CA1339935C - Dna and rna molecules stabilized by modifications of the 3'-terminal phosphodiester linkage and their use as nucleic acid probes and as therapeutic agents to block the expression of specifically targeted genes - Google Patents

Abstract

The use of oligodeoxynucleotides modified at the 3'-terminal internucleotide link as therapeutic agents by a method of hybridizing the modified oligonucleotide to a complementary sequence within a targeted mRNA and cleaving the mRNA within the RNA-DNA helix by the enzyme RNaseH to block the expression of the corresponding gene.

Description

133~93S

GRANT REFEI~ENCE
The invention described herein was made in part in the course of work under grants from the tJational Insti-tutes of Health, Grant Nos. Hl-33555 and AM-25295.

BACKGROUND OF TEIE INVENTION
It is well known that nucleic acids, i.e., deoxyribo-nucleic acids (DNA) and ribonucleic acids (RNA) are essential building blocks in living cells. These compounds are high molecular weight polymers which are made up of many "nucleotide" units, each such nucleotide unit being composed of a base (a purine or a pyrimidine), a sugar ~which is either ribose or deoxyribose) and a molecule of phosphoric acid. DNA contains deoxyribose as the sugar moiety and the bases adenine, guanine, cytosine, and thymine (which may be represented as A, G, C and T, respectively). RNA contains ribose instead of deoxyribose and uracil (U) in place of thymine.
The nucleotide units in DNA and RNA are assembled in de~inite linear sequences which determine specific biological functions. In the normal mammalian cell, D~A replicates itself, and also serves as the template ~or the synthesis of RNA molecules whose nucleotide sequences carry the inforlnation encoded by the ~NA.
RNA molecules serve several different functions within the cell. Messenger ~NA (mRNA) directs protein synthesis.
According to the well known ~atson-Crick model, DNA
molecules consist of two polynucleotide strands coiled about a common axis. The resulting double helix is held together by hydrogen bonds between complementary base pairs in each strand. ~ydrogen bonds are formed 'k 1339'~3~

only between adenine (A) and thymine (T) and between guanine (G) and cytosine (C). ~lence within the double helix, adenine (A) and thymine (T) are viewed as comple-mentary bases which bind to each other as A-T. The same is true for guanine (G) and cytosine (C), as G-C.
In a single polynucleotide strand, any sequence of nucleotides is possible. ~lowever, once the order of bases within one strand of a DNA molecule is specified, the exact sequence of the other strand is simultaneously determined due to the indicated base pairing. Accordingly, each strand of a DNA molecule is the complement of the other. In the process of DNA replication, the two strands act as templates for the synthesis of two new chains with complementary nucleotide sequences. The net result is the production of two new DNA molecules each containing one of the original strands plus a newly synthesized strand that is complementary to it.
This process is referred to as semiconservative repli-cation and allows the genetic information encoded with-in the DNA, specified by the nucleotide sequence, to be transmitted from one generation to the next.
Collectively, the genetic information of an organ-ism is termed the genome. The genome of bacteria and of all higher species is composed of DNA. The genome of viruses may be either DNA or ~NA. In any case, the genome of any particular species, whether a virus, bacteria, or higher organism has a characteristic nucleo-tide sequence wllich, stated simply, can he viewed as the "fingerprint" of that species.
In the process of transcription, DNA is copied, or transcribed, into RNA. T~le RNA molecule that is syn-thesized is single stranded. Its nucleotide sequence ~3t33~ 3rj is complementary to a segment of one cf the two strands of the DNA molecule, and hence it is an exact copy of the opposite strand except that thymine is replaced with uracil. The length of DNA that is copied to form a single RNA molecule represents one gene. Ihe human genome contains approximately 50,000 genes.
Messenger RNAs (mRNAs) carry the coding information necessary for protein synthesis. The sequence of nucleo-tides in the mRNA molecule specifies the arrangement of amino acids in the protein whose synthesis will be directed by that mRNA. Each set of three nucleotides, a unit called a codon, specifies a particular amino acid. The process by which mRNAs direct protein synthesis is termed translation. Translation of mRNAs occurs on ribosomes within the cytoplasm of the cell.
The flow of information frorn gene to protein can thus be represented by the following scheme:

Gene > mRNA > Protein For each protein made by an organism there exists a corresponding gene.
It is generally known in the field of molecular biology that oligodeoxyribonucleotides, short single-stranded DNA fragments, havin~ a nucleotide sequence complementary to a portion of a specified mRNA may be used to block the expression of the corresponding gene.
This occurs by hybridization ~binding) of the oligo-nucleotide to the mRNA, according to the rules of base pairing described above, which then prevents translation of the m~NA. It has now been discovered that the inhibition of translation observed is due to cleavage of the mRNA

1339~35 by the enzyme RNaseH at the site of the RNA-DNA double helix formed with the oligonucleotide, ~3ybridization of the oligonucleotide to the mRNA is generally not sufficient in itsel~ to significantly inhibit translation. Evidently an oligonucleotide bound to mRNA can be stripped off in the process of translation. An important corollary of this result is that for any modified oligonucleotide to efficiently block expression of a gene, the hybrid formed with the selected mRNA must be recognized by RNase~ and the mRNA
cleaved by the enzyme.
The process oE hybrid-arrested translation outlined herein has obvious implications for the use of oli~onucleotides as therapeutic agents. By selectively blocking the expression of a particular gene essential for the replication of a certain virus or bacteria an oligonucleotide could serve as an antimicrobial agent.
Similarly, an oligonucleoti~e targeted against a gene responsible or required for the uncontrolled prolifera-tion of a cancer cell would be useful as an anticancer agent. There are also applications in the treatment of genetic diseases such as sickle cell disease and thal-assemia for which no adequate treatment currently exists.
Any gene can be targeted by this approach. The problern of drug specificity, a major hurdle in the development of conventional chemotherapeutic a~ents, is immedia-tely solved by choosing the appropriate oligonucleotide sequence complementary to the selected mRr~A. This cir-cumvents toxicity resulting frorn a lack of selectivity of the drug, a serious limitation of all existing anti-viral and anticancer agents.

' ' ;41 1~3~ 3~

In addition to the use of oligonucleotides to block the express-on o~ selected genes for the treatment of diseases in man, other applications are also recognized.
Additional applications include but are not limited to _ ~ ~"
the use of such oligonucleotides in veterinary medicine, ~ D f~.
as pesticides or fungicides, and in industrial or agricul-tural processes in which it is desirable to inhibit the expression of a particular gene by an organism utilized in that process.
A major problem limiting the utility of oligonucleo-tides as therapeutic agents is the rapid degradation of the oligonucleotide in blood and within cells. Enzymes which degrade DNA or RNA are termed nucleases. Such enzymes hydrolyze the phosphodiester bonds joining the nucleotides within a DNA or R~A chain, thereby cleaving the molecule into smaller fragments.
In the past there has been some progress made in the development of oligonucleotide analo~s that are resistant to nuclease degradation, but the use of such derivatives to block the expression of specifically targeted genes has met with limited success. See, for example, Ts'o et al. U.S. Patent 4,469,863 issued Septem-ber 4, 1984; Miller et al. U.S Patent 4,507,433 issued March 26, 1985; and Miller et al. U.S. Patent 4,511,713 issued April 16, 1985. Each of the above patents have in common the objective of blocking the expression of selected genes, but the approach taken has been less than successful judging fr~m the high concentrations of oligonucleotide required and a corresponding lack of selectivity. The work described in each of the three 1~ ~9i3 ~

prior patents mentioned involves the u5e of oligonucleo-tides in which all of the phosphate groups have been modified in the form of methylphosphonates:
HO o~ B1 0 I - Cl~3-P~~
O ~ 2 o ~ B7 Oil Orl Normal Nucleotide Methylphosphonate Phosphodiester Diester B, and B2 are nucleic acid bases (either A, G, C or T).
In particular, U.S. Patent 4,469,863 involves inter alia, methylphosphonate modification of oligonucleotides.
U.S. Patent 4,507,433 involves a process for synthesiz-ing deoxyribonucleotide rnethylpllosphonates on polysty-rene supports; and U.S. Patent 4,511,713 involves deter-mining the base sequence of a nucleic acid and hybridiz-,ng to it an appropriately synthesized oligonucleotide methylphosphonate to interfere with its function.
The low level of activity observed usin~3 these rully modified methylphosphonate analogs led us to suspect that they would not form effective substrates for RNase~l when hybridized with mRNAs. Results present-ed in Example 2 show this to be the case, In an efFort to design new oligonucleotide analogs 1~3933S

o~ greater use as therapeutic agents, we undertook a study of the pathways by which oligonucleotides are degraded within blood and within cells. We have dis-covered that the sole pathway of degradation in blood (Example 3) and the predominant pathway in cells (Example 4) is via sequential de~radation from the 3'-end to the 5'-end of the DNA chain in which one nucleotide is removed at a time. This pathway of degradation is illustrated. These results provided for the first time a rational basis for the design of oligo-nucleotide analogs modified so as to inhibit their degradation without interfering with their ability to form substrates for RNaseli.
It has now been discovered that oligonucleotides modified at only the 3'-most internucleotide link are markedly protected from degradation within blood and within cells (Examples 5 and 6). Moreover, we have found that such derivatives have normal hybridization properties and do form substrates with mRNAs that are recognized and cleaved by RNaseH, thereby preventing expression oE the targeted gene (Example 7).
- The accomplishlnent oE the inhibition of expression of selected genes by oligonucleotides that are resist-ant to degradation, and that are, therefore, more effect-ive when used therapeutically, is the primary objective of this invention.
Other objectives oE the present invention include the preparation of oligonucleotides modified at the 3'-phosphodiester linkage and the formulation of the modified oli~onucleotide in a manner suitable for therapeutic 1 ~ 3 9 r~ 3 ~i use. It is also apparent that DN~ or RNA molecules so modified would be useful as hybridization probes for diagnostic applications, in which case, the modification of the 3'-internucleotide link would inhibit degrada-tion of the probe by nucleases which may be present within the test sample.

B~IEF DESCRIPTION OF TIIE DRAWINGS
Figure 1 is a schematic illustration to show how the expression of a selected gene is blocked in accord-ance with the process of this invention by hybridiza-tion of an oligonucleotide to its complementary sequence within the mRNA and cleavage of the mRNA by the enzyme RNaseH at the site of the RNA-DNA double helix.
Figure 2 illustrates the degradation of a single stranded DNA molecule by the action of a nuclease that removes sequentially one nucleotide at a time from the 3'- to the 5'-end of the chain.
Figure 3 is a drawing of an autoradio~ram demon-strating that the exclusive pathway of degradation of DNA fragments in hurnan blood is by exonucleolytic cleavage from the 3'- to the 5'-end of the DNA chain. The oligo-nucleotide utilized in this experiment, MA-15, has the sequence: 5'- GAGCACCATGGTTTC. In lanes 1-5, the oligo-nucleotide was radiolabeled at the 5'-end of the mole-cule with 32p. In lanes 6-10, the 32P-label is at the 3'-terminal internucleotide link. The labeled oligo-nucleotide was incubated in human blood at 37~C. Ali-quots were removed at the times shown and analyzed by polyacrylamide gel electrophoresis. The labeled bands ~339~35 --1 o--within the gel were visualized by autoradiography.
Figure 4 is a drawing o~ an autoradiogram demon-strating that an oligonucleotide in which only the 3'-most internucleotide link is modified is completely resistant to degradation in human blood. The oligo-nucleotide is a derivative of MA-15 in which the 3'-terminal internucleotide link is a 2,2,2-trichloro-1,1-dimethylethyl phosphotriester. Lane 1 is the oligonucleo-tide prior to incubation. Lanes 2-5 are aliquots taken after incubation of the oligonucleotide in blood for 15 minutes, 1 hour, 4 hours, and 18 hours, respectively.
The decrease in intensity of the original band is due simply to removal of the 5'-end label rather than de-gradation of the oligonucleotide. Pi is inorganic phosphate.
Figure 5 is a drawing of an autoradiogram of a Northern blot demonstratinq that oligonucleotides m~di-fied at the 3'- internucleotide link form substrates with mRNAs that are acted upon by RNaseH resulting in cleavage of the mRNA at the site of the RNA-DNA double helix. Lane 1 is mouse globin mRNA alone. Lane 2 is mouse globin mRNA incubated with RNasell in the absence of oligonucleotide. Lane 3 is (~lobin m~NA hybridlzed with MA-l5 and cleaved with RNasel~. In lane 4, the substrate for nNasel~ is ~ormed with the derivative of MA-l5 in which the 3'- terminal internucleotide link is a tri-chlorodimethylethyl phosphotriester. In lane 5, the oligonucleotide used was further modified with naphthylisocyanate at the 5'-0~l group of the molecule.
Samples were electrophoresed on a l.5S agarose gel.
After electrophoresis the gel was blotted onto (*)Gene A ( * ) Trade Mark 1~

(*) Screen Plus (DuPont), and bands were localized using P labeled MA-15 as a probe for a-globin mRNA.

SUMMARY OE' THE INVENTION
The in~ention comprises the method, means and composi-tion which together enable the use of oligonucleotides that are modified at the 3'-terminal phosphodiester linkage, and are thereby rendered resistant to degradation within cells and body ~luids, to selectively block the expression of a particular gene. The method involves the hybridization of the modified oligonucleotide to the corresponding mRNA to fonn a substrate fully capable of being recognized by the enzyme RNasel~, followed by the cleavage of the mRNA at the site of the RNA-DNA
double helix such that the expression of the targeted gene is blocked. The invention further details the use of DNA and RNA molecules modified at the 3'-terminal phosphodiester linkage as nucleic acid probes for diagnostic applications.

DETAILE~ DESCRIPTION O~ TI~E INVENTION
In the broadest terms, the objectives of the pre-sent invention are accomplished by modifying oligonucleo-tides at just the 3'-terminal phosphodiester linkage.
When this is accomplished it has been found that the oligonucleotide is markedly resistant to degradation by nucleases within cells and within blood. Equally important, such modified oligonucleotides hybridize normally with complementary nucleic acid sequences and do form substrates with mRNAs that are recognized and cleaved by RNaseH.

(*)Trade Mark 1~39935 As a result of the cleave of the mRNA, expression of the corresponding gene is selectively blocked as is desired in various therapeutic applications.
Recent advances in molecular biology employing recombinant DNA techniques have led to new diagnostic and therapeutic strategies. In DNA diagnostics, a DNA
or RNA probe is used to detect the presence of a comple-mentary nucleic acid (DNA or RNA) sequence. In this process the probe hybridizes to its complementary se-quence if it is ~resent within the sample. The target se~uence may be a DNA or RNA molecule of a bacteria or virus, or an altered gene leading to a genetic abnor-mality such as sickle cell disease. Frequently, the sample to be analyzed is immobilized on a solid support such as nitrocellulose or a nylon membrane. The probe carries a label, e.g., a radioactive, fluorescent, or enzyme marker to permit its detection. The field of DNA
diagnostics is currently under very active inYeStigation~
having applications in the diagnosis of infectious diseases, cancer, and genetic disorders. E~owever, there are as yet no products in widespread clinical use.
Existing metl-ods, although extremely useful as research tools, lack adequate sensitivity for many ~ractical applications, and are often too complicated to be carried out routinely and reliably in a clinical laboratory.
In DNA therapeutics, an oligonucleotide is used to block the expression o~ a specifically tar~eted gene hy hybridizing to a complementary sequence within the corresponding mRNA and preventing the mR~A from being translated. In this manner, synthesis of the protein 133~33~

encoded by the gene is blocked. Such is the basic con-cept employed in each of the above three mentioned prior patents.
However, recent studies which formed the essence of the present invention have shown that blocking of the expression of the targeted gene occurs not simply by the hybridization of the oligonucleotide to the mRNA, but requires that the RNA-DNA double helix so formed serve as a substrate for the enzyme RNase~l. RNaseH, a ubiquitous enzyme required for DNA replication, then digests the m~NA at the RNA-DNA duplex. Only the mRNA
is cleaved. The oligonucleotide is released intact and can react repeatedly with other mRNA molecules as indicated by the dashed line in Figure l. The effect of each molecule of the oli~onucleotide is thereby amplified, increasing the efficiency by which the expression of the targeted ~ene is blocked.
RNaseH acts only on RNA-DNA double helices. RNA-RNA duplexes do not serve as substrates for the enzyme.
Therefore, oligodeoxyribonucleotides (short single-stranded DNA fragments) are preferred as therapeutic agents, rather than ~NA molecules. Either RNA or DNA
molecules can be used as probes for diagnostic purposes.
One problem faced in both DNA diagnostics and DNA
therapeutics is the degradation of the DNA or RNA mole-cule used either as a probe or tllerapeutic agent by nydrolytic enzymes, termed nucleases, present in body fluids and tissues. This problem is of far greater concern in the case of DNA therapeutics. Oligonucleo-tides are degraded in blood, and even more rapidly 1~3!~93~

within cells, the rate of degradation being suffici-ently fast to completely abrogate the effect of the drug. In preparing clinical samples for DNA diagnostic tests, efforts are made to remove contalninating nuc-leases. However, low levels of activity may remain. In certain instances, particularly when applied to genetic tests, the detection method is based on specific cleav-age of the probe when it is hybridized to the target sequence. Non-specific degradation of the probe can give rise to an increased level of background signal, obscuring detection o~ the target-dependent cleavage of the probe.
Nucleases catalyze the degradation of DNA or RNA
strands by hydrolysis of the phosphodiester bonds which join the nucleotide units within the chain, thereby cleaving the molecule into smaller fragments. DNA or RNA molecules useful as hybridization probes range from approximately lS nucleotide5 in length to over several thousand nucleotides. DNA fragments useful as therapeutic agents range from about 10 to about 75 nucleotides in - length, preferably from 15 to 35 nucleotides. Thus, any DNA or RNA molecule used as a probe or therapeutic asent would have many potential sites at which cleav-eage by nucleases may occur.
It is ~enerally known in the field of molecular biolo~y that phosphodiester linkages normally subject to enzymatic cleavage by nucleases can be protected by modifying the phosphate ~roup. Earlier efforts, such as the three mentioned patents, have focused on deri-vatives in which all of the phosphate groups within the X

~3933~

strand are modified, or on derivatives in which the phosphate groups are extensively modified in a random fashion following the synthesis of the oligonucleotide (Tullis, R.H., P.C.T. W083/01451, 1983). Such modifica-tions, however, often interfere with hyhridization of the DNA to its target sequence, and consequently limits its utility for both diagnostic and therapeutic applica-tions. Moreover, it has been found that D~A fragments in which the phosphate groups are extensively modified do not form substrates for RNaseH (Example 2). This abrogates their use as therapeutic agents.
While not wishing to be bound by any theory, the important discoveries underlying the present invention are: (1) that the mechanism by which oli~onucleotides used as therapeutic a~ents block the expression of specifically targeted genes is to hybridize to the corresponding mRNA, forming a substrate for RNaseH
which then cleaves the m~NA; and (2) that the exclusive pathway of degradation of oligonucleotides in blood, and the predominant pathway within cells, is via sequential degradation from the 3'- to the 5'- end of the chain as outlined in Figure 2. It has thus proven possible to greatly increase the stability of DNA fragrnents by modification of only the first internucleotide link from the 3'- end of the DNA molecule. ~hen this is done, cleavage of the first nucleotide unit is blocked, and no further degradation can proceed. A variety of substitutions at the 3'- terminal internucleotide link, as explained below, will serve this purpose, the position of the modification being the important factor. As demonstrated in Example 7, such modified derivatives . ~

133933~

have normal hybridization properties and do form sub-strates with mRNAs that are recognized and cleaved by ~NaseH. Such modified oligonucleotides, thus, are useful for both diagnostic and therapeutic applications.
Oligonucleotides, althou~h highly negatively charged molecules, do enter cells at a ~inite rate. This was first suggested by a study carried out by Zamecnik and Stephenson (Proceedings of the National Academy of Sciences USA (1978) 75, 280-2~4 and 285-28~), in which an oligonucleotide ~omplementary to the 5'- and 3'-repeat sequences of Rous sarcoma virus were found to inhihit ' viral replication and the synthesis of viral proteins in chick embryo fibroblasts. In these studies, the mechanism o~ action of the oligonucleotide was not established, nor was the possiblity that the effects observed were due to degradation prod~ucts of the oligo-nucleotide ruled out. Nonetheless, ihis study, with more recent work, Heikkila et al., Nature (1987) 328, 445-449, indicates that oligonucleotides do enter cells.
Zamecnik and Stephenson also showed that the activity of the oligonucleotide could be substantially increased by modifying the 3'-OEI and 5'-OEI groups with large apolar substituents, i.e., phenyl and naphthyl g~oups.
These large apolar groups enhance the binding of the oligonucleotide to the cell membrane and transport into the cell. The introduction of these groups may also have inhibited the degradation of the oligonucleotide by e~onucleases. E~owever, only by directly modifying the phosphodiester group is it possible to precluAe cleavage of the internucleotide link within a DNA or RNA strand. Neither the degradation of the native 133~335 oligonucleotide nor the modified~ derivatives wer~ studied in this work. Even the most recent study of'the degradation of oligonucleotides is completely silent as to the pathway by which degradation occurs ~Wickstrom, E.
(1986) Journal o~ Biochemical and Biophysical Methods 3, 97-102).
Thus, while there has been some study of the degrada-tion of DNA and RNA by endogenous nucleases within human and other mammalian tissues previously, it has not hithertofore been appreciated that degradation of single-stranded DNA molecules occurs predominantly by a 3'- to 5'- exonucleolytic mechanism. Hence the unique value of the modified derivatives claimed herein could not have been predicted. Put another way, because the pathway of degradation had not been established, no basis existed for the design o~ DNA molecules in which only selected phosphate groups within the chain were modified to effect an increase in stability, such that the resulting oligonucleotide still hybridized to mRNAs to fonn substrates that are reco~nized and cleaved by RNaseH, as is required for use of the oli~onucleotide in therapeutic applications.
As heretofore mentioned, the exact chemical modi-fication at the 3'-phosphodiester linkage is not as important as that the modification be at this precise location. A limited number of other positions on the molecule may also be modified without departing from the scope of this invention. In the case of DNA diagnos-tics, it is necessary to attach a label onto the nucleic acid probe to enable its detection. A radioactive, fluorescent, electrochemical, chemilumenescent or enzyme 133~

marker may serve this purpose. For oligonucleotide probes ranging in length from 15 to about 35 nucleotides in length generally only one label is appended to the molecule. ~nger DNA or RNA molecules use~ as probes may be up to several thousand nucleotides in length. In this case the label may be introduced at a density of up to 1 for every 15 nucleotide units in the chain.
For applications in DNA therapeutics, the oligo-nucleotide may be modified at positions in addition to the 3'-phosphodiester linkage to increase its stability further or to facilitate its transport into cells.
Transport of the oligonucleotide into the target cell may be enhanced by attachment to a carrier molecule normally taken up by the cell such as transferrin the plasma iron binding protein, or epidermal growth factor, or the oligonucleotide may be attached to an antibody against a surface protein of the target cell as in the case of the so-called im~unotoxins. Alternatively, additional phosphate groups of the oligonucleotide may be modified to reduce the ne~ative charge on the molecule and thereby facilitate its transport across the lipid bilayer of the cell membrane. With such modifications, apolar substituents such as hydrocarbon chains or aro-matic groups may also be incorporated into the molecule to further enhance intracellular uptake. Modification of additional phosphate groups within the chain may also increase the stability of the oligonucleotide further by blocking the activity of 5'- to 3'- exonucleases and endonucleases that are present in certain cells (see Example 4). Although degradation by these pathways 133~1335 is much less rapid than degradation from the 3'- end of the chain, the added effect of blocking these activities as well, particularly 5'- to 3'-exonucleolytic cleavage, may be beneficial in many cases. In these various applications, however, it is important that the modifica-tions be restricted to a limited number of the phospho-diester groups. If most or all of the phosphodiester linkages within the chain are modified this will frequently interfere with the hybridization properties of the oligonucleotide, and more importantly, it will preclude the use of such derivatives as therapeutic agents because they will no longer form substrates with targeted mRNAs that are acted upon by RNase~l. It is essential that there be one or more continuous stretches of the oligonucleotide chain in which the phosphodiester linkages are unmodified. The length of this unmodified portion of the chain must be at least 4 nucleotides, and preferably greater than 7 nucleotides in length.
When this is accomplished, the oligonucleotide will be protected from degradation due to the modification of the 3'-terminal phosphodiester linkage, and will also hybridize with mRNAs to form substrates for RNaseH, which will cleave the m~NA across from the unmodified portion of the oligonucleotide, enabling the use of such derivatives as therapeutic agents to block the expression of selected genes. The method of modifying the 3'-phosphodiester linkage will now be described.
The modifying moiety is not critical. It can be selected from any of the following functionalities:

:~339~335 R-O-P=O ~ O-P=O alkyl or aryl O O phosphotriesters R-P=O ~ P=O alkyl or aryl O O phosphonates ~_p=o hydrogen O phosphonates R O
R-N-P=O ~ N-P=O alkyl or aryl Il O 11 I phosphoramidates S-P=O phosphorothioates o Se-P=O phosphoroselenates o ~39935 In these structures, R may vary in length from 1 to about 20 carbon atoms, and may contain halogen substi-tuents (as in Examples 5 and 6) or other heteroatoms, e.g., N, O, or S, attached to the chain. The methods required to prepare these derivatives are simple, straight-forward reactions well known to those skilled in the art. See for example, the following literature refer-ences, each of which discloses modifications of the type expressed herein: Letsinger, et al., Journal of the American Chemical Society, t1982), 104, 6805-6806;
Stec et al., Journal of the American Chemical Society (1984), 106, 6077-6079; Miller et al., ~iochemistry (1986), 25, 5092-5097; and Froehler, et al., Tetrahedron Letters, (1986), 27, 469-472.
The preferred modification is a phosphotriester, for simplicity of synthesis and the diversity of func-tional groups which may be attached to the phosphorus atom. Incorporation of a large l-ydrophobic substituent at this position, for example, may be utilized to faci-litate intracellular uptake. As demonstrated in the present work, phosphotriesters are also resistant to hydrolysis by nucleases, and when incorporated at only a limited number of positions within an oligonucleotide chain, the molecule remains able to hybridize normally with m~NAs to form suhstrates which are cleaved by RNaseH.
In all methods now commonly used, oligonucleotides are synthesized from the 3'- to the 5'- end of the chain. The first residue is coupled to a solid support, such as controlled pore glass, through the 3'-OH group.

1~39935 ~f the synthesis is carried out in solution, rather than on a solid support, the 3'-0~1 of the first residue is blocked by a protecting group which is removed at the culmination of the synthesis. Usually, one nucleo-tide unit is added at a time to the 5'-0~1 group o~ the growing chain. Alternatively, a block of several resi-dues may be added in a single reaction. There are several methods by which the internucleotide link to the 5'-0~ group may be formed. These procedures are reviewed in the ~ollwing monograph, "Oligonucleotide Synthesis: A Practical Approach" (Gait, ~.J., ed) IRL
Press, Oxford (l9~4), The following reaction scheme illustrates the phosphoramidite method, the preferred chemistry at present.

~_O ~ O B2 DMT-O ~ O 2 + ~ ~I
coupling 1~

3 ~ o~ ~ 1 S . S . C=O
~2 oxidation V

~3~5 ~IT-O-~ o ~ 2 o ~-0- P=O
~1~ O ~ 1 ~1 o C=O
~2 DMT is the dimethoxytrityl protecting group; S.S. is the solid support on which the synthesis is carried out. Bl and B2 are nucleic acid bases (either A, G, C or T).

~3~9~35 The immediate product of this reaction is a phospho-triester. At the end of the synthesis, the protecting group R is removed to yield a phosphodiester linkage.
Typically, the ~-cyanoethyl group is used for this purpose. Alternatively, R may be a group which is not removed, in which case the final product is a phos-photriester, the group R remaining permanently attached to the phosphorus atom. It is in this manner that a modified phosphotriester may be incorporated at the first internucleotide linkage from the 3'-end of the chain. By essentially the same chemistry an alkyl or aryl phosphonate may be introduced specifically at this position, in which instance the group R is attached directly to the phosphorous atom rather than through oxygen. Similarly, a hydro~en phosphonate may be intro-duced at this position. Such derivatives may be used directly, or oxidized to give phosphotriesters or phos-phoramidates incorporated at the 3'-terminal internucleotide linkage. The following reference discloses modifications of the hydrogen phosphonate group, Froehler, B.C., Tetrahedron Letters, ~1986), 27, 5575-5578. A phosphoro-thioate or phosphoroselenate group may be incorporated at the 3'-terminal internucleotide link by carrying out the oxidation step in the phosphoramidite method with sulfur, or YSeCN, respectively, Stec. et al., Journal of the American Chemical Society, (1984), 106, 6077-6079. At the completion of the synthesis, following the removal of all protecting groups, the final prol~uct may be used directly or further purified by chromato~raphic methods well known to those skilled in the art, see for 1~3'-3~

example "Oligonucleotide Synthesis: A Practical Approach"
(Gait, M.J., ed.) IRL Press, Oxford (1984).
The procedures just described are generally appli-cable for the synthesis of oli~onucleotides up to approxi-mately 100 residues in length. ~n order to synthesize a longer DNA or RNA molecule in which the 3'-terminal phosphodiester group is modified, as may be required for use as a hybridization probe in diagnostic tests, a combination of both chemical and enzymatic methods is used. To accomplish the synthesis of such a sequence, a short oligonucleotide in which the 3'-terminal phospho-diester linkage was modified, prepared by chemical methods, is enzymatically 1isated onto the 3'-OH group of a longer DNA or RNA chain.
Methods of formulation and administration of the modified oligonucleotides described herein are obYious to those of ordinary skill in the medical arts, see for example, Avis, K. (1985) in Remington's Pharmaceutical Sciences, ~Gennaro, A.R., ed) pp. 151~-1541, Mack Publishing Company, Easton, Pa.
Such method9 of administration may include but are not limited to t~pical, oral, or parenteral routes depending on the disease state.
Appropriate vehicles for parenteral administration include 5~ dextrose, normal saline, Ringer's solution and Ringer's lactate. The material, of course, must be sterile and substantially endotoxin free. It may be possible in certain cases to store the product as a lyophilized powder which would then be reconstituted when needed by addition of an appropriate salt solution.
Ihe following examples are offered to further illus-trate, but not limit, the process, product and medical .
~'t': ~, r ¦l 3 5~ 3 3 ~

techniques of the invention, and serve to point out the unique features of the invention which enable it to over-come limitations of earlier contributions to the field.

EXAMPLE
Demonstration that the Inhibition of Translation of a Targeted mRNA by a Complementary Oli~onucleotide is Dependent on the Cleavage of the mRNA by RNaseH
For the purpose of these experiments two oligonucleo-tides were synthesi~ed: 5'-GAGCACCATGGTTTC, MA-15;
ands 5'-TGTCCAAGTGATTCAGGCCATCGTT, CODMB-25. Both were prepared on a Beckman automated DNA synthesizer using the phophoramidite method. MA-15 is complementary to a se~uence within mouse ~-globin mRNA spanning the initiation codon. CODMB-25 hybridizes to a sequence within the coding region of mouse ~-globin mRNA. Mouse globin mRNA was isolated from reticulocytes obtained from mice rendered anemic by treatment with phenylhydrazine as described by Goosens and Kan (Methods in Enzymology (Antonini, E., Rossi-~ernadi, L., and Chiancone, E., eds.) 76, 805-817, 19~1). In vitro translation reactions were carried out in the rabbit reticulocyte lysate system. The reticylocyte lysates used were purchased from ProMega Biotec, and contain RNase~l. Translation reactions were programmed with 1 u~ of total mouse reticulocyte RNA. The reactions were allowed to proceed for 1 hour at 30~C. Each sample contained 45 uCi of [35S~methionine (Amersham) in a total volume of 25 ul.
Protein synthesis was measured by incorporation of [35S]methionine into material precipitatable in 10%

~39935 trichloracetic acid. In control reactions in the ahsence of oligonucleotide, synthesis of ~ globin accounts for 60~ of the incorporated counts; ~ globin synthesis accounts for 40% of the radioactivity incorporated into protein. Reactions were carried out in both the presence and absence of poly rA/oligo dT t500 ug/ml). Poly rA/oligo dT is a competitive inhibitor of RNaseH, and at the concentration used almost completely blocks the activity of the enzyme. Both MA-15 and CODMB-25 very effectively inhibited translation of the targeted mRNA
(see Table 1). MA-15 is complementary to 12 of 15 residues in the corresponding sequence of ~-globin mRNA, and, therefore, also inhibits ~-globin synthesis to some extent. Poly rA/oligo dT completely blocked the inhibitory effect of the oligonucleotides, indicating that the arrest of translation is mediated entirely by the cleavage of the mRNA by RNaseH.

Table 1. Role of RNaseEI in ~ybrid-Arrested Translation % Inhibition of Globin Synthesis Oligonucleotide - poly rA/oligo dT +poly rA/oligo dT
MA-15 (8 M) 51 3 CODM~-25 (4 uM) 61 5 Oligodeoxynucleotide Methylphosphonates Do Not Form Substrates for RNaseH
Two related substrates for RNaseE~ were compared:
p y rA/oligO dTl2_18 and poly rA/oligo dT18 methyl-phosphonate (MP). Oligo dT12 18 was purchased from ~3~9~3~

Pharmacia PL Biochemicals; the c}-ain length varies from 12 to 18 residues. Oligo dT18MP was synthesized on a Beckman automated DNA synthesizer using the phosphon-amidite method. The 5'-terminal internucleotide link is a normal phosphodiester; the remaining 16 phosphate groups in the chain are methylphosphonates. Poly rA
labeled with tritium was purchased from Amersham.
Each reaction was conducted in a final volume of 100 ul containing 44 picomoles of the substrate, either H-2~1Y rA/oligo dT12 18 or H-poly rA/oligo dT18MP, in 50 mM Tris-HCl buffer pH 8.0, 20 mM KCl, 9 m~ MgC12, 1 mM 2-mercaptoethanol and 25 ug of bovine serum albumin.
The reaction was initiated by the addition of 2.3 units of E.coli RNase~ (Bethesda Research Laboratories) and allowed to proceeed for 30 minutes at 37~C. After the reaction, 0.2 ml of carrier DNA solution (0.5 mg/ml of salmon sperm DNA, 0.1 M sodium pyrophosphate, and 1 mM
ethylenediamine tetraacetic acid) plus 0.3 ml of 10%
trichloracetic acid was added to each sample. The tubes were then incubated on ice for 20 minutes, and afterwards centrifuged at 16,000 X g for 15 minutes.
Cleavage of the poly rA strand by RNase~ releases smaller fragments that are not precipitated by trichloroacetic acid, and hence remain in the supernatant. The super-natant was carefully removed, mixed with liquid scin-tillation cocktail, and counted for radioactivity. The reaction with poly rA/oligo dT proceeded nearly to completion - 85% of the radioactive counts were solu-bilized. In contrast, less than 1~ of poly rA/oligo dT18MP was digested by the enzyme. Thus, when much or all of the phosphate backbone o~ an oligodeoxynucleotide ~39~35 -2~-analogue is modified, it does not form a substrate with the complementary RNA sequence that is recognized and cleaved by RNase~.

Oligonucleotides are Degraded in Llood Exclusively by a 3'- to 5'-Exonucleolytic Activity The purpose of this study was to establish the pathways by which oligonucleotides are degraded in blood. Surprisingly, a single pathway was observed:
sequential degradation from the 3'-to the 5'- end of the DNA molecule.
The oligonucleotide MA-15 described in Example 1 was radioactively labeled with 32p at the 5'- end of the molecule using y32P-adenosine triphosphate (y 32p_ ATP, Amersham) and T4 polynucleotide kinase (Bethesda Research Laboratories):
5-*GAGCACCATGGTTTC
where the asterisk represents the radiolabel. Separ-ately, a radiolabeled derivative was prepared in which 32p was introduced at the 3'most internucleotide link usiny ~3 P-deoxycytidine triphosphate (dCTP) and CNA
polymerase I ~5ethesda Research Laboratories):
5'-GAGCACCATGGTTT*C
The reaction catalyzed by DNA polymerase I is between 5'- GAGCACCATGGTTT and dCTP and requires the comple-mentary DNA strand as a template. A 50 fold excess of unlabeled MA-15 was added after the reaction to insure that the labeled derivative is single stranded. These ~rocedures are standard methods for radiolabeling DNA
fragments, see for example, Maniatis, T., Fritsch, E.F., and Sambrook, J., Molecular Cloning: A Laboratory 1339.~3~

r1anual, Cold Spring Harbor Laboratory, New York, pp.
114-115 and 122-126, lg82~

The radiolabeled oligonucleotide (5 X 10 counts/min) plus 2.5 nanomoles o~ unlabeled MA-15 were added to 48 ul of freshly ~rawn human blood anticoagulated with heparin. Samples were incubated at 37~C, body temperature.
Aliquots of 6 ul were removed at 15 minutes, 30 minutes, 1 hour, and 2 hours. A fraction of each aliquot was loaded onto a polyacrylamide gel and electrophoresed to separate the products on the basis of size. To visual-ize the radiolabeled products, the gel was exposed to x-ray film overnight. A drawing o~ the autoradiogram is shown in Figure 3.
Degradation of the oligonucleotide la~eled at the 5'- end of the molecule (lanes 1-5) gave rise to a ladder of discrete products of decreasing size indicative of sequential exonucleolytic cleavage of the strand from the 3'- toward the 5'- end of the chain. Had cleavage occurred within the middle o~ the strand, products of intermediate length would have appeared throughout the time course. When ~1A-15 was labeled at the 3'-end of the molecule, the resulting autoradiogram, lanes 6-10, showed, over time, disappearance of the full-length 15-mer, and a corresponding increase in the mononucleotide pC. This result likewise demonstratcs a 3'- to 5'- exonuc-leolytic activity. The lack of products o~ intermediate size rules out the presence of a 5'- to 3'- exonucleolytic activity, or an endonucleolytic (internal) cleavage activity.

i339i335 By the same methods as just described, we showed that in blood from other mammalian species, specifically mouse, rabbit, and rat, oligonucleotides are also degraded solely by sequential cleavage of mononucleotides from the 3'-end of the chain.

The Principal Pathway of ~gradation of Oligonucleoti~es in E~uman Cells is by Sequential Exonucleolytic Cleavage of the Strand From the 3'- to the 5'- End of the Chain The purpose of these studies was to determine the pathways by which oligonucleotidec are degraded by nuc-leases present within human cells. Radiolabeled deri-vatives of MA-15 were prepared as in Example 3. Degrada-tion of the oligonucleotide was studied in extracts prepared from the human erythroleukemia cell line K562.
The K562 cell line was first isolated by Lozzio, C.B., and Lozzio, B.B., Blood (1975) 45, 321-334, and has been studied extensively as a mo~el for red cell differ-entiation. The cells were suspended in 10 mM tris buffer p~3 7.4 with 5 mM MgC12, 1 mM CaC12, 100 mM NaCl, and 0.1% triton X-100 and homogenized with a cold motor driven teflon pestle homogenizer for 30 seconds. Cell membranes and organelles were furt3~er disrupted by sonication by five 15 second bursts using a Biosonic sonifier set cn full power. Remaining cell debris was removed from the extracts by cerltrifugation for 5 minutes at 12,000 X g.
~ Trade Mark 133993~

The radiolabeled oligonucleotide was incubated in KS62 cell extracts under the same conditions as decribed in Example 3 and the degradation products analyzed by polyacrylamide gel electrophoresis. Degradation o~ the oligonucleotide was much faster than that in blood.
The primary pathway of degradation was again a 3' to 5' exonucleolytic activity. The half-life for cleavage of the 3'- tenminal ' internucleotide link was 2 minutes.
A slower 5'- to 3'-exonucleolytic activity was also observed. The half-life for cleavage of the S'- terminal nucleotide unit from the chain by this mechanism was 20 minutes.
Using the same methods as above, the pathways of degradation of oligonucleotides in extracts prepared from other mammalian cells have also been studied.
These systems include African green monkey kidney cells tCOS cells) and mouse liver cells. The results suggested a low level of endonucleolytic activity (cleavage in the middle of the strand), as well as a S'- to 3'-exo-nuclease, but again the major pathway of degradation of the oligonucleotide in each of these systems was by a 3'- to 5'-exonuclease.

EXAMPLE S
Modification of Oligonucleotides at the 3'- Terminal ~nternucleotide Link Completely E310cks Their Degradation in Eluman Blood A derivative of the oligonucleotide MA-lS was prepared in which the first internucleotide link from the 3'-end of the molecule was modi~ied as a 2,2,2-trichloro-1,1- dimethylethyl phosphotriester. The phosphotri-ester was introduced into the chain using the corres-13 39~33~

ponding phosphorodichloridite:
CH3 Cl Cl C--C-O-P
CH3 Cl To 20 micromoles o~ 5'-deoxythymidine dissolved in 1 cc of 10~ pyridine in acetonitrile was added 19 micromoles of the phosphorodichlorodite. The reaction was allowed to proceed for 5 minutes at room temperature. The result-ing monochlorodite was added to 1 micromole of deoxycy-tidine attached to controlled pore glass beads (American ~ionetics). The coupling reaction to produce the dinucleo-tide phosphotriester was allowed to proceed for lS
minutes at room temperature. Excess reagents and by-products of the reaction were washed away with anhydrous acetonitrile. The remaining portion of the DNA strand was synthesized by the ~- cyanoethyl phosphoramidite ~ethod on a Beckman automated DNA synthesizer. Only the 3'- terminal internucleotide link is modified. The remaining 13 phosphate groups in the chain are normal phos~hodiesters.
The modified oligonucleotide was labeled with 32p at the 5'-terminal OH group using y P-ATP and T4 polynucleotide kinase as descri~ed in Example 3. The radiolabeled oligonucleotide was incubated in human blood at 37~C. Aliquots were removed at various times and analyzed by polyacrylamide gel electrophoresis. A
drawing of the autoradiogram of the gel is shown in Figure 4. ~o degradation of the oligonucleotide occurred over a period of 18 hours. The gradual decrease in the intensity of the band due to the full length molecule was due to removal of the 32p label from the olisonucleotide 13 3~3~

by a phosphatase present in blood, not to cleavage of the strand.
Similar studies were carried out in which the modi-fied oligonucleotide was incubated in blood from the mouse. Again, no degradation of the oligonucleotide was observed.

Modification of Oligonucleoti~es at the 3'-Terminal Internucleotide Link Markedly ~nhihits Their Degradation by Nucleases Present in Human Cells The derivative of MA-15 described in Example 5 in which the 3'-ter~ina~ internucleotide link was modi~ied as a 2,2,2-trichloro-1,1-dimethylethyl phosphotriester was also used in this study. Extracts from K562 cells were prepared as in Example 4. The modified oligonucleotide was incubated in the cell extracts at 37~C. Aliquots were removed at various times, and the products of degradation were analyzed by polyacrylamide gel electrophoresis.
Degradation of this modified oligonucleotide by the rapid 3'- to 5'-exonuclease present in the cell extracts waS completely blocked. This led to a marked increase in stability of the modified derivative compared to the native oligonucleotide. The half-life was prolonged by 10-fold. Degradation occurred by the slower 5'- to 3'-exonucleolytic activity present in the K562 cell extracts (see Example 4). The half-life for cleavage of the 5'-terminal nucleotide from the chain was about 20 minutes.

133g33~

Similar studies were carried out in which the modi-fied oligonucleotide was incubated in extracts prepared from COS cells (African ~reen monkey kidney) and mouse liver cells. In both cases, the half-life of the deri-vative modified at the 3'- terminal internucleotide link was increased 10- to 15-fold compared to the un-modified oligonucleotide.

Oli~onucleotides Modified at the 3'-Terminal Internucleotide Linkage Form Substrates with mRNAs That are Cleaved by RNaseH
Mouse globin mRNA was isolated from mice rendere~
anemic by treatment with phenylhydrazine as described in Example 1. Three oligonucleotides were compared in this study; MA-15, the derivative of MA-15 utilized in Examples 5 and 6 in which the 3'- terminal internucleo-tide link is modified as a trichlorodimethylethyl phos-photriester (MA-15-TCDM), and a derivative further modified at the 5'-terminal 0ll group with naphthyli-socyanate (NPH-MA-15-TCDM). Mouse globin m~NA ~2 micro-grams) was added to l.l nanomoles of the oligonucleotide in a final volume of 22 microliters containing 10 mM
tris-HCl buffer at p~l 7.5, 5 mM MgC12, 2S mM NaCl, and 1 mM dithiothreitol. The reaction~ were initiated by the addition of 2 units of E.coli RNaseH, and were allowed to proceed for 30 minutes at 37~C. The reactions were stopped by extraction of the samples with a 1:1 mixture of phenol and chloroform. The mRNA species remaining after ~,~

$ ~

digestion with RNaseH were isolated by precipitation wi~h ethanol and analyzed on Northern blots.
The RNA samples were reacted with 1 M slyoxal in a ~ volu~e/volume) mixture of 10 m~l sodium phosphate buffer at pll 6.5 and dimethylsulEoxide for 1 hour at 50~C, and then electrophoresed on a 1.5~ agarose gel.
The RNAs were transferred from the gel to a Gene Screen Plus filter ~DuPont-NEN) by capillary blotting accord-ing to the method described by the manufacturer. Pre-hybridization of the filter was done in 1~ sodium do-decylsulfate, 1 M NaCl, 10% dextran sulfate, 50 mM
sodium phosphate pH 6.5 for 2 hours at 35~C. Hybridi-zation was conducted in S ml o~ the prehybridization buffer containing 200 microgram/ml denatured salmon sperm DNA 21us 2 x 106 counts/min of MA-15, radiolabeled with 32p at the 5'-end of the molecule, for 24 hours at 35~C. After the hybridization, the filter was washed in the following order: once in 2 x SSC ~O.lS M NaCl, 0.015 M sodium citrate) for S minutes at room tempera-ture, twice in 2 X SSC plus 1~ sodium dodecylsulfate for 30 minutes at 35~C, and once in 0.1 X SSC for 5 minutes at room temperature. The filter was blotted dry and exposed to t~)~dak XAR-S film for autoradiography.
A drawing of the autoradiogram is shown in Figure 5.
Using radiolabeled MA-15 as a ~robe, a single band due to ~-globin mRNA was detected on the Northern blot. In the absence of oligonucleotide, no cleavage of the mRNA occurred during the ~ncubation with RNase~l (lane 2~. For the reaction in the presence o~ MA-lS
llane 3), a marked decrease in the intensity of the band for ~-globin mRNA is observed due to the cleavage of the mRNA by ~NaseH at the site o~ hybridization (*)Trade Mark '3 3 5 with the probe. A similar decrease in intensity of the ~-globin mRNA band was observed for the reactions with each o~ the modified oligonucleotides (lanes 4 and 5), indicating that they also hybridize with the mRNA to form a substrate that is recognized and cleaved by RNaseH.

Inhibition of Expression of the C-MYC Oncogene in the Murine Plasmacytoma Cell Line MOPC 315 with Complementary Oligonucleotides Modified at the 3'-Terminal Internucleotide Link C-myc is a cellular oncogene aberrantly expressed by a number of human and animal tumor cells both in vivo and in cell culture. In the case of the murine plasmacytoma cell line MOPC 315, the c-myc gene is translocated from its normal position on chromosome 15 to the immunoglobulin locus on chromosome 12 and is expressed at an abnormally high level.
Two oligonucleotide analogs complementary to the sequence of c-myc mRNA coding for amino acid residues 191 to 1~8 were synthesized:
5'-G~TAGGGAAAGACCACTGAGGGT~C
5'-(PN)-GTAGGGAAAGACCACTGAGGGT~C
where (~) represents a trichlorodimethylethyl phosphotriester and (PN) -N~l2 2 2 ~-The TCDM phosphotriester was incorporated into the chain by the procedure described in Example 5. The i339~3~

remaining portion of the sequence containing unmodified phosphodiester groups was synthesized by the phos-phoramidite method. The 5'-terminal group (PN) was introduced using the reagent(*)Aminolink from Applied ~iosystems according to the procedure described by the manufacturer. For comparison, the unmodified oligo-nucleotide having the same base sequence was also prepared, as well as an oligonucleotide complementary to human albumin mRNA:
5'-T~CCCTTCATCCCGAAGTT~C
where (~) again represents the TCDM group.
MOPC 315 cells were maintained in RPMI 1640 culture media containing 10% fetal calf serum at 37~C with 7~
C02. Log ~hase MOPC 315 cells were harvested by centri-fugation (250 X g for 19 minutes at 23~C) and resus-pended at a density of 5 X 106/ml in HB101 serum-free medium (DuPont). One ml aliquots of this cell suspen-sion were incubated in the presence of oligonucleotide for 4 hours at 37~C and 7~ C02. Additional one ml aliquots were incubated concommitantly without oligo-nucleotide. Cell~ from each group were then collected by centrifugation. Total cellular RNA was prepared as described by Chir~win et al., Biochemistry 18, 5294-5299 (197g), and analyzed by Northern blots as described in Example 7. The filters were probed with oligo-nucleotide complementary to c-myc mRNA labeled with 32p using polynucleotide kinase and Y32P-ATP as in previous examples.
Both of the anti-c-myc oligonucleotides modified at the 3'-terminal internucleotide link with the TCDM
group, at a ~o~entration of l micromolar, decrea~ed the level of e-n~c mRNA ~y ~5% when (*)Trade Mark 1~3~3~

compared to the steady-state level of the mRNA in MOPC
315 cells incubated in the absence of oligonucleotide.
The unmodified oligonucleotide against c-myc and the anti-albumin oligonucleotide had no effect on c-myc mRNA levels at a concentration of either 1 or 5 micro-molar. This result is consistent with recent studies of another lymphoid cell line in which no inhibition of expression of the c-myc gene was found at concentrations of an unmodified anti-c-myc oligonucleotide below a concentration of 15 micromolar (Heikkila et al. t1987) ~ature 328, 445-449). The modified derivatives described herein, thus, are between 20- and 50-fold more effective at inhibitins the expression of a targeted gene in intact cells than is the corresponding unmodified oligonucleotide.
In summary, the results presented show that oligo-nucleotides in which the 3'-terminal phosphodiester linkage is modified are resistant to nuclease digestion within cells and body fluids, retain normal hybridization properties with complementary nucleic acid sequences, and when hybridized to specifically targeted mRNA species form substrates which are recognized and cleaved by the enzyme ~NaseH, enabling the oligonucleotide to be used to block the expression of the corresponding gene as is desired in various therapeutic applications.
Other modifications of the oligonucleotide may be made without departing from the scope and .spirit of the invention, the important factor and contribution being the discovery that modification of the 3'-terminal phosphodiester linkage alone is sufficient to markedly X

1~ ~>3~'~35 increase the resistance of the oligonucleotide to degradation, and that such d~rivatives are able to form substrates with RNAs that are recognized and clea~ed by RNase~!.
Additional substituents may be added to further increase the stability o~ the oligonucleotide against degradation, to facilitate ntracellular transport, or to attach labels, e.g., radioactive, fluorescent, or enzy~e markers, for use of the oligonucleotide as a hybridization probe in diagnostic tests. It is lmportant that such further modifications be restricted to a li~ited number of other sites on the molecule. If most or all of the phos~hate grcups within the chain are modified, the hybridization properties of the oligcnucleotide may be adversely affected, and it will not form substrates with mRNAs that are acted upon by RNase~. This latter fact defeats the use of high]y modified oligonucieotide analogs as therapeutic a~ents to block the expresslon of specifically targeted genes.
It therefore can be seen that the invention accom-plishes at least all of the objectives heretofore stated.

Claims (30)

1. CLAIMS:
   
   An in vitro method of blocking expression of a targeted gene in a eukaryotic cell which expresses the targeted gene comprising:
   contacting the cell with an oligodeoxynucleotide, which oligodeoxynucleotide has (a) complementarity for and is hybridizable to an mRNA
   region within mRNA transcribed from the targeted gene, such complementarity being sufficient to ensure specificity of hybridization of the oligodeoxynucleotide to the region within the mRNA transcribed from the targeted gene;
   (b) a modified 3'-terminal internucleotide phosphodiester linkage, which linkage is resistant to 3' to 5' exonuclease degradation;
   (c) one or more additional modified internucleotide phosphodiester linkages, which additional modified internucleotide phosphodiester linkage(s) is/are resistant to nuclease degradation; and (d) a continuous stretch of at least four internucleotide phosphodiester linkages which are unmodified;
   under conditions such that said oligodeoxynucleotide enters the cell in sufficient quantity to block the expression of the targeted gene.
2. 2. The method of Claim 1 wherein the eukaryotic cell is in cell culture or tissue culture.
3. 3. The method of Claim 1 wherein the modified 3'-terminal internucleotide phosphodiester linkage of the oligodeoxynucleotide is a phosphotriester, phosphonate, phosphoramidate, phosphorothioate, or phosphoroselenate linkage.
4. 4. The method of Claim 1 wherein the additional modified internucleotide phosphodiester linkage of the oligodeoxynucleotide is at the 5'-terminal internucleotide phosphodiester linkage.
5. 5. The method of Claim 1 wherein the additional modified internucleotide phosphodiester linkage of the oligodeoxynucleotide is a phosphotriester, phosphonate, phosphoramidate, phosphorothioate, or phosphoroselenate linkage.
6. 6. The method of Claim 1 wherein the oligodeoxynucleotide further contains one or more additional modified internucleotide phosphodiester linkages, which additional modified internucleotide phosphodiester linkage(s) enhance(s) membrane crossing ability of said oligonucleotide, thus facilitating transport of said oligodeoxynucleotide into the cell.
7. 7. The method of Claim 6 wherein the additional modified internucleotide phosphodiester linkage of the oligodeoxynucleotide is at the 5'-terminal internucleotide phosphodiester linkage.
8. 8. The method of Claim 1 wherein the oligodeoxynucleotide is from about 10 to about 75 nucleotides in length.
9. 9. The method of Claim 1 wherein the oligodeoxynucleotide is attached to a carrier molecule, which carrier molecule is capable of being taken up by a cell.
10. 10. The method of Claim 1 wherein the oligodeoxynucleotide continuous stretch is greater than seven internucleotide phosphodiester linkages.
11. 11. The method of Claim 1 wherein all oligodeoxynucleotide phosphodiester linkages outside the continuous stretch are modified.
12. 12. An oligodeoxynucleotide having: (a) a modified 3'-terminal internucleotide phosphodiester linkage, which modified 3'-terminal internucleotide phosphodiester linkage is resistant to 3' to 5' exonuclease degradation; (b) one or more additional modified internucleotide phosphodiester linkages, which additional modified phosphodiester linkage(s) is/are resistant to nuclease degradation; and (c) a continuous stretch of at least four internucleotide phosphodiester linkages that are unmodified.
13. 13. The oligonucleotide of Claim 12 further having one or more additional modified internucleotide phosphodiester linkages, which additional modified internucleotide phosphodiester linkage (5) facilitate(s) intracellular transport of said oligodeoxynucleotide.
14. 14. The oligodeoxynucleotide of Claim 12 wherein said oligodeoxynucleotide is from about 10 to about 75 nucleotides in length.
15. 15. The oligodexoynucleotide of Claim 12 wherein said oligodeoxynucleotide is attached to a carrier molecule, which carrier molecule is capable of being taken up by a cell.
16. 16. The oligodeoxynucleotide of Claim 12 wherein the additional modified internucleotide phosphodiester linkage is at the 5'-terminal internucleotide phosphodiester linkage.
17. 17. The oligodeoxynucleotide of Claim 12 wherein the continuous stretch of internucleotide phosphodiester linkages which are unmodified is greater than seven internucleotide phosphodiester linkages.
18. 18. The oligodeoxynucleotide of Claim 12 wherein all phosphodiester linkages outside the continuous stretch are modified.
19. 19. The oligodeoxynucleotide of Claim 12 wherein the modified 3'-terminal internucleotide phosphodiester linkage is a phosphotriester linkage, a phosphonate linkage, a phosphoramidate linkage, a phosphorothioate linkage, or a phosphoroselenate linkage.
20. 20. The oligodeoxynucleotide of Claim 12 wherein the additional modified internucleotide phosphodiester linkage(s) is/are a phosphotriester linkage(s), phosphonate linkage(s), phosphoramidate linkage(s), phosphorothioate linkage(s), or a phosphoroselenate linkage(s).
21. 21. The oligodeoxynucleotide of Claim 12 having: (a) a phosphotriester linkage at the 3'-terminal internucleotide position; (b) one or more additional phosphotriester linkage(s) at the 5' end of the oligodeoxynucleotide; and (c) a continuous stretch of at least four internucleotide phosphodiester linkages which are unmodified.
22. 22. An oligodeoxynucleotide of Claim 12 having: (a) a phosphoramidate linkage at the 3'-terminal internucleotide position; (b) one or more phosphoramidate linkage(s) at the S' end of the oligodeoxynucleotide; and (c) a continuous stretch of at least four internucleotide phosphodiester linkages which are unmodified.
23. 23. The oligodeoxynucleotide of Claims 12-22, the nucleotide sequence of which has complementarity for and is hybridizable to a nucleotide sequence within the mRNA
    transcribed from a targeted gene.
24. 24. The oligodeoxynucleotide of Claims 12-23 for use in blocking expression of a targeted gene.
25. 25. The oligodeoxynucleotide of Claims 12-23 for therapeutic use.
26. 26. The oligodeoxynucleotide of Claim 24 wherein the RNA-DNA duplex is a substrate for RNaseH and the mRNA within the mRNA-DNA duplex is cleaved by RNaseH to block expression of the targeted gene.
27. 27. An oligodeoxynucleotide of seven or greater internucleotide linkages having: (a) complementarity for and being hybridizable to a nucleotide sequence within the mRNA
    sequence of a targeted gene; (b) a modified 3'-terminal internucleotide phosphodiester linkage, which modified 3'-terminal internucleotide phosphodiester linkage is resistant to 3' to 5' exonuclease degradation; (c) one or more additional modified internucleotide phosphodiester linkages, which additional modified phosphodiester linkage(s) is/are resistant to nuclease degradation; and (d) a continuous stretch of at least four internucleotide phosphodiester linkages that are unmodified.
28. 28. The oligodeoxynucleotide of Claim 16 for use in blocking expression of a targeted gene.
29. 29. The oligodeoxynucleotide of Claim 27 for therapeutic use.
30. 30. The oligodeoxynucleotide of Claim 28 wherein the RNA-DNA duplex is a substrate for RNaseH and the mRNA within the mRNA-DNA duplex is cleaved by RNaseH to block expression of the targeted gene.