +' ~ O O O O ~
~'+l +l +l +1 ~' ~ ~ ~ ~ ~n &

O
O O O O
+l +l +l +l ~ O -- ~ ~
3 ~a C C~ O
3 ~- ~ o o - ~ c V ~ _ :~
C~
~.

CA 02227867 l998-0l-26 o ~ ~ oO
~, +l * * +l +l * * +l +l O ~ r, ~ O

_~ -- o ~,. o ~~ V~ , ~ 'D ~ E o +l +l +l +l +l +l +l +l +l a ~ ~ O ~ .~O O
~ _ ~t ~ ~ O
O ~
C

O CO ~ ~
o O
+l +l +l +l +l +l +l +l +l O ~ O ~ oo O t_ V~
E

~ ~ ~ _ ~J~ _ V- o o ~
~Q
~t oo o ~ +l * * +l +l * * +l +l a o ~ O O

1, +1 +1+1 +1 _ +1 +1 +1 +1 --~ +l ~ o ~ c ~
~ ~ o ~ o o _ ~ ~ o -- +l +l +l' +l +l +l +l +l -- oo o ~ -- -- o c~ o . - ~ o o CA 02227867 l998-0l-26 W~) 97M~997 = P~TLLL~Qfi/l I 7~fi W O 97/03997 PCT~US96/11786 C
o U~ o '~c _ o _ o +l +l +1 ~ 3 -- C
o o o -,_ u~ ~ ~ ~e o _ C
o o o .~
+l +l +l _ C~
~~ ,_ t-- -- C
-~ ~~t ~ ~o o * * ~ U~

C
~t C

+l +l +l -- ~ , ' o ~ ~ ~ ~ 3 +l +l +I c~
O -- ' Q>
o~ ~D ' E--oo o ~,.
C ~
V~ +l CA 02227867 l998-0l-26 W O 97/03997 PCTtUS96tl1786 o ~ ~ -H * * ~ -H * * -H ~ # *
~ ~ _ '~ o -~ " ~ ~ ~ ~' -- ~o o C~ -- S _ C~ o C U~ o~

~, ~ ~ ~ o, o _ o, o~ - ~ o ~ oc, ~ ~ o o, ~ oo o~
C.

. --' _ o. ~o X o. C' ~ ~ o ~o '' o = ~ ~ ~ +, c~ ~ E ~ ~ ~ -- E o ~, _ V
o ~ ~ ~~. ~ ~. ~, o o ~

- - - - o o o v c ~ _t-- -- c ~_ ~r ~ ~ ~ O g 3 ~ ~ ~ - o ~ ~ ~ ~ ~ ~ X
~ _ o O

LL
D
~ ~ O ~ ~ ~ X ~~C
~_ ~ ~ ~ ~ -- ~ ~ ~ O o O
~, , . . _ o o ,~ W C~ oo -- ~ ~ ~

X)~ '-- ~ ~ ~ ~,, ~ o ~ ~
-- O ~ , O ~ ~
-- o o o -~'~~ ~ e~r --'n O ~0 W097/03997 PCT~US96/11786 E

..
x ~ x o = c"
~ ~ ~ c n o ~ ~~ _ .

~, x ~o ~ x O x 1-- x x o o ~ ~.~ er ~ , ~ 'D Xu~ X
o ~ ~ o o +~ ~ -H -H --o c~ ~~.~ ~ ._ ~ ~ ,c O l_ _ ~ x t l -H -H -H -H ~
~ n ~ E ~
o L~ 'O
Oc~ r-- O ~~ ~ n O~L X C~
+l ~ ~ c ~ n o o ~ ~ ~.
o~ O

r- ~~ x ~ ~_ ;
O o ~~' ~ ~ _.
-H -H -H -H -H ~
o ~ ~ ~ c .
O

O x _ o ~ ~ ~ ~
-~ E
o ._ ~ o o x er = ~ _ ~ . O

-- --W O 97/03997 PCT~US96/11786 o ~
-- o o o o ~
+, * * +~ +, * * +, +
o o o C C'~ ~ ~ _ X
o o o o o o o o -- ~
:.~ +l +l +l +l +, +, +, +, +l +
2 X ~ x ~
oooo ----C.~

~ , o o ,-- , o o o o o o o o ~_ +l +l +l +l +l +l +l +l +l +l E ~ ~ ~ ~ o o o o ~, ~ .'-, _ o o e +, +, +, +, _ o o E ~ ~ ~ ~ - o o o o _ ~ ~
~" _ o o -~ o o o o ; o o o o ~ ~ ~
+~ +~ +~ +, +, +l +l +l +l +l ~D ~ u~ ~ ~ ~ ~ ~~ ~ ~
-~~ ~oo ~ooo C +l +l +l +l +l +l +l +l +l +l U~ ~ ~ ~ X ~ o ~ o o o o V~

Q~
a ~ x ~o ~ ~ x ~' ~ '~

O O O O O O O O O
+l +l +l +l +l +l +l +l +l +l -O _ V~ o o _ ~ o ~ _ V~

W097/03997 PCT~US96/11786 o *

r o. ~o o _ ~oC~ ~ o o o C,, ~.
-o~ ~ o, ~,~, ~, ,,, ",, ~, _ _ oo .
~ _ ~ ~ C

-~o _ o, o _ ,., ~ ~ ~ .=
o o~ =
o o o~
.

o ~ ~ ~ ~ ~ ~ o o _", ~ ~ o. ~
~ ~ ~ ~ ~ ~ o ~n o -- ~ ~ ~ ~ E ~
~ ~ ~ ~ ~ +i o "
_ ~ ~ ~ ~ o c ' o _ ~ ~ O~ _ ~ ~ c O
o~--E
er _ V~ O ~r U ~
D E
y o CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 G. Human Clinical Trials Four HIV-infected patients/group were dosed with AR177 at 0.75 mg/kg and 1/5 mg/kg, and two HIV-infected patients were dosed so far with AR177 at 3.0 mg/kg by intravenous infusion over two hours.
s Methods Blood was collected in EDTAp coated tubes at 0.25, 0.5, 1, 2, 2.05, 2.5, 3, 3.5, 4, 6, 8, 11, 14, 26, 48, 98, and 122 hours following initi~tinn of drug ~.1 " ,i "i~l . alion. Plasma was obtained by low speed centrifugation of the blood, and was stored frozen until analyzed by HPLC for AR177 c.J~Irr~ dlion. The conrçntration of AR177 was ~1~t~rrnin~d in patient plasma using a validated anion-exchange HPLC method at the Division of Clinical Pharmacy of the Univ~ ,ily of California, San Francisco. This method has a limit of qu~ntit~tion of 15 ng/mL in human plasma.

Pharmacokinetic analvsis ph~rmzlr-~kint-tic pa.dlllel~l~ were c~lrnl~t~od using PKAnalyst software (MicroMath, Salt Lake City, UT). The ph~rm~rokin~tir data best fit a two Cvlll~dlllllclll model for all of the patients. The alpha and beta half-lives were almost i~l~ntir~l in each of the patients, based on the software hllt~ ldLion of the AR177 plasma cul~r~ lion versus time plot (Figures G14). For this reason, only one half-life is reported. (Note that in monkeys, a third half-life of d~lv~hlldL~ly 24 hours was observed at a dose of 5 mg/kg given as an intravenous infusion over two hours. A
third half-life was not evident in human data, except perhaps for patient #10.) For each ph~rm~rf~kin~tic paldlllt:L~l, the mean _s.d. of n=4 was r~lrnl~tr-l for the 0.75 and 1.5 mg/kg groups and the mean +s.d. of n=2 was r.~lrl-l~trd for the 3.0 mg/kg group.

Results The plasma concontrations of AR177 following intravenous infil~ion are shown in Figure G-1 (0.75 mg/kg), Figure G-2 (1.5 mg/kg), Figure G-3 (3.0 mg/kg) and Figure G4 (all doses).
Analysis of this data indicate that the plasma pl.,., .,.~rnkin.otirs of AR177 are not directly plv~vlLional to the dose (Table G-1). The increase in the CmaX and AUC were pLu~ulLionally much greater than the increase in the dose from 0.75 to 3.0 mg/kg. The increase in the Cmax and AUC
were much greater than the increase in the dose. The CmaX value in the 0.75 mg/kg group was 5.1 _ 1.4 ~g/mL and the Cma,~ value in the 3.0 mg/kg group was 37.5 _ 0.1 ,Ibg/mL, d~ nxi~ ly a seven-fold increase (Figure G-5). The AUC value in the 0.75 mg/kg group was 703.6 _ 154.7 ~g-min/mL and the AUC value in the 3.0 mg/kg group was 8,277.8 _ 2.937.4 ~g-min/mL, dWl~Xilll~ ly a 12-fold increase (Table 1).

The plasma clearance and Vd values reflected the Cm~ and AUC data. The plasma clearance in the 0.75 mg/kg group was 1.1 + 0.2 mL/min/kg and the clearance in the 3.0 mg/kg group was 0.4 + 0.2 mL/min/kg, appro~im~t~ly a 65 % decrease (Figure 6) . the initial and steady-state volumes of (ii.ctrihntion in the 0.75 mg/kg group were 0.16 + 0.05 L/kg and 0.14 + O.OS v L/kg, lespe~;liv~ly, whereas the initial and steady-state volumes of distribution (Vd) in the 3.0 mg/kg group were 0.08 + 0.00 L/kg and 0.05 + 0.03 L/kg, respectively (Table l).
In a~ lllGl-l with the above data, the plasma half-life in the 0.75 mg/kg group was 28.0 + 12.7 minutes, and the half-life in the 3.0 mg/kg group was 120.1 + 60.7 minutes, approximately a four fold increase (Figure 5).

Cnn~ lciqn~
These results indicate that the plasma ph~rm~r~kin~fi~s of AR177 are non-linear and suggest that there is a saturable m--~ h~nicm for the elimin~rion of the drug.

CA 02227867 l998-0l-26 W O 97/03997 PCT~US96/11786 ~ _, ~ ~ _ o "' o ''' o +l +l +l +l +l +l +l +l +l +l +l CO o _ ~ ~ ~ ~ o U') C~ C~ _ 1-- ~ o ~ o ~ C~ ~
o ~ ~ o -- oo o o o o o . '' ~

o o ~ o ~ cr~
C~ o _ ~ ~ ~"
~,, ~ o o ~ ~ ~ ~ U~ o _ o ,~
¢ ~
~s ~ oo o -- ~ o o a~ ~o ~ _ ~ ~ o ~
-~ o ~:~

~,5 ~ -- a~ c~ ~ C~ --~ O O ~ O ~ ~
O ~ oo ~ ~ oo ~ O -- o cr.

E ~ R y _ S E

PAGE I~TENTIONALLY LEFT BLANK

CA 02227867 l998-0l-26 o _ ~ ~o ~ -- ~ o ~ ~ ~
+l +l +l +j +l +l+l +l +l +l +l ~ ~t ~, o~ _ _ o o~

ooU~~D ~t ~ ~ ~ V~ ~ OO _C~ O
O _ ~ O v~ et ~ 00 0 . . . . 00 -- C~ ~ ~ oo o oo o ~ _ ~ _ ._ ~

~ ~ ~ ~ o ~ ~ oo o ~ ~ ~ o~
'-- O _ O ~ -- ~~~D O ~
_ _ ~ C~ U~ ~ o CO o oo _ o V ~

. -- ~ O ~ oo ~0 0 c~ ~D O
O ~ O

O ~ O Cr~ --. ~ ~ O ~ ~ ~ O
~ ~ -- ~
,~ _ _ '~

E c ~ c ;~ E
~ x ~ ~ _, ~a-2 G ~ ~ ~ ¢ E~ V ~ > ¢

CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 PAGE INTENTIONALLY LEFT BLANK

_ CA 02227867 l998-0l-26 WO 97/03997 PCT~US96/11786 C" o ~ o ~ o o ~o ~ ~~ ~
+l +l +l +l +l +l +l +l +l +l +l o ~' ~ ~, o ~ ~ ~
oo .~

~ ~ ~ ~ ~ ~ ~ ~ ~o ~ ~ ~ ~ oo _ ~ tn ~ ~ ~D O ~ O
C>

~ ~ ~ ~ ~ ~ ~ o o, o ~, .

,~

~o E C ~ 3 E

m V

CA 02227867 l998-0l-26 P ~ ~ S 9 6 / 1 I ? ~
lPE~IS 14 OCT i~7 SEQUENCE LISTING
(1) GENERAL INFORMATION:
(i) APPLICANT: Rando, Robert F.
Fennewald, Susan Zendegui, Joseph G.
Ojwang, Joshua O.
Hogan, Michael E.
Pommier, Eyves Mazumder, Abhijit (ii) TITLE OF INVENTION: Anti-Viral Guanosine-Rich Oligonucleotides (iii) NUMBER OF SEQUENCES: 87 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: Conley, Rose & Tayon, P.C.
(B) STREET: 600 Travis, Suite 1850 (C) CITY: Houston (D) STATE: Texas (E) COUNTRY: U.S.A.
(F) ZIP: 77002-2912 ~' (v) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release #1.0, Version #1.25 (vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER: PCT/US96/11786 (B) FILING DATE: 17-JULY-1996 (C) CLASSIFICATION:
(vii) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US 08/535,168; 60/001,505; 60/014,007;
60/013,688; 60/015,714; 60/016,271 (B) FILING DATE: 23-OCT-95; 17-JULY-96; 25-MARCH-96; 19-MARCH-96;
23-APRIL-96; 17-APRIL-96 ..~
(viii) ALLOR~Y/AGENT INFORMATION:
(A) NAME: McDaniel, C. Steven (B) REGISTRATION NUMBER: 33,962 (C) REFERENCE/DOCKET NUMBER: 1472-06215 (ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: 713/238-8010 (B) TELEFAX: 713/238-8008 (2) INFORMATION FOR SEQ ID NO:1:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 38 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
= 134 AM~ED S~OET

CA 02227867 l998-0l-26 PCT~S 9 6 lPE~J~ 14 OCT'9 (A) NAME/KEY: misc_~eature (B) LOCATION: 38 (D) OTH3R INFORMATION: /note= "Amine moiety attached to 3' end"
(XI) SEQUENCE DESCRIPTION: SEQ ID NO:1:

(2) INFORMATION FOR SEQ ID NO:2:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 38 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

(2) INFORMATION FOR SEQ ID NO:3:
~Qu~ CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

(2) INFORMATION FOR SEQ ID NO:4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 38 base pairs ---. (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

(2) INFORMATION FOR SEQ ID NO:5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 36 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

A~D 9ffET

CA 02227867 1998-01-26 ~ ~ S 9 6 ~ 1 ~ 7 86 lPEAIUS'~ f 4 QCT '~7 (2) INFORMATION FOR SEQ ID NO:6:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 36 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:
GTGGTTGGTG GTG~L~l~lG GGTTTGGGGT GGGGGG 36 (2) INFORMATION FOR SEQ ID NO:7:
(i) S~QU~N~ CHARACTERISTICS:
(A) LENGTH: 36 base pairs (B) TYPE: nucleic acid _~ (C) STRANDEDNESS: single .3 (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 36 (D) OTHER INFORMATION: /note= "phosphorothioate backbone"
(xi) S~YU~N~ DESCRIPTION: SEQ ID NO:7:

(2) INFORMATION FOR SEQ ID NO:8:
(i) ~QU~N~ CHARACTERISTICS:
.- (A) LENGTH: 36 base pairs ,. (B) TYPE: nucleic acid ~ (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 36 (D) OTHER INFORMATION: /note= "phosphorothioate backbone"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:8:
GTGGTTGGTG GTG~l~l~lG GGTTTGGGGT GGGGGG 36 (2) INFORMATION FOR SEQ ID NO:9:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 31 base pairs ~ CA 02227867 1998-01-26 P C T ~ S 9 6 / 1 1 7 8 6 ~pE~JS ~ 4 OCT '97 (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:

(2) INFORMATION FOR SEQ ID NO:10:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) sTR~Nn~nN~ss single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:10:
,_~ GGTGGTTGGG GGGTGGGGGG G 21 -~ (2) INFORMATION FOR SEQ ID NO:11:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRA~ ~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (Xi) ~U~N~ DESCRIPTION: SEQ ID NO:11:

(2) INFORMATION FOR SEQ ID NO:12:
.- (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 30 base pairs ~~ (B) TYPE: nucleic acid (C) STR~Nn~nN~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:12:
GGTGGGTGGT ll~l~lGGTT GGTGGGTTTT 30 (2) INFORMATION FOR SEQ ID NO:13:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 31 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) AM~9~0 9~

~ CA 02227867 1998-01-26 P ~ T ~ S 9 6 / 1 I 7 8 6 T'97 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:13:

(2) INFORMATION FOR SEQ ID NO:14:
(i) S~Qu~N~ CHARACTERISTICS:
(A) LENGTH: 27 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:14:

(2) INFORMATION FOR SEQ ID NO:15:
(i) ~U~N~ CHARACTERISTICS:
_ (A) LENGTH: 29 base pairs (B) TYPE: nucleic acid ~- (C) STR~NnRn~R~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) ~Qu~N~ DESCRIPTION: SEQ ID NO:15:
TGGG~LllGG GTGGGGGGTT GGGTGGTTG 29 (2) INFORMATION FOR SEQ ID NO:16:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 24 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) ~Qu~N~ DESCRIPTION: SEQ ID NO:16:
GGGTGGTGGT ~-L-L~-l~-L-lG TGTG 24 (2) INFORMATION FOR SEQ ID NO:17:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 22 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:17:

CA 02227867 l998-0l-26 p ~ ~ ~ 9 6 / 1 1 78 - IPEAIUS 1 4 O~T '97 (2) INFORMATION FOR SEQ ID NO:18:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 26 base pairs (Br TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) ~Qu~ DESCRIPTION: SEQ ID NO:18:

(2) INFORMATION FOR SEQ ID NO:19:
(i) ~QU~N~ CHARACTERISTICS:
(A) LENGTH: 26 base pairs (B) TYPE: nucleic acid (C) STRPNn~n~.~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) -~ (xi) SEQUENCE DESCRIPTION: SEQ ID NO:19:

(2) INFORMATION FOR SEQ ID NO:20:
(i) ~QU~N~ CHARACTERISTICS:
(A) LENGTH: 26 base pairs (B) TYPE: nucleic acid (C) STRAN~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature . (B) LOCATION: 26 . (D) OTHER INFORMATION: /note= "Amine moiety a~tached to 3' end"
(xi) ~yu~N~ DESCRIPTION: SEQ ID NO:20:

(2) INFORMATION FOR SEQ ID NO:21:
U~N~ CHARACTERISTICS:
(A) LENGTH: 45 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 45 ~AMEN~6D 9HE~

CA 02227867 l998-0l-26 PCT~ 96/11 786 ~S 14 OCT'97 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end"
(xi) S~Qu~N~ DESCRIPTION: SEQ ID NO:21:

(2) INFORMATION FOR SEQ ID NO:22:
( i ) S~U~N~ CHARACTERISTICS:
(A) LENGTH: 45 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 45 (D) OTHER INFORMATION: /note= "cholesterol moiety attached to 3' end"
,_.
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:22:
,_ _ (2) INFORMATION FOR SEQ ID NO:23:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 45 base pairs (B) TYPE: nucleic acid (C) STR~n~n~RSS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEAluKE:
(A) NAME/KEY: misc_feature (B) LOCATION: 45 - (D) OTHER INFORMATION: /note= "cholesterol moiety . attached to 3' end"
(xi) ~Qu~N~ DESCRIPTION: SEQ ID NO:23:

(2) INFORMATION FOR SEQ ID NO:24:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 45 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc feature (B) LOCATION: 45 (D) OTHER INFORMATION: /note= "Amine moiety ~ED 9HE~

CA 02227867 l998-0l-26 PCT/U~ 9 6 ~ I ~ 7~a pE~JS 14 O~T 'g~
attached to 3' end"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:24:

(2) INFORMATION FOR SEQ ID NO:25:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 26 base pairs (B) TYPE: nucleic acid (C) STR~Nn~N~-~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 26 (D) OTHER INFORMATION: /note= "cholesterol moiety attached to 3' end~
~ (Xi) ~VU~N~ DESCRIPTION: SEQ ID NO:25:
~~ GTTGGGGGTT GTTGGTGGGG TGGTGG26 (2) INFORMATION FOR SEQ ID NO:26:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 45 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 45 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end"
; (xi) SEQUENCE DESCRIPTION- SEQ ID NO:26:

(2) INFORMATION FOR SEQ ID NO:27:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 26 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 26 (D) OTHER INFORMATION: /note= I'phosphorothioate backbone and amine moiety attached to backbone"

AMEN~E{~ SHEET

~ CA 02227867 l998-0l-26 PCT/U~ 96~1786 \PEP'IU~ 14 0CI '9~
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:27:
GTTGGGGGTT ~llG~lGGGG TGGTGG 26 (2) INFORMATION FOR SEQ ID NO:28:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 26 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:28:
GGTGGTGGGG TG~ll~LlGG GGGTTG 26 (2) INFORMATION FOR SEQ ID NO:29:
(i) SEQUENCE CHARACTERISTICS:
.~ (A) LENGTH: 45 base pairs (B) TYPE: nucleic acid (C) STR~Nn~nN~S single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:29:
GGTGGTGGGG TG~ll~llGG GG~ll~llGG GG~l~l~lGG GTGGT 45 (2) INFORMATION FOR SEQ ID NO:30:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STR~n~n~-~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) S~U~N~ DESCRIPTION: SEQ ID NO:30:

(2) INFORMATION FOR SEQ ID NO:31:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 18 (D) OTHER INFORMATION: /note= "Amine moiety 1~2 .

CA 02227867 l998-0l-26 ~ CT/U~ 6 ~ ~ 78~
~p~ S 14 OCT'97 attached to 3' end"
(xi) ~QD~N~ DESCRIPTION: SEQ ID NO:31:
GG~lGG~lGG GTGGGTGG 18 (2) INFORMATION FOR SEQ ID NO:32:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 18 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end and phosphothioate backbone"
(Xi) S~U~N~ DESCRIPTION: SEQ ID NO:32:
GG~lGG~lGG GTGGGTGG 18 (2) INFORMATION FOR SEQ ID NO:33:
(i) S~U~N~'~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 17 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end"
(Xi) ~:QU~N~ DESCRIPTION: SEQ ID NO:33:

(2) INFORMATION FOR SEQ ID NO:34:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 27 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 27 (D) OTHER INFORMATION: /note= "Amine moiety =
~ CA 02227867 l998-0l-26 P~S 9 6/I 1 ?8h s ~40~r~q~
attached to 3' end"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:34:

(2) INFORMATION FOR SEQ ID NO:35:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 37 base pairs (B) TYPE: nucleic acid (C) STR~Nn~nN~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 37 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:35:
- ~ GTGGTGGGTG GGTGGGTGGT GGGTGGTGGT TGTGGGT 37 (2) INFORMATION FOR SEQ ID NO:36:
(i) ~Qu~N~ CHARACTERISTICS:
(A) LENGTH: 16 base pairs (B) TYPE: nucleic acid (C) STRAN~N~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 16 (D) OTHER INFORMATION: /note= "Amine moiety --. attached to 3' end~
~ (xi) SEQUENCE DESCRIPTION: SEQ ID NO:36:

(2) INFORMATION FOR SEQ ID NO:37:
(i) ~yU~N~ CHARACTERISTICS:
(A) LENGTH: 29 base pairs (B) TYPE: nucleic acid (C) sTR~Nn~nN~s single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 29 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' endll NaED 9H~r -CA 02227867 l998-0l-26 PCT~ 9 6 ~ 1 1 78 ~p-~S 14 OCT'9 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:37:

(2) INFORMATION FOR SEQ ID NO:38:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 38 base pairs (B) TYPE: nucleic acid (C) STR~NDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 38 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:38:
. GTGGGTGGGT GGTGGGTGGT GGTTGTGGGT GGGTGGTG 38 (2) INFORMATION FOR SEQ ID NO:39:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 45 base pairs (B) TYPE: nucleic acid (C) STR~Nn~nN~.~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 45 (D) OTHER INFORMATION: /note= "phosphorothioate backbone and amine moiety attached to 3' end"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:39:

(2) INFORMATION FOR SEQ ID NO:40:
( i ) S~U~N~ CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) sTRANn~nN~s: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 18 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end"

CA 02227867 l998-0l-26 ~ ~ t~
\Q~ S 1 4 0CT '97 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:40:

(2) INFORMATrON FOR SEQ ID NO:41:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 18 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end and phosphorothioate backbone"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:41:
.~

(2) INFORMATION FOR SEQ ID NO:42:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEA~l~u~E:
(A~ NAME/KEY: misc_~eature (B) LOCATION: 18 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end"
_ (xi) S~Qu~NC~ DESCRIPTION: SEQ ID NO:42:

(2) lN~-O~TION FOR SEQ ID NO:43:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 18 (D) OTHER INFORMATION: /note= "Amine moiety attached to 3' end and phosphorothioate backbone"

-- r CA 02227867 l998-0l-26 PCT~S 96/1 1 786 ~p~S 14 OCT'97 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:43:

(2) INFORMATION FOR SEQ ID NO:44:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 47 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:44:
TTCATTTGGG A~ACCCTTGG AACCTGACTG ACTGGCCGTC GTTTTAC 47 (2) INFORMATION FOR SEQ ID NO:45:
(i) SEQUENCE CHARACTERISTICS:
-~ (A) LENGTH: 15 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:45:

(2) INFORMATION FOR SEQ ID NO:46:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear - (ii) MOLECULE TYPE: DNA (genomic) (Xi) S~U~N~ DESCRIPTION: SEQ ID NO:46:

(2) INFORMATION FOR SEQ ID NO:47:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 16 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:47:

~Iv1EN~lED SHEE~

~ CA 02227867 l998-0l-26 ~r~$ 9 6 / l 1 7 8 6 ~p~S 14 OCT'9i (2) INFORMATION FOR SEQ ID NO:48:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 16 base pairs (B)- TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) ~Qu~N~ DESCRIPTION: SEQ ID NO:48:

(2) INFORMATION FOR SEQ ID NO:49:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 13 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear ~~ (ii) MOLECULE TYPE: DNA (genomic) (xi) ~yu~N~ DESCRIPTION: SEQ ID NO:49:

(2) INFORMATION FOR SEQ ID NO:50:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) ~yu~N~ DESCRIPTION: SEQ ID NO:50:
.-~t_ GTGGTGGGT g (2) INFORMATION FOR SEQ ID NO:51:
(i) ~yU~N~ CHARACTERISTICS:
(A) LENGTH: 14 base pairs (B) TYPE: nucleic acid (C) STRAN~N~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:51:

(2) INFORMATION FOR SEQ ID NO:52:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: lO base pairs (B) TYPE: nucleic acid AM~~

CA 02227867 l998-0l-26 P ~ ~ ~ S 9 6 / 1 1 7 8 6 4 a~r ~s7 (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:52:
GTGGGTGGGT lO

(2) INFORMATION FOR SEQ ID NO:53:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 15 ba~e pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:53:
~~ GGTTGGTGTG GTTGG 15 ~ _, (2) INFORMATION FOR SEQ ID NO:54:
(i) ~u~N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STR~Nn~nNE.qS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:54:

(2) INFORMATION FOR SEQ ID NO:55:
.,' (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) sTR~Nn~nN~qs single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:55:

(2) INFORMATION FOR SEQ ID NO:56:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear AMENDED Sl/~T

CA 02227867 l998-0l-26 ~ 9 6 / 1 ~ 786 ~o~\~ 14 OCT'97 (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:56:

(2) INFORMATION FOR SEQ ID NO:57:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTE: 18 base pairs (B) TYPE: nucleic acid (C) STR~NnRnNE~S single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:57:

(2) INFORMATION FOR SEQ ID NO:58:
~Qu~:N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STR~NnRnNE.~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEAluKE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 11 (D) OTHER INFORMATION: /note= "the base is removed ~rom this nucleoside"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:58:

..-~
J (2) INFORMATION FOR SEQ ID NO:59:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STR~NnRn~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: RNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:59:

(2) INFORMATION FOR SEQ ID NO:60:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 16 base pairs (B) TYPE: nucleic acid CA 02227867 l998-0l-26 ~T~ 96/1178 14 ~CT '-9 (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 2 (D) OTHER INFORMATION: /note= "the base is removed ~rom this nucleosidell (xi) SEQUENCE DESCRIPTION: SEQ ID NO:60:

(2) INFORMATION FOR SEQ ID NO:61:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear - (ii) MOLECULE TYPE: DNA (genomic) (Xi) ~U~N~ DESCRIPTION: SEQ ID NO:61:
~LGG~LGGTG GGTGGGT 17 (2) INFORMATION FOR SEQ ID NO:62:
( i ) ~QU~N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STR~Nn~nNR-~S: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:62:
~s,'~.
~ GTGGTGGGGT GGTGGGT 17 (2) INFORMATION FOR SEQ ID NO:63:
(i) ~U~N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:63:

(2) INFORMATION FOR SEQ ID NO:64:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs AMEW~ED 9Hffl CA 02227867 l998-0l-26 PCT~ 96/1~ 786 IPE~S ~ Jl ~T 'g7 ~A/~ g7 (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6g:

(2) INFORMATION FOR SEQ ID NO:65:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEAluKE:
(A) NAME/KEY: misc_~eature ~~~. (B) LOCATION: 5 (D) OTHER INFORMATION: /note= "C-5 propynl dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:65:

(2) INFORMATION FOR SEQ ID NO:66:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) sTR~Nn~nN~s: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
~ (A) NAME/KEY: misc_~eature _, (B) LOCATION: 13 (D) OTHER INFORMATION: /note= "C-5 propynl dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:66:

(2) INFORMATION FOR SEQ ID NO:67:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 5,13 AMENDED St~E~

CA 02227867 l998-0l-26 P ~ ~ ~ 9 6 / 1 ~ 7 8 6 IPEA/Us 14~GT'-47 (D) OTHER INFORMATION: /note= "C-5 propynl dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:67:

(2) INFORMATION FOR SEQ ID NO:68:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 16 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 2 (D) OTHER INFORMATION: /note= "the base is removed ~rom this nucleoside"
(ix) FEATURE:
s (A) NAME/KEY: misc_feature (B) LOCATION: 5,13 (D) OTHER INFORMATION: /note= "C-5 propynl dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:68:

(2) INFORMATION FOR SEQ ID NO:69:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 16 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) _- (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 2 (D) OTHER INFORMATION: /note= "the base is removed ~rom this nucleoside"
(ix) FEA~l~u~E:
(A) NAME/KEY: misc ~eature (B) LOCATION: 6,13 (D) OTHER INFORMATION: /note= "C-5 propynl dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:68:

(2) INFORMATION FOR SEQ ID NO:70 (i) SEQUENCE CHARACTERISTICS:

AN~

CA 02227867 l998-0l-26 ~r~v~ 9 6 / l ~ 78 6 1 4 0~J '97 (A) LENGTH: 16 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 2 (D) OTHER INFORMATION: /note= "the base is removed from this nucleoside"
(ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 1,5,6,9,10,13,14,17 (D) OTHER INFORMATION: /note= "deoxyinosine"
(xi) S~Qu~N~ DESCRIPTION: SEQ ID NO:70:

~~ (2) INFORMATION FOR SEQ ID NO:71:
,.
(i) ~h~U~N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 6,13 (D) OTHER INFORMATION: /note= "C-5 propynl dU"
(Xi) S~OU~N~ DESCRIPTION: SEQ ID NO:71:

., ~.
(2) INFORMATION FOR SEQ ID NO:72:
(i) S~QU~N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 13 (D) OTHER INFORMATION: /note= "3' cholesterol via triglycyl linker"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:72:

GTGGTGGGTG GGTGGGT j 17 AMEN~EO ~1~
-CA 02227867 l998-0l-26 ~T~ 9 b / 1 } ? 8 6 IPE~u~ 14 o,GT'97 (2) INFORMATION FOR SEQ ID NO:73:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc feature (B) LOCATION: 13 (D) OTHER INFORMATION: /note= "5-bromo dU"
(xij SEQUENCE DESCRIPTION: SEQ ID NO:73:

(2) INFORMATION FOR SEQ ID NO:74:
-~ (i) SEQUENCE CHARACTERISTICS:
.~ (A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 5,9,13 (D) OTHER INFORMATION: /note= "5-bromo dU"
(xi) ~U~N~ DESCRIPTION: SEQ ID NO:74:

(2) INFORMATION FOR SEQ ID NO:75:
.. _ (i) ~Qu~N~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 5 (D) OTHER INFORMATION: /note= "5-iodo dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:75:

(2) INFORMATION FOR SEQ ID NO:76:

(i) SEQUENCE CHARACTERISTICS:

AM~ ~

CA 02227867 l998-0l-26 PG~/~$ 9 6 / 1 1 786 ~P~AtUS 14 OCT'97 (A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 9 (D) OTHER INFORMATION: /note= "5-iodo dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:76:

(2) INFORMATION FOR SEQ ID NO:77:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single ~-~ (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 13 (D) OTHER INFORMATION: /note= "5-iodo dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:77:

(2) INFORMATION FOR SEQ ID NO:78:
(i) ~Qu~NC~ CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STR~Nn~nN~.5S: single ~_, (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_~eature (B) LOCATION: 5,9,13 (D) OTHER INFORMATION: /note= "5-iodo dU"
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:78:

(2) INFORMATION FOR SEQ ID NO:79:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear lS6 ~IE~IDEI) SHEEt -~ CA 02227867 l998-0l-26 PCTlU~ 9 6 /~ ~ 786 -lPEA/US 14 O~T '97 (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:79:

(2) INFORMATION FOR SEQ ID NO:80:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 ba~e pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:80:

(2) INFORMATION FOR SEQ ID NO:81:
.~ (i) ~Qu~N~ CHARACTERISTICS:
~ (A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRAN~h~SS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:81:

(2) INFORMATION FOR SEQ ID NO:82:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid .~ (C) STRANDEDNESS: single j (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) ~u~N~ DESCRIPTION: SEQ ID NO:82:

(2) INFORMATION FOR SEQ ID NO:83:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 15 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:83:

,~MENE)Eo 9H~

~ CA 02227867 1998-01-26 PC~fU~ 9~/11786 4 0CT'g;

(2) INFORMATION FOR SEQ ID NO:84:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 15 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) S~Q~N~ DESCRIPTION: SEQ ID NO:84:

(2) INFORMATION FOR SEQ ID NO:85:
(i) ~Qu~N~ CHARACTERISTICS:
(A) LENGTH: 15 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear J (ii) MOLECULE TYPE: DNA (genomic) (Xi) ~U~N~ DESCRIPTION: SEQ ID NO:85:

(2) INFORMATION FOR SEQ ID NO:86:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 20 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) . (xi) SEQUENCE DESCRIPTION: SEQ ID NO:86:
-' GCGGGGCTCC ATGGGGGTCG 20 (2) lN~O~L~TION FOR SEQ ID NO:87:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:87:

A~EN~ED 51H~
-CA 02227867 1998-01-26 ~T~ 6/ 11 78 ~PE~JS ~ 4 ac~ ~g7 WHL~TIS CLAIMEDIS:

1. A method of inhibiting an enzyme responsible for integrating nucleic acid of a virus into a host genome comprising contacting said enzyme with an oligonucleotide which is capable of forming at least one guanosine tetrad.

2. The method of claim 1 wherein said oligonucleotide inhibits said enzyme through interaction at a zinc-finger protein motif of said enzyme.

3. The method of claim 1 wherein said oligonucleotide comprises a sequence selected from the group consisting of SEQ. ID NOS2-87.

' 4. The method of claim 1 wherein said oligonucleotide comprises a sequence selected -from the group con~i~tin~ of T30177, T30038, T30340, T30341, T30659, T30673, T30674, T30675, T30676, T30677, T30678, T30679, T30666, T30661, T30695, T30696, T30397, T30698, T30699, T30700, T30701, T30702, T30719, T30720, T30721, T30722, T30075, T30570, T30571, T30576, T30577, T30578, T30579, T30743, T30744, T30745, T30746, T30748, T20747A, T30747B, T30754 and S935833 (SEQ ID NO 87, (SEQ ID NO 87), (SEQ ID NO 53, (SEQ ID NO 53), (SEQ
ID NO 53), (SEQ ID NO 54), (SEQ ID NO 55), (SEQ ID NO 56), (SEQ ID NO 57, (SEQ ID NO 55), (SEQ ID NO 56), (SEQ ID NO 58), (SEQ ID NO 59), (SEQ ID NO
59), (SEQ ID NO 60), (SEQ ID NO 61), (SEQ ID NO 62), (SEQ ID NO 63), (SEQ ID
_ NO 64), (SEQ ID NO 65), (SEQ ID NO 66), (SEQ ID NO 67), (SEQ ID NO 68), (SEQ ID NO 69), (SEQ ID NO 70), (SEQ ID NO 71), (SEQ ID NO 72), (SEQ ID NO
73), (SEQ ID NO 74), (SEQ ID NO 75), (SEQ ID NO 76), (SEQ ID NO 77), (SEQ ID
NO 78), (SEQ ID NO 79), (SEQ ID NO 80), (SEQ ID NO 81), (SEQ ID NO 82), (SEQ ID NO 83), (SEQ ID NO 84), (SEQ ID NO 85), (SEQ ID NO 86), (SEQ ID NO
86).

5. The method of claim 1 wherein said oligonucleotide comprises a sequence selected from the group consisting of T30l77 (SEQ ID NO 87) and T30695 (SEQ ID NO 87).

AMEN~ED SH~

~ CA 02227867 1998-01-26 PCTll~ q6/1178 ~ ~PE.WS ~ 4 ~CT '9 6. The method of claim 1, wherein said oligonucleotide further comprises a 3' terminus modifier. selected from the group consisting of propylamine, polyamine, poly-L-lysine and cholesterol.

7. The method of claim 6, wherein the modifier is propylamine.

8. The method of claim 1, wherein said oligonucleotide further comprises a 5' terminlle modifier for the purpose of increasing cellular uptake, or of modifying the tissue or subcellular distribution, or of increasing the biological stability of said oligonucleotide.

9. The method of claim 8, wherein said modifier is selected from the group consisting of propylamine, polyamine, poly-L-lysine, cholesterol, fatty acid chains of length 2 to 24 carbons, and vitamin E.

10. The method of claim 9, wherein the modifier is propylamine.

11. The method of claim 1, wherein said oligonucleotide further compri~ec at least one phosphodiester linkage.

12 The method of claim 1, wherein said oligonucleotide further comprises at least one phosphorothioate linkage.

13. The method of claim 1, wherein said virus is a retrovirus.
.

14. The method of claim 13, wherein said virus is a human immunodeficiencyvirus.

15. The method of claim 14, wherein said human immunodeficiencyvirus is HIV-l .

16. A method of inhibiting the production of a virus compri~in~ cont~cting a cell or an organism infected with said virus, and/or cont~rting said virus or a protein encoded by said virus with an oligonucleotide which is capable of forming at least one guanosine AM~

CA 02227867 1998-01-26 P~T~ ~ 6~1 1 786 U'E~JUS 14 ~CT'9 tetrad, said oligonucleotide comprising a sequence selected from the group con~i~tin~
of SEQ. ID NOS 2-27, 29,31-39 and 53-87.

17. The method of claim 15 wherein said oligonucleotide is selected from the group consisting of T30177 (SEQ ID NO 87) and T30695 (SEQ ID NO 87).

18. The method of claim 15, wherein said oligonucleotide further comprises a 5' and/or 3' terminus modifier for the purpose of increasing cellular uptake, or of modifying the tissue or subcellular distribution, or of increasing the biological stability of said oligonucleotide.

19. The method of claim 18, wherein said modifier is selected from the group con~i~ting of propylamine, polyamine, poly-L-lysine, cholesterol, fatty acid chains of length 2 to 24 carbons, and vitamin E.

20. The method of claim 18, wherein the modifier is propylamine.

21. The method of claim 15, wherein said oligonucleotide further compri~e~ at least one phosphodiester linkage.

22. The method of claim 15, wherein said oligonucleotide further comprises at least one phosphorothioate linkage.
. .
-- 23. The method of claim 15, wherein said virus is selected from the group consisting of herpes simplex virus, human papilloma virus, Epstein Barr virus, human immlmodeficiency virus, adenovirus, respildloly syncytial virus, hepatitis B virus and human cytomegalovirus.

24. The method of claim 23, wherein said virus is human immunodeficiency virus 25. A method of inhibiting the production of a human imml-nndeficiency virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or AM~N~ED

a protein encoded by said virus, with an oligonucleotide capable of forming at least one guanosine tetrad.

26. A method of inhibiting the production of a human immunodeficiency virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide selected from the groupconsisting of oligonucleotides in SEQ. ID NOS 2-27, 29, 31 -39 and 53-87.

27. A method of inhibiting the production of a human immllnodeficiency virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with the oligonucleotideT30177 (SEQ ID NO 87).

28. A method of inhibiting the production of a human immunodeficiency virus comprising __ cont~rting a cell or an organism infected with said virus, and/or cont~cting said virus or a protein encoded by said virus, with an oligonucleotide selected from the groupcon~i~ting of SEQ. ID NOS 2-27, 29, 31-39 and 53-87, wherein said oligonucleotide is capped at the 3' terminll~ and/or at the 5' terminllc with a modifier.

29. The method of claim 16, wherein said oligonucleotide sequence is chosen from the group consisting of T30177 (SEQ ID NO 87) and T30695 (SEQ ID NO 87).

30. A method of inhibiting the production of a virus compri~ing cont~tin~ a cell or an -- organism infected with said virus, and/or cont~ctin~ said virus or a protein encoded by - said virus, with an oligonucleotide, wherein said oligonucleotide inhibits a viral enzyme responsible for integrating nucleic acid of said virus into a host genome.

31. A method of inhibiting the production of a virus comprising cont~tin~ a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide capable of forming a guanosine tetrad, wherein said oligonucleotide inhibits a viral enzyme responsible for integrating nucleic acid of said virus into a host genome through interaction at a zinc-finger protein motif of said enzyme responsible for integrating nucleic acid of said virus into a host genome.

CA 02227867 1998-01-26 PCT~ 96/11 78 lP~WS ~ 4 OCT'97 32. A method of inhibiting the production of a virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virtls, with an oligonucleotide cont~ining at least about 50% guanosine bases, wherein said oligonucleotide forms a guanosine tetrad, and wherein said oligonucleotide inhibits a viral enzyme responsible for integrating nucleic acid of said virus into a host genome through interaction at a zinc-finger protein motif of said enz,vme responsible for integrating nucleic acid of said virus into a host genome.

33. An oligonucleotide comprising a sequence chosen from the group conei.~ting of SEQ.
ID NOS 2-27,29,31-39 and 53-87.

34. An oligonucleotidecomprisingthe sequence SEQ ID NO 87, varied, however, such that thymidine 17 is either present or omitted, and thymidines 2, 5, 9 and 13 are, indepenc~ently, present, omit the thymine base, or have another pyrimidine or modified pyrimidine~ul.~ uL~d therefor.

35. An inhibitor of an enzyme responsible for integrating viral nucleic acid into a host genome comrri ~ing the oligonucleotide of claim 33 or 34.

36. The inhibitor of claim 35 wherein said oligonucleotide interacts with said viral enz,vme at a zinc-finger protein motif of said enzyme.

~ 37. The oligonucleotide of claim 35, wherein said oligonucleotide further comrri~es a 5' ~~ and/or 3' ~ "~ modifier for the purpose of increasing cellular uptake, or of modifying the tissue or subcellular distribution, or of increasing the biological stability of said oligonucleotide.

38. The oligonucleotide of claim 39, wherein said modifier is selected from the group consisting of propylamine, polyamine, poly-L-lysine, cholesterol, fatty acid chains of length 2 to 24 carbons, and vitamin E.

39. The oligonucleotide of claim 38, wherein the modifier is propylamine.

AMENIIED Slt~Er J CA 02227867 1998-01-26 PCT/U~ 9 6 ~ 1 ~ 786 IPEA~Js 14 OCT'97 40. The oligonucleotideof claim 35 further comprising at least one phosphodiesterlinkage.

41. The oligonucleotide of claim 35 further comprising at least one phosphorothioate linkage.

42. The oligonucleotide of claim 35 further comprising a sugar backbone having at least one occurrence of a 2'-0-methyl RNA sugar.

43. The oligonucleotide of claim 35 further comprising at least one occurrence of a backbone sugar having no base ~tt~'.h~fl ~ 44. A ph~rm~ce~ltical composition comrri~in~ at least one guanosine tetrad forming i _; oligonucleotidehaving a sequence chosen from the group conci~ting of SEQ ID NOS 2-27, 29,31-39 and 87, and a ph~rm~cologicallyacceptable carrier.

45. A ph~rm~ceutical composition comprising at least one guanosine tetrad forming oligonucleotide having a sequence chosen from the group consisting of SEQ ID NOS53-87, and a ph:~rm~ologicallyacceptable carrier.

46. The ph~rm~se~ltical composition of claim 44 or 45 wherein said' oligonucleotide further comprises a backbone at least partially substituted with a phosphorothioatemoiety.

'~ 47. The ph~rrn~ce~ltical composition of claim 44 or 45 wherein said oligonucleotide further comprises an end termin~l capped with a modifier selected from the group consisting of propylamine, polyamine, poly-L-lysine and cholesterol.

48. The ph~rm~ceutical composition of claim 44 or 45 wherein said modifier is propylamine.

49. The ph~rm~eutical composition of claim 44 or 45 wherein said oligonucleotide further comprises a sugar backbone having at least one occurrence of a 2'-0-methyl RNA
sugar.

t~T

CA 02227867 1998-01-26 P ~ ~ S 9 6 1 1 1 7 8 6 lPEAllJS 14 OCT'97 50. The ph~rm~eutical composition of claim 44 or 45 wherein said oligonucleotide further comprises at least one occurrence of a backbone sugar having no base attached.

51. A method of treating a viral tii~e~e, comprising ~-lmini~tering a ph~rm~cological dose of an oligonucleotide having a sequence chosen from the group consisting of SEQ ID
NOS 2-27, 29, 31-39 and 87 together with a ph~rm~-~ologically acceptable carrier, said dose being sufficient to inhibit the production of said virus.

52. A method of treating a viral di~e~e, comprising ~imini~tering a pharmacological dose of an oligonucleotide having a sequence chosen from the group consisting of SEQ ID
NOS 53-87 together with a ph~rm~rologically acceptable carrier, said dose being sufficient to inhibit the production of said virus.

53. The method of claim 51 or 52, wherein said virus is selected from the group Con~i~ting of herpes simplex virus, human papilloma virus, Epstein Barr virus, human immllnodeficiency virus, adenovirus, le~ dlory syncytial virus, hepatitis B virus and human cytomegalovirus.

54. A method of treating an infection by a human immllnodeficiency virus, comprising the step of ~-lmini~terin~ a ph~rm~cological dose of a guanosine tetrad forming oligonucleotide to a human, said dose being sufficient to inhibit the replication of said -- virus.
~ . .
55. A method of treating an infection by a human immnnodeficiency virus, comprising ~lmini~rin~ to a human a ph~rm~ological dose of the oligonucleotide of claim 33 or 34 together with a ph~rm~ologically acceptable carrier, said dose being sufficient to inhibit the replication of said virus.

56. A method of treating a viral disease comprising the method of claim 52 or 54 wherein said dose is at least 3.0 mg/kg of patient body weight.

Ah~ D ~T

PCT/V~ 96/11786 IPEA~Js 14 OCT g7 57. A method of keating a viral disease comprising the method of claim 52 or 54 wherein said dose is at least 3.0 mg/kg of patient body weight ~lmini.~tered in seven equal doses over 14 days.

58. A method of treating a viral disease compri~ing the method of claim 52 or 54 wherein said inhibition of production of said virus results in a reduction in viral load in a patient's serum of at least 12-55%.

. ~

~MFN~ED 9 (2) INFORMATION FOR SEQ ID NO:49:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 13 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:49:

(2) INFORMATION FOR SEQ ID NO:50:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:50:

(2) INFORMATION FOR SEQ ID NO:51:

(i) SEQUENCE CHARACTERISTICS:

WO 97/03997 PCTrUS96/11786 (A) LENGTH: 14 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:51:

(2) INFORMATION FOR SEQ ID NO:52:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 10 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:52:

Claims (69)

WHAT IS CLAIMED IS:

1. A method of inhibiting an enzyme responsible for integrating nucleic acid of a virus into a host genome comprising contacting said enzyme with an oligonucleotide which is capable of forming at least one guanosine tetrad.

2. The method of claim 1 wherein said oligonucleotide inhibits said enzyme through interaction at a zinc-finger protein motif protein motif of said enzyme.

3. The method of claim 1 wherein said oligonucleotide is selected from the group consisting of SEQ. ID Nos. 1-52 and Table C-1.

4. The method of claim 1 wherein said oligonucleotide is selected from the group consisting of T30177 and T30695.

5. The method of claim 1, wherein said oligonucleotide is capped at the 3' terminus with a modifier selected from the group consisting of propylamine, polyamine, poly-L-lysine and cholesterol.

6. The method of claim 5, wherein the modifier is propylamine.

7. The method of claim 1, wherein said oligonucleotide is capped at the 5' terminus with a modifier for the purpose of increasing cellular uptake, or of modifying the tissue or subcellular distribution, or of increasing the biological stability of said oligonucleotide.

8. The method of claim 7, wherein said modifier is selected from the group consisting of propylamine, polyamine, poly-L-lysine, cholesterol, fatty acid chains of length 2 to 24 carbons, and vitamin E.

9. The method of claim 7, wherein the modifier is propylamine.

10. The method of claim 1, wherein said oligonucleotide comprises at least one phosphodiester linkage.

11. The method of claim 1, wherein said oligonucleotide comprises at least one phosphorothioate linkage.

12. The method of claim 1, wherein said virus is a retrovirus.

13. The method of claim 12, wherein said virus is a human immunodeficiency virus.

14. The method of claim 13, wherein said human immunodeficiency virus is HIV-1.

15. A method of inhibiting the production of a virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus with an oligonucleotide which is capable of forming at least one guanosine tetrad.

16. The method of claim 15 wherein said oligonucleotide is selected from the group consisting of SEQ. ID Nos. 1-52.

17. The method of claim 15 wherein said oligonucleotide is selected from the group consisting of T30177 and T30695.

18. The method of claim 15, wherein said oligonucleotide is capped at the 5' terminus with a modifier for the purpose of increasing cellular uptake, or of modifying the tissue or subcellular distribution, or of increasing the biological stability of said oligonucleotide.

19. The method of claim 18, wherein said modifier is selected from the group consisting of propylamine, polyamine, poly-L-lysine, cholesterol, fatty acid chains of length 2 to 24 carbons, and vitamin E.

20. The method of claim 18, wherein the modifier is propylamine.

21. The method of claim 15, wherein said oligonucleotide is capped at the 5' terminus with a modifier selected from the group consisting of propylamine, polyamine, poly-L-lysine and cholesterol.

22. The method of claim 19, wherein the modifier is propylamine.

23. The method of claim 15, wherein said oligonucleotide comprises at least one phosphodiester linkage.

24. The method of claim 15, wherein said oligonucleotide comprises at least one phosphorothioate linkage.

25. The method of claim 15, wherein said virus is selected from the group consisting of herpes simplex virus, human papilloma virus, Epstein Barr virus, human immunodeficiency virus, adenovirus, respiratory syncytial virus, hepatitis B virus and human cytomegalovirus.

26. The method of claim 25, wherein said virus is human immunodeficiency virus

27. A method of inhibiting the production of a human immunodeficiency virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide capable of forming at least one guanosine tetrad.

28. A method of inhibiting the production of a human immunodeficiency virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide selected from the group consisting of oligonucleotides in SEQ. ID Nos. 1-_.

29. A method of inhibiting the production of a human immunodeficiency virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with the oligonucleotide T30177.

30. A method of inhibiting the production of a human immunodeficiency virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide selected from the group consisting of SEQ. ID Nos 1-52, wherein said oligonucleotide is capped at the 3' terminus and/or at the 5' terminus with a modifier.

31. The method of claim 16, wherein said oligonucleotide T30177 and T30695.

32. A method of inhibiting the production of a virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide, wherein said oligonucleotide inhibits a viral enzyme responsible for integrating nucleic acid of said virus into a host genome.

33. A method of inhibiting the production of a virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide capable of forming a guanosine tetrad, wherein said oligonucleotide inhibits a viral enzyme responsible for integrating nucleic acid of said virus into a host genome through interaction at a zinc-finger protein motif of said enzyme responsible for integrating nucleic acid of said virus into a host genome.

34. A method of inhibiting the production of a virus comprising contacting a cell or an organism infected with said virus, and/or contacting said virus or a protein encoded by said virus, with an oligonucleotide containing at least about 50% guanosine bases, wherein said oligonucleotide forms a guanosine tetrad, and wherein said oligonucleotide inhibits a viral enzyme responsible for integrating nucleic acid of said virus into a host genome through interaction at a zinc-finger protein motif of said enzyme responsible for integrating nucleic acid of said virus into a host genome.

35. An oligonucleotide of any of SEQ. ID Nos. 2-_.

36. The oligonucleotide of claim 35 wherein said oligonucleotide is capable of forming a guanosine tetrad.

37. The oligonucleotide of claim 35 wherein said oligonucleotide is capable of inhibiting a viral enzyme responsible for integrating nucleic acid of said virus into a host genome.

38. The oligonucleotide of claim 35 wherein said oligonucleotide is capable of inhibiting said viral enzyme responsible for integrating nucleic acid of said virus into a host genome through interaction at a zinc-finger protein motif of said enzyme responsible for integrating nucleic acid of said virus into a host genome.

39. The oligonucleotide of claim 35, wherein said oligonucleotide is capped at the 5' terminus with a modifier for the purpose of increasing cellular uptake, or of modifying the tissue or subcellular distribution, or of increasing the biological stability of said oligonucleotide.

40. The method of claim 39, wherein said modifier is selected from the group consisting of propylamine, polyamine, poly-L-lysine, cholesterol, fatty acid chains of length 2 to 24 carbons, and vitamin E.

41. The oligonucleotide of claim 40, wherein the modifier is propylamine.

42. The oligonucleotide of claim 35, wherein said oligonucleotide is capped at the 5' terminus with a modifier selected from the group consisting of propylamine, polyamine, poly-L-lysine and cholesterol.

43. The oligonucleotide of claim 42, wherein the modifier is propylamine.

44. The oligonucleotide of claim 35, wherein said oligonucleotide comprises at least one phosphodiester linkage.

45. The oligonucleotide of claim 35, wherein said oligonucleotide comprises at least one phosphorothioate linkage.

46. An oligonucleotide capable of forming a guanosine tetrad and capable of inhibiting a viral enzyme responsible for integrating nucleic acid of a virus encoding said enzyme into a host genome.

47. An oligonucleotide capable of forming a guanosine tetrad and capable of inhibiting a viral enzyme responsible for integrating nucleic acid of a virus encoding said enzyme into a host genome through interaction at a zinc-finger protein motif of said enzyme.

48. A composition for inhibiting viral production comprising at least one oligonucleotide capable of forming a guanosine tetrad.

49. The composition of claim 48 wherein said oligonucleotide has a nucleotide sequence substantially non-homologous and substantially non-complementary to a genome of a virus being produced.

50. A composition for inhibiting HIV-1 production comprising at least one oligonucleotide capable of forming a guanosine tetrad.

51. A composition for inhibiting HIV-1 production comprising at least one oligonucleotide having at least two runs of at least two guanosine bases.

52. A non-sequence specific guanosine-rich oligonucleotide, wherein said oligonucleotide inhibits viral production by forming a guanosine tetrad.

53. A non-sequence specific composition for inhibiting HIV-1 production comprising at least one oligonucleotide having at least two runs of at least two guanosine bases capable of forming a guanosine tetrad.

54. The oligonucleotide of any of claims 48 through 53, wherein said oligonucleotide contains at least about 40% guanosine bases.

55. The oligonucleotide of any of claims 48 through 53 selected from the group consisting of SEQ. ID. Nos. 1 to _.

56. The oligonucleotide of any of claims 48 through 53 having backbone at least partially substituted with a phosphorothioate moiety.

57. The oligonucleotide of any of claims 48 through 53 wherein said oligonucleotide has an end terminal capped with a modifier selected from the group consisting of propylamine, polyamine, poly-L-lysine and cholesterol.

58. The oligonucleotide of any of claims 48 through 53 wherein said oligonucleotide has an end terminal capped with a propylamine modifier.

59. A method of treating a viral disease, comprising administering a pharmacological dose of an oligonucleotide, said dose being sufficient to inhibit the production of said virus.

60. A method of treating a viral disease, comprising administering a pharmacological dose of an oligonucleotide, said dose being sufficient to inhibit the production of said virus, wherein said oligonucleotide inhibits an enzyme responsible for integrating nucleic acid of said virus into a host genome of a virus agent causing said disease.

61. A method of treating a viral disease, comprising administering a pharmacological dose of an oligonucleotide, said dose being sufficient to inhibit the production of said virus, wherein said oligonucleotide forms at least one guanosine tetrad.

62. A method of treating a viral disease, comprising the step of administering apharmacological dose of an oligonucleotide, said dose being sufficient to inhibit the production of said virus, wherein said oligonucleotide contains at least two runs of at least two guanosine bases.

63. The method of claim 59, wherein said virus is selected from the group consisting of herpes simplex virus, human papilloma virus, Epstein Barr virus, human immunodeficiency virus, adenovirus, respiratory syncytial virus, hepatitis B virus and human cytomegalovirus.

64. The method of claims 59-63, wherein said oligonucleotide is selected from the group of sequences consisting of SEQ ID Nos. 1-_.

65. A method of treating an infection by a human immunodeficiency virus, comprising the step of administering a pharmacological dose of an oligonucleotide to a human, said dose being sufficient to inhibit the replication of said virus, wherein said oligonucleotide forms guanosine tetrads.

66. A method of treating an infection by a human immunodeficiency virus, comprising the step of administering a pharmacological dose of an oligonucleotide to a human, said dose being sufficient to inhibit the replication of said virus, wherein said oligonucleotide contains at least two runs of at least two guanosine bases.

67. The method of any of claims 59-66 wherein said dose is at least 3.0 mg/kg of patient body weight.

68. The method of any of claims 59-67 wherein said dose is at least 3.0 mg/kg of patient body weight administered in seven equal doses over 14 days.

69. The method of any of claims 59-68 wherein said inhibition of production of said virus results in a reduction in viral load in a patient's serum of at least 12-55%.