WO1999061071A2 - Assay in vivo of labeled triplex-forming oligonucleotides - Google Patents

Abstract

Triplex-forming oligonucleotides (TFOs) can form triplex complexes with target nucleotide sequences in a nucleic acid in vitro which, upon introduction into cultured eukaryotic cells, remain stable inside the cells for at least 48 hours. The stability of triplex complexes formed by binding of 125I-labeled TFOs to target nucleic acids can be quantitatively monitored by radioprinting, wherein the number of 125I-induced DNA strand breaks introduced into the target nucleic acid is determined, and is proportional to the number of triplex complexes present in the nucleic acid. According to the present invention, TFOs are labeled in vitro with imaging agents detectable in vivo by imaging methods such as gamma camera imaging, rectilinear scanner imaging, magnetic resonance imaging (MRI), positron emission tomography (PET), and single photon emission computed tomography (SPECT). The labeled TFOs are allowed bind to and form triplex complexes with their target nucleic acids, and a pharmaceutical composition comprising the labeled triplexes is administered to a multicellular organism. Imaging methods that detect the labeled nucleic acid are then employed to non-invasively monitor the biodistribution of the nucleic acids in the subject organism. The present invention thereby provides an important new tool for developing nucleic acid delivery methods for gene therapy protocols.

Description

METHODS FOR LABELING NUCLEIC ACID VECTORS

WITH TRIPLEX-FORMING OLIGONUCLEOTIDES AND

MONITORING VECTOR DISTRIBUTION IN VIVO

TECHNICAL FIELD

The present invention pertains to methods for labeling nucleic acids such as recombinant double-stranded DNA (dsDNA) expression vectors and, following a(irninistrauon of the labeled nucleic acid molecules to a living animal or human, non- invasively measuring the distribution of the labeled nucleic acid molecules in organs and tissues of the subject's body. The present invention provides a rapid, non-invasive means for assessing the effectiveness of techniques for delivering therapeutic gene expression vectors to their target tissues in vivo. The present invention also pertains to methods in which nucleic acids such as dsDNA molecules are labeled and are introduced into cells, and the intracellular disposition of the labeled DNA molecules is determined in vitro.

According to the methods of the present invention, (i) triplexΛbrming oligonucleotides (TFOs) labeled with detectable chemical groups are allowed to bind to native dsDNA molecules to form triple-stranded nucleic acid complexes (triplexes) that are stable inside of cells under normal growth conditions, (ii) the resulting labeled triplex-contairiing DNA molecules are introduced into a multicellular organism or into cultured cells, and (iii) the distribution of the labeled DNA molecules in the treated organism or cells is quantitatively monitored by detecting the labeled triplex DNA molecules in the subject organism or cells. Non-invasive imaging methods such as

Nuclear Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) are used to detect the distribution of the labeled DNA molecules in vivo. Intracellular distribution of the labeled DNAs is carried out by commonly used methods for detecting labeled molecules in cells; e.g., by visual detection of fluorescent labels fluorescence microscopy, or by detection of radiolabels by autoradiography. BACKGROUND ART

Efforts to treat disease by administering recombinant nucleic acid expression vectors carrying therapeutic genes as pharmaceutical agents (gene therapy) to human patients who would benefit from expression of the genes in specific cells and tissues of their bodies are generally confounded by the inability of currently available methods to deliver therapeutically effective amounts of the nucleic acid expression vectors to target cells and tissues of the patients. Medical researchers attempting to introduce therapeutic DNA expression vectors into cells in diverse tissues, such as brain (Weyerbrock et al., Current Opinions in Oncology, 1999, 11 (3): 168-73), liver (DiCampli et al., European Journal of Gastroenterology, 1999, l l (4):421-429), skin

(Deng et al., Nature Biotechnology, 1887, 15(13):1388-1391), and blood (Engel et al., Frontiers in Biosciences, 1999, 4:e26-33), have called for development of new and more effective methods lor delivering DNA vectors to their target cells in vivo. In order to develop improved techniques for delivering DNA expression vectors to target tissues in vivo, methods for quantitatively monitoring the biodistribution of the DNA molecules in experimental subjects are needed. Currently available methods for monitoring biodistribution and stability of DNA expression vectors in vivo, such as detection by Southern analysis or Polymerase Chain Reaction, or by measuring expression of reporter genes present in the vectors, necessitate killing the experimental animals and removing the tissues for analysis (Thierry et al., Gene Therapy, 1997,

4(3):226-7; Li et al., Gene Therapy, 1997, 4(9):891-900; Norman et al., Vaccine, 1997, 15(8):801-3; Baru et al., Journal of Drug Targeting, 1998, 6(3):191-9). Monitoring methods which require killing the experimental subjects are disadvantageous because such methods cannot be used on human subjects, and the expense of supplying the experimental animals may limit the number of experimental conditions that can be tested, particularly if the subjects are costly transgenic disease model animals. Biodistribution of DNA expression vectors carrying the firefly luciferase reporter gene has been monitored non-invasively in mice and zebrafish by detecting biolumincsccnce in tissues where the luciferase gene is expressed (Contag et al., Photochem. Photobiology, 1997, 66(4):523-31; Patil et al., Zoological Science, 1994, 11 (l):63-68); however, it is unclear whether bioluminescence can be detected non-invasivcly in internal tissues of large animals such as humans. Monitoring methods based on detection of expression of a reporter gene present on the DNA vectors in tissues in vivo are incapable of identifying those tissues which successfully take up the DNA vectors but fail to express the reporter gene; hence, such methods do not quantitatively assay the physical biodistribution of the DNA vectors. Monitoring the intracellular distribution of recombinant DNA plasmids that have been introduced into cells is problematic, because such plasmids are commonly used in circular form, as circular plasmids are more stable than linearized ones; however, labeling moieties are not readily attached to covalently closed, circular DNA molecules. Kayyem et al. found that cells treated with DNA expression vectors bound to polylysine to which are conjugated transferrin molecules and chelated gadolinium ions showed specific contrast enhancement with MRI, and they proposed using MRI to non-invasively monitor the biodistribution of such labeled DNA molecules in vivo for developing gene therapy protocols (Chem. Biology, 1995, 2(9):625-20).

DNA oligonucleotides and their analogs conjugated to imaging agents such as fluorine-18 ("¥), technetium-99m CTc) and indium-I l l ('"-In) have been administered to experimental animals, and their biodistribution in the animals has been monitored by PET (Tavitian et al., Nature Medicine, 1998, 4(4):467-71) and whole-body gamma camera imaging (Dewanjee et al., The Journal of Nuclear Medicine, 1994, 35(6):1054-63; Hnatowich et al., Journal of Pharmacology and Experimental Therapeutics, 1996, 276(l):326-34). Three general motifs of triplex-forming oligonucleotides (TFOs) have been described. Pyrimidine motif TFOs are composed of pyrimidines comprising thymine (T) and cytosine (C), and bind in parallel orientation to a run of purines in duplex DNA by Hoogsteen base-pairing in the major groove of the DNA, with T in the TFO pairing with adenine (A) in the target DNA, and C in the TFO pairing with guanine (G) in the target DNA. Purine motif nucleotides are composed of purines comprising A and G, and bind to a run of purines in duplex DNA by reverse Hoogsteen base- pairing in the major groove, with G in the TFO pairing with G in the target DNA, and A in the TFO pairing with A in the target DNA (Beal et al., Science, 1991, 251:1360-3; Debin et al., Nucleic Acids Research, 1997, 25(10):1965-74; Dervan et al., U.S. Patent No. 5,874,555, in entirety). GT-type TFOs comprise G and T bases, and bind in the major groove to a run of at least about 65% purines in duplex DNA, with G in the TFO opposite a GC pair in the target DNA, and T in the TFO opposite an AT pair in the target DNA; the binding may be in either parallel or anti-parallel orientation, depending on the target nucleeotide sequence (Hogan et al., U.S. Patent No. 5,176,996, in entirety; Debin et al., Nucleic Acids Research, 1997, 25(10):1965). Debin et al. allowed unmodified, deoxynucleotide, purine motif TFOs to bind to their target sequences in dsDNA in vitro to form triplex complexes, introduced the preformed triplexes into cultured cells, and showed by DMS footprinting that the triplex complexes were stable in the cells for at least three days (Nucleic Acids Research, 1997, 25(10):1965).

DISCLOSURE OF THE INVENTION The present invention provides methods in which TFOs are labeled with chemical moieties that are readily detected by imaging systems, the labeled TFOs are bound to target sequences in nucleic acid molecules in vitro to form triplex complexes, the labeled triplex complexes are introduced into a living organism, and non-invasive imaging means are used to quantitatively monitor the biodistribution of the labeled nucleic acid molecules in the organism. To obtain results relevant to developing methods for delivering nucleic acid expression vectors that provide therapeutic benefit, the nucleic acid molecules of the present method are selected to have the same size and structure as nucleic acid gene expression vectors, e.g., dsDNA plasmids, that are designed for gene therapy. The TFOs are labeled with detectable chemical groups, for example, with paramagnetic metal ions detectable by MRI, with gamma- or positron- emitting radionuclides detectable by a gamma camera or PET, or with a lluorophore detectable by its fluorescence in ultraviolet light. The labeled TFOs are complexed with nucleic acids molecules to form triplex complexes, and the triplexes are administered to a living organism. The biodistribution of the triplex nucleic acid molecules in the subject organism after one or more time intervals subsequent to administration of the triplexes is then monitored by a device that detects the labeled TFOs.

The present invention permits monitoring the intracellular distribution of nucleic acid vectors such as dsDNA plasmids by allowing the nucleic acid vectors to form triplex complexes with TFOs that are labeled with a fluorophore, introducing the labled triplexes into cells, and viewing the cells in vitro using fluorescent and confocal microscopy. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a map of the pCR3HPRT plasmid. The 833 base pair (bp) PCR fragment cloned into pCR3 vector (pos. 724-1556) is shown in bold. The amplified inset shows the target sequence along with TFO. Positions of '"I dC's are marked with stars. The broad line between the Hind III and Nde sites represents the fragment cut from the plasmid by Hind III and Nde I restriction enzymes for use as probe for Southern hybridization.

Figure 2 (A-D) shows the rate of triplex formation in vitro (Figs. 2 A and 2B) and the dependence of the rate of binding on plasmid concentration (Figs. 2C and 2D).

Figs. 2A and 2B: 1 nM ¹²⁵I-TFO and 10 nM pCR3HPRT plasmid were incubated at 37°C for 0, 1, 5, and 24 hrs. in TMSp buffer, the DNA molecules were electrophoresed in 2% agarose, the distribution of radioactivity in the gel was measured

(Fig. 2 A) and the fraction of TFO bound to plasmid as a function of time was graphed (Fig. 2B).

Figs. 2C and 2D: different concentrations of plasmid (0-50 nM) were incubated with a constant concentration of ¹²⁵I-TFO overnight, the DNA molecules were electrophoresed in 2% agarose, the distribution of radioactivity in the gel was measured

(Fig. 2C), and the fraction of TFO bound to plasmid as a function of plasmid concentration was graphed (Fig. 2D).

Figure 3 (A & B) shows the dependence of triplex stability in vitro on temperature (Fig.

3 A) and on the concentration of Mg^*'2 ions (Fig. 3B) .

Figure 3 A: pre-formed triplex was diluted in a buffer containing 0.5 mM MgC and aliquots were incubated at room temperature, 37°C, and 50°C for 30 min, and at

90°C for 1 min., the DNA molecules were electrophoresed in 2% agarose, and the distribution of radioactivity (^12sI-TFO) in the gel was measured.

Figure 3B: pre-formed triplex was diluted in buffer to ImM Mg^*2 and aliquots were incubated for 10 min at 37°C with 0.5, 1.0, 2.5, and 5.0 mM EDTA, which chelates Mg^*2, the DNA molecules were electrophoresed in 2% agarose, and the distribution of radioactivity (^12,I-TFO) in the gel was measured. Sixty percent of the triplexes incubated with 0.5 mM EDTA dissociated, and all the triplexes incubated with higher concentrations of EDTA dissociated completely. Figure 4 (A-C) shows that the number of breaks introduced by ¹ T-TFO in dsDNA in vitro is directly proportional to the fraction of plasmid to which ¹²T-TFO is bound. Triplexes were formed in the presence of different concentrations of ¹²ⁱI-TFO, and then were frozen and incubated at -70°C to permit strand breaks to occur.

Figure 4A shows the distribution of ethidium bromide-labeled DNA molecules in an agarose gel following electrophoresis. The percentage of the plasmid carrying TFO is shown at the top.

Figure 4B shows an autoradiograph made following Southern blot of the gel, and hybridization with a radio-labeled probe.

Figure 4C shows the percentage of DNA breaks plotted against the percentage of plasmid bound to TFO. Percent of breaks was calculated as the ratio of the sum of the intensities of the two shorter bands in the case of ethidium stained gel (•) and the intensity of the single shorter band in the case of Southern blot (ϋ) to the total intensities of the bands in lanes.

Figure 5 (A &B): Triplex radioprinting in cells.

Fig. 5A: Southern blot of Hirt extracts from HeLa cells that have been transformed with preformed triplexes (¹²T-TFO/pCR3HPRT). Cells were incubated with triplexes for 5 hr, postincubated for 17-43 hr, then frozen and stored at -70°C to accumulate ^12SI decays. The recovered plasmid was cut with Pvul, and the fragments were separated by electrophoresis and transferred to a hybridization blot. The blot was hybridized with ³²P-labeled Hind III-Nde I probe. The positive control was triplex that was stored in a test tube without delivery into cells (Triplex). The negative control was pCR3HPRT plasmid delivered into cells and Hirt-extracted (Plasmid).

Fig. 5B: Percent double-strand plasmid DNA breaks vs. time of incubation of the cells with preformed triplexes.

Figure 6 (A-D) shows the distribution of triplexes inside cells. HeLa cells were transfected with pre-formed triplexes (FITC-TFO/pCR3HPRT plasmid) complexed with DMRIE liposomes for 5 hr. Confocal (Figure 6A) and fluorescent (Figure 6C) microscope analysis showed that FITC-labeled triplexes were equally distributed in cytoplasm and nuclei of transfected cells. Confocal (Figure 6B) and fluorescent (Figure 6D) microscopy of control cells transfected for 5 hr with FITC- TFO/DMRIE shows that fluorescence is concentrated in bright grains in cytoplasm. Magnification: lOOx.

Figure 7 (A-C) compares assay of plasmid uptake based on physical detection of triplexes to assay based on detecting reporter gene expression.

Fig. 7A shows HeLa cells transfected with pCMV-sport-β-gal plasmid, stained with β-gal staining kit, and counterstained with nuclear fast red.

Fig. 7B shows HeLa cells transfected with preformed FITC-TFO/ pCR3HPRT triplexes and monitored for flourescence.

Fig. 7C shows HeLa calls transfected with preformed ¹²⁵I-TFO/pCR3HPRT triplexes (autoradiography); and counterstained with hematoxylin and eosin. Magnification: 40x.

MODES FOR CARRYING OUT THE INVENTION

The method of the present invention operates with any nucleic acid molecule that contains a target nucleotide sequence to which a labeled TFO stably binds to form a triplex complex. The nucleic acid molecules which bind the labeled TFOs to form triplexes can be single- or double-stranded DNA or RNA molecules that are linear or circular, native or recombinant, and they can be plasmids, viral genomes, episomes, or artificial chromosomes (see Huxley, Gene Therapy, 1994, 1(1):7-12). In a preferred embodiment, the TFO-binding nucleic acids of the present invention are circular dsDNA molecules such as plasmids. Methods for identifying sites in a dsDNA molecule to which TFOs stably bind to form triplex complexes are well known (for example, see Beal et al., Science, 1991, 251 :1360-3; Helenc,C, Anti-Cancer Drug Design, 1991, 6:569-584; Frank-Kamenetskii et al., 1995, Annual Review of Biochemistry, 64:65-69; Lacoste e al., Nucleic Acids Research, 1997, 25(10):1991-8; Delporte et al., Antisense Nucleic Acid Drug Development, 1997, 7(5):523-9; Xodo et al., European Journal of Biochemistry, 1997, 248(2): 424-32; Giovannangeli et al.,

Antisense Nucleic Acid Drug Development, 1997, 7(4):413-21 (review)). Skilled artisans recognize that DNA and RNA oligonucleotides having a number of different structures function effectively as TFOs to form stable triplex complexes with specific sequences in target DNA and RNA molecules. For example, one can prepare DNA and RNA TFOs that bind to single-stranded DNA and RNA molecules, such as fold- back TFOs (Hiratou et al., Nucleic Acids Symposium Series, 1997, 37:221-2; Kandimalla et al., Nucleic Acids Research, 1995, 23 (6): 1068-74), tethered TFOs (Moses et al., Bioorganic Med. Chemistry, 1997, 5(6):1123-9), dsDNA probes

(Noonberg et al, Biotechniques, 1994, 16(6) 1070-2, 1074); and circular DNA and RNA TFOs (Wang et al., Nucleic Acids Research, 1994, 22(12):2326-33; Wang et al., Biochemistry, 1995, 34(30):9774-84). Those skilled in the art recognize that one can also prepare TFOs that are structurally modified to have increased triplex stability or resistance to nucleases; e.g., by introducing nucleoside analogs (Wang et al., Bioorganic

Med. Chemistry, 1997, 5(6):1043-50; and Wang et al., 1995, Nucleic Acids Research, 1994, 23(7):1157-64, for example), or by attaching chemical moieties such as intercalating groups (Kukreti et al., Nucleic Acids Research, 1997, 25(21)4264-70; Garbesi et al., Nucleic Acids Research, 1997, 25(l l):2121-8, for example). The methods of the present invention can be practiced successfully with any of the diverse types of TFOs known in the art that stably bind to their target nucleic acid molecules to form triplex complexes. In a preferred embodiment, the TFOs of the present invention are single-stranded, polypurine, deoxynucleotide TFOs.

As used herein, "stably bound" means that the labeled TFO remains bound to the dsDNA under physiological conditions for a time period of sufficient duration that monitoring of biodistribution can be carried out. The time period of stability of the triplex required for monitoring biodistribution be from 10 minutes or less to about one hour, or to as long as two or more days, depending on the goal and design of the assay protocol. As used herein, "physiological conditions" refers to chemical and physical conditions in cells or in tissues in a living organism in which dsDNA biodistribution is to be monitored.

The detectable labeling agents of the present invention are compounds that arc detected in vivo by accepted non-invasive techniques in the art of diagnostic imaging.

A number of radionuclide atoms can be used to label detectably the TFOs in accordance with the present invention. These include both gamma-emitters and positron emitters; examples of which include, but are not limited to, fluorine-18, copper-64, copper-65, gallium-67, gallium-68, bromine-77, ruthenium-95, ruthenium- 97, ruthenium- 103, ruthenium- 105, technetium-99m, mercury 107, mercury-203, iodine-123, iodine-125, iodine-126, iodine-131, iodine-133, indium-I l l, indium-113m, rhenium-99m, rhenium-105, rhenium-101, rhenium-186, rhenium-188, tellurium- 121m, telurium-122m, tellurium-125m, thulium-165, thulium 167, thulium-168, and nitride or oxide forms derived therefrom. Metal atoms detectable by MRI, such as ions of manganese, iron, gadolinium, terbium, dysprosium, holmium, erbium, ytterbium, lutetium, holmium, bismuth, lead and hafnium, can also be used as labeling agents. Those skilled in the art of diagnostic labeling recognize that metal ions such as those listed above can be bound by chelating moieties, which in turn can be conjugated to the TFOs of the present invention. For example, gadolinium ions are chelated by diethylenetriaminepentaacetic acid (DTP A), and a number of lanthanide ions, including gadolinium and dysprosium, are chelated by terraazacyclododocane compounds (Klaveness et al., U.S. Patent No. 5,738,837, in entirety; and Meade et al., U.S. Patent No. 5,707,605, in entirety); "Tc is chelated by an N-hydroxysuccinimide derivative of mercaptoacetyltriglycine (NHS-MAGs, see Mardirossian et al., The Journal of Nuclear Medicine, 1997, 38(6):908); and '"In is chelated by DTPA isothiocyanate (Dewanjee et al., The Journal of Nuclear Medicine, 1994, 35(6):1055).

The chelating groups can be attached to the TFOs by known methods, e.g., via a nitrogen atom introduced at one of the TFO termini (as in Dewanjee et al., The

Journal of Nuclear Medicine, 1994, 35(6):1055). ATFO that is detectable by PET can be produced, for example, by attaching a chemical moiety containing a positron- emitting ¹⁸F atom to either TFO terminus (Tavitian et al., Nature Medicine, 1998, 4(4):467-71), or by incorporating a nucleoside analog containing a positron-emitting "C atom into the TFO (Conti et al., 1995, Nucl. Med. Biol. 22(6):783-789). For monitoring DNA distribution in tissues and cells in vitro, a number of visually detectable fluorescent labels that can be conjugated to TFOs, and that emit light of different wavelengths, are available (Tyagi et al., Nature Biotechnology, 1998, 16:49- 53).

Labeled TFOs are incubated in the presence of the target nucleic acids and divalent cations to obtain formation of stable triplex complexes using routine methods known by those skilled in the art.

The composition comprising the labeled triplex complexes prepared for use in monitoring biodistribution of the triplexes in a subject animal in accordance with the present invention comprises a pharmaceutically acceptable carrier. In this context, "pharmaceutically acceptable" means acceptable for use in the pharmaceutical and veterinary arts; i.e., a carrier which is non-toxic and which does not adversely affect the activity of the composition in its function to deliver the labeled nucleic acids to their target tissues, or the monitoring of the biodistribution of the labeled nucleic acids. It is within the knowledge of those skilled in the art of compositions comprising nucleic acid vectors for administration in vivo to select and include a pharmaceutically acceptable carrier in the compositions of the present invention. Those skilled in the art of preparing pharmaceutically acceptable compositions comprising nucleic acid vectors for administration in vivo recognize drat such compositions can also comprise chemical agents that assist in delivering the nucleic acids into their targeted cells in vivo, such as anionic or cationic lipids, polycations, and compounds that bind to specific cell-surface receptors and promote introduction of the nucleic acids into the cells bearing the receptors on their surface (see, for example, Alino, Biochem. Pharmacology, 1997, 54(1):9-13; Liu et al., Journal of Biological Chemistry, 1995, 270(42) :24864-70; Hong et al., FEBS Letters, 1997, 400(2):233-7; Thierry et al.,

Proceedings of the National Academy of Sciences U.S.A., 1995, 92(21):9742-6; Dzau et al., Proceedings of the National Academy of Sciences U.S.A., 1996m 93(12) 11421- 5). When the labeling moiety attached to the TFO is a radionuclide, stabilizers to prevent or minimize radiolytic damage, such as ascorbic acid, gentisic acid, or other appropriate antioxidants, may be added to the composition comprising die labeled triplexes that is administered to the subject organism.

Those skilled in the art of preparing and administering pharmaceutically acceptable compositions comprising nucleic acid vectors in vivo recognize that the compositions comprising labeled triplex complexes for use in monitoring biodistribution of the triplexes in a subject animal in accordance with the present invention can be administered by many of the same routes that are commonly used to administer conventional drugs; for example, by intravenous, intraperitoneal, or intramuscular injection, by aerosol inhalation, by intratracheal installation, and by injection direcdy into a target tissue (e.g., a tumor) (for example, sec Canonico et al., Journal of Applied Physiology, 1994, 77(l):415-9; Cooper, Seminars in Oncology,

1996, 23(l):172-87; Trainer et al., Human Molecular Genetics, 1997, 6(10):1761-7; Li et al., Journal of Molecular and Cellular cardiology, 1997, 29(5): 1499-504; Griesenbach et al., 1998, 5(2):181-8). The dosages of the compositions comprising detectably labeled triplexes of the present invention are established in controlled trials, and correspond to an amount sufficient to allow detection of the labeled triplexes in tissues of the subject organism following administration, as compared to the background signal obtained upon administration of an appropriate control composition, without causing intolerable side effects and without unacceptable exposure to radioactivity. It will be appreciated that the dosages required to obtain a desired measure of biodistribution will vary according to the specific organism or individual used as the subject (i.e., species, age, sex, and general health), the chemical make-up of the triplex-containing composition (liposomes, cationic lipids, polyanions, targeting peptides, etc.), the route of administration, the type of labeling moiety, and the imaging method (MRI, PET, SPECT, etc.) that are used.

Those skilled in die diagnostic art appreciate that die instrumentation that may be used in the detection of die labeled TFOs depends on the type of label attached to the triplexes, and on the type of target tissue or cells being imaged. For imaging TFOs linked to paramagnetic labels, one can use, for example, a superconducting quantum interference device magnetometer (SQUID, see Klaveness et al., U.S. Patent No. 5,738,837). A gamma camera and a rectilinear scanner each represent instruments useful to detect radioactivity in a single plane. Single Photon Emission Computed Tomography (SPECT) and PET devices represent instruments that are capable of detecting radioactivity in more than one dimension. Imaging instruments suitable for practicing the methods of the present invention are readily available from commercial sources in the U.S. (for example, for PET: ADAC, Milpitas, CA; Siemens, Hoffman Estates, IL; Concorde Microsystems, Inc, Knoxville, TN; for MRI: Picker International, Inc., Cleveland, OH; Siemens, Iselin, NJ; GE, Waukesha, WI; and for SPECT: Toshiba America/USA, Tustin, CA; Siemens, Hoffman Estates, IL; ADAC,

Milpitas, CA).

METHOD FOR DETERMINING STABILITY OF A TRIPLEX COMPLEX IN A CELLFREE SOLUTION IN VITRO AND IN A CELL Before administering a composition containing labeled triplex complexes and determining die biodistribution of die triplexes in vivo according to die present invention, the stability of the triplexes under physiological conditions should be assayed in order to verify that the triplex complexes will remain intact for the duration of the biodistribution analysis. The following examples demonstrate radioprinting, a method for detecting triplex complexes comprising '"I-labeled TFOs based on measurement of DNA strand breaks at sites in the target duplex DNA proximal to the decay site (Panyutin et al., Nucleic Acids Research, 1994, 22(23):4979-82; Panyutin et al., Nucleic Acids Research, 1997, 25(4):883-7). Radioprinting may be used to determine the stability of a triplex complex comprising a TFO and a nucleic acid of interest under physiological conditions.

Nucleic acid to be labeled pCR3HPRT plasmid containing the triple helix-forming polypurine- polypyrimidine region of human hypoxanthine-guanine phosphoribosyl transferase (HPRT) gene was constructed by inserting the PCR-amplified fragment into pCR3 vector (Invitrogene, Carlsbad, CA) (Panyutin et al., Nucleic Acids Research, 1994, 22(23):4979-82). A map of the pCR3HPRT plasmid containing the 832 bp insert from the human HPRT gene intron A (Panyutin et al., Acta Oncol., 1996, 36:817-824;

Panyutin et al., Nucleic Acids Research, 1997, 25(4):883-7) along with the triplex forming sequence and purine motif TFO, are shown in Fig.l . Prior to use, supercoiled plasmid was relaxed with topoisomerase I (Promega, Madison, WI). pCMV-sport-β- gal plasmid containing β-galactosidase gene was purchased from Gibco BRL. All plasmids were purified by centrifugation through CsCl gradient.

Triplex-forming oligonucleotides

Oligonucleotides were synthesized on an ABI-394 DNA syndiesizer (Applied Biosystems, Foster City, CA) followed by purification from a polyacrylamide gel (PAG). The template oligonucleotide was biotinylated using BioTEG modifiers (Glen

Research, Sterling, VA). Phosphodiester TFOs were labeled with ¹² -dCTP at the C5 position of three cytosines (marked with stars on Fig. 1) by primer extension method (Panyutin et al., Acta Oncol., 1996, 36:817-824; Panyutin et al., Nucleic Acids Research, 1997, 25(4):883-7). The product TFOs were estimated to have 1.5 ^1MI per oligonucleotide. Fluorescein (FITC) labeled TFOs were synthesized using FITC modifiers (Glen Research). Triplex formation and stability

^,2T-TFOs or FITC-TFOs were mixed with relaxed pCR3HPRT plasmid in a TMSp buffer containing 50 mM Tris-HCl, pH 8, 10 mM MgCL, 0.1 mM Spermidine and 16 μM Coralyne (15-17). Aliquots of the samples were analyzed by 2% Agarose gel at 10°C using TAE buffer containing 3 mM MgAc as an electrode buffer for 1 hr at 60 volts. Gels were analyzed with a BAS 1500 Bio-Imaging Analyzer (Fuji, Tokyo, Japan) or Fluo-Imager (Molecular Dynamics, Sunnyvale, CA).

Cell line and culture conditions The human cervical epithelial carcinoma HeLa cell line (ATCC, Manassas,

VA; CCL2) was grown in monolayer in Dulbecco's modified Eagle's minimum essential medium (DMEM), supplemented with 0.1 mM non-essential amino acids and 10% fetal bovine serum (FBS), purchased from BioWhittaker (Walkersville, MD). All procedures were performed according to ATCC recommendations.

DNA/Liposomes complexes

For cell transfection experiments, DMRIE liposomes (Gibco BRL, Gaithersburg, MD) were used. Five μl of DMRIE (initial concentration 2mg/ml) was diluted in 45 μl OptiMEM medium (Gibco BRL, Gaithersburg, MD). Three μg of plasmid DNA (or plasmid/TFO triplex) was diluted in OptiMEM to 50 μl. Fifty μl of the liposomes solution were mixed with DNA solution, and the mixture was left for 20- 30 min at room temperature before further use.

Cell transfection Cells were grown in 6-well plates to 70-80% confluency and washed with

OptiMEM medium. Then 900 μl of OptiMEM was added to each well and the cells were put in 37°C CO_a incubator until further use. DNA/DMRIE complexes (100 μl) were added to each well. The cells were incubated for 5 hrs, washed with DMEM containing 3 mM EDTA, and then postincubated in DMEM medium for die desired time. Alter the incubation was completed, the cells were trypsinized, collected, frozen, and stored at -70°C to accumulate ""I decays. Cells precipitated by ccntrifugation and medium were counted in a gamma scintillation counter (Packard Auto-Gamma 5650, Downers Grove, IL) to determine cellular uptake ol initial input dose of radioactivity (Sedelnikova et al, Journal of Nuclear Medicine, 1998, 39:1412-1418).

To determine the transfection efficiency with pCMV-sport-β-gal /DMRIE, the transfected cells were fixed in fixative solution (2% formaldehyde, 0.2% glutaraldehyde in PBS), and stained with β-gal staining kit (Invitrogen). The amount of blue cells was counted with a microscope (Karl Zeiss, Oberkochen, Germany). To check the viability of transfected cells, cells were stained with trypan blue and counted the number of dead and survived cells with a microscope. The cells were then analyzed with fluorescent and confocal microscopy, and by autoradiography (Sedelnikova et al., Journal of Nuclear Medicine, 1998, 39:1412-1418).

Analysis of breaks in DNA pCR3HPRT plasmid DNA was extracted from cells and nuclei, following the modified Hirt extract procedure (Hirt, Journal of Molecular Biology, 1967, 36:365- 369). The cell pellet was then resuspended in 0.5 ml Hirt buffer (0.6% SDS, lOmM EDTA, 10 mM Tris-HCl) and let stand for 10 min with occasional agitation. The suspension was then put into a microtube containing 150 μl of 5M NaCl, incubated on ice overnight, and spun 10 min to precipitate chromosomal DNA. One hundred μg of yeast RNA and 100 μg of proteinase K were added to the supernatant, and the tube was incubated for 1-2 hr at 50°C. DNA was extracted twice with equal volumes of phenol/chloroform, and then once with chloroform. The supernatant from the last extraction was placed into a microtube containing 150 μl of 10 M ammonium acetate and 700 μl of isopropanol. The tube was inverted several times and spun for 30 min in microcentrifuge. The pellet was dried and resuspended in 50 μl of TE buffer, then precipitated with ethanol, washed with 70% cold ethanol, and spun again. The dried pellet was resuspended in TE buffer.

Plasmid DNA was electrophoresed through 1% agarose gel and transferred to a Genescreen hybridization membrane (NEN Life Science Products, Boston, MA). The 405 base pair Hindlll/Ndel fragment of pCR3HPRT plasmid (see Fig. 1) was labeled with ³²P using an oligolabeling kit (Pharmacia Biotech, Piscataway, NJ), and the ³²P labeled oliginucleotides were incubated with the hybridization membrane, following die manulacturer's instructions, and then were visualized with a BAS 1500 Bio-Imaging Analyser. Kinetics of triplex formation

Plasmid (10 nM) was incubated with ^raι-TFO (1 nM) at 37°C for 1 hr, 5 hr and 24 hr in TMSp buffer. As a control, an aliquot of the mixture was not incubated and kept on ice (0 hrs). Formation of triplex complexes was monitored by gel shift assay (Panyutin et al., Nucleic Acids Research, 1994, 22(23) :4979-82; Panyutin et al., Nucleic

Acids Research, 1997, 25(4):883-7), and the results of the kinetics of binding experiment are shown in Figs. 2A and 2B. At 10 nM concentration 50% binding of ¹²⁵I- TFO to the plasmid occur only after 4 hrs of incubation at 37°C. As expected, no binding occurred in the "0 hrs" control (Fig. 2B). Such slow kinetics of formation for the purine-motif triplexes has been observed before (Vasquez et al., Biochemistry,

1995, 34:7243-7251).

Equilibrium binding experiments were carried out with different concentrations of d e plasmid and overnight incubation at 37C; the results are shown in Figs. 2 C and 2D. At 1 nM concentration of both plasmid and TFO, about 50% the TFOs are bound to the plasmids. This shows that binding of the TFOs to the target DNA molecules occurs even at subnanomolar concentrations, which favors use of low concentrations of ^I2SI-TFO that would minimize radiodamage to the cells.

Triplex stability in vitro The triplex complexes were found to be stable alter dilution and incubation at elevated temperatures. To determine the stability of the triplex complexes in vitro, an aliquot of the preformed triplex was diluted 10 times in 20 mM Tris-HCl, 20 mM NaCl, 0.5 mM MgCL buffer and kept overnight at room temperature. Aliquots of the diluted sample were also incubated at 37°C, 50°C for 30 min, and at 90°C for 1 min. Fig. 3A shows that the triplexes remained stable at all the conditions, and complete dissociation of the triplexes was observed only after heating at 90°C .

Next, aliquots of the preformed triplex complexes were diluted in 20 mM Tris-HCl, 20 mM NaCl, buffer without MgCL (the final concentration of magnesium was ImM), and incubated with 0.5, 1, 2.5, and 5 mM EDTA for 10 min at 37°C. As shown in Fig. 3B, 60% of the triplexes incubated with 0.5 mM EDTA dissociated, and the triplexes incubated with higher concentrations of EDTA dissociated completely.

The dependence of triplex stability on the time of incubation (0 - 60 min) at 37°C was then determined in solution containing 0.5 mM EDTA. About 60% of the triplexes dissociated immediately alter EDTA was added, and this amount remained stable during 1 hr incubation (data not shown). This result shows that magnesium is required for stability of the triplex complexes; addition of an equimolar equivalent of EDTA results in immediate dissociation of the triplexes.

Radioprinting in vitro

Radioprinting was first demonstrated using conventional analytical methods which could monitor triplex formation, such as the gel shift assay. A series of samples with different percent of triplex formation was prepared. All the samples contained 80 nM of pCR3HPRT plasmid and the following concentrations of ^,25I-TFO; 8nM, 20nM,

40 nM, 64 nM, and 80 nM. Gel shift analysis showed that all the added ¹²T-TFOs were bound to the plasmids alter overnight incubation at 37°C (data not shown). Therefore, the percent of the plasmid carrying TFO in the samples was 10%, 25%, 50%, 80%, and 100%, respectively. The samples were frozen for 60 days to accumulate ¹²⁵I decays, then thawed, cut with Pvul restriction enzyme and analyzed in 1% agarose gel stained with ethidium bromide (Fig.4 A). Breaks at the triplex forming sequence of pCR3HPRT produced by decay of ¹²⁵I in the bound ^12SI-TFOs, together with breaks introduced by Pvul cleavage, gave rise to two fragments; 3.74 kb and 2.16 kb (Figs.l and 4A). The intensity of the bands corresponding to these fragments increases in proportion to the amount of ^12SI-TFOs present in triplex complexes in the plasmids.

DNA-containing bands from a gel similar to one shown in Fig. 4A, were transfered to a nylon membrane and hybridized with the ³²P-labeled Hindlll-Ndel probe (Figs. 1 and 4B). The probe hybridized only with the 2.15 kb fragment. The intensity of the band corresponding to that fragment increases with increase of the percentage of the triplex. Fig. 4C shows the percent of breaks, calculated from the ratio of the intensity of the decay-produced bands to the total intensity of all bands in the lanes, plotted against the percentage of triplexes. The ¹²ⁱⁱI-TFO produced breaks are direcdy proportional to the percent of the triplexes on the plasmid. Thus, quantitation of breaks introduced into a nucleic acid by bound ¹²⁵I-TFOs by the radioprinting assay provides an accurate measure of the amount of the nucleic acid that is present in triplex form. The relative number of breaks in the plasmid, and therefore the fraction of the plasmid in triplex form, can be quantitatively determined either by direct staining with ethidium bromide or by Southern hybridization, as shown in Fig. 4C. The slighdy lower values of the percentage of breaks in the hybridization experiments as compared with the ethidium bromide sterining can be attributed to the weaker probe hybridization with the shorter fragment.

The maximum amount of DNA breaks in the samples widi 100% triplex formation was 23.7% as measured with ethidium bromide staining and 20.6% as determined by Southern hybridization. Since each TFO contained on average 1.5 ¹²⁵I and for 60 days only half on them decayed, the efficiency in terms of double strand breaks (dsb) per decay can be estimated as 40%, which is in good agreement with previously published data for the purine motif triplex (Panyutin et al., Acta Oncol., 1996, 36:817-824; Panyutin et al., Nucleic Acids Research, 1997, 25(4):883-887).

Triplex delivery into cells

Plasmid DNAs were delivered into HeLa cells using DMRIE liposomes. Transfection conditions were optimized with pCMV-sport-β-gal plasmid to obtain 60% transfection efficiency, i.e. 60% of transfected cells expressed β-galactosidase gene and became blue after staining with "β-gal Staining Kit" (Fig. 7A). Staining with trypan blue showed that almost 90% cells survived transfection (not shown). When preformed ¹²T- TFO/pCR3HPRT triplexes were delivered into cells, 30% of radioactivity was associated with cell pellet after 5-hr incubation.

Triplex radioprinting to assay triplex stability in cells

The intracellular stability of triplex complexes formed from equal amounts of pCR3HPRT plasmid and ^12,I-TFO was analyzed by the radioprinting assay. The preformed triplex complexes were transfected into HeLa cells by incubating the cells for 5 hours with DMREI liposomes containing the triplexes, and then washing the cells with DMEM/3mM EDTA. The concentration of EDTA in the wash solution is considered to be sufficient to dissociate any triplex complexes remaining outside of the cells. Alter the DMEM/EDTA wash, the cells were post-incubated in DMEM for 20, 31, 38, and 48 hrs., collected, and frozen for 60 days. After thawing, plasmid DNA was extracted, cut with Pvul restriction enzyme and analyzed in 1% agarose gel, followed by Southern hybridization with Hindlll-Ndel probe to detect ^12'I-TFO induced breaks. The distribution of radioactivity in die gel is shown in Fig. 5A, with lane 1 containing the control triplex that had been frozen in a test tube for the same amount of time. As shown in Fig. 5B, all of the analyzed samples contained approximately the same amount of breaks; about 20%, which means that the triplexes remained stable inside the living cells for at least 48 hrs, with no significant dissociation of TFO from the plasmid during that time.

Distribution of TFO/plasmid complexes inside cells

FITC-labeled TFOs were used to visualize the distribution of plasmid/TFO complexes inside cells. HeLa cells were transfected with preformed FITC- TFO/pCR3HPRT triplexes using DMRIE liposomes. As a control, the cells were transfected with FITC-TFO alone using the same liposomes. Fluorescent (Figs. 6A and 6B) and confocal (Figs. 6C and 6D) microscopy showed that FITC labeled triplexes were uniformly distributed in the cytoplasm and nuclei of the cells, whereas the unbound FITC-labeled TFOs were not released from the liposomes, and remained concentrated in bright grains in the cytoplasm. Fluorescent signal from the FITC-TFO/pCR3HPRT triplexes was observed in almost all of the cells; however, measurement of expression of the β-galactosidase reporter gene in the cells revealed expression of the gene in only half of the cell population (Figs. 7 A and 7B). Autoradiography confirmed introduction of the ¹²⁵I- TFO/pCR3HPRT triplexes into nearly all the cells (Fig. 7C). This illustrates that physical detection of triplexes containing labeled TFOs permits quantitative measurement of the presence of the labeled triplexes that is not possible by assaying for expression of reporter genes.

The radioprinting assay described above allows detection of the interaction of TFOs with their target sequences both in vitro and in vivo. The method is based on a unique property of Auger electron emitters, such as 'T, ^{1 3}I, "'In and others, to produce DNA breaks within close proximity to the decay site. Radioprinting can be considered as a general form of radioprobing, a method that allows one to detect conformational changes in DNA structure by measuring distribution of breaks induced by Auger electron emitters with single nucleotide resolution (Panyutin et al., Nucleic Acids Research, 1997, 25(4):883-887). The assays of triplex stability described above were performed with commercially available ¹²ⁱI-dCTP and can be easily repeated in any laboratory. The drawback of using ¹²ⁱI is its 60 days half life, that requires the freezing of samples for decay accumulation. The use of Auger electron emitters with shorter half lives, for example, ^,23I (13 hrs), '"In (2.6 days), ⁶⁷Ga (78 hr) would allow considerably shortening of the time required for the radioprinting procedure, and would avoid the need for incubating the frozen material for extended time periods while strand breaks accumulate. (Martin et al., 1988, in Baverstock,K.F. and Charlton,D.E. (eds), DNA damage by Auger emitters, Taylor & Francis, New York, pp. 55-68; Hnatowich et al., J. Pharmacol. Exper. Therap., 1996, 276:326-334). By carrying out precise analysis of the distribution of breaks in the DNA to which die labeled TFO is bound, for example, using ligation-mediated PCR (Pfeifer et al., Methods: A Companion to Methods in Enzymology, 1997, 11:189-196) and high- resolution gel electrophoretic methods such as those used in DNA sequencing, one skilled in the art can also use radioprinting methodology to obtain information about the conformation of DNA triplexes inside cells. The radioprinting methods can be extended to detect interaction of a target DNA or RNA sequence with other types of ligands labeled with Auger electron emitters, such as a proteins, non-triplex forming nucleic acids, antibiotics and other low molecular weight compounds.

All publications and patents mentioned in the above specification are incorporated herein by reference. Nothing herein is to be construed as an admission that the invention is not entided to antedate such disclosure by virtue of prior invention. Various modifications and variations of the described methods and system of the invention will be apparent to those skilled in the art without departing from the scope and spirit of the invention. Although the invention has been described in connection with specific embodiments, it should be understood that the invention as claimed is not limited to such specific embodiments, and that various modifications of the described modes for carrying out the invention which are obvious to those skilled in the arts to which this invention pertains are intended to be within the scope of the following claims.

Claims

WE CLAIM:

1. A method for determining the biodistribution of a nucleic acid in a living multicellular organism comprising:

(i) obtaining a triplex-forming oligonucleotide (TFO) that binds to and forms a triplex with said nucleic acid, said TFO being covalendy attached to a detectable label moiety that does not prevent triplex formation; (ii) allowing the labeled TFO to bind to the nucleic acid to form a triplex; (iii) administering the triplex to a multicellular organism; and (iv) detecting the distribution of the labeled TFO in said organism, thereby determining the distribution of said nucleic acid in the organism.

2. The method of claim 1 wherein said TFO is an oligodeoxynucleotide.

3. The method of claim 1 wherein said TFO is a polypurine oligonucleotide.

4. The method of claim 1 wherein said nucleic acid is a double-stranded DNA molecule.

5. The method of claim 1, wherein said label moiety is detectable in a living multicellular organism by at least one of gamma camera imaging, rectilinear scanner imaging, magnetic resonance imaging (MRI), positron emission tomography (PET), and single photon emission computed tomography (SPECT), and wherein said step of detecting the distribution of the labeled TFO in said organism comprises using one of gamma camera imaging, rectilinear scanner imaging, MRI, PET, and SPECT.

6. A method for determining the intracellular distribution of a nucleic acid comprising: (i) obtaining a TFO that binds to and forms a triplex widi said nucleic acid, said

TFO being covalendy attached to a detectable label moiety diat does not prevent triplex formation; (ii) allowing the labeled TFO to bind to the nucleic acid to form a triplex; (iii) introducing the triplex into a eukaryotic cell; and (iv) detecting the intracellular distribution of die labeled TFO in the cell, thereby determining the distribution of the nucleic acid in the cell.

7. The method of claim 6 wherein said TFO is an oligodeoxynucleotide.

8. The method of claim 6 wherein said TFO is a polypurine oligonucleotide.

9. The method of claim 6 wherein said nucleic acid is a double-stranded

DNA molecule.

10. The method of claim 6, wherein said label moiety is detectable in a living cell by at least one of fluorescence microscopy, confocal microscopy, and autoradiography, and wherein said step of detecting the distribution of the labeled TFO in said cell comprises using one of fluorescence microscopy, confocal microscopy, and autoradiography.

11. A method for measuring the stability, under selected conditions, of triplex complexes formed by binding of TFOs to a nucleic acid, comprising:

(i) obtaining TFOs that bind to and form triplexes with said nucleic acid, each of said TFOs being covalendy attached to at least one label moiety comprising an Auger electron emitter; (ii) allowing said labeled TFOs to bind to copies of the nucleic acid to form triplexes; (iii) incubating said triplexes under said selected conditions for at least one period of time during which stability is to be measured; (iv) freezing and storing the triplexes for a time period of such duration that decay of said Auger electron emitters occurs and causes strand breaks in said nucleic acids; (v) measuring the number of strand breaks in said nucleic acids, wherein the number of strand breaks is proportional to the number of intact triplex complexes present at the time the triplexes were frozen and stored.

12. The method of claim 11 wherein said TFO is an oligodeoxynucleotide.

13. The method of claim 11 wherein said TFO is a polypurine oligonucleotide.

14. The method of claim 11 wherein said nucleic acid is a double-stranded DNA molecule.

15. The method of claim 11 wherein step (iii) comprises incubating said triplexes in a cell-free solution for at least one period of time during which stability is to be measured.

16. The method of claim 11 wherein step (iii) comprises introducing said triplexes into cells in vitro, and incubating said cells containing said triplexes in vitro for at least one period of time during which stability is to be measured.

17. The method of claim 11 wherein step (iii) comprises administering said triplexes to a multicellular organism, allowing said triplexes to remain in said organism for at least one period of time during which stability is to be measured, and then recovering said nucleic acids from said organism.