WO1997011085A1

WO1997011085A1 - Inhibited biological degradation of oligodeoxynucleotides

Info

Publication number: WO1997011085A1
Application number: PCT/US1996/014812
Authority: WO
Inventors: Robert H. Singer; Joan C. Politz; Krishan L. Taneja
Original assignee: University Of Massachusetts
Priority date: 1995-09-19
Filing date: 1996-09-16
Publication date: 1997-03-27
Also published as: AU7073096A

Abstract

Disclosed is a method for inhibiting the biological degradation of a synthetic oligodeoxynucleotide or oligodeoxynucleotide analog placed into a cell or extracellular environment. Biological degradation is inhibited by the covalent attachment of a protection moiety, e.g., a fluorochrome label such as FITC or Texas red, to a modified base incorporated into the oligodeoxynucleotide or oligodeoxynucleotide analog.

Description

INHIBITED BIOLOGICAL DEGRADATION OF OLIGODEOXYNUCLEOTIDES

Field of the Invention This invention relates to cell biology and nucleic acid biochemistry.

Background of the Invention Oligodeoxynucleotides ("oligos") complementary to a particular gene transcript or a viral nucleic acid ("target nucleic acid") are commonly known as antisense oligos. An antisense oligo hybridizes with its target nucleic acid by means of Watson-Crick base pairing. Such hybridization can inhibit the function of a particular mRNA or inhibit viral replication. In principle, therefore, by virtue of being able to inhibit selectively the expression of specific genes, oligos have potential as therapeutic agents and research tools.

Oligos with naturally-occurring phosphodiester backbones ("O-oligos") are susceptible to rapid degradation (i.e., within minutes to hours) by endonucleases and exonucleases found within cells and in extracellular environments, e.g, serum and cell culture media. A practical limitation on the usefulness of synthetic oligos, especially O-oligos, as therapeutic agents and research tools has been this lack of stability against nucleases. To overcome this limitation, much research over the past decade has gone toward developing nuclease-resistant oligodeoxynucleotide analogs, such as methylphosphonates, phosphorothioates, and phosphoroamidates (see, e.g., Stein et al. (1993) Science 261:1004-1012; Agrawal et al. (1991) Proc . Natl . Acad . Sci . USA 88:7595-7599; Wickstrom (1986) J. Biochem. Biophys . Meth . 13:97-102; Eckstein (1985) Ann . Rev. Biochem . 54:367-402; U.S. Patent No. 5,264,423; U.S. Patent No. 5,276,019; U.S. Patent No. 4,511,713). In a phosphorothioate oligo ("S-oligo") , one of the non-bridging oxygen atoms in each phosphate group of the phosphodiester backbone is replaced by a sulfur atom. The substitution results in an oligo that retains its net charge, aqueous solubility, and ability to hybridize with mRNA targets. Phosphorothioate oligos display limited nuclease resistance in vitro and in vivo . They have been isolated virtually intact from some cell types after several hours (Stein et al., supra) . Phosphorothioate oligos are, however, slowly digested by SI and Pl nucleases. A potential disadvantage of phosphorothioate- based antisense oligo therapy is that such digestion can lead to the liberation of phosphorothioate mononucleotides, which can be reincorporated into cellular DNA, causing mutagenesis.

The intracellular stability of a synthetic oligo may depend on the way in which it enters the cell, as well as the chemical structure of the oligo. Horishita et al. have reported that oligos internalized via HVJ (hemagglutinating virus of Japan) liposomes display significantly longer half-lives than oligos passively taken up in a naked state. (Morishita et al. (1994) Gene 149:13-19; Morishita et al. (1994) OligoTechniques 11:1- 5).

Summary of the Invention

We have discovered that if an O-oligo has a fluorochrome moiety covalently attached to a modified base near the oligo's 3' end, the oligo has a significantly increased stability in living cells (as compared to the corresponding unlabeled O-oligo) . We have also discovered that fluorochrome labeled O-oligo hybrids are more stable in living cells than labeled S- oligo hybrids or unlabeled O-oligo hybrids. Based on these discoveries, the invention features a method for inhibiting the biological degradation of a synthetic oligo. The method comprises the steps of:

(a) incorporating at least one modified base into a synthetic oligo;

(b) covalently attaching a protection moiety onto the modified base in the oligo, thereby creating a biological degradation-resistant oligo; and

(c) placing the biological degradation-resistant oligo into a cell or an extracellular environment.

The invention also features a method for producing an intracellular nucleic acid hybrid with enhanced stability against biological degradation. The method comprises the steps of: (a) incorporating at least one modified base into a synthetic oligo that is capable of hybridizing with a target nucleic acid;

(b) covalently attaching a protection moiety onto the modified base in the synthetic oligo, thereby creating a biological degradation-resistant oligo;

(c) allowing the biological degradation-resistant oligo to enter a target cell; and

(d) allowing the biological degradation-resistant oligo to hybridize with the target nucleic acid.

The oligo used in this invention can be either an O-oligo or an S-oligo. The oligo used in this invention can be an antisense oligo or a sense oligo, e.g., a negative experimental control.

When this method is used with an oligo complementary to a target nucleic acid, the target nucleic acid can be DNA or RNA. In addition, the target nucleic acid can be nuclear or cytoplasmic. A target nucleic acid can be a target cell's own nucleic acid, i.e., native to the target cell. Alternatively, a target nucleic acid can be a viral nucleic acid that has infected a target cell.

Biological degradation of an oligo can occur inside cells or in an extracellular environment, such as serum, cell culture media, or interstitial spaces in intact tissue. Accordingly, the method of this invention can be used to stabilize synthetic oligos in a target cell, in an extracellular environment, or both. The target cell can be prokaryotic or eukaryotic. If the target cell is eukaryotic, it can be in culture or in intact tissue.

Examples of protection moieties that can be used in this invention are fluorescein isothiocyanate ("FITC"), Texas red, Cy-3 (Biological Detection Systems, Pittsburgh, PA) , tetramethylrhodamine isothiocyanate ("TRITC") ,

7-aminomethylcoumarin hydroxysuccini ide ("AMCA") , biotin, and digoxigenin. Preferably, the protection moiety used in this invention is FITC or Texas red. As used herein, "biological degradation" means enzymatic or nucleolytic cleavage of an oligo inside a cell or in an extracellular environment.

As used herein, "extracellular environment" means the immediate physical surroundings of a cell. The plasma membrane is the boundary separating the extracellular environment from the inside of a cell, i.e., the intracellular environment. Examples of extracellular environments include cell culture media, serum, interstitial spaces in intact animal tissues, and cell wall matrices of plant, fungal and bacterial cells. As used herein, "synthetic oligo" means a pre¬ assembled oligodeoxynucleotide or oligodeoxynucleotide analog synthesized for uptake by, or injection into, a target cell. A synthetic oligo is to be distinguished from an oligo transcribed in vivo from a transgene in a transformed cell.

As used herein, "fluorochrome" means a molecule that emits light of a second predetermined wavelength upon absorption of light of a first predetermined wavelength.

As used herein, "oligo" means an oligodeoxy¬ nucleotide or oligodeoxynucleotide analog. As used herein, the term "oligo" includes O-oligos, S-oligos, unlabeled oligos and labeled oligos.

As used herein, "nucleolytic cleavage" means cleavage catalyzed by a ribozyme.

As used herein, "O-oligo" means an oligodeoxynucleotide whose constituent deoxynucleotide residues are linked (5' to 3') by phosphodiester bonds. The phosphodiester backbone of an O-oligo is the same as that in naturally-occurring DNA.

As used herein, "protection moiety" means a moiety which, when attached to a modified base in a synthetic oligo, inhibits biological degradation of the oligo.

As used herein, "S-oligo" means a phosphorothioate oligodeoxynucleoside, which is an analog of a phosphodiester oligodeoxynucleotide. In the phosphorothioate analog, a sulfur atom replaces a non- bridging oxygen atom at each phosphodiester bond in the backbone of the molecule.

As used herein, "target cell" means a cell in which a target nucleic acid is located.

As used herein, "target nucleic acid" means a nucleic acid to which an oligo is complementary.

Other features and advantages of the invention will be apparent from the following description of the preferred embodiments thereof, and from the claims. Brief Description of the Drawings Fig. IA is the chemical formula of a preferred modified base.

Fig. IB is the chemical formula of a preferred deoxynucleotide (dT) analog comprising the modified base of Fig. IA.

Fig. 2A is a graph of uptake of fluorescently- labeled S-oligo dT and dA by L6 cells. The plot shows fluorescence/cell after incubation for two hours, as a function of oligo concentration. Cells were incubated with ST30tr (closed squares) or SA30tr (open squares) . The intracellular fluorescence was quantitated after the cells were fixed in formaldehyde. ST30tr is a 30- nucleotide thymidine homopolymer wherein sulfur replaces one oxygen at every phosphorodiester bond, and a Texas red label is attached to a modified base at positions 2 and 29. SA30tr is a 30-nucleotide adenosine homopolymer wherein sulfur replaces one oxygen at every phosphorodiester bond, and a Texas red label is attached to a modified base at positions 2 and 29.

Fig. 2B is a graph of uptake of fluorescently- labeled S-oligo dT and dA by L6 cells. The plot shows fluorescence/cell (incubated with 0.1 μM oligo), as a function of incubation time. Cells were incubated with ST30tr (closed squares) or SA30tr (open squares) . The intracellular fluorescence was quantitated after the cells were fixed in formaldehyde.

Fig. 3A is a graph of in vivo hybridization as a function of incubation time. The plot shows silver stain absorbance/cell after incubation for varying times with 0.1 μM ST30tr (closed squares) or T43tr (closed circles) and their dA analogs (open squares and open circles, respectively) . After the incubation with an oligo, cells were extracted, fixed, and subjected to 1ST. T43tr is a 43-nucleotide thymidine homopolymer wherein sulfur replaces one oxygen at every phosphodiester bond, and a Texas red label is attached to a modified base at positions 2, 12, 22, 32, and 42.

Fig. 3B is a graph of in vivo hybridization as a function of oligo concentration. The plot shows silver stain absorbance/cell after incubation for 2 hours with varying concentrations of S-oligo ST30tr (closed squares) or O-oligo T43tr (closed circles) and their dA analogs (open squares and open circles, respectively) . After the incubation with an oligo, cells were extracted, fixed, and subjected to 1ST.

Fig. 4A is a graph comparing total intracellular S-oligo and intracellular S-oligo hybridized to RNA, as a function of time. Cells were incubated with oligo ST30tr for two hours, washed and then allowed to grow in oligo- free medium for varying lengths of time. Parallel populations were either fixed directly for fluorescence quantitation or Triton extracted, fixed, and subjected to 1ST. Levels of silver stain in cells incubated with control oligo dA were close to zero (data not shown) . Total intracellular S-oligo amount is indicated by open circles, and hybridized S-oligo amount is indicated by closed squares.

Fig. 4B is a graph comparing intracellular hybridized S-oligo and intracellular hybridized O-oligo, as a function of time. Cells were incubated with S-oligo ST30tr or O-oligo T43tr for two hours, washed and then allowed to grow in oligo-free medium for varying lengths of time. Cells were Triton extracted, fixed, and subjected to 1ST. Levels of silver stain in cells incubated with control oligo dA were close to zero (data not shown) . Hybridized O-oligo amount is indicated by closed circles, and hybridzed S-oligo amount is indicated by closed squares. Detailed Description The invention provides a convenient and effective method for significantly inhibiting the biological degradation of a synthetic oligo. Inhibiting its biological degradation increases the biological potency of an oligo, thereby enhancing the oligo's usefulness as a therapeutic agent, prophylactic agent, diagnostic agent, or research tool. This invention is useful in any situation where: (1) a synthetic oligo is placed into a cell or into an extracellular environment, and (2) it is desirable to stabilize the oligo against biological degradation.

The inhibition of the biological degradation of the oligo is achieved by covalently attaching a protection moiety, e.g. , a fluorochrome label, onto one or more modified bases in the oligo, by conventional methods, before the oligo is placed into a cell or into an extracellular environment.

While the protection moiety on an oligo used in this invention can be used for in situ visualization of the oligos (if the protection moiety is a fluorochrome) , the advantages of the invention relate directly to the increased stability of the labeled oligo, not its visualization. Accordingly, the practice of this invention does not require visualization of the fluorochrome-labeled oligo. Of course, where visualization of the oligo is desired, e.g., in diagnostic methods or biological research, the increased stability of the oligo will facilitate its visualization. This invention can be practiced by conjugating the protection moiety to a modified base in an O-oligo or an S-oligo. The use of an O-oligo is preferred. The use of an O-oligo avoids the intracellular liberation of phosphorothioate nucleotide monomers during nuclease degradation of the S-oligo. Such avoidance is advantageous when the oligos are used in vivo for therapeutic, prophylactic, or diagnostic applications. Liberated phosphorothioate nucleotide monomers potentially could be reincorporated into cellular DNA, which could result in mutagenesis. Oligo Design and Synthesis

The basic principles of oligo design, synthesis and use are well known. Those basic principles apply generally to the design, synthesis and use of the biological degradation-resistant oligos of this invention.

This invention includes embodiments comprising an essentially unlimited number of different oligos. For example, an oligo used according to this invention can vary considerably in length. The preferred length of the oligo will depend on considerations such as target cell type, method of oligo introduction into the target cell, oligo concentration used, target nucleic acid type (e.g. , mRNA, double-stranded DNA) , target nucleic acid length, target nucleic acid copy number, target nucleic acid G-C content, and target cell temperature. Preferably the length of the oligo used in this invention is in the range of 5 to 200 nucleotides. More preferably, the length is in the range of 10 to 100 nucleotides. Most preferably it is in the range of 15 to 50 nucleotides. For a general discussion of oligo length and factors relating thereto, see: Goodchild, "Inhibition of Gene Expression by Oligonucleotides," in Topics in Molecular and Structural Biology,. Vol . 12 : Oligodeoxynucleotides (Cohen, ed.), MacMillan Press, London, pp. 53-77, which is herein incorporated by reference.

Regardless of oligo length, the oligo can vary in nucleotide sequence. The nucleotide sequence of an oligo used in this invention will depend on the sequence of the target nucleic acid. The oligo's nucleotide sequence must have sufficient complementarity to the target nucleic acid to allow oligo hybridization with the target nucleic acid, under conditions inside the target cell. Preferably, base pair matching between the oligo and target nucleic acid is at least 80%. More preferably, the base pair matching is approximately 100%.

Methods of synthesizing DNA generally, including oligos used in this invention, are well known. For a general discussion of oligo synthesis, see Caruthers, "Synthesis of Oligonucleotides and Oligonucleotide

Analogs," in Topics in Molecular and Structural Biology, Vol . 12 : Oligodeoxynucleotides (Cohen, ed.), MacMillan Press, London, pp. 9-24. Apparatuses for automated DNA synthesis are commercially available. Preferably automated DNA synthesis is employed in obtaining oligos used in the practice of this invention.

Typically, a biological degradation-resistant oligo used in this invention is obtained in a two step process. The first step is synthesis of an oligo which comprises a modified base at each position in the oligo's nucleotide sequence where a protection moiety is desired. The second step is covalent attachment of the protection moiety to the modified base.

The purpose of the modified base used in the first step is to provide a functional group through which the protection moiety is covalently attached to the oligo, in the second step. Preferably, the functional group provided by the modified base is a primary amino group. Preferably, the functional group is at the end of a spacer arm.

During synthesis of an oligo, the functional group provided by the modified base (for attachment of the protection moiety) typically bears a protecting (blocking) group, e.g. , a trifluoroacetamide group. One of skill in the art will recognize that the protecting group must be removed by a suitable chemical reaction before the functional group can be used for attachment of the protection moiety.

For preparation of amino modified bases, see Jablonski et al. ((1986) Nucleic Acids Res . 14:6115-6128) and Ruth ((1984) DNA 3:123). A particularly preferred modified base is a thymine analog with the chemical structure shown in Fig. IA. The thymine analog depicted in Fig. IA can be conveniently incorporated into an oligo by means of a dT analog whose structure is shown in Fig. IB. The dT analog depicted in Fig. IB is available commercially as "Amino-Modifier C6 dT" (Glen Research, Sterling, VA) . "Amino-Modifier C6 dT" is designed for use in conventional automated DNA synthesis. The trifluoroacetamide group on "Amino-Modifier C6 dT" is a protecting group. It is removed by hydrolysis during deprotection, to expose a primary amine group for use in attachment of a protection moiety.

The total number, and the spacing, of the modified bases (and covalently attached protection moieties) in the oligo can vary, in the practice of this invention. Preferably, a modified base is incorporated within five bases from the 3' end of the oligo. More preferably, a modified base is incorporated in the ultimate or penultimate base position, at the 3' end of the oligo.

If visualization of a fluorochrome labeled oligo is desired, it is preferable to incorporate a modified base near the 3' end of the oligo and at approximately every tenth base position in the nucleotide sequence of the oligo. Incorporation of modified bases, and thus fluorochrome moieties, closer than every ten bases causes quenching of fluorescence and concomitant loss of visual signal strength. Attachment of Protection Moieties to Oligos

Various protection moieties can be covalently attached to the modified base in the practice of this invention. Examples of protection moieties useful in this invention are FITC (Molecular Probes, Inc. , Eugene, OR), Texas red (Molecular Probes, Inc., Eugene, OR), Cy-3 (Biological Detection Systems, Pittsburgh, PA) , TRITC, AMCA, biotin, and digoxyigenin. Preferably, the protection moiety is a fluorochrome. More preferably, the protection moiety is FITC or Texas red.

FITC and Texas red comprise an isothiocyanate or sulfonylchloride functional group, respectively, which reacts with primary amines. Thus, in a preferred method of making oligos for use in this invention, FITC or Texas red is allowed to react with the primary amino group of a modified base. For methods of attaching fluorochromes onto amino groups, see Agrawal et al. (1986) Nucleic Acids Res . 14:6227-6245. A preferred protocol for covalent attachment of FITC or Texas red to the primary amino group of a modified base is as follows:

1. Dissolve 0.1 μM of amino-modified oligonucleotide (i.e., 0.1 μM of free primary amines) in 0.7 ml of sterile water.

2. Add 0.1 ml of 10X buffer (0.1 M carbonate, pH 9.0) .

3. Freshly prepare a 10 mg/ml solution of FITC or Texas red in DMF. Add 0.2 ml of the solution to the reaction mixture.

4. Allow the mixture to stand at least 2 hours. (Note: Overnight reaction may be more convenient.)

5. Purify the reaction mixture on a POLY-PAK~ cartridge (Glen Research) , or a column containing SEPHADEX™ G-25 or G-50, to remove the excess label. Kits containing reagents prepared for this protocol are commercially available (e.g.. Glen Research, Sterling, VA) .

Target Nucleic Acids

The target nucleic acid can be DNA or RNA. The target nucleic acid can be located in the target cell's nucleus or cytoplasm. Examples of target nucleic acids include target cell mRNA, target cell pre-mRNA, target cell chromosomal DNA, viral RNA present in the target cell, or viral DNA present in the target cell. For a general discussion of target nucleic acids, see, Goodchild (supra) . Typically, the target nucleic acid is mRNA native to the target cell or single-stranded viral RNA present in the target cell. It is possible, however, for the target nucleic acid to be single-stranded DNA, e.g., a region of chromosomal DNA whose base pairing has been disrupted for any reason. It is also possible for the target nucleic acid to be double-stranded DNA. When the target nucleic acid is double stranded DNA, the oligo can act by forming a triple helix, such as described by Cooney et al. ((1988) Science 241:456-459). Introduction of Oligos into Target Cells

A fluorochrome-labeled oligo used according to this invention can be introduced into target cells by any method. Numerous methods for introducing DNA, including synthetic oligos, into cells are known in the art. For a general discussion of cellular uptake of antisense oligos, see Akhtar et al. (1992) Trends in Cell Biology 2:139-144.

The preferred method for introducing biological degradation-resistant labeled oligos into target cells according to this invention will depend on various factors, including the type of target cell, e.g., animal, plant, or bacterial. The choice of method will also depend on whether the target cell is in culture or in intact tissue (e.g., in a mammal). Selection of methods suitable for introducing oligos into cells of a particular type, in culture or in intact tissue, is within ordinary skill in the art.

A method particularly suited for topical delivery of oligos into vascular walls in the mammalian body is the pluronic gel method. Antisense oligos have been administered against c-myc (at concentrations in excess of 150 μM) using pluronic gels applied to the adventitial layer, to inhibit vascular smooth muscle accumulation following angioplasty injury (Simons et al. (1992) Nature 359:67-70).

Cells in culture or in intact tissue can take up naked DΝA. Oligos have been shown to be passively taken up by cultured cells following addition of naked oligos to the culture medium (e.g. , at a concentration in the range of 10-200 μM) . Also, mammalian muscle cells have been shown to take up naked DΝA dissolved in aqueous solution and injected into muscle tissue.

Additional methods for introducing oligos into target cells include the following: microinjection (see, e .g. , Leonetti et al. (1991) Proc . Natl . Acad. Sci . USA 88:2702-2706); electroporation (see, e .g. , Sambrook et al., Afolecular Cloning - A Laboratory Manual (2d Ed.) , Cold Spring Harbor Laboratory Press (1989), at pages 16.54- 16.55); bombardment with high velocity tungsten microprojectiles (see, e . g. , BioRad Technical Bulletin #1687, BioRad, Hercules, CA; Johnston (1990) Nature 346:776) ; transfection of coprecipitates of calcium phosphate and DNA (see, e .g. , Sambrook et al., supra , at pages 16.32-16.40); transfection mediated by DEAE-dextran (see, e .g. , Sambrook et al., supra, at pages 16.41-16.46); and

HVJ-liposome mediated delivery (see, e .g. , Morishita et al. (1994) Gene 149:13-19).

Use of the Invention

This invention can be used in any situation where it is desirable to stabilize a synthetic oligo against biological degradation. Specific examples of therapeutic and prophylactic uses are presented below.

An oligo according to this invention can be used for antiviral therapy. More particularly, a fluorochrome-labeled antisense oligo having a sequence complementary to HIV-l rev mRNA can be used therapeutically to treat an infection by human immunodeficiency virus. See, e.g., Matsukura et al. (1989) Proc . Natl . Acad. Sci . USA 86:7790. Also see, Agrawal (1992) TIBTECH 10:152-158.

In another example, oligos according to this invention can be used prophylactically to inhibit neointimal hyperplasia. More particularly, a combination of antisense oligos directed against mRNAs encoding cdc2 kinase and proliferating cell nuclear antigen ("PCNA") can inhibit neointimal hyperplasia following angioplasty injury. See, e.g., Morishita et al. (1993) Proc . Natl . Acad . Sci . USA 90:8474-8478. Alternatively, to inhibit neointimal hyperplasia, antisense oligos according to this invention can be directed against c-myb mRNA. See, e.g, Simons et al. (1992) Nature 359:67-70.

Oligos directed against non-coding regions of mRNA, e.g., the 3' untranslated region, can affect the stability, translatability, or localization of the mRNA. A stable RNA/DNA hybrid formed between a synthetic oligo of this invention and a target nucleic acid in a target cell is a useful research tool. For example, a fluorescently-labeled oligo stably hybridized to an mRNA can be used as a tag for tracking the movement of the mRNA within the living cell.

Experimental Information Oligodeoxynucleotides

Oligos were synthesized using an Applied Biosystems 394 DNA Synthesizer. Fluorochromes were covalently linked to oligos at modified thymidine residues (see Fig. IB) . Fluorochrome-conjugated oligos were separated from low molecular weight components of the conjugation reaction mixture by two rounds of gel filtration chromatography using SEPHADEX™ G-50 columns. ST30 is a 30 nucleotide ("nt") homopolymer of thymidine, wherein sulfur replaces one oxygen at every phosphodiester bond (i.e., it is an S-oligo). ST30 was labeled at positions 2 and 29 using Texas red (Molecular Probes, Inc., Eugene, OR), to yield a fluorochrome- labeled oligo designated ST30tr. Deoxythi idine homopolymers were chosen for use in these experiments, because the dT oligos should hybridize with the poly A "tails" normally present on mRNA molecules. A 30 nt S- oligo homopolymer of adenosine, designated SA30, was similarly labeled to yield SA30tr. SA30 and SA30tr cannot hybridize with mRNA poly A tails. Therefore, SA30 and SA30tr were used as negative experimental controls for ST30 and ST30tr, respectively. The O-oligo, T43tr, is a 43 nt homopolymer of thymidine with Texas red covalently linked to positions 2, 12, 22, 32 and 42. A43tr is the comparable oligo dA control for T43tr; T43f and A43f are identical to T43tr and A43tr, respectively, except that fluorescein (Molecular Probes) is substituted for Texas red. Unlabeled O-oligos used to test the effect of the fluorochrome moiety included a 40 nt poly dT (T40a) and poly dA (A40a) with an amino group at identical positions to T43f and unmodified 55 nt poly dT (T55) and 55 nt poly dA (A55) . Oligo concentrations were calculated assuming 1 OD₂₆₀ unit = 33 μg/ml. To take into account the differences between oligo dA and oligo dT molar extinction coefficients, oligo dT concentrations should be increased 30%. Cell Culture and Oligo Uptake

L6 rat muscle cells (American Tissue Culture) were cultured using standard techniques in Dulbecco's modified Eagle's media ("DMEM") containing heat-inactivated 10% fetal bovine serum ("FBS") . Cells to be treated with oligos were first trypsinized and then plated onto 12 mm round glass coverslips (boiled previously in 0.1 N HCl, rinsed and autoclaved) in 24-well dishes (Falcon) containing DMEM plus 10% FBS. Cells were plated at a density of 25,000-35,000 cells per coverslip and allowed to grow 20-28 hrs at 37 °C, 5% C0₂. The medium was then replaced with 250 μl fresh DMEM containing oligos for varying times at 37°C, 5% C0₂. For efflux experiments, oligo-containing medium was removed, cells were washed 3 times and then incubated with 250 μl serum-free DMEM. Cell Extraction and Fixation

In some experiments, cells on coverslips were fixed directly in 4% formaldehyde in 1 mM KH P0₄, 10 mM Na₂HP0₄, 0.137 M NaCl, 2.7 mM KCl, pH 7.0 ("PBS") and 5 mM MgCl₂. For most in situ transcription ("1ST") experiments, cells were TRITON™ extracted as follows before fixing. Cells were washed in CSK buffer (0.3 M sucrose, 0.1 M KCl, 5 mM MgCl₂ 10 mM PIPES, pH 6.9, 2 mM EGTA, 1 μg/ml leupeptin (Sigma Chem. Co. , St. Louis, MO) and 1 μg/ml trypsin inhibitor (Sigma)) at 4°C and then extracted with 0.5% Triton in CSK buffer for 90-120 seconds. Extractions were routinely done at 4°C, where maximal RNA retention (and hybridization signal) is observed (see, Bassell et al. (1994) J. Cell . Biol . 126:863-876). At this temperature, microtubules may dissociate, so cells were also extracted at room temperature and analyzed as described below. No obvious difference in the distribution of signal was observed. After washing with CSK buffer, cells were fixed with 4% formaldehyde, 0.1% glutaraldehyde, 5 mM MgCl₂ in PBS for 15 minutes at room temperature. Coverslips were then rinsed twice with 70% ethanol and stored in same at 4°C for up to three months. In Situ Transcription (1ST)

Cells that had been exposed to various oligos in vivo and then fixed as described above were washed in PBS for 10 min., and then twice in 150 mM NaCl, 15 mM sodium citrate, pH 7.0 (SSC) for 10 minutes each time. Samples were then treated with AMV reverse transcriptase (Promega, Madison, WI) essentially as described by Bassell et al. (supra), and Eberwine et al. ((1988) Meth . Enzymol . 216:80-100), except that the commercially available buffer (Promega) was used. Digoxigenin ("dig") was used at 50 μM, in addition to all four unlabeled deoxynucleotides (Pharmacia) at 250 μM, in the reaction mixture. Label was visualized using sheep antidigoxigenin antibodies linked to 1 nm gold particles ("SAD1") followed by silver enhancement (Amersham) . Control cells in which primers were added in situ were also subjected to 1ST. Because S-oligo/RNA hybrids have a lower Tm than O-oligo/RNA hybrids, formamide was omitted from all hybridization solutions when S-oligos were present to stabilize the ST30tr/RNA hybrids. Microscopy and Image Analysis

Coverslips containing cells to be viewed were mounted in Vectashield (Vector Labs) containing 0.1 μg/ml 4', 6-diamidino-2-phenylindole (DAPI) . Silver stained or fluorescent cells were viewed and photographed using a Nikon Microphot SA equipped with bright field and fluorescence optics and a 35 mm camera.

Cells were visualized on a cathode ray monitor. Digitized areas of the signal were individually segregated for quantitation, using software provided by the Biomedical Imaging Center at the University of Massachusetts Medical School. We also used a DISCOVERY™ instrument to automate image digitation and quantitation (Becton Dickinson Cellular Imaging Systems, San Jose, CA) .

To quantify silver-staining, a microscopic field image was captured with a Xillix Microimage 1400 gray¬ scale camera using a 500 ran interference filter in absorbance mode. In real time, cells were automatically identified using standard image processing routines to segment objects and mophological filters to eliminate artifacts. Image thresholds were selected automatically based on the standard deviation of the field image pixel intensity histogram. The instrument verified segmented objects were cells by superimposing a similarly segmented image that was captured using a fluorescent light source with a DAPI filter set. Optical density of the silver stained cell was then automatically calculated and images captured for later inspection. Fluorescence in cells containing Texas red or fluorescein labeled oligos was also measured automatically using the appropriate filters and similar segmentation programs. Raw data (mean fluorescence/cell or A₅₀₀/cell) were converted to percent maximal signal/cell in any given experiment so that data from separate experiments could be averaged. In each of the figures, error bars represent standard error of the mean. Where not visible, the error bars fall underneath plot symbol. Oligonucleotide Uptake

Initial experiments used digital imaging microscopy to characterize the uptake of fluorescently- labeled oligos by L6 cells. Varying amounts of Texas red labeled poly dT (ST30tr) or dA (SA30tr) S-oligos were added to cells growing on glass coverslips and cells were incubated for varying amounts of time. Cells were washed, formaldehyde-fixed and intracellular fluorescence digitized.

As shown in Figs. 2A and 2B, intracellular fluorescence increased with either increasing extracellular concentration or increasing incubation times. Intracellular fluorescence was saturated when incubated at oligo concentrations of 10 μM for 2 hours. S-oligo dT showed primarily a punctate perinuclear distribution with additional diffuse fluorescence throughout the cell. Nuclei were labeled more intensely than cytoplasm in about 30% of the cells. When S-oligo dA was incubated with cells under these same conditions, intracellular labeling was about half of that observed with S-oligo dT (Figs. 2A and 2B) . No nuclear labeling was observed, but a similar punctate cytoplasmic distribution of label was present.

S-oligos labeled with different fluorochromes (fluorescein or cy3) showed labeling patterns similar to that described above and similar results were obtained in a different cell line (human fibroblasts) exposed to ST30tr or SA30tr. We conclude that S-oligo dT may be sequestered more effectively than S-oligo dA in these cell types and that the nature of the fluorochrome label does not appreciably affect the cellular compartments labeled.

As a result of these experiments, a standard incubation time of 2 hours, and concentration of 0.1 μM, were chosen for further experimentation. Under these conditions, extracellular probe concentration was at least 1000-fold excess of the target poly (A) RNA concentration, intracellular fluorescence was reproducible and cell morphology and growth characteristics were not detectably affected by oligo treatment. At high (50 μM) concentrations, vacuolation and cell death sometimes occurred, most frequently with S-oligo dA. Antisense Oligos Added to Living Cells Prime Reverse Transcription In Situ

To determine directly whether S-oligo dT probe taken up by live cells hybridizes to cellular poly (A) , we exploited the fact that the resulting oligo-RNA hybrid should be capable of priming elongation by reverse transcriptase ("RT") in situ . Because RT extends a DNA primer hybridized to an RNA template, only hybridized oligos will prime incorporation of labeled nucleotide into new DNA to give a measurable signal in situ . This detection procedure modifies previous in situ reverse transcription ("1ST") assays (i.e., those of Eberwine et al. (supra), and Mogensen et al. (1991) Exper. Cell . Res . 196:92-9818) by allowing the primer to hybridize in vivo first and then assaying the hybridization in situ after permeabilization and fixation of the cell. Digoxigenin labeled nucleotides were used in the RT reaction mixture. After incorporation, they were detected by SAD1 antibodies linked to 1 nm gold particles. The gold was then enhanced by silver deposition.

Our experiments show that ST30 taken up by cells in vivo can act as a primer for the reverse transcription reaction in situ . Cytoplasmic signal was reproducibly observed when cells were incubated with S-oligo dT for 2 hours in vivo, and then Triton extracted, fixed and subjected to 1ST. Signal resulting from elongation of the transcripts was detected primarily in the perinuclear region of the cytoplasm, with about 30% of cells also showing nuclear signal. This intracellular distribution was similar to the pattern of transcripts detected when the primer was added after, instead of before, cell fixation. Controls showed that no reverse transcription was seen in the absence of primer, or in the presence of primer when RT was heat inactivated before 1ST. Likewise, use of S-oligo dA (SA30) as a primer, either before or after fixation, did not result in signal (data not shown) . These experiments indicated that ST30 formed hybrids with cellular poly (A) RNA in vivo and that these hybrids could be detected specifically by exploiting their ability to prime synthesis of labeled transcripts by RT in situ . Substantial silver stain was also observed when the above experiments were repeated using labeled dCTP, rather than dUTP, as substrate for 1ST (data not shown) , indicating that RT could copy mRNA sequences (containing G residues) upstream of the poly A tail. The intracellular distribution of hybridization signal observed using digoxigenin dUTP as label.

Fluorescently-Labeled Phospodiester Poly dT Oligos Also Hybridize to RNA In Living Cells

We have tested other variously modified poly dT oligos for their ability to act as primers for 1ST after their uptake by living cells. Fluorescent derivatives of both S-oligo and O-oligo dTs (directly conjugated to either Texas red or FITC at positions internal to the 3' and 5' ends) were as effective as ST30 at priming the reverse transcriptase. We obtained signal detection after treating cells with T43tr, a Texas red labeled 0- oligo, and detecting hybridization using the modified 1ST technique (data not shown) . ST30tr, the Texas red labeled analogue of ST30, also supports a high level of reverse transcription. The intracellular distribution of signal obtained after using the fluorescently labeled O- oligo or S-oligo appeared similar to the pattern seen after cells were treated with unlabeled ST30. Cells treated with the oligo dA analogues of these probes showed very low signal after 1ST. Results with unlabeled O-oligo dT probes were variable. Low levels of silver staining were usually observed, even though these oligos were added to cells in serum-free media to reduce nuclease exposure (not shown) . These results are consistent with observations that O-oligos are susceptible to intracellular nucleases. We conclude that the covalently-bonded fluorochrome molecules protect the oligos from nuclease digestion.

Detection of Hybridization in Cells Not Subjected to Triton Extraction Step As has been reported previously for 1ST (Bassell et al., supra), signal from the modified 1ST performed here was highest when cells were permeabilized by Triton extraction and then fixed in paraformaldehyde/glutaraldehyde. The majority of poly (A) RNA was strongly associated with the cytoskeleton and is detected in the Triton insoluble fraction of the cell using 1ST or in situ hybridization ("ISH") techniques. 1ST or ISH detection of poly (A) RNA gives similar patterns in extracted and unextracted cells. Likewise, priming of the RT reaction was also detected after in vivo hybridization in unextracted cells, although the signal was sometimes less intense. We obtained data showing a field in which cells exposed to ST30tr and fixed immediately in 4% formaldehyde were subjected to 1ST as described. Silver stained cells exhibit an intracellular distribution of signal that correlates with the perinuclear distribution pattern observed in cells that were Triton extracted before fixation. Therefore, we concluded that Triton extraction did not alter the distribution of poly (A) RNA in a way that affects the oligo hybridization pattern, as compared to unextracted cells.

Because these cells were not fixed with glutaraldehyde, which induces autofluorescence, we were able to visualize the fluorescent signal from ST30tr in this same population of cells. This signal represented the cellular distribution of total ST30tr, both hybridized and unhybridized. In general, cytoplasmic fluorescent signal overlaps with the perinuclear location of silver stain, less fluorescent signal or silver stain is seen in the extreme periphery of the cells. However, some cells and/or nuclei that contained oligo (as detected by fluorescence) did not show hybridization (silver stain) at all. A simple explanation for this may be that the 1ST reagents cannot diffuse into unextracted cells or nuclei as readily as cells permeabilized by Triton extraction. Kinetics of Hybridization In Living Cells

Using the modified 1ST technique described here, we quantitated the amount of oligo dT hybridization to poly A RNA in living cells as a function of both incubation time and oligo concentration during uptake. Living cells were exposed to varying amounts of fluorescent S-oligo or O-oligo dT for varying amounts of time and then extracted and fixed. 1ST was performed with silver enhanced signal detection as before and average silver stain present per cell was quantitated using a digital imaging workstation, DISCOVERY™. Fig. 3A shows that intracellular hybridization was reproducibly detected after cells had been incubated with ST30tr for 30 minutes. Signal resulting from hybridization was detected as early as 5 minutes after ST30tr oligo treatment began (not shown) . Similar kinetics were observed when O-oligo dT was used, although saturation was not reached as early. These results correlated with intracellular quantitation of fluorescence; oligo that entered the cell rapidly (as measured by fluorochrome detection) also hybridized rapidly (represented by silver stain) . Fig. 3B shows that the maximal detectable hybridization in the cell occurs at external oligo concentrations of about 0.5 μM for both T43tr and ST30tr. Therefore, hybridization appears to saturate at lower concentrations than those required for internalized S-oligo to reach a steady state (concentration greater than 1 μM, see Fig. 2B) . The intracellular distribution of hybridization does not change at the light microscopy level as the concentration or time of incubation increases (not shown) . Hybrid Stability In Living Cells

The amount of fluorescently-labeled S-oligo dT in the cell was next measured as a function of efflux time. Cells incubated with ST30tr for 2 hours were washed with media and allowed to grow in oligo free media for varying lengths of time. Fluorescence in the formaldehyde-fixed cells was quantitated using DISCOVERY™. At most, a 20% decrease in the amount of total fluorescent ST30tr in the cell was seen after 12 hours of efflux (Fig. 4A) . The intracellular distribution of the fluorescent oligo did not change detectably over the efflux periods examined (not shown) and resembled that shown in Fig. 2A. Levels of fluorescent SA30tr also remained high after long efflux times (not shown) . It was not possible to accurately quantitate fluorescent label in individual cells treated with O-oligo dT or dA because of high levels of extracellular fluorescence. It may be these oligos interact with extracellular components or simply stick to the coverslips. However, diffuse fluorescence remained detectable in both T43tr and A43tr treated cells up to 18 hours after efflux.

Hybrid half-life was next measured after efflux times similar to those described above. After incubation with S- or O-oligo dTs (ST30tr or T43tr) and appropriate efflux times, cells were extracted, fixed and subjected to 1ST as described above. The amount of silver stain in each cell was quantitated using DISCOVERY™. Hybridization was still detected after a 30 minute efflux time in cells treated with either ST30tr or T43tr (Fig. 4B) . Surprisingly, however, cells treated with ST30tr showed no hybridization by 6 hours efflux time, even though fluorescent levels of ST30tr remained high (compared in Fig. 4A) . This indicates that the fluorescent signal represented unhybridized S-oligos retained by the cells.

In contrast to the loss of hybridization signal with ST30tr, signal representing hybridization of T43tr could still be detected in cells after 18 hours of efflux time (Fig. 4B) . The amount of this hybridization varied from cell to cell, ranging from 10 - 75% of the initial hybridization. This might be expected of an unsynchronized cell population such as used here; cells that have divided would contain less hybridized oligo than undivided cells. Given an 18 hour division time, about 50% of maximal signal would be expected. Taken together, these experiments show that cells treated with either S- or O-oligo dT retain oligo dT/poly (A) RNA hybrids when transferred to oligo free media and additionally, that fluorescently-labeled O-oligo dT/poly (A) RNA hybrids are more stable in cells than S-oligo dT/poly (A) RNA hybrids. Other Embodiments Other embodiments of the invention are within the following claims.

Claims

Claims We claim:

1. A method for inhibiting the biological degradation of a synthetic oligodeoxynucleotide or oligodeoxynucleotide analog (oligo) in a cell or an extracellular environment, said method comprising the steps of:

(a) incorporating at least one modified base into a synthetic oligo; (b) covalently attaching a protection moiety onto said modified base in said oligo, thereby creating a biological degradation-resistant oligo; and

(c) placing said biological degradation-resistant oligo into a cell or an extracellular environment, where its biological degradation is inhibited.

2. The method of claim 1, wherein said oligo is an O-oligo.

3. The method of claim 1, wherein said oligo is an S-oligo.

4. The method of claim 1, wherein the length of said oligo is from 5 to 200 nucleotides.

5. The method of claim 4, wherein the length of said oligo is from 10 to 100 nucleotides.

6. The method of claim 5, wherein the length of said oligo is from 15 to 50 nucleotides.

7. The method of claim 1, wherein said modified base comprises a functional group through which a fluorochrome is covalently attached to said oligo.

8. The method of claim 7, wherein said functional group is a primary amino group.

9. The method of claim 7, wherein said functional group is at the end of a spacer arm.

10. The method of claim 1, wherein said synthetic oligo comprises a modified base within five bases from the 3'end of the oligo.

11. The method of claim 10, wherein said modified base is in the ultimate or penultimate base position at the 3'end of the oligo.

12. The method of claim 1, wherein said synthetic oligo comprises a multiplicity of modified bases with covalently attached fluorochromes.

13. The method of claim 12, wherein said modified bases are incorporated at approximately every tenth base position in the nucleotide sequence of said oligo.

14. The method of claim 1, wherein said oligo is complementary to a target nucleic acid.

15. The method of claim 14, wherein said target nucleic acid is DNA.

16. The method of claim 14, wherein said target nucleic acid is an RNA.

17. The method of claim 16, wherein said RNA is a messenger RNA.

18. The method of claim 14, wherein said target nucleic acid is a target cell nucleic acid.

19. The method of claim 18, wherein said target cell nucleic acid is an mRNA selected from the group consisting of:

(a) cdc2 kinase mRNA;

(b) proliferating-cell nuclear antigen (PCNA) mRNA; and

(c) c-myb mRNA.

20. The method of claim 14, wherein said target nucleic acid is a viral nucleic acid.

21. The method of claim 20, wherein said viral nucleic acid is HIV-l rev mRNA.

22. The method of claim 14, wherein said target nucleic acid is in a prokaryotic target cell.

23. The method of claim 14, wherein said target nucleic acid is in a eukaryotic target cell.

24. The method of claim 23, wherein said eukaryotic target cell is in culture.

25. The method of claim 23, wherein said eukaryotic target cell is in intact tissue.

26. The method of claim 1, wherein said protection moiety is Texas red.

27. The method of claim 1, wherein said protection moiety is fluorescein isothiocyanate.

28. A method for producing an intracellular nucleic acid hybrid with enhanced stability against biological degradation, said method comprising the steps of: (a) incorporating at least one modified base into a synthetic oligodeoxynucleotide or oligodeoxynucleotide analog, said synthetic oligodeoxynucleotide or oligodeoxynucleotide analog being capable of hybridizing with a target nucleic acid; (b) covalently attaching a protection moiety onto said modified base in said synthetic oligodeoxynucleotide or oligodeoxynucleotide analog, thereby creating a biological degradation-resistant oligo;

(c) allowing said biological degradation-resistant oligo to enter a target cell; and

(d) allowing said biological degradation-resistant oligo to hybridize with said target nucleic acid, thereby producing an intracellular nucleic acid hybrid with enhanced stability against biological degradation.

29. The method of claim 28, wherein said protection moiety is a fluorochrome and said target nucleic acid is an mRNA.

30. A method for producing an intracellular nucleic acid hybrid with enhanced stability against biological degradation, said method comprising the steps of:

(a) incorporating at least one modified base, comprising a spacer arm which terminates in a functional group, into a synthetic oligodeoxynucleotide whose constituent deoxynucleotide residues are linked (5' to 3') by phosphodiester bonds, said synthetic oligodeoxynucleotide being capable of hybridizing with a target nucleic acid; (b) covalently attaching a protection moiety onto said spacer arm of said modified base in said synthetic oligodeoxynucleotide, thereby creating a biological degradation-resistant oligodeoxynucleotide; (c) allowing said biological degradation-resistant oligodeoxynucleotide to enter a living target cell in vitro;

(d) allowing said biological degradation-resistant oligodeoxynucleotide to hybridize with said target nucleic acid, thereby producing an intracellular nucleic acid hybrid with enhanced stability against biological degradation; and

(e) measuring the stability of said intracellular nucleic acid hybrid in said living cell in vitro, whereby inhibited biological degradation of said intracellular nucleic acid hybrid is demonstrated.