Polynucleoside chain with 3'.fwdarw.5' guanidyl linkages

Abstract

The invention provides a cationic polynucleoside chain having multiple nucleosides, the nucleosides being coupled together by positively charged guanidyl linkages.

Description

BACKGROUND OF THE INVENTION

Oligonucleotide technology utilizes short sequences of nucleic acids as highly specific probes to assay the presence of DNA and/or RNA sequences of interest. It also has applications for treating diseases as it can be used to selectively inhibit the expression of genes of interest. As such, oligonucleotide technology has great potential in the analysis and treatment of genetic disorders.

As an example, genetic disorders account for approximately 30 percent of admissions to pediatric hospitals. Therefore, the ability to provide a comprehensive diagnostic molecular genetic service is an important aspect of a modern health care system.

In order to solve problems attendant in some applications using oligonucleotides, such as degradation by host cell nucleases, or transfer across cell membranes, oligonucleotide analogs have been constructed by replacing a non-bridging oxygen in the phosphate moiety with a sulfur, methyl or ethyl group (reviewed in Stein, C. A., and Cheng, Y. -C., Science (1993) 261:1004). A second example of an oligonucleotide analog is one using a formacetal/ketal type linkage replacing one or more phosphodiester linkages between adjacent nucleosides. (U.S. Pat. No. 5,264,564, issued Nov. 23, 1993).

Dimeric nucleosides have also been constructed with uncharged guanidyl linkages replacing some phosphodiester bonds. Vandendriessche et al. provide a method for synthesizing thymidine nucleosides coupled together with substituted guanidyl linkages, J. Chem. Soc. Trans. 1:1567 (January, 1993); Bio. Org. Chem. 3:193 (February, 1993). The electron-withdrawing properties of the substituent groups make the guanidyl moiety uncharged at physiological Ph. These thymidine dimers can be incorporated into oligonucleotide chains with standard phosphate backbones.

Vandendriessche et al. developed their method using neutral guanidyl groups for coupling nucleosides because they reasoned that positively charged groups may hamper cellular uptake of oligonucleosides containing them and may lead to non-specific binding.

However, the resulting method having uncharged guanidyl units at physiological pH is limited in its applications to oligonucleoside technology. It does not allow for construction of oligonucleoside chains containing two or more successive guanidyl linkages. Instead, each nucleoside connected to one nucleoside via a guanidyl linkage must be attached to a second with a phosphodiester linkage. Thus, many of the advantages provided by guanidyl linkages are undermined.

For example, the necessity for phosphodiester bonds following guanidyl bonds exposes oligonucleosides to cellular nucleases that recognize phosphodiester bonds.

A second limitation of the Vandendriessche method is that the electrically neutral guanidyl bridges will not interact with the phosphates on the backbones of naturally occurring nucleic acids.

The present invention overcomes both of these shortcomings. First, it provides an alternative approach to constructing oligonucleosides that utilizes guanidyl linkages having a positive charge at physiological pH. In addition, the method provides a way to construct a oligonucleoside chain built with successive positively charged guanidyl linkages. It is thus possible to use the present invention to construct a positively charged oligonucleoside chain that will bind with high affinity to negatively charged phosphate backbone of cellular and viral DNA or RNA.

The higher affinity binding reduces the incidence of false positives and facilitates earlier detection of a diseased state. Early detection of disease improves the patient's prognosis and aids in determining an optimal therapy regimen. Therefore, the diagnostic agents of the invention provide a means to determine disease states early, efficiently and predictably.

SUMMARY OF THE INVENTION

The present invention provides a polynucleoside chain having multiple nucleosides coupled together by positively charged guanidyl linkages. Assays for using such polynucleoside chains are also provided.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1. The general synthetic strategy for the formation of thymidyl DNG oligomers.

FIG. 2. General synthetic strategy for chain-extending intermediate thymidyl S-methylisothiourea derivative (9) of FIG. 1.

FIG. 3. DNG coupling reaction.

FIG. 4. Protocol for synthesis of 5'-OH terminal unit of RNG.

FIG. 5. Protocol for synthesis of chain-extending units of RNG.

FIG. 6. Protocol for synthesis of 3'-OH terminal unit of RNG.

FIG. 7. Protocol for synthesis of 5'-OH terminal unit of DNG.

FIG. 8. Protocol for synthesis of chain-extending unit of DNG.

FIG. 9. Protocol for synthesis of 3'OH terminal unit of DNG.

FIG. 10. Stereo view of a ball and stick model of 3'-Deoxythymidylyl-�3'-5'guanidyl!-3'-azido-3',5'-deoxythymidine Iodide generated from the X-Ray coordinates of thymidylyl-(5'3')-thymidylate-(5') hydrate.

FIG. 11. Stereo view of the duplex hybrid of (d(gT).sub.10 .multidot.d(Ap).sub.10). The structure was computer generated from the structure of B-(d(pT).sub.10 .multidot.d(Ap).sub.10) by replacing the phosphate linkages by guanidyl linkages and energy minimization via ABNR in CHARMM.

FIG. 12. Stereo view of the triple helical structure of (d(Tp).sub.10 .multidot.d(Ap).sub.10 .multidot.d(gT).sub.10). The model was created from the coordinates of (d(Tp).sub.10 .multidot.d(Ap).sub.10 .multidot.d(Tp).sub.10) after Ragunathan et al. Proc. Nat. Acad. Sci. USA 90:9542-9546.

FIG. 13 is a graphic representation of base-paired DNG-DNA hybrids.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a polynucleoside chain having multiple nucleosides. Each of the nucleosides in the chain are coupled together by positively charged guanidyl linkages.

Each nucleoside in the chain is made up of a sugar component and a purine or pyrimidine base. The sugar is 2'-deoxyribose for DNG and ribose for RNG. The base may be a guanine (G), cytosine (C), adenine (A), or thymine (T) in the case of DNG, or uracil (U) in the case of RNG. Each sugar component of the chain is linked to a purine or pyrimidine base at the C1' position of the sugar component. Each nucleoside of the chain is connected to another at the C5' and C3' positions of the sugar component by a positively charged guanidyl linkage replacing a phosphodiester linkage or other linkages.

The guanidyl linkage has the following formula: ##STR1##

The R group represents a hydrogen molecule or a lower alkyl group that does not alter the positive charge on guanidine. Examples of suitable lower alkyl groups include, but are not limited to, CH.sub.3, CH.sub.2 CH.sub.3, CH.sub.2 (CH.sub.2).sub.n CH.sub.3, CH.sub.2 X, CH.sub.2 (CH.sub.2).sub.n X, wherein X is any functional group that does not alter the positive charge on guanidine. Examples include, but are not limited to, --OH or--NH.sub.2.

The polynucleoside backbone of the invention may be composed entirely of guanidyl linkages, in which case the backbone is said to be homomeric with respect to guanidyl linkages, thereby resulting in homomers of the invention. Alternatively, some sugar components within one nucleoside chain are linked to one another through guanidyl linkages, and others are linked through a phosphate moiety. In this situation the polymer is said to be heteromeric, thereby resulting in heteromers of the invention.

A description of how to synthesize homomers and heteromers of the invention is given in Example 1 and further herein.

For assembling an oligonucleoside chain, monoucleosides may be appended to the 5' and/or 3' end of the nucleosides of a homomer or heteromer. Alternatively, the mononucleosides may be inserted within the nucleosides of the homomer or heteromer.

In accordance with the practice of the invention, the polynucleoside chain may be a DNG chain which is substantially complementary to a gene sequence of interest. Alternatively, the polynucleoside chain may be a RNG chain which is substantially complementary to a gene sequence of interest. The polynucleoside may additionally be a chimera of RNG and DNG subunits.

The gene sequence of interest may be the DNA of any known sequence, a known genomic RNA such as an RNA virus or viroid, or an RNA transcribed from a DNA sequence of interest.

For example, the gene sequence of interest may be a segment of the nucleotide sequence of HTLV-I (U.S. Pat. No. 4,999,421, issued Mar. 12, 1991). Probes of the invention may be made and used for detecting HTLV-I DNA or RNA in samples suspected of being infected with HTLV-I.

In another example, the gene sequence of interest may be the nucleotide sequence of pregnancy-specific beta-glycoprotein (PS.beta.G) cDNA that corresponds to a mRNA that is expressed at high levels in certain gestational diseases (U.S. Pat. No. 5,141,849, issued Aug. 25, 1992). Probes of the invention may be used for detecting Ps.beta.g RNA sequences in a sample of placental tissue.

In one embodiment, the polynucleoside chain of the invention is a component in a composition comprising an effective amount of the polynucleoside and a suitable pharmaceutical carrier. Suitable pharmaceutical carriers include water, saline, Ringer's solution, Hank's solution or nonaqueous carriers such as fixed oils. The carrier may also contain substances that enhance isotonicity and chemical stability of the chain such as buffers or preservatives.

The compositions of the invention also preferably include conventional pharmaceutically acceptable carriers and adjuvants known in the art such as human or bovine serum albumin, ion exchangers, alumina, lecithin, buffer substances such as phosphates, glycine, sorbic acid, potassium sorbate, and salts or electrolytes such as protamine sulfate.

In accordance with the practice of the invention, the pharmaceutical carrier may be a lipid carrier. The lipid carrier may be a phospholipid. Further, the lipid carrier may be a fatty acid. Also, the lipid carrier may be a detergent. As used herein, a detergent is any substance that alters the surface tension of a liquid, generally lowering it.

In one example of the invention, the detergent may be a nonionic detergent. Examples of nonionic detergents include, but are not limited to, polysorbate 80 (also known as Tween 80 or (polyoxyethylenesorbitan monooleate), Brij, and Triton (for example Triton WR-1339 and Triton A-20).

Alternatively, the detergent may be an ionic detergent. Examples of ionic detergent includes, but are not limited to, alkyltrimethylammonium bromide and sodium dodecyl sulfate.

Additionally, in accordance with the invention, the lipid carrier may be a liposome. As used in this application, a "liposome" is any membrane bound vesicle which contains any molecules of the invention or combinations thereof.

USES OF THE PRESENT INVENTION

The oligonucleosides of the invention may be used to identify specific nucleic acids either as probes in nucleic hybridization assays, or via amplification techniques such as the polymerase chain reaction (PCR). Because the identification of specific DNA and/or RNA sequences is often indicative of a disease state, the ability to detect specific nucleic acids with the invention has both diagnostic and prognostic value.

The molecules of the invention may be used in various assays utilizing antibodies raised against them. These anti-DNG or anti-RNG antibodies can then be used in assays such as radioimmunoassay (RIA), enzyme-linked immunosorbent assay (ELISA), enzyme immunoassay (EIA), enzyme-multiplied immunoassay technique (EMIT), substrate-labeled fluorescent immunoassay (SLIFA) and the like. The assays may be homogeneous (without a separation step between free reagent and receptor-ligand complex) or heterogeneous (with a separation step between free reagent and receptor-ligand complex).

The reagents may be supplied in a kit, so as to facilitate ease of application of the invention and also to optimize the invention's reproducibility and sensitivity.

For the subject invention, depending upon the nature of the assay, the protocol, and the label, either labeled or unlabeled polynucleoside chain is provided, usually in conjunction with other additives, such as buffer, stabilizers, materials necessary for signal production, e.g., substrates for enzymes, and the like.

DNG and RNG oligonucleosides containing guanidyl linkages will be labeled by DNA modification enzymes routinely used to label nucleic acids. Because the invention allows the addition of nucleosides via either guanidyl or phosphate linkages, it is possible to construct chains with conventional phosphate linkages at the extreme 5' or 3' termini.

A DNG or RNG oligonucleoside can be radiolabeled by kinasing its 5' end with gamma-.sup.32 P ATP and T4 polynucleotide kinase. Alternatively, DNG oligonucleosides can be radiolabeled using terminal deoxynucleotidyl transferase and alpha-.sup.32 P-dNTP.

The oligonucleosides can also be labeled using non-radioactive labeling methods. For example, a DNG oligonucleoside can be labeled by using terminal transferase to add digoxigenin-11-dUTP to its 3' end using a method for labeling oligonucleotides developed by Bohringer-Mannheim Corporation, Indianapolis, Ind.

Following hybridization, the probe can be detected with an antibody specific for the digoxigenin moiety. The antibody is conjugated to an easily detectable reporter molecule such as alkaline phosphatase.

Depending on the particular application, an oligonucleoside probe should have at least about 14 nucleosides and may range upward in size to over one hundred nucleosides.

The reagents are provided as a dry powder, where the reagents may be reconstituted in an aqueous medium having appropriate concentrations for performing the assay. Alternatively, the reagents are in a liquid medium.

The molecules described herein may be in a variety of dosage forms which include, but are not limited to, liquid solutions or suspensions, tablets, pills, powders, suppositories, polymeric microcapsules or microvesicles, liposomes, and injectable or infusible solutions. The preferred form depends upon the mode of administration and the therapeutic application.

The most effective mode of administration and dosage regimen for the molecules of the present invention depends upon the location of the disease being treated, the severity and course of the disease, the subject's health and response to treatment and the judgment of the treating physician. Accordingly, the dosages of the molecules should be titrated to the individual subject.

The interrelationship of dosages for animals of various sizes and species and humans based on mg/m.sup.2 of surface area is described by Freireich, E. J., et al. Cancer Chemother., Rep. 50 (4):219-244 (1966). An approximation of the in vivo dosage of oligonucleotides is about 30 mg/kg of body weight, Agrawal, S. et al. Proc. Natl. Acad. Sci. USA .88 (17):7595Adjustments in the dosage regimen may be made to optimize the tumor cell growth inhibition and killing, e.g., doses may be divided and administered on a daily basis or the dose reduced proportionally depending upon the situation, e.g. several divided doses may be administered daily or proportionally reduced.

It would be clear that the dose of the composition of the invention required to achieve cures may be further reduced with schedule optimization.

The use of probes to bind the target may be carried out in any conventional techniques such as nucleic acid hybridization, plus and minus screening to identify sequences differentially expressed in different RNA populations, recombinational probing, hybrid released translation (HRT), hybrid arrested translation (HAT), and amplification techniques such as polymerase chain reaction (PCR). Probes constructed by the invention may also be used to detect proteins that bind to specific nucleic acid sequences.

The present invention additionally provides a method for inhibiting the expression of a cellular gene. The method comprises constructing a DNG or RNG oligonucleoside with a sequence complementary to a portion of the known sequence of a transcribed sequence of interest; delivering said oligonucleoside to the target tissue; contacting said oligonucleoside with the transcribed sequence of interest; forming a complex between the said oligonucleoside and the transcribed sequence; and the complex then capable of preventing or inhibiting proper folding of the transcribed sequence or translation of the transcribed sequence.

The present method further provides a method for inhibiting the replication of a virus. The method comprises contacting the nucleic acid comprising the genome of the virus with the polynucleoside of the invention thereby permitting the formation of a double-stranded complex. The complex obstructs replication transcription, and translation of the viral genome. Production of progeny viral particles is thereby prevented.

The present invention also provides a method for inhibiting viral replication by preventing viral integration into the genome of a host cell. The method comprises determining the presence of a virus by detecting the presence of a viral nucleic acid sequence; constructing a polynucleoside of the invention which is substantially complementary to at least a portion of the viral sequence required for integration in to the host cell genome; and contacting the polynucleoside of the invention with the viral sequence so as to form a nucleoside-viral sequence complex, the complex so formed capable of preventing viral integration into the genome of the host cell.

METHODS OF MAKING POLYNUCLEOSIDE CHAINS OF THE PRESENT INVENTION

The following synthetic methodology results in the formation of the 3',5'-guanidyl linkage of the polynucleosides of the invention. Oxidation of 3'-azido-5'-thiourea-3',5'-deoxythymidine with peracetic acid in methanol (FIG. 3) provided the 5'-aminoiminomethanesulfonic acid derivative 22 which proved superior as an intermediate (faster reaction time) in the DNG coupling reaction in place of 3'-azido-5'-S-methylisothiouronium-3',5'-deoxythimidine Iodide. Thus the reaction of 22 with 3'-amino-2',3'-deoxythymidine in the presence of 4-dimethylaminopyridine (DMAP) produces the thymidine DNG dimer in yields greater than 90%.

General synthetic methodology for the formation of RNG (ribonucleic guanidine) is illustrated in FIGS. 4-6. Synthesis of the 5'-terminal unit of RNG is shown in FIG. 4. The reaction of D-xylose with 2-methoxypropene results in the selective formation of the 1,3-O-isopropylidene product 1 (Fanton E.; Gelas, J.; Horton, D., J. Chem. Soc., Chem. Commun., (1980), 21). Acetylation of the 1- and 2-hydroxyl's with acetic anhydride followed by cleavage of the isopropylidene moiety with trifluoracetic acid provides intermediate 3. Selective 5-protection is accomplished by reaction of 3 with trityl chloride in pyridine. Functionalization of the 3-position as the azide moiety involves two steps starting with the mesylation of 4 followed by nucleophilic substitution with sodium azide. The acetate moieties protecting the 1- and 2-positions of 5 are reportedly stable in the presence of azide ion (Greene, T. W.; Wuts, P. G.; In Protecting Groups in Organic Synthesis, Wiley-Interscience Publication: New York (1991); p. 418).

Compound 6, now containing the ribose ring configuration, can be alkylated at the C-1' position (with beta stereochemistry) utilizing standard Vorbruggen reaction conditions (Vorbruggen, H.: Nieballa, V.; Krolikiewicz, K.; Bennua, B.; Hofle, G. In Chemistry and Biology of Nucleosides and Nucleotides, Harmon, R. E., Robins, R. K., Townsend, L. B., Eds.; Academic Press: New York (1978); p. 251; Vorbruggen, H.; Hofle, G., Chem. Ber. (1981), 114, 1256).

Any of the natural RNA or DNA bases can be applied to this reaction. The resulting nucleoside 7 is deprotected with sodium methoxide followed by trifluoroacetic acid. The 3'-azido nucleoside 8 is then converted to the amino analogue by catalytic hydrogenation, producing the 5'-OH coupling intermediate 9.

Chain extension of the RNG polynucleoside of the invention is as follows. The chain-extending RNG intermediate 15 (FIG. 5) begins with the conversion of 3 to the 5-azido analogue 10 by a two-step process involving tosylation followed by nucleophilic substitution. Compound 10 is then reduced to the resulting amine and protected with the t -butyl carbamate protecting group. After mesylation, 11 is converted to the 3'-azido compound 12 by treatment with sodium azide. Conversion of 12 to the desired nucleoside is followed by deprotection of the 5'-amino group with trifluoroacetic acid. Compound 14 is converted to the sulfonic acid derivative 16 by a three-step process involving formation of the 5'-thiourea followed by oxidation of the sulfur atom with peracetic acid.

FIG. 6 illustrates the synthesis of the 3'-OH terminal coupling unit of RNG. Any of the natural nucleosides are easily converted to the 5'-azido derivatives by selective 5'-tosylation followed by nucleophilic substitution. Catalytic reduction is followed by conversion of the resulting amine to the thiourea analogue 19. Oxidation of 19 to the sulfonic acid coupling intermediate 20 is accomplished with peracetic acid in methanol.

The methodology described above allows for the synthesis of RNG homomers, or by incorporation of a 3'- or 5'- hydroxyl group, the synthesis of RNG heteromers (RNG with RNA units interspersed throughout the polymer backbone).

The general synthesis for DNG homomers or heteromers is similar to that described for RNG polymers. However, DNG (deoxyribonucleic guanidine) oligomer synthesis includes a chemical step resulting in the dehydroxylation at the 2'-position. For example, synthesis of the 5'-OH terminal unit of DNG (FIG. 7) utilizes intermediate 7. Deacetylation is followed by conversion of the 2'-OH group to the thiocarbonyl ester which is reduced to the 2'-deoxynucleoside 23 with tributyltin hydride (Singh, J.; Wise, D.; Townsend, L. In Nucleic Acid Chemistry, Part 4. Townsend, L. B., Tipson, R. S., Eds; Wiley-Interscience Publication: New York (1991); p. 96). Deprotection of the 5'-position with trifluoroacetic acid followed by reduction of the azide moiety to the amine by catalytic hydrogenation provides the coupling intermediate 24.

Chain extension of the DNG polynucleoside of the invention is as follows. FIG. 8 sets forth the synthesis of the chain-extending unit of DNG. Thus, intermediate 13 is converted to compound 22 which then converted into the coupling intermediate 23 by the same sequence of reactions used for the synthesis of 16 for 13 (FIG. 5) .

The synthesis of the 3-OH terminal unit of DNG (FIG. 9) starts with any of the natural DNA nucleosides. The 5'-position is functionalized to the sulfonic acid derivative by a similar set of reactions as depicted in FIG. 5.

The synthetic methodology described above allows for the synthesis of DNG or RNG homomers or heteromers of the invention capable of targeting any gene or RNA sequence of interest.

ADVANTAGES OF THE PRESENT INVENTION

The oligonucleosides of the present invention can be synthesized easily and in large quantities. Therefore, methods for large-scale synthesis of homomers or heteromers of the present invention can be commercially pursued.

The oligonucleosides of the present invention are stable. In the past, significant effort has been expended by chemists to develop nuclease-resistant oligonucleotides. Perhaps the greatest successes in doing so have been achieved with the phosphorothioate (PS) and methylphosphoroate (MP) oligos. Because nucleosides of the present invention are coupled by guanidyl linkages, they will be resistant to cellular nucleases that specifically recognize phosphodiester linkages.

The ability of oligonucleosides to penetrate the cell membrane and their mechanism of penetration are critical considerations in developing these agents as therapeutic and diagnostic agents. While cell membranes are generally impermeable to charged molecules, this problem can be overcome for DNG or RNG oligonucleosides. These positively charged oligonucleosides can be annealed to a complementary oligonucleotide having a standard phosphate backbone prior to delivery into a host cell. The duplex then crosses the membrane. Once the oligonucleotide having a standard phosphate backbone is degraded by host cell nucleases, the DNG or RNG oligonucleoside can bind to its target sequence.

In order that the invention described herein may be more fully understood, the following examples are set forth. It should be understood that these examples are for illustrative purposes only and are not to be construed as limiting the scope of this invention in any manner.

EXAMPLE 1

Materials and Methods

Synthesis

General procedures. All thin layer chromatography (TLC) was run with Merck silica gel 60 (F.sub.254) plates. .sup.1 H NMR chemical shift assignments were referenced to dimethylsulfoxide (DMSO) (2.49 ppm). High resolution mass spectra were obtained from the U.C.L.A. mass spectrometry laboratory.

5'-t-butyl carbamoyl-5'-deoxythymidine (2 of FIG. 2). To a suspension of 5'-amino-5'-deoxythymidine (Horwitz, J. P.; Tomson, A. J.; Urbanski, J. A.; Chua, J.; (1962) J. Org. Chem. 27, 3045-3048) (4.0 g, 16.6 mmol) in 100 Ml of dry dimethylformamide (DMF) was added triethylamine (4.4 mL, 18.2 mmol) followed by dropwise addition of di-t-butyl dicarbonate (4.2 mL, 31.6 mmol).

The structure of 5'-t-Butyl carbamoyl-5'-deoxythymidine is as follows. ##STR2##

The resulting solution was stirred at room temperature for 30 minutes and then evaporated to dryness.

The residue was crystallized from water: 5.5 g (97%) yield; m.p.=164.degree. C.; TLC (5% methanol in ethyl acetate), R.sub.f =0.43; infrared (IR) (KBr) 3381, 1727, 1690, 1660, 1530, 1480, 1370, 1278, 1180, 1094 and 611 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 7.52 (1H, s, 6-H), 7.05 (1H, br t, carbamoyl N--H), 6.13 (1H, t, J=6.9 Hz, 1'-H), 5.28 (1H, d, J=4.0 Hz, 3'-OH), 4.13 and 3.73 (2H, 2 x m, 3'-H and 4'-H, no assignments made), 3.13 (2 H, m, 5'-Hs), 2.04 (2H, m, 2'-Hs), 1.80 (3H, s, methyl), 1.37 (9H, s, t-butyl). high resolution mass spectroscopy (HRMS) (EI) m/e 341.1592(M.sup.+), calcd for C.sub.15 H.sub.23 N.sub.3 O.sub.6 341.1587.

5'-t-butyl carbamoyl-5'-deoxy-3'-o-mesyl thymidine (3 of FIG. 2). To a suspension of 2 (5.5 g, 16.11 mmol) in 100 mL of dry CH.sub.2 Cl.sub.2 at 0.degree. C. was added dry triethylamine (9.3 mL, 66.7 mmol) followed by careful dropwise addition of methanesulfonyl chloride (9.3 mL, 118 mmol). After warming to room temperature and stirring and stirring for 30 minutes, the completed reaction solution was washed with 100 mL of water.

The structure of 5'-t-butyl carbamoyl-5'-deoxy-3'-O-mesyl-thymidine is as follows (T represents thymidine). ##STR3##

The CH.sub.2 Cl.sub.2 layer was separated and the aqueous layer was extracted with CH.sub.2 Cl.sub.2 (3.times.75 mL). The CH.sub.2 Cl.sub.2 extracts were pooled, dried over Na.sub.2 SO.sub.4 and evaporated. The residue was dissolved in 25 mL of ethyl acetate and filtered through a slurry of silica gel (3.times.7 cm) eluting with ethyl acetate.

The filtrate was evaporated and the residue was precipitated from H.sub.2 Cl.sub.2 -hexane: 5.83 g (86%) yield; m.p=141.degree. C. (dec); TLC (ethyl acetate), R.sub.f =0.65; IR (Kbr) 3420, 3030, 1712, 1690, 1360, 1175, 1049, 960, 916 and 530 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d. 7.56 (1H, s, 6-H), 7.16 (1H, br t, carbamoyl N--H), 6.13 (1H, t, J=7.3 Hz, 1'-H), 5.22 (1H, m, 3'-H), 4.06 (1H, m, 4'-H), 3.31-3.18 (5 H, s and m, mesyl methyl and 5'-Hs), 2.43 (2H, m, 2'-Hs), 1.80 (3 H, s, C(5)-methyl), 1.38 (9H, s, t-butyl).

5'-tt-Butyl carbamoyl-5'-deoxy-2,3'-anhydrothymidine (4 of FIG. 2). To a solution of 3 (5.70 g, 13.6 mmol) in 60 Ml of DMF was added a solution of potassium phthalimide (10.3 g, 55.6 mmol) in 17.0 mL of water. The resulting solution was heated to 100.degree. C. with stirring and then allowed to cool to room temperature and diluted with 300 mL of water.

The structure of 5'-t-Butyl-carbamoyl-5'-deoxy-2,3'-anhydrothymidine is as follows. ##STR4##

The phthalimide precipitate was filtered off and the filtrate was reduced to a .about.100 mL and then extracted with ethyl acetate (8.times.150 mL). The pooled ethyl acetate extracts were dried over Na.sub.2 SO.sub.4 and evaporated.

The residue was precipitated from ethyl acetate-hexane: 4.1 g (93%) yield; m.p.=82.degree. C. (dec); TLC (n-butanol-water-acetic acid, (5:3:2)), R.sub.f =0.54; IR (KBr) 3222, 1688, 1662, 1632, 1531, 1482, 1311, 1140, 880, 790 and 720 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 7.56 (1H, s, 6-H), 7.07 (1H, br t, carbamoyl N--H), 5.81 (1H, d, J=2.9 Hz, 1'-H), 5.24 and 4.24 (2H, 2 x m, 2'-Hs), 3.12 (2H, m, 5'-Hs), 1.75 (3H, s, methyl), 1.36 (9H, s, t-butyl). HRMS (EI) m/e 324.1575 (M+H).sup.+, calcd for C.sub.15 H.sub.22 N.sub.3 O.sub.5 324.1559.

3'-Azido-5'-tt-butyl carbamoyl-3',5'-deoxythymidine (5 of FIG. 2). To a solution of 4 (5.0 g, 15.5 mmol) in 50 mL of DMF was added a solution of NaN.sub.3 (5.0 g, 76.9 mmol) in 10 mL of water. The resulting solution was stirred at 100.degree. C. for 20 h, evaporated to dryness, and the residue was redissolved in 30 mL of water and extracted with ethyl acetate (4.times.70 mL). The pooled extracts were dried over Na.sub.2 SO.sub.4 and evaporated.

The structure of 3'-Azido-5'-t-butyl carbamoyl-3',5'-deoxy-thymidine is as follows. ##STR5##

The residue was precipitated from ethyl acetate-hexane: 4.90 g (86%) yield; m.p.=66.degree. C. (dec); TLC (10% hexane in ethyl acetate), R.sub.f =0.56; IR (KBr) 3345, 2980, 2110, 1690, 1518, 1470, 1370, 1272, 1159 and 1090 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 7.52 (1 H, s, 6-H), 7.17 (1H, br t, carbamoyl N--H), 6.05 (1H, t, J=6.6 Hz, 1'-H) 4.29 and 3.75 (1 H 2 x m, 3'-H and 4'-H, no assignments made), 3.23 (2H, m, 5'-Hs), 2.32 (2H, m, 2'-Hs), 1.81 (3H, s, methyl), 1.38 (9H, s, t-butyl). HRMS (CI) m/e 367.1733 (M+H).sup.+, calcd for C.sub.15 H.sub.23 N.sub.6 O.sub.5 367.1730.

5'-Amino-3'-azido-3',5'-deoxythymidine hydro-chloride salt (6 of FIG. 2). A solution of 5 (0.95 g, 2.60 mmol) dissolved in 30 mL of 30% trifluoroacetic acid in CH.sub.2 Cl.sub.2 was stirred at room temperature for 20 minutes. The completed reaction was evaporated and the residue was dissolved in 30 mL of absolute ethanol. Concentrated HCl (0.30 mL) was added and the solution was evaporated to dryness to afford a brown solid: 752 mg (96%) yield; m.p=78.degree. C. (dec); TLC (n-butanol-water-acetic acid, (5:3:2)), R.sub.f =0.39; IR (KBr) 3400, 3000, 2090, 1675, 1462, 1263 and 1062 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 8.27 (3H, br s, ammonium protons), 7.63 (1H, s, 6-H), 6.13 (1H, t, J=6.7 Hz, 1'-H), 4.57 and 3.96 (2H, 2 x m, 3'-H and 4'-H, no assignments made), 3.16 (2H, m, 5'-Hs), 2.62-2.30 (2H, 2 x m, 2'-Hs), 1.80 (3H, s, methyl). HRMS (CI) m/e 267.1205 (M+H).sup.+, calcd for C.sub.10 H.sub.15 N.sub.6 O.sub.3 267.1206.

The structure of 5'-Amino-3'-azido-3',5'-deoxythymidine hydrochloride salt is as follows. ##STR6##

3'-Azido-5'-N-Benzoylthiourea-3',5'-deoxythymidine (7 of FIG. 2). To a suspension of 6 (1.40 g, 4.62 mmol) in 60 mL of dry CH.sub.2 Cl.sub.2 was added 0.65 mL of dry triethylamine followed by dropwise addition of benzoylisothiocyanate (0.67 mL, 5.00 mmol).

The structure of 3'-Azido-5'-N-Benzoylthiourea-3',5'-deoxythymidine is as follows. ##STR7##

The suspension turned to a clear solution after a few minutes and the reaction was stirred at room temperature for a total of 30 minutes and washed with 40 mL of water. The aqueous layer was extracted with CH.sub.2 Cl.sub.2 (2.times.100 mL).

The pooled extracts were dried over Na.sub.2 SO.sub.4 and evaporated to dryness to afford a brown solid: 1.95 g (98%) yield; m.p=84.degree. C. (dec); TLC (ethyl acetate), R.sub.f =0.79; IR (KBr) 3180, 3045, 2106, 1695, 1550, 1518, 1272, 1176, 1072 and 712 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 7.93 and 7.56 (6H, d and m, J=6.7 Hz for the doublet, phenyl and 6-H), 6.16 (1H, t, J=6.7 Hz, 1'-H) 4.46 and 4.04 (4H, 2 x m, 3'-H, 4'-H and 5'-Hs, no assignments made), 2.41 (2H, m, 2'-Hs), 1.73 (3H, s, methyl). HRMS (CI) m/e 430.1295 (M+H).sup.+, calcd for C.sub.18 HN.sub.7 O.sub.4 S 430.1297.

3'-Azido-5'-thiourea-3',5'-deoxythymidine (8 of FIG. 2). To a suspension of 7 (1.0 g, 2.33 mmol) in 40 mL of methanol was added a solution of 1.16 g of potassium carbonate in 16.0 mL of water. The mixture was stirred at room temperature for 30 minutes. The methanol was evaporated and the remaining aqueous solution was diluted to 50 mL with water and adjusted to pH 8 with acetic acid. The solution was extracted with ethyl acetate (3.times.100 mL). The pooled extracts were dried over Na.sub.2 SO.sub.4 and evaporated.

The structure of 3'-Azido-5'-thiourea-3',5'-deoxythymidine is as follows. ##STR8##

The residue was precipitated from ethyl acetate-hexane: 619 mg (82%) yield; m.p=113.degree. C. (dec); TLC (n-butanol-water-acetic acid, (5:3:2)), R.sub.f =0.78; IR (KBr) 3330, 3205, 2106, 1678, 1473, 1277 and 1085 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 8.91 (1H, br s, 5'-NH), 7.41 (3H, br s, 6-H and CONH.sub.2), 6.13 (1H, t, J=6.6 Hz, 1'-H), 4.47 and 3.91 (2H, 2 x m, 3'-H and 4'-H, no assignments made), 3.74 (2H, m, 5'-Hs), 2.31 (2H, m, 2'-Hs), 1.78 (3H, s, methyl). HRMS (CI) m/e 326.1034 (M+H).sup.+ calcd for C.sub.11 H.sub.16 N.sub.7 O.sub.3 S 326.1035.

3'-Azido-5'-S-methylisothiouronium-3',5'-deoxythymidine Iodide (9 of FIG. 2). To a solution of 8 (280 mg, 0.86 mmol) in 12 mL of absolute ethanol was added methyl iodide (0.080 mL, 1.27 mmol). The reaction vial was sealed and the solution was allowed to stir for 14 h at room temperature.

The structure of 3'-Azido-5'-S-methylisothiouronium-3',5'-deoxythymidine iodide is as follows. ##STR9##

The completed reaction was evaporated to dryness and the residue was triturated with ether to afford a white solid: 326 mg (81%) yield; m.p.>80.degree. C. (slow darkening); TLC (n-butanol-water-acetic acid, (5:3:2)), R.sub.f =0.78; IR (KBr) 3160, 3050, 2118, 1698, 1650, 1475, 1280, 1094, 780 and 570 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 9.4-7.9 (br peak, N--H protons), 7.55 (1H, s, 6-H), 6.14 (1H, t, J=6.7 Hz, 1'-H), 4.42 and 3.93 (2H, 2 x m, 3'-H and 4'-H, no assignments made), 3.51(2H, m, 5'-Hs), 2.38 (2H, m, 2'-Hs), 1.82 (3H, s, methyl).

3'-Deoxythymidylyl-((3'-5')guanidyl)-3'-azido-3',5'-deoxythymidine Iodide (1 of FIG. 2). The trifluoroacetate salt of 3'-amino-3'-deoxythymidine (76 mg, 0.213 mmol) and 9 (100 mg, 0.213 mmol) were dissolved in 0.50 mL of dry pyridine. Anhydrous triethylamine (300 .mu.L) was added and the solution was stirred at 100.degree. C. for 40 h. The reaction solution was cooled to room temperature and then diluted with 6.0 mL of isopropanol. The tan colored crystals of 1 were collected (48 mg). Hexane was added to the filtrate to produce another crop of crystals (57 mg).

The structure of 3'-Deoxythymidylyl-((3'-5')guanidyl)-3'-azido-3',5'-deoxythymidine Iodide is as follows. ##STR10##

Total yield of 1: 105 mg (75%); m.p=115.degree. C. (dec); TLC (isopropanol-water-ammonium hydroxide, (7:2:1)), R.sub.f =0.43; IR (KBr) 3400, 3210, 3060, 2120, 1695, 1478, 1280, 1100, 785 and 570 cm.sup.-1 ; .sup.1 H NMR (DMSO-d.sub.6) d 11.37 and 11.32 (2H, 2 x s, thymidine ring N--Hs), 7.76, 7.56 and 7.45 (4H, 3 x br s, guanidinium protons), 7.67 and 7.46 (2H, 2 x s, 6-Hs), 6.15 and 610 (2H, 2 x t, J=6.0 and 6.5 Hz for the respective triplets, 1'-Hs), 5.40 (1H, t, J=4.5 Hz, 5-OH), 4.32 and 4.20 (2H, 2 x m, 3'-Hs), 3.82 (2H, m, 4'-Hs) 3.70-3.47 (4H, 3 x m, 2'-Hs), 1.80 and 1.78 (6H, 2 x s, methyl protons).

Addition of D.sub.2 O resulted in collapse of the thymidine ring N--H protons, the guanidyl protons and the 5'--OH. HRMS (FAB) m/e 533.2220 (M+H).sup.+, calcd for C.sub.21 H.sub.29 N.sub.10 O.sub.7 533.2221.

Molecular modeling

Model building and computations were performed on a Silicon Graphics 4D/320GTX workstation using CHARMM (Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J.; Swaminathan, S.; Karplus, M. (1983) J. Comput. Chem. 4, 187-217) v 22 (Adopted-Basis Newton Raphson (ABNR), steepest descents (SD) minimization, DNAH.RTF topology file) and QUANTA v. 3.3 (Nucleic Acid Builder, Molecular editor) programs (MSI, Waltham, Mass.).

Atom types C, NC and HC were used for the carbon, nitrogens and hydrogens of the guanidines, respectively. The X-ray structure for sodium thymidylyl-(5',3')-thymidylate-(5') hydrate (pTpT), found in the Cambridge structural database (Medical Foundation of Buffalo, Buffalo N.Y.) as refcode THYTHY10 (Camerman, N.; Fawcett, J. K.; Camerman, A. (1976) J. Mol. Biol. 107, 601-621) was used to build a model for the dimeric DNG 1.

The phosphate PO.sub.4 .sup.- of the nucleotide linkage of pTpT was overlapped with a molecule of guanidine ((H.sub.2 N).sub.3 C.sup.+). Two nitrogens overlap with O3' and O5', while the carbon assumes a position near the phosphorus. The third ---NH.sub.2 lands in between the two charged oxygens of >PO.sub.4 .sup.-.

The 5' terminal phosphate was truncated to an --OH, and an azide (--N.sub.3) was placed onto the position of the O3' terminal atom. Coordinates for the azide moiety were taken from an X-ray structure of tri-O-acetyl-b-D-xylopyranosyl-azide (refcode ACXPAZ in the Cambridge structural database).(Luger, P.; Paulsen, H. (1976) Acta Cryst. Sect. B 32, 2774).

ABNR minimization was subsequently applied, with atom constraints on the --N.sub.3 group to give the model of 1.

A Watson-Crick base-paired decameric duplex of B-DNA (d(Ap).sub.10 .multidot.d(Tp).sub.10) was made in Nucleic Acid Builder.(Almarsson, O.; Bruice, T. C.; Kerr, J.; Zuckermann, R. N. (1993) Proc. Natl. Acad. Sci. USA 90, 7518-7522; Almarsson, O.; Bruice, T. C. (1993) Proc. Natl. Acad. Sci. USA 90, 9542-9546) Hybrids of DNA.multidot.DNG were built by the overlapping procedure described for 1, on the d(Tp).sub.10 strand of d(Ap).sub.10 .multidot.d(Tp).sub.10 (for Watson-Crick duplex), or onto d(Tp).sub.10 strands in the structure of DNA d(Ap).sub.10 .multidot.(d(Tp).sub.10).sub.2 triple helix (Raghunathan, G.; Miles, H. T.; Sasisekharan, V. (1993) Biochemistry 32, 455-462) (for Watson-Crick and Hoogsteen base paired triplexes).

The modified backbone was adjusted by SD minimization with atom constraints on the positions of nucleobases and ribose rings. Constraints were then removed (ribose rings first, then nucleobases) with added distance constraints for base pairs. ABNR algorithm was used in subsequent minimizations. A CHARMM.RTF for the DNG unit was written by comparison with DNAH.RTF.

RESULTS

The general synthetic strategy for the formation of thymidyl DNG oligomers with a 5-OH terminal begins with the condensation reaction between 3'-amino-3'-deoxy-thymidine 10 and the thymidyl S-methylisothiourea derivative 9 (FIG. 1).

Chain extension is then possible by a two-step process, involving reduction of the 3'-azido group with Pd/C, in the presence of hydrogen, followed by condensation of the resulting amino dinucleoside with another equivalent of 9.

The chain-extending intermediate 9 was synthesized by an eight step pathway (FIG. 2) starting with 5'-amino-5'-deoxythymidine, which is conveniently prepared by the method of Horwitz et al. (J. Org. Chem 27:3045-3048 (1962)).

Functionalization of the 3'-position as the azide first required protection of the 5'-amino group. Thus, treatment of 5'-amino-5'-deoxythymidine with di-tert-butyl dicarbonate in dry DMF afforded the Boc-protected amino derivative 2, which crystallized from water after evaporation of the DMF.

The reaction of 2 with methanesulfonyl chloride in CH.sub.2 Cl.sub.2 produced the 3'-O-mesyl compound 3 which was then converted to the 2,3'-anhydronucleoside (Glinski, R. P.; Khan, M. S.; Kalamas, R. L.; (1973) J. Org. Chem. 38, 4299-4305) 4 upon treatment with potassium phthalimide in DMF. Ring opening of 4 with sodium azide in DMF-water (9) afforded the desired 3'-azido compound 5.

The IR spectrum of 5 shows an absorbance at 2110 cm.sup.-1, confirming the presence of the azido functionality. Compound 5 was converted to the S-methylisothiourea derivative by a four-step process beginning with the acid-catalyzed deprotection of 5 with trifluoroacetic acid in CH.sub.2 Cl.sub.2.

The product from this reaction, the ammonium trifluoroacetate salt of 6, proved difficult to precipitate and was converted to the hydrochloride salt, which solidified upon evaporation from ethanol.

The N-benzoylthiourea derivative 7 was prepared by the treatment of 6 with benzoyl isothiocyanate in CH.sub.2 Cl.sub.2 (Buschauer, A.; (1989) J. Med. Chem. 32, 1963-1970). Debenzoylation was achieved under alkaline conditions (K.sub.2 CO.sub.3 in aqueous methanol (Buschauer et al., supra) to produce the thiourea 8.

The reaction of 8 with methyl iodide in absolute ethanol (Buschauer et al., supra) afforded the S-methylated derivative 9.

The trifluoroacetate salt of 3'-amino-3'-deoxythymidine 10 was prepared by a two step procedure starting with 3'-azido-3'-deoxy-5'-O-tritylthymidine. Reduction of the azido group with H.sub.2 S in aqueous pyridine followed by removal of the 5'-trityl group with 30% trifluoroacetic acid in CH.sub.2 Cl.sub.2 afforded 10 in good yield. However, this procedure offered no advantage over previous methods (Saneyoshi, M.; Fujii, T.; Kawaguchi, T.; Sawai, K.; Kimura, S.; in "Nucleic Acid Chemistry", part 4, New York, 1991; pp. 67-72) for the synthesis of 10.

Condensation of 9 with 10 in a solution of dry pyridine-triethylamine at 100.degree. C. resulted in a 75% yield of dimeric thymidyl DNG 1.

The .sup.1 H NMR spectrum of 1 identifies two thymidine moieties present in a 1:1 ratio. The four guanidinium N--H protons are present in the region of the spectrum between 7.76-7.45 ppm. An exchangeable triplet corresponding to the 5'-OH proton appears at 5.40 ppm.

The presence of the 3'-azido moiety was confirmed by an IR absorbance at 2120 cm.sup.-1. High resolution mass spectral analysis of 1 provided the correct mass within 0.13 ppm. The HRMS analysis also provided an isotopic ratio of mass elements predicting the correct molecular formula: C.sub.21 H.sub.29 N.sub.10 O.sub.7.

Models of 1 and the hybrid complexes of decameric DNG with DNA strands. A structure for 1 (0) was modeled using the X-ray crystal structures for the thymidylyl-(5'>3')-thymidylate-(5') dinucleotide (pTpT) and tri-O-acetyl-b-D-xylopyranosyl-azide (Camerman et al., supra; Luger et al., Acta Cryst. Sect. B 32:2774 (1976).

Replacement of the phosphate of the nucleotide linkage with a guanidyl moiety mandates adjustment of the torsion angle C5'-N5'-C-N3' of the new linkage between the thymidine nucleosides. This angle is near -180.degree. in the model of 1, and is required for planarity of the guanidine.

For comparison, the dihedral counterpart in pTpT, C5'--O5'--P--O3' or a, is between -39.degree. and -73.degree. in typical conformations of DNA and RNA (Saenger, W. "Principles of Nucleic Acid Structure" Springer Verlag, New York, 1984) and is -66.3.degree. in the structure of pTpT. The change in the C5'--N5'--C--N3' torsion, brought about by energy minimization, causes the juxtaposition of the two thymine rings of 1 to alter, such that the bases are further apart than in the X-ray structure of pTpT.

Several options exist for the composition of base-paired oligomeric DNG-DNA hybrids. The following are combinations of decameric DNG and DNA strands which we consider at the present time (Scheme I): (i) A hybrid duplex of thymidyl DNG d(gT).sub.10 with a d(Ap).sub.10 complement of DNA (d(gT).sub.10 .multidot.d(Ap).sub.10); (ii) A hybrid triplex of dsDNA with a d(gT).sub.10 as a Hoogsteen paired strand in the major groove of the DNA (d(Tp).sub.10 .multidot.d(Ap).sub.10 .multidot.d(gT)10); and (iii) A hybrid triplex of d(Ap).sub.10 with two strands of d(gT).sub.10 (d(gT).sub.10 .multidot.d(Ap).sub.10 .multidot.d(gT)10).

(i) FIG. 11 shows a model of d(gT).sub.10 .multidot.d(Ap).sub.10 after prolonged energy minimization (2000 steps) in CHARMM. The backbones of the two strands approach one another due to electrostatic interactions between repeating --O--(PO.sub.2 .sup.-)--O-- and --NH--(C.dbd.NH.sub.2 .sup.+)--NH-units. It is interesting to note that it is the minor groove that contracts.

In gas-phase calculations this contraction is from .about.11.5.ANG. to .about.4.ANG. with retention of Watson-Crick base-pairing. These electrostatic interactions should be attenuated to some extent in water. The sugar puckering of the DNA portion remains the characteristic C2'-endo (as in B-DNA), whereas the ribose rings of the DNG strands resemble an O4'-endo conformation, with the five-membered rings approaching flat geometries. In the gas phase calculations a repeating hydrogen bond between N(5')H to O(4') in the same residue is observed. The latter may be responsible for the conversion of ribose sugar puckering from C(2')-endo toward O(4')-endo.

(ii) Building of d(Tp).sub.10 .multidot.d(Ap).sub.10 .multidot.d(gT).sub.10 ensued from the structure proposed for the d(Tp).sub.10 .multidot.d(Ap).sub.10 .multidot.d(Tp).sub.10 triple helix of DNA (Raghunathan et al., supra). The guanidinium ions of the Hoogsteen base-paired d(gT).sub.10 strand in the major groove of the decameric dsDNA reside quite near the phosphates of the purine strand, such that d(Ap).sub.10 and d(gT).sub.10 experience electrostatic attractions between the backbones of one another.

Energy minimization resulted in the model of 2. The structure reveals enforced d(Ap).sub.10 and d(gT).sub.10 backbone interactions. The Watson-Crick paired d(Tp).sub.10 is remote from the guanidine backbone of d(gT).sub.10 and is not involved in electrostatic interactions in the complex to any significant extent.

(iii) We finally consider the possible hybridization of d(gT).sub.10 with a purine strand of ssDNA, as well as strand invasion into dsDNA by d(gT).sub.10. A precedent from the recent antisense/antigene literature for antagonistic behavior toward dsDNA is found in the case of peptide-nucleic acids (PNA), Almarsson et al., Proc. Natl. Acad. Sci. USA 90, 9542 (1993), Almarsson and Bruice, Proc. Natl. Acad. Sci. USA 90, 9542 (1993).

In the final complex, a central triple helix is produced, leaving an extruded strand of ssDNA. Additional electrostatic interactions would be gained, due to the presence of two anionic DNA strands and two cationic DNG strands in the same complex.

The presented models for decameric duplexes and triplexes serve to indicate the possible modes of interaction of DNG polymers with DNA.

Attenuation of the strong electrostatic effect should be resolved by interspersing nucleotide sequences into a DNG oligomer. The lengthening of the base separations in A-RNA vs. B-DNA (Saenger, supra) suggest that DNG should exhibit specificity for DNA over RNA.

The absence of the 2'-0H in DNG structures also makes A-conformations of the ribose rings less likely to occur in the polymer.

EXAMPLE 2

This example is a prophetic example which teaches detection of HTLV-I anti-sense RNA in a HTLV-I infected cell using the polynucleoside chain of the invention.

Transcripts from HTLV-1 proviral genomes can be detected from the "anti-sense" strand as well as the "sense" strand that encodes the gag-pol-env genes. Labeled DNG oligonucleoside probes may be used to determine whether HTLV-I anti-sense RNA accumulates in infected cells using poly(A)+RNA blots of SLB1 (an HTLV-I positive human T cell line) and Molt4 (an HTLV-I negative human T cell line) RNA.

RNA can be isolated from the T cell lines and poly(A)+RNA selected using standard procedures (Sambrook, Fritsch, and Maniatis, Eds. Molecular Cloning, Cold Spring Harbor Press, 1989). One microgram of poly(A)+enriched RNA from each cell line may be loaded in a 1.2% agarose/2.2M formaldehyde gel and processed for blot analysis. The blot may be cut into strips having one lane of each RNA, Molt4 and SLB1. The blots will be hybridized at a temperature, ionic strength, probe concentration, and for a characteristic time. The precise parameters will be empirically determined as they are based on a the length and base composition, in particular the percentage of the molecule comprised of G's and C's, of the probe molecule (Maniatis, Sambrook, Fritsch, and Maniatis, Molecular Cloning: A Laboratory Manual, p. 11.45-49, Cold Spring Harbor Press (1989)). However, these parameters can be easily and routinely determined by one skilled in the art. As an example of conditions that can be used, an oligonucleotide 24 nucleotides in length could be hybridized for 6-8 hours at 42.degree. C. in 6.times.SSC (1.times.SSC:0.15M NaCl, 0.015M NaCitrate), 0.1% sodium dodecyl sulfate (SDS), and 0.5% (w/v) non-fat dry milk.

The oligonucleodides containing DNG can be labeled with gamma-.sup.32 P-ATP by kinasing the oligonucleoside molecules at their 5' ends. Alternatively, the oligonucleosides can be radiolabeled by using terminal deorynucleotidyl transferase to attach a labeled alpha-.sup.32 PdNTP to the 3'OH end of the oligonucleoside.

Oligonucleosides containing RNGs and DNGs could also be labeled by using non-radioactive methods. For example, 11-digoxigenin-11-DUTP, a nucleotide analog, can be added to the 3' end of DNG's using terminal deoxynucleotidyl transferase using a procedure developed by Bohringer-Mannheim (Indianapolis, Ind.). The digoxigenin-labeled probe can then be detected using an antibody specific for digoxigenin moiety. The antibody is conjugated to a reporter molecule, such as alkaline phosphatase that is detected upon incubation with an appropriate substrate.

Following hybridization, strips can be washed to remove non-specifically bound probe molecules using empirically determined conditions. For example, the blots may be washed three times each for five minutes at 42.degree. C. in 2.times.SSC, 0.1% SDS.

The strips will then be processed for detection of the probe molecule. In the case of radiolabeled probes, the strips will be exposed to XRay film.

As one skilled in the art will appreciate, DNG and RNG probes utilizing these methods can be also be used to recover cDNA and genomic clones containing these sequences.

EXAMPLE 3

Prophetic example of hybridization analysis to PS.beta.G RNA from placental tissue using polynucleoside chains of the invention

PS.beta.G is a glycoprotein expressed during pregnancy and at particularly high levels in certain gestational trophoblastic diseases, such as hydatiform moles. A sensitive assay to detect PS.beta.g RNA levels in placental tissue is a useful tool for diagnosing such diseases.

RNG or DNG oligounucleosides of the invention that are specific for the PS.beta.G transcript can be used to detect PS.beta.G RNA in placental tissue. Total RNA from two placental samples, one from a healthy term pregnancy and the other carrying a hydatiform mole will be isolated and poly(A)+RNA selected using standard protocols (Sambrook, Fritsch, and Maniatis). One microgram of poly(A)+enriched RNA from the sample of infected cells may be loaded in a 1.4% agarose/2.2M formaldehyde gel and processed for blot analysis. The blot may be cut into strips having one lane of each source of RNA.

The blots containing the may then be hybridized at a temperature, ionic strength, probe concentration, and for a specified time. The precise parameters will be empirically determined as they are based on a the size and base composition of the probe molecule (Maniatis, Sambrook, supra). However, these parameters can be easily and routinely determined by one skilled in the art. As an example of conditions that can be used, an oligonucleotide 24 nucleotides in length could be hybridized for 6-8 hours at 42.degree. C. in 6.times.SSC (1.times.SSC: 0.15M NaCl, 0.015M NaCitrate), 0.1% sodium dodecyl sulfate (SDS), and 0.5% (w/v) non-fat dry milk.

The probe is a polynucleoside version of the invention containing a portion of the published PS.beta.G sequence of U.S. Pat. No. 5,141,849. It is synthesized following the protocol described herein.

The probe can be labeled using gamma.sup.32 P-ATP by kinasing molecules at their 5' ends. Alternatively, oligonucleosides containing DNG can be radiolabeled by using terminal deoxynucleotidyl transferase to attach alpha-.sup.32 PdNTPs to the 3'OH ends of the molecule.

Oligonucleosides containing RNGs or DNGs could also be labeled by using non-radioactive methods. For example, 11-digoxigenin-11-DUTP, a nucleotide analog, can be added to the 3' end of oligonucleosides containing DNGs using terminal deoxynucleotidyl transferase using a procedure developed by Bohringer-Mannheim (Indianapolis, Ind.). The digoxigenin-labeled probe can then be detected using an antibody specific for digoxigenin moiety. The antibody is conjugated to a reporter molecule, such as alkaline phosphatase that is detected upon incubation with an appropriate substrate.

Following hybridization, strips can be washed to remove non-specifically bound probe molecules using empirically determined conditions. For example, the blots may be washed three times each for five minutes at 42.degree. C. in 2.times.SSC, 0.1% SDS.

The strips will then be processed for detection of the probe molecule. In the case of radiolabeled probes, the strips will be exposed to XRay film.

Claims

1. A cationic DNG polynucleoside chain consisting of multiple nucleosides, each nucleoside having a sugar component and a purine or pyrimidine base, the nucleosides being connected each to the other at the C5' and C3' positions of the sugar component of each of the nucleosides with a positively charged guanidyl linkage.

2. A cationic DNG polynucleoside chain consisting of multiple sugar components, each sugar component being connected to another sugar component by a positively charged guanidyl linkage and each sugar component being linked to a purine or pyrimidine base at the C1' position of the sugar component, the nucleosides being connected each to the other at the C5' and C3' positions of the sugar component.

3. An antisense molecule of claim 1 or 2 which is substantially complementary to at least a portion of a nucleic acid sequence.

4. The polynucleoside of claim 1, wherein the polynucleoside is a homomer.

5. The polynucleoside of claim 2, wherein the polynucleoside is a homomer.

6. The DNG molecule of claim 3, wherein the nucleoside sequence is complementary to a gene sequence.

7. A composition comprising an effective amount of the polynucleoside of claim 1 and a suitable carrier.

8. A composition comprising an effective amount of the polynucleoside of claim 2 and a suitable carrier.