US 20040229833 A1
The invention provides for the use of oligodeoxynucleotide decoys for the prophylactic or therapeutic treatment of diseases associated with the binding of endogenous transcription factors to genes involved in cell growth, differentiation and signalling or to viral genes. By inhibiting endogenous trans-activating factors from binding transcription regulatory regions, the decoys modulate gene expression and thereby regulating pathological processes including inflammation, intimal hyperplasia, angiogenesis, neoplasia, immune responses and viral infection. The decoys are administered in amounts and under conditions whereby binding of the endogenous transcription factor to the endogenous gene is effectively competitively inhibited without significant host toxicity. The subject compositions comprise the decoy molecules in a context which provides for pharmacokinetics sufficient for effective therapeutic use.
13. A method for treating an NFκB-associated inflammatory disease or condition is a mammal, comprising introducing an NFκB decoy into a cell of said mammal.
14. The method of
15. The method of
16. The method of
17. The method of
18. The method of
 The field of this invention is therapeutic treatment of disease with double stranded nucleic acids which bind transcription factors.
 A wide variety of diseases result from the over- or under-expression of one or more genes. Given cells may make insufficient amounts of a protein (e.g. insulin) or too much of a protein, be it a normal protein (e.g. TNF), a mutant protein (e.g. an oncogene), or a non-host protein (e.g. HIV tat). The ultimate goal of therapeutic intervention in such diseases is a selective modulation of gene expression.
 A variety if methods of transcriptional modulation in vitro have been reported including the use of anti-sense nucleic acids capable of binding nascent message, intracellular immunization with dominant negative mutants.
 With the broad potential therapeutic applications, massive efforts have been extended by prominent laboratories and clinics around the world to extend these methods in vivo. To date, the transcription factor decoy strategy has never been successfully adopted in vivo.
 Description of the roles of transcription factors may be found in Nevins, Science 258, 424-429 (1992); Dalton, EMBO J. 11, 11797 (1992); Yee et al. ibid. 6, 2061 (1987), Weintraub et al., Nature 358, 259-261 (1992), Pagano et al., Science 255, 1144-1147 (1992). Viral coat protein-liposome mediated transfection is described by Kaneda et al., Science 243, 375 (1989). Ritzenthaler et al. (1991) Biochem J 280, 157-162; Ritzenthaler et al (1993) J Biol Chem 268, 13625-13631; Bielinska et al., Science 16, 997-1000 (1990) and Sullenger et al., Cell 63, 601-608 (1990) describe inhibition of transcription with double stranded nucleic acids.
 A general discussion concerning the mechanism of restenosis may be found in Libby et al., Circulation 86, III-47 (1992) and Clowes et al., J. Cardiovasc. Pharmacol. 14, S12-15 (1989).
 The invention provides for the therapeutic treatment of diseases associated with the binding of endogenous transcription factors to genes involved in cell growth, differentiation and signalling or to viral genes. Methods and compositions are provided for blocking the capacity of endogenous trans-activating factors to modulate gene expression and thereby regulating pathological processes including inflammation, intimal hyperplasia, angiogenesis, neoplasia, immune responses and viral infection.
 The methods comprise administering to a patient double stranded nucleic acid “decoys” in a form such that the decoys are capable of entering target cells of the patient and specifically binding an endogenous transcription factor, thereby competitively inhibiting the transcription factor from binding to an endogenous gene. The decoys are administered in amounts and under conditions whereby binding of the endogenous transcription factor to the endogenous gene is effectively competitively inhibited without significant host toxicity. Depending on the transcription factor, the methods can effect up- or down-regulation of gene expression. The subject compositions comprise the decoy molecules in a context which provides for pharmacokinetics sufficient for effective therapeutic use.
FIG. 1. Effect of NRE “decoy” on renin gene expression in cultured SMG cells. SCA-9 cells expressed renin as shown by immunohistochemistry (panel A). Primer extension analysis demonstrated that this was exclusively Ren 2 (panel B). These cells effectively took up FITC labeled double stranded decoy oligomer corresponding to the NRE (panel C). RNA was prepared 24 hours after transfection from control and “decoy” treated SCA-9 cells. Note that Ren 1 mRNA could be observed after exposure of the cells to the NRE decoy (panel B, lane 4).
FIG. 2. Detection of the NRE binding protein in cultured SMG cells. Nuclear extracts were prepared from SCA-9 cells and incubated with 32P end-labeled mouse renin NRE oligonucleotide. Competition analysis was performed with 50- and 100-fold excess of unlabeled NRE oligonucleotide. Note that a specific NRE:NRE binding protein complex formation was observed.
FIG. 3. In Vivo Expression of CAT in the Mouse SMG. Ten ug of renin gene CAT construct was transfected directly into the SMG of DBA/2J mice using the HVJ-DNA-Liposome technique. Four days after transfection, the SMG was removed, cell extracts prepared and CAT activity measured.
FIG. 4. Schematic diagram of the factors influencing renin gene expression. The hatched bar represents the CRE/NRE region present in the renin gene. The CRE binding protein and the NRE binding protein compete for binding to this region. The triangle represents the 150 base pair insert which is present in the Ren 2 gene.
 Ren 1 expression: in the kidney, the CRE binding protein binds tighter, blocking the binding of the NRE binding protein, and allowing expression of the Ren 1 gene. In the SMG, an inhibitory protein forms an inactive complex with the CRE binding protein, allowing the NRE binding protein to bind, silencing expression of Ren 1.
 Ren 2 expression: The 150 bp insertion interferes with the NRE function in the Ren 2 gene, resulting in Ren 2 expression in the SMG and kidney.
 Methods and compositions are provided for modulating gene expression in vivo. The methods involve administering a composition to a patient so as to introduce into a target cell molecular decoys comprising double-stranded nucleic acid, usually DNA, to which transcription factors endogenous to the target cell bind. Various methods are employed for in vivo administration of the decoys such that sufficient decoys enter into the target cells to competitively inhibit transcription factor binding to an endogenous gene regulatory region.
 The compositions which are employed comprise “decoys”: double-stranded nucleic acid molecules with high binding affinity for the targeted transcription factors. The targeted transcription factors are endogenous, sequence-specific double-stranded DNA binding proteins which modulate (increase or decrease) the rate of transcription of one or more specific genes in the target cell. Essentially any such transcription factor (henceforth, “transcription factor”) can be targeted so long as a specific decoy capable of competitively inhibiting binding to the endogenous gene can be identified. Numerous transcription factors and their binding sequences are known in the art as are methods for identifying such complements, see e.g. Wang and Reed (1993) Nature 364, 121 and Wilson et al. (1991) Science 252, 1296. As used herein, endogenous means that the gene or transcription factor is present in the target cell at the time the decoy is introduced.
 The transcription factors will, for the most part and depending on the clinical indication, regulate the transcription of genes associated with cell growth, differentiation and signalling or viral genes resident in the target cell. Examples include genes necessary for mitosis, particularly going from Go to S, such as proteins associated with check points in the proliferative cycle, cyclins, cyclin dependent kinases, proteins associated with complexes, where the cyclin or cdk is part of the complex, Rosenblatt et al., Proc. Natl. Acad. Sci. 89, 2824 (1992) and Pagano et al., Science 255, 1144 (1992). Often such genes or the transcription factors themselves will be oncogene products or cellular counterparts, e.g. fos, jun, myc, etc. Other examples include genes encoding secreted proteins and peptides such as hormones e.g. growth factors, cytokines, e.g. interleukins, clotting factors, etc. Target transcription factors also include host and host-cell resident viral transcription factors which activate viral genes present in infected host cells.
 Preferred target transcription factors are activated (i.e. made available in a form capable of binding DNA) in a limited number of specifically activated cells. For example, a stimulus such as a wound, allergen, infection, etc may activate a metabolic pathway that is triggered by the transient availability of one or more transcription factors. Such transcription factors may be made available by a variety of mechanisms such as release from sequestering agents or inhibitors (e.g. NFκB bound to IκB), activation by enzymes such as kinases, translation of sequestered message, etc. Desirably, the target transcription factor(s) will be associated with genes other than genes whose lack of expression results in cytotoxicity. For the most part, it is desirable not to kill the cell, but rather to inhibit or activate specific gene transcription.
 Exemplary transcription factors and related cis elements, the cellular processes impacted and therapeutic indication include:
 The length, structure and nucleotide sequence of the decoy will vary depending on the targeted transcription factor, the indication, route of administration, etc. For example, targeted transcription factors frequently bind with different degrees of affinity to a variety of sequences, normally sharing a high degree of homology. Accordingly, one may choose to use a sequence associated with a particular target gene or use a consensus sequence based on the nucleotide at each site which occurs most frequently in the binding sequences to which the particular transcription factor binds. For example, when targeting a host transcription factor involved in viral transcription, it may be possible to minimize undesirable effects on host transcriptions preferable by employing the viral-specific binding sequence. Similarly, where transcription is mediated by a multimeric complex, it is often desirable to target a single transcription factor to minimize effects on non-targeted genes. For example, in the case of Herpes virus transcription, one may target the viral VP16 without concomitant targeting of the promiscuous host Oct protein.
 In addition to binding affinity, the decoys are also selected for binding specificity. Desirably, the decoys will be highly specific for the target transcription factor(s) such that their effect on nontarget cells and nontargeted metabolic processes of target cells are minimized. Such selection is accomplished in vitro by comparative binding to known transcription factors and nuclear extracts and in culture and in vivo by assaying nontargeted cell function and effects on nontargeted cell types.
 The decoys contain sufficient nucleotide sequence to ensure target transcription factor binding specificity and affinity sufficient for therapeutic effectiveness. For the most part, the target transcription factors will require at least six base pairs, usually at least about eight base pairs for sufficient binding specificity and affinity. Frequently, providing the decoys with flanking sequences (ranging from about 5 to 50 bp) beside the binding site enhance binding affinity and/or specificity. Accordingly, cis element flanking regions may be present and concatemer oligonucleotides may be constructed with serial repetitions of the binding and/or cis element flanking sequences.
 In one embodiment, the decoys are non-replicative oligonucleotides fewer than 100 bp, usually fewer than 50 bp and usually free of coding sequence, being primarily from the non-coding 5′ region of a gene. Alternatively, the decoys may comprise a portion of a larger plasmid, including viral vectors, capable of episomal maintenance or constitutive replication in the target cell to provide longer term or enhanced intracellular exposure to the decoy sequence. Plasmids are selected based on compatibility with the target cell, size and restriction sites, replicative frequency, copy number maintenance, etc. For example, plasmids with relatively short half-lives in the target cell are preferred in situations where it is desirable to maintain therapeutic transcriptional modulation for less than the lifetime of the target cell. Exemplary plasmids include pUC expression vectors driven by a beta-actin promoter and CMV enhancer, vectors containing elements derived from RSV or SV40 enhancers, etc. The adeno-associated viral vector preferentially integrates in chromosome 19 and may be utilized for longer term expression.
 The oligonucleotides which are employed may be naturally occurring or other than naturally occurring, where the synthetic nucleotides may be modified in a wide variety of ways, see e.g. Bielinska et al (1990) Science 250, 997. Thus, oxygens may be substituted with nitrogen, sulfur or carbon; phosphorus substituted with carbon; deoxyribose substituted with other sugars, or individual bases substituted with an unnatural base. In each case, any change will be evaluated as to the effect of the modification on the binding of the oligonucleotide to the target transcription factor, as well as any deleterious physiological effects. These modifications have found wide application for “anti-sense” oligonucleotides, so that their safety and retention of binding affinity are well established in the literature. See, for example, Wagner et al., Science 260, 1510-1513 (1993). The strands may be synthesized in accordance with conventional ways using phosphoramidite synthesis, commercially available automatic synthesizes, and the like.
 The administered compositions may comprise individual or mixtures of decoys. Usually the mixture will not exceed 4 different decoys usually not exceed 2. The decoys are administered to a host in a form permitting cellular internalization of the decoys in an amount sufficient to competitively inhibit the binding of the targeted transcription factor to an endogenous gene. The host is typically a mammal, usually a human. The selected method of administration depends principally upon the target cell, the nature of the decoy, the host, the size of the decoy. Exemplary methods are described in the examples below; additional methods including transfection with a retrovirus, viral coat protein-liposome mediated transfection, lipofectin etc. are described in Dzau et al., Trends in Biotech 11, 205-210 (1993).
 Where administered in liposomes, the decoy concentration in the lumen will generally be in the range of about 0.1 uM to 50 uM per decoy, more usually about 1 uM to 10 uM, most usually about 3 uM. For other techniques, usually one will determine the application rate empirically, using conventional techniques to determine desired ranges.
 In some situations it may be desirable to provide the decoy source with an agent which targets the target cells, such as an antibody specific for a surface membrane protein on the target cell, a ligand for a receptor on the target cell, etc. For example, for intervention in HIV infection, cells expressing HIV gene products or CD4 may be specifically targeted with gene product or CD4-specific binding compounds. Also, where liposomes are involved, one may wish to include proteins associated with endocytosis, where the proteins bind to a surface membrane protein associated with endocytosis. Thus, one may use capsid proteins or fragments thereof tropic for a particular cell type, antibodies for proteins which undergo internalization in cycling, proteins that target intracellular localization and enhance intracellular half-life.
 The application of the subject therapeutics are preferably local, so as to be restricted to a histological site of interest e.g. localized inflammation, neoplasia or infection. Various techniques can be used for providing the subject compositions at the site of interest, such as injection, use of catheters, trocars, projectiles, pluronic gel, stents, sustained drug release polymers or other device which provides for internal access, or the like. Where an organ or tissue is accessible because of removal from the patient, such organ or tissue may be bathed in a medium containing the subject compositions, the subject compositions may be painted onto the organ, or may be applied in any convenient way. Alternatively, systemic administration of the decoy using e.g. lipofection, liposomes with tissue targeting (e.g. antibody), etc. may be practiced. Systemic administration is most applicable where the distribution of the targeted transcription factor is primarily limited to targeted cell types, e.g. virus-specific transcription factors limited to infected cells, mutant oncogenic transcription factors limited to transformed cells, etc.
 Optimal treatment parameters will vary with the indication, decoy, clinical status, etc., and are generally determined empirically, using the guidance provided herein. Several exemplary indications, routes and vehicles of administration and decoy combinations are disclosed in the following table.
 A wide variety of indications may be treated, either prophylactically or therapeutically with the subject compositions. For example, prophylactic treatment may inhibit mitosis or proliferation or inflammatory reaction prior to a stimulus which would otherwise activate proliferation or inflammatory response, where the extent of proliferation and cellular migration may be undesirable. Similarly, a therapeutic application is provided by a situation where proliferation or the inflammatory response is about to be initiated or has already been initiated and is to be controlled. The methods and compositions find use, particularly in acute situations, where the number of administrations and time for administration is relatively limited.
 Conditions for treatment include such conditions as neoproliferative diseases including inflammatory disease states, where endothelial cells, inflammatory cells, glomerular cells may be involved, restenosis, where vascular smooth muscle cells are involved, myocardial infarction, where heart muscle cells may be involved, glomerular nephritis, where kidney cells are involved, hypersensitivity such as transplant rejection where hematopoietic cells may be involved, cell activation resulting in enhancement of expression of adhesion molecules where leukocytes are recruited, or the like. By administering the decoys to the organ ex vivo prior to implantation and/or after implantation, upregulation of the adhesion molecules may be inhibited. Adhesion molecules include homing receptors, addressins, integrins, selectins, and the like.
 The following examples are offered by way of illustration and not by way of limitation.
 For the nuclear extracts, vascular smooth muscle cells (“VSMCs”) were stimulated by serum until confluent. After confluent, the cells were made quiescent by placing in serum free medium for 48 h. After the transfection of decoy oligodeoxynucleotides (“ODN”; 14 bp) essentially as described in Morishita et al. (1993) Proc. Natl. Acad. Sci. USA, 90, 8474-8478, cells were stimulated by 5% serum. After 6 h, RNA was extracted by RNAzol (Tel-Test Inc, Texas) Chomczynski and Sacchi (1987) Anal Biochem 162, 156-159. Levels of PCNA, cdc2 and beta-actin mRNAs were measured by RT-PCR (Morishita et al. (1993) supra). The PCNA primer (nucleotides 150-177 of rat PCNA cDNA) and the cdc2 5′ primer (nucleotides 54-75 of human cdc2 cDNA) were previously described (Morishita et al. (1993) supra). The primers complementary to the rat beta-actin gene were obtained from Clontech Laboratories Inc. (Palo Alto, Calif.). Aliquots of RNA were amplified simultaneously by PCR (30 cycles) and compared with a negative control (primers without RNA). Amplification products were electrophoresed through 2% agarose gels stained with ethidium bromide. A gel mobility shift assay was performed as previously described (Horiuchi et al., J. Biol. Chem. 266, 16247-16254 (1991).
 The 14 bp double-strand ODN effectively abolished the binding of the E2F transcription factor to a specific binding site in serum stimulated VSMCs. See also, Hiebert et al. (1989) PNAS 86, 3594. Transfection of the E2F decoy ODN markedly inhibited the induction of c-myc, cdc2 and PCNA mRNA expression in response to serum stimulation. The E2F decoy ODN had no effect on beta-actin mRNA expression. Furthermore, the control missense E2F element ODN containing two base pair substitutions that eliminate E2F binding failed to inhibit the induction of c-myc, cdc2 and PCNA RNA expression in response to serum stimulation. In association with effective inhibition of cell cycle regulatory gene expression, transfection of the 14 bp E2F decoy also abolished serum-stimulated VSMC proliferation. In contrast, the missense E2F element ODN had no effect on mitogenesis induced by serum.
 To further confirm the specificity of this response to the E2F decoy, a 30 bp double-stranded ODN which contained two 8 base pair E2F cis elements capable of specific binding to E2F was employed (Weintraub et al., Nature 358, 259-261 (1992)). In the 30 bp E2F decoy the fifth nucleotide of the 8 bp E2F cis elements was changed from C to G. Despite these differences in flanking sequences and nucleotide composition, both E2F decoys effectively bind E2F and inhibit serum-stimulated vascular smooth muscle cell (VSMC) proliferation. Moreover, a 30 bp missense E2F element ODN with 5 bp substitutions with the 8 bp E2F consensus elements fails to bind E2F and also fails to inhibit serum-stimulated VSMC proliferation. Thus, these in vitro studies documented that transfection of the E2F cis element decoy ODN binds the E2F transcription factor, blocked the induction of cell cycle regulatory gene expression and inhibited VSMC proliferation in a sequence specific manner.
 Liposomes were prepared as follows: Phosphatidylserine phosphatidylcholine, and cholesterol were mixed in a weight ratio (1:4.8:2) to create a lipid mixture. Lipid was hydrated in a balanced salt solution containing ODN (110 nmol). Purified HVJ(Z) strain was inactivated by UV radiation just before use. The liposome suspension was mixed with HVJ (Z strain) (20,000 hemagglutinating units), incubated, then free HVJ removed by sucrose density gradient centrifugation. The final concentration of encapsulated DNA was calculated as previously reported (Morishita et al. (1993) supra). This method results in a more rapid cellular uptake and nuclear concentration, leading to a 100-fold higher transfection efficiency of ODN than lipofection or passive uptake methods.
 The sequences of the phosphorothioate ODN utilized:
 We also examined another set of decoy ODNs containing two binding sites:
 Rat aortic VSMC (passage 4-10) were studied in a confluent, quiescent state in serum-free media (Morishita et al, J. Clin. Invest. 91, 2580-2585 (1993)). The cells were incubated with hemagglutinating virus Japan (HVJ) liposomes (3 μM) at 4° C. for 5 min and 37° C. for 30 min. Three days after transfection in either calf serum (CS) or serum-free media, cell number was determined by Coulter-Counter (Coulter, Fla.).
 A 2 French Fogarty catheter was used to induce vascular injury in male Sprague-Dawley rats (400-500 g; Charles River Breeding Laboratories) (Hanke et al., Circ. Res. 67, 651-659 (1990)). These rats were anesthetized, and a cannula introduced into the left common carotid via the external carotid artery. After vascular injury of the common carotid, the distal injured segment was transiently isolated by temporary ligatures. The HVJ complex was infused into the segment and incubated for 10 min at room temperature. No adverse neurological or vascular effects were observed in any animal undergoing this procedure.
 For RNA analysis, vessels were harvested at 6 h, (c-myc and beta-actin) and one day (cdc2 kinase, PCNA and beta-actin) after transfection. RNA was extracted from mismatched or E2F decoy ODN (3 μM) treated injured or untreated intact vessels by RNAzol (Tel-Test Inc., Texas). RT-PCR was performed as described above. For BrdU staining, BrdU was injected into rats (100 mg/kg subcutaneous and 30 mg/kg intraperitoneal at 18 h prior, and then 30 mg/kg intraperitoneal at 12 h prior to sacrifice (Hanke et al., supra)). Rats were sacrificed after day four after transfection. The carotid artery was removed after perfusion-fixation with 4% paraformaldehyde, and processed for immunohistochemistry in a standard manner using anti-BrdU antibodies (Amersham). The proportion of BrdU positive cells was determined by cell counts under light microscopy in a blinded fashion.
 Transfection procedures were described above. HVJ-ODN complex (3 μM) was administered into the rat carotid injured arteries. At two weeks after transfection, rats were sacrificed and vessels were perfusion-fixed with 4% paraformaldehyde. Three individual sections from the middle of transfected segments were analyzed. In addition, three sections from the middle section of the injured untransfected region were also analyzed. Animals were coded so that the operation and analysis were performed without knowledge of which treatment individual animals received. Intimal and medial areas were measured on a digitizing tablet (Southern Micro Instruments, Georgia). Analysis of variance with subsequent Duncan's test was used to determine significant differences in multiple comparisons. P<0.05 was considered significant.
 The scrambled and progesterone responsive element (PRE) decoy sequences utilized as control ODNs are as follows: