CA2228977A1 - Localized oligonucleotide therapy for preventing restenosis - Google Patents

Abstract

Antisense oligonucleotide gene therapy selective for the 5' region of PDGFR-.beta.
subunit mRNA was used in attempt to prevent intimal thickening following rat carotid arterial injury. Sustained perivascular application of the antisense oligomers for 14 days reduced PDGFR-.beta. protein overexpression and prevented neointima formation by 80%. Specificity was verified by the absence of effects on the expression of a non-targeted gene PDGFR-.alpha.. These data demonstrated that antisense oligonucleotide sequences can effectively suppress a growth factor receptor, and the reduction of intimal hyperplasia after injury correlates with the extent to which these oligomers inhibited PDGFR-.beta. protein expression. Methods and materials useful for preventing restenosis are described and claimed.

Description

TITLE OF THE INVENTION
LOCALIZED OLIGONUCLEOTIDE THERAPY FOR PREVENTING RESTENOSIS

BACKGROUND OF THE INVENTION
This invention relates to a method of delivery of antisense oligonucleotide to a5 presellected locus in vivo, useful in the treatment of disease.
In the last several years, it has been demonstrated that oligonucleotides are capable of inhibiting the replication of certain viruses in tissue culture systems. For example, Zamecnik and Stephenson, Proc. Natl. Acad. Sci. U.S.A., 75: 280-284 (1978), showed oligonucleotide-mediated inhibition of virus replication in tissue culture, 10 using Rous Sarcoma Virus. Zamecnik et al., Proc. NaU. Acad. Sci. U.S.A., 83:
4145 4146 (1986), demonstrated inhibition in tissue culture of the HTLV-III virus (now HIV-1) which is the etiological agent of AIDS. Oligonucleotides also have been used to suplpress expression of selected non-viral genes by blocking translation of the protein encoded by the genes. Goodchild, et al., Arch. Biochem. Biophys., 264:
401-4()9 (1988) report that rabbit-globin synthesis can be inhibited by oligonucleotides in a cell-free system. Treatment with antisense c-myb has been shown to block proliferation of human myeloid leukemic cell lines in vitro. G. Anfossi, et al., Proc. Nafl.
Acad. Sci. USA, 86: 3379 (1989).
A drawback to this method is that oligonucleotides are subject to being 20 degraded or inactivated by cellular endogenous nucleases. To counter this problem, some researchers have used modified oligonucleotides, e.g., having altered internucleotide linkages, in which the naturally occurring phosphodiester linkages have been replaced with another linkage. For example, Agrawal et al., Proc Natl. Acad. Sci.
U.S.A., 85: 7079-7083 (1988) showed increased inhibition in tissue culture of HIV-1 25 using oligonucleotide phosphoramidates and phosphorothioates. Sarin et al., Proc.
NaU. Acad, Sci. U.S.A., 85: 7448-7451 (1988) demonstrated increased inhibition of HIV-1 using oligonucleotide methylphosphonates. Agrawal et al., Proc. Natl. Acad. Sci.
U.S.A., 86: 7790-7794 (1989) showed inhibition of HIV-1 replication in both early-infected and chronically infected cell cultures, using nucleotide sequence-specific :30 oligonucleotide phosphorothioates. Leither et al., Proc. Natl. Acad. Sci U.S.A., 87:
3430-:3434 (1990) report inhibition in tissue culture of influenza virus replication by oligonucleotide phosphorothioates.

Oligonucleotides having artificial linkages have been shown to be resistant to degradation in vivo. For example, Shaw et al., in Nucleic Acids Res., 19: 747-750 (1991)l, report that otherwise unmodified oligonucleotides become more resistant to nucleases in vivo when they are blocked at the 3' end by certain capping structures and that uncapped oligonucleotide phosphorothioates are not degraded in vivo.
While antisense oligonucleotides have been shown to be capable of interfering selectively with protein synthesis, and significant progress has been made on improving their intracellular stability, the problem remains that oligonucleotides must reach their intended intracellular site of action in the body in order to be effective.
Where the intended therapeutic effect is a systemic one, oligonucleotides may beadministered systemically. However, when it is necessary or desirable to administer the oligonucleotide to a specific region within the body, systemic administration typically will be unsatisfactory. This is especially true when the target mRNA is present in norrnal cells as well as in the target tissue, and when antisense rRNA binding in normal cells induces unwanted physiological effects. Stated differently, the dosage of antisense oligonucleotide administered systemically that is sufficient to have the desired effect locally may be toxic to the patient. An example of a treatment strategy which could greatly benefit from development of a method of limiting the effect of antisense oligonucleotide to a target tissue is the inhibition of smooth muscle cell ;20 proliferation which leads to restenosis following vascular trauma.
Smooth muscle cell proliferation is a poorly understood process that plays a major role in a number of pathological states including atherosclerosis and hypertension. It is the leading cause of long-term failure of coronary and peripheral angioplasty as well as of coronary bypass grafts.
;25 Vascular smooth muscle cells in adult animals display a well defined phenotype charac:terized by an abundance of contractile proteins, primarily smooth muscle actin and mlyosins, as reviewed by S.M. Schwartz, G.R. Campbell, J.H. Campbell, Circ.
Res., 58: 427 (1986), and a distinct lack of rough endoplasmic reticulum. When subjected to injury in vivo or placed in an in vitro cell culture, adult smooth muscle cells :30 (SMC) undergo a distinct phenotypic change and lose their "differentiated" state. The cells acquire large amounts of endoplasmic reticulum and gain actively synthesizing extracellular matrix, and they begin expressing a number of new proteins.
U.S. Patent N~ 5,593,974 describes the therapeutic effect of antisense oligonucleotides against the PCNA, c-myb and NMMHC mRNAs, when locally adminiistered in damaged vascular tissue. It is inferred to, in this reference, that smooth muscle growth is stimulated by PDGF (platelet-derived growth factor). Further, in the patent publication WO 93/08845, it is also mentioned that antisense could be made agains,t the messengers of PDGF and its vascular receptor. These two references do 5 not teach that antisense oligonucleotides to these molecules would effectively present restenosis.
It is now widely accepted that within the first 2 days following vascular injurydamaged and dying medial vascular smooth muscle cells (vSMC) release growth promoters such as bFGF. This induces vSMC proliferation for the next 3-5 days, l O delineating the first wave of the vascular healing process (1-3). The second and third waves rely on migration of medial vSMC and their proliferation within the neointima (4).
It is thought that half of the migrating vSMC will undergo 3 rounds of cell cycle proliferation in the intima, ultimately representing nearly 90% of the final cell count in the neointima. The other half of the migrating vSMC do not divide, and account for the 15 remaining 10% of the intimal cell count (1). vSMC are observed within the neointima as soon as 3 days after the injury. Their number peaks within 2 weeks of injury and remains relatively constant for up to 1 year (5). Several molecules such as angiotensin Il, TGI--13, bFGF and PDGF-BB might act as vSMC chemotactic factors during the second wave of cellular events (4). PDGF-BB has received particular attention 20 because it is both mitogenic for cultured vSMC through activation of either PDGF
receptors (PDGFR-aa or PDGFR-1313), and chemotactic through the activation of PDGFR-1313 (6). In vivo, however, PDGF-BB acts predominantly as a chemotactic factor on vSMC. Injection of this growth factor increased vSMC migration by 10-20 fold, but proliferation by no more than 2 fold (7), and polyclonal anti-PDGF antibodies blocked 25 the migration of vSMC migration, but not their proliferation (8). It is therefore, reasonable to postulate that PDGF-BB plays a critical role in intimal thickening during the first 2 weeks after a vascular lesion.
PDGFR-13 subunit is specifically expressed in mesenchymal cells, such as vSMC and fibroblasts (22). Basal expression in the medial vSMC of the normal artery :30 increases within days of injury (23). What is not known is whether PDGF receptor expression is directly related to the extent of neointimal hyperplasia. Antisense oligonucleotide gene therapy enables us to examine this question (10-17). Antisense oligonucleotide sequences hybridize (18-20) with targeted mRNA or gene regions at ribosomic or nuclear sites preventing mRNA translation into protein (21). To date, antisense oligonucleotides directed against growth-regulatory or cell-cycle genes (c-myb, c-myc, PCNA, cdc2, cdk2) involved in vSMC proliferation after injury have successfully altered intimal hyperplasia (10-17). Yet, to the best of our knowledge no one has used antisense sequences to prevent the expression of chemotactic proteins or their receptors. We examined these issues by examining the effect of antisense phosphorothioate- oligodeoxyribonucleotide sequences complementary to PDGFR-13 mRNA on PDGFR-~ protein expression and intimal thickening after vascular injury.The sustained release of PDGFR-,B mRNA antisense oligonucleotide reduced PDGFR-~ protein expression and intimal thickening in injured rat carotid arteries in an exponentially correlative fashion. Thus, myointimal proliferation depends on both PDGFR-~ subunit overexpression and its activation by platelet-derived PDGF-BB.
Removal of either one of these two elements can suppress neointima formation.

SUMI~lARY OF THE INVENTION
The present invention relates to a method for inhibiting translation or transcription of a target nucleic acid sequence preferentially at a locus in vivo. The invention involves application directly to the target tissue through a surgical or catheterization procedure of specific oligonucleotides having a nucleotide sequence compllementary to at least a portion of the target nucleic acid, i.e., antisenseoligonucleotides. The oligonucleotides are preferably antisense sequences specific for the messenger RNA (mRNA) transcribed from the gene whose expression is to be inhibited. The antisense oligonucleotides hybridize with the target mRNA therebypreventing its translation into the encoded protein. Thus, the present method prevents the protein encoded by a selected gene from being expressed. Furthermore, animalexperiments have demonstrated dramatic local therapeutic effects in vivo.
The present oligonucleotides preferably are modified to render them resistant to degradation and/or extension by cellular nucleases or other enzymes present in vivo. This can be accomplished by methods known in the art, e.g., by incorporating one or more internal artificial internucleotide linkages, such as replacing the phosphate in the linkage with sulfur, and/or by blocking the 3' end of the oligonucleotide with capping '30 structures. Oligonucleotides of the present invention are preferably between about 14 and 3~ nucleotides in length, more preferably between 15 and 30 nucleotides.
The oligonucleotides are applied locally in order to suppress expression of the protein of choice in a circumscribed area. In a preferred embodiment, the antisense oligonucleotide is applied to the surface of the tissue at the locus disposed within a biocompatible matrix or carrier. The matrix or carrier can be a hydrogel material such as a ploly(propylene oxide-ethylene oxide) gel, e.g., one which is liquid at or below room lemperature, and is a gel at body temperature and above. In this embodiment, 5 the oli!3onucleotides are mixed with the hydrogel material, and the mixture is applied to the desired location during surgery or by catheter. The oligonucleotides also can be applied in solution by liquefying the gel, i.e., by cooling, and are retained at the area of application as the gel solidifies. Carriers which can be used also include, for example, liposomes, microcapsules, erythrocytes and the like. The oligonucleotides 10 also can be applied locally by direct injection, can be released from devices such as implanted stents or catheters, or delivered directly to the site by an infusion pump.
The methods of the present invention are useful in inhibiting the expression of protein encoding genes, as well as regulating non-encoding DNA such as regulatory sequences. Since the antisense oligonucleotides are delivered to a specific defined 15 locus, they can be used in vivo when systemic administration is not possible. For example, systemically administered oligonucleotides may be inactivated by endonucleases rendering them ineffective before they reach their targets. Large doses of the oligonucleotide may be necessary for successful systemic treatment systemically, which may have harmful or toxic effects on the patient. The present 20 method provides a means for treating a large number of specific disorders using oligonucleotide therapy by delivering an antisense sequence to the specific location where it is needed.
The elucidation of molecular mechanisms of vascular cell biology has markedly influenced our thinking on the pathophysiology of vascular disease. Antisense 25 oligonucleotide gene therapy have helped identify proteins critical to cell cycle progression and proliferation and possible therapeutic strategies to combat human disease. This approach, however, has not yet been employed to examine the contribution of chemotactic proteins and/or their receptors. PDGF-BB released from activated platelets adherent to subendothelial connective tissue is one of the principal 30 smooth muscle cell chemotactic factor.
A series of experiments were performed to assess: 1 ) the capacity of antisense oligonucleotides to reduce PDGFR-~B subunit expression and 2) the contribution of PDGFR-~ subunit in neointimal formation. Sustained, direct and local perivascular administration of two different antisense oligonucleotide sequences complementary to PDGFR-~ subunit mRNA almost completely abolished the expression of PDGFR-~protein in the intima and media of injured carotid arteries, and decreased neointima formalion by 80 and 60% respectively. Furthermore, neointima formation correlated precisl_ly with PDGFR-,B subunit expression in an exponential fashion.
5 Thus, myointimal proliferation depends on both PDGFR-,B subunit overexpression and its acl:ivation by platelet-derived PDGF-BB. Removal of either one of these two elements can suppress neointima formation.

DETAILED DESCRIPTION OF THE INVENTION
A method for inhibiting expression of protein encoding genes using antisense 10 oligonucleotides is described. The method is based on the localized application of the oligonucleotides to a specific site in vivo. The oligonucleotides preferably are applied directly to the target tissue in mixture with an implant or gel, or by direct injection or infusion. In one aspect, the oligonucleotides are treated to render them resistant in vivo to degradation or alteration by endogenous enzymes.
The Oligonucleotides The therapeutic approach using antisense oligonucleotides is based on the principle that the function of a gene can be disrupted by preventing transcription of the gene or translation of the protein encoded by that gene. This can be accomplished by providing an appropriate length oligonucleotide which is complementary to at least a 20 portion of the messenger RNA (mRNA) transcribed from the gene. The antisense strand hybridizes with the mRNA and targets the mRNA for destruction thereby preventing ribosomal translation, and subsequent protein synthesis.
The specificity of antisense oligonucleotides arises from the formation of Watson-Crick base pairing between the heterocyclic bases of the oligonucleotide and 25 complementary bases on the target nucleic acid. For example, a nucleotide sequence sixteen nucleotides in length will be expected to occur randomly at about every 4'6, or 4 x 109 nucleotides. Accordingly, such a sequence is expected to occur only once in the human genome. In contrasts a nucleotide sequence of ten nucleotides in length would occur randomly at about every 4'~ or 1x1 o6 nucleotides. Such a sequence might 30 be present thousands of times in the human genome. Consequently, oligonucleotides of greater length are more specific than oligonucleotides of lesser length and are less likely lto induce toxic complications that might result from unwanted hybridization.
Therelore, oligonucleotides of the present invention are preferably at least 14 nucleotide bases in length. Oligonucleotides having from about 14 to about 38 bases are pn~ferred, most preferably from about 15 to 30 bases.
The oligonucleotide sequence is selected based on analysis of the sequence of the gene to be inhibited. The gene sequence can be determined, for example, by 5isolation and sequencing, or if known, through the literature. The sequence of the oligonucleotide is an "antisense" sequence, that is, having a sequence complementary to the coding strand of the molecule. Thus, the sequence of the oligonucleotide is substalntially identical to at least a portion of the gene sequence, and is complementary to the mRNA sequence transcribed from the gene. The oligonucleotide therapy can be 10used l:o inhibit expression of genes from viruses or other microorganisms that are essential to infection or replication, genes encoding proteins involved in a disease process, or regulatory sequences controlling the expression of proteins involved in disease or other disorder, such as an autoimmune disorder or cardiovascular disease.
Oligonucleotides useful in the present invention can be synthesized by any 15art-rec:ognized technique for nucleic acid synthesis. See, for example, Agrawal and Goodc:hild, Tetrahedron Letters, 28: 3539 (1987), Nielsen, et al., Tetrahedron Letters, 29: 2'311 (1988); Jager et al., Biochemistry, 27: 7237 (1988); Uznanski et al., Tetrahedron Leffers, 28: 3401 (1987); Bannwarth, Helv. Chim. Acta., 71:1517 (1988);
Crosstick and Vyle, Tetrahedron Letters, 30: 4693 (1989); Agrawal, et al., Proc. Natl.
20Acad. Sci. USA, 87: 1401-1405 (1990), the teachings of which are incorporated herein by reference. Other methods for synthesis or production also are possible. In a preferred embodiment the oligonucleotide is a deoxyribonucleic acid (DNA), although ribonucleic acid (RNA) sequences may also be synthesized and applied.
The oligonucleotides useful in the invention preferably are designed to resist 25degradation by endogenous nucleolytic enzymes. In vivo degradation of oligonucleotides produces oligonucleotide breakdown products of reduced length.
Such breakdown products are more likely to engage in non-specific hybridization and are less likely to be effective, relative to their full-length counterparts. Thus, it is desirable to use oligonucleotides that are resistant to degradation in the body and 30which are able to reach the targeted cells. The present oligonucleotides can be rendered more resistant to degradation in vivo by substituting one or more internal artificial internucleotide linkages for the native phosphodiester linkages, for example, by replacing phosphate with sulfur in the linkage. Examples of linkages that may be used include phosphorothioates, methylphosphonate, sulfone, sulfate, ketyl, phosphorodithioates, various phosphoramidates, phosphate esters, bridged phosphorothioates and bridged phosphoramidates. Such examples are illustrative, rather than limiting, since other internucleotide linkages are known in the art. See, e.g., Cohen, Trends in Biotechnology (1990). The synthesis of oligonucleotides having one 5 or more of these linkages substituted for the phosphodiester internucleotide linkages is well known in the art, including synthetic pathways for producing oligonucleotides having mixed internucleotide linkages.
Methods of Application of the Oligonucleotides In accordance with the invention, the inherent binding specificity of antisense 10 oligonucleotides characteristic of base pairing is enhanced by limiting the availability of the antisense compound to its intended locus in vivo, permitting lower dosages to be used and minimizing systemic effects. Thus, oligonucleotides are applied locally to achieve the desired effect. The concentration of the oligonucleotides at the desired locus is much higher than if the oligonucleotides were administered systemically, and 15 the therapeutic effect can be achieved using a significantly lower total amount. The local high concentration of oligonucleotides enhances penetration of the targeted cells and effectively blocks translation of the target nucleic acid sequences.
The oligonucleotides can be delivered to the locus by any means appropriate for localized administration of a drug. For example, a solution of the oligonucleotides 20 can be injected directly to the site or can be delivered by infusion using an infusion pump. The oligonucleotides also can be incorporated into an implantable device which when placed at the desired site, permits the oligonucleotides to be released into the surrounding locus.
The oligonucleotides can be administered by means of numerous implants that 25 are cornmercially available or described in the scientific literature, including liposomes, microcapsules and implantable devices.
The oligonucleotides may be administered via a hydrogel material as well. The hydrogel is noninflammatory and biodegradable. Many such materials now are known, including those mode from natural and synthetic polymers. In a preferred embodiment, 30 the method exploits a hydrogen which is liquid below body temperature but gels to form a shape-retaining semisolid hydrogel at or near body temperature. Preferredhydroqel are polymers of ethylene oxide-propylene oxide repeating units. The properties of the polymer are dependent on the molecular weight of the polymer and the relative percentage of polyethylene oxide and polypropylene oxide in the polymer.

_ 9 _ Preferred hydrogels contain from about 10 to about 80% by weight ethylene oxide and from about 20 to about 90% by weight propylene oxide. A particularly preferred hydrogel contains about 70% polyethylene oxide and 30% polypropylene oxide.
Hydrogels which can be used are available, for example, from BASF Corp., 5 Parsippany, NJ, under the tradename Pluronic(g).
In this embodiment, the hydrogel is cooled to a liquid state and the oligonucleotides are admixed into the liquid to a concentration of about 1 mg oligonucleotide per gram of hydrogel. The resulting mixture then is applied onto the surfacle to be treated, e.g., by spraying or painting during surgery or using a catheter 10 or endoscopic procedures. As the polymer warms, it solidifies to form a gel, and the oligonucleotides diffuse out of the gel into the surrounding cells over a period of time defined by the exact composition of the gel.
Implants made of biodegradable materials such as polyanhydrides, polyorlthoesters, polylactic acid and polyglycolic acid and copolymers thereof, collagen, 15 and protein polymers, or non-biodegradable materials such as ethylenevinyl acetate (EVAc), polyvinyl acetate, ethylene vinyl alcohol, and derivatives thereof can be used to locally deliver the oligonucleotides. The oligonucleotides can be incorporated into the material as it is polymerized or solidified, using melt or solvent evaporation techni~ques, or mechanically mixed with the material. In one embodiment, the 20 oligonucleotides are mixed into or applied onto coatings for implantable devices such as dextran coated silica beads, stents, or catheters.
As described in the following examples, the dose of oligonucleotides is dependent on the size of the oligonucleotides and the purpose for which is it administered. In general, the range is calculated based on the surface area of tissue 25 to be l:reated. The effective dose of oligonucleotide is somewhat dependent on the length and chemical composition of the oligonucleotide but is generally in the range of about 30 to 3000 ,ug per square centimeter of tissue surface area. Based on calculations using the application of antisense myb in a hydrogen to blood vessel that has been injured by balloon angioplasty in a rat model, a dose of about 320 ,ug 30 oligonucleotide applied to one square centimeter of tissue was effective in suppressing expression of the c-myb gene product.
The oligonucleotides may be administered to the patient systemically for both therapeutic and prophylactic purposes. For example, antisense oligonucleotides specific for PDGFR-~ may be administered to a patient who is at risk for restenosis due to angioplasty or other procedure. The oligonucleotides may be administered by any efFective method, for example, parenterally (e.g., intravenously, subcutaneously, intramuscularly or by oral, nasal or other means which permit the oligonucleotides to access and circulate in the patient's bloodstream.
Oligonucleotides administered systemically preferably are given in addition to locally administered oligonucleotides, but also have utility in the absence of local administration. A dosage in the range of from about 0.1 to about 10 grams per administration to an adult human generally will be effective for this purpose.
Therapeutic Applications The method of the present invention can be used to treat a variety of disorders which are linked to or based on expression of a protein by a gene. The method isparticularly useful for treating vascular disorders, particularly vascular restenosis. The following non-limiting examples demonstrate use of antisense oligonucleotides toprevent or very significantly inhibit restenosis following vascular injury such as is 15 induced by balloon angioplasty procedures. This has been already accomplished by using antisense, delivered locally, to inhibit expression of genes encoding proteins determined to be involved in vascular restenosis, including c-myb, non-muscle myosin heavy chain (NMMHC) and proliferative cellular nuclear antigen (PCNA). Particularly, this invention describes the use of antisense oligonucleotides against the messenger 20 RNA rnolecules encoding the ~-subunit of the receptor for platelet-derived growth factor (PDGFR~
Expression of specific genes in specific tissues may be suppressed by oligonucleotides having a nucleotide sequence complementary to the mRNA transcript of the target gene. PDGFR-,B protein appears to be critically involved in the initiation 25 of migration and/or proliferation of smooth muscle cells. The inhibition of the production of this protein by antisense oligonucleotides offers a means for treating post-angioplasty restenosis and chronic processes such as atherosclerosis, hypertension, primary pulmonary hypertension, and proliferative glomerulonephritis, which involve proliferation of smooth muscle cells.
Illustrative of other conditions which may be treated with the present method are pulmonary disorders such as acute respiratory distress syndrome, idiopathic pulmonary fibrosis, emphysema, and primary pulmonary hypertension. These conditions may be treated, for example, by locally delivering appropriate antisense incorporated in an aerosol by inhaler. These disorders are induced by a complex overlalpping series of pathologic events which take place in the alveolus (air side), the underlying basement membrane and smooth muscle cells, and the adjacent endotl1elial cell surface (blood side). It is thought that the alveolar macrophage recognizes specific antigens via the T cell receptor, become activated and elaborates a variety of substances such as PDGF which recruit white blood cells as well as stimuliate fibroblasts. White cells release proteases which gradually overwhelm the existing antiproteases and damage alveolar phneumocytes; fibroblasts secrete extracMlular matrix which induce fibrosis. Selected growth factors such as PDGF and the subsequent decrease in blood oxygen, which is secondary to damage to the 10 alveol.3r membrane, induce smooth muscle growth. This constricts the microvascular blood vessels and further decreases blood flow to the lung. This further decreases the transport of oxygen into the blood. The molecular events outlined above also induce activation of the microvascular endothelial cell surface with the appearance of selectins and integrins as well as the appearance of tissue factor which initiates blood 15 coagulation. These selectin and integrin surface receptors allow white blood cells to adhere to microvascular endothelial cells and release proteases as well as othermolecules which damage these cells and allow fluid to accumulate within the alveolus.
The above events also trigger microvascular thrombosis with closure of blood vessels.
The end result of this process is to further impede oxygen exchange.
Antisense oligonucleotides, locally delivered to the alveolar/microvascular area, could Ibe directed against the following targets to intervene in the pathology outlined above, since the cDNA sequences of all of the targets selected are known. Thus, antisense oligonucleotides specific for mRNA transcribed from the genes would inhibit produc:tion of the alveolar macrophage T cell receptor to prevent initiation of the above 25 events; inhibit product of a protein to prevent activation of alveolar white cells, or inhibit produc:tion of elastase to prevent destruction of alveolar membrane; inhibit production of PD(,F to prevent recruitment of white cells or resultant fibrosis; inhibit production of c-myb to suppress SMC proliferation; inhibit production of p-selectin or e-selectin or various integrins to prevent adhesion of blood white cells to pulmonary microvascular 30 endothelial cells; or inhibit the production of tissue factor and PAI-1 to suppress microvascular thrombosis.
As additional examples, Tissue Factor (TF) is required for coagulation system activation. Local application of antisense targeting the mRNA or DNA of a segment of TF in the area of clot formation can prevent additional coagulation. This therapy can be employed as an adjunct to or as a substitute for systemic anticoagulant therapy or after fibrinolytic therapy, thereby avoiding systemic side effects.
Plasminogen activator inhibitor (PAI-1) is known to reduce the local level of tissue plasminogen activator (TPA). The human cDNA sequence for PAI-1 is known.
Local application of antisense targeting the mRNA or DNA of PAI-1 should permit a buildup of TPA in the targeted area. This may result in sufficient TPA production to naturally Iyse the clot without systemic side effects.
A combination of antisense-TF and antisense-PAI-I may be utilized to maximize the efficacy of treatment of several disorders, including local post thrombolytic therapy 10 and preventative post-angioplasty treatment.
Many other vascular diseases can be treated in a manner similar to that described above by identifying the target DNA or mRNA sequence. The treatment ofdiseases which could benefit using antisense therapy include, for example, myocardial infarction, peripheral muscular disease and peripheral angioplasty, thrombophlebitis, 15 cerebro-vascular disease (e.g., stroke, embolism), vasculitis (e.g., temporal ateritis) angina and Budd-Chiari Syndrome.
This method can be used against a variety of targets in addition to those detailed above. For example, DNA or mRNA encoding the following proteins could be used as target sequences: growth factors and receptors, including: PDGF-M, 20 PDGF-AB, PDGF-BB, PDGF-alpha Receptor, PDGF-beta Receptor.
This invention will be described hereinbelow by reference to the following preferred embodiments and appended figures which purpose is to illustrate rather than to limit its scope.
Figure 1: Effects of mRNA PDGFR-,B subunit antisense oligonucleotides on neointima formation: Following balloon denuding carotid arterial injury antisense oligonucleotide sequences corresponding to the fragment 4-21 (AS1) or 22-39 (AS2) or scrambled oligonucleotide of fragment 4-21 (SCR1) or 22-39 (SCR2) of 5'-region of PDGFR-~ subunit mRNA
were released into the perivascular space of injured vessels from implanted EVAc matrices. The rats were sacrificed 14 days later and the extent of neointimal hyperplasia expressed as the mean intima:media area ratio i SE from 5-6 animals per group. ~ P < 0.001 as compared to normal rats subject to balloon injury (Bl).

Figure 2: Quantitative ~ssessn,~nt of antisense oligonucleotide regulation of PDGFR-~ subunit expression in injured carotid arteries: In the absence of injury (No injury) basal expression of PDGFR-~ subunit reached 26.5 i 2.5% of all medial cells. Balloon denuding injury (Bl) led to overexpression of PDGFR-~ in both the media (black bars) and neointima (stippled bars). Both antisense oligonucleotide sequences (AS1 and AS2) to PDGFR-,B subunit mRNA reduced the PDGFR-,B
subunit expression 14 days after the vascular injury. * P < 0.05 and ***
P < 0.001 as compared to noninjured rats (No injury), and tttP < 0.001 as compared to normal rats subject to balloon arterial denudation (Bl).
Figure 3: Antisense oligonucleotide regulation of PDGFR-~ subunit expression on representative cross sections of injured carotid arteries: In the absence of injury (A) basal expression of PDGFR-~
subunit reached 26.5 i 2.5% of all medial cells. Balloon denuding injury led to overexpression in both the media and neointima (B). Both antisense oligonucleotide sequences complementary to PDGFR-~
subunit mRNA reduced receptor subunit expression 14 days after the vascular injury (C-D), magnification (400X).
Figure 4: CGr, eldlion of antisense regulation of PDGFR-~ subunit expression and neointimal hyperplasia in injured carotid arteries: The upper and lower panels show respectively the expression of PDGFR-~ subunit in the media and the intima versus the intima:media area ratio 14 days after balloon carotid arterial injury. Data was obtained from rats subject to balloon injury (Bl) but not to antisense oligonucleotide sequences treatment (~), and from rats that were treated either with AS1 -PDGFR-~
(0) or AS2-PDGFR-~ (o) Exponential fits were obtained in both cases.
Figure 5: Quantitative assessment of antisense oligonucleotide regulation of PDGFR-a subunit expression in injured carotid arteries: In the absence of injury (No injury) basal expression of PDGFR-a subunit reached 32.8 i 4.6% of all medial cells. Balloon denuding injury (Bl) led to overexpression of PDGFR-a in both the media (black bars) and neointima (doted bars). Both antisense oligonucleotide sequences (AS1 and AS2) to PDGFR-,B subunit mRNA did not reduce the PDGFR-a subunit expression 14 days after the vascular injury. *** P < 0.001 as compared to noninjured rats (No injury).
Figure 6: Antisense oligonucleotide regulation of PDGFR-a subunit expression on representative cross sections of injured carotid arteries: In the absence of injury (A) basal expression of PDGFR-a subunit reached 32.8 + 4.6% of all medial cells. Balloon denuding injury led to overexpression of PDGFR-a in both the media and neointima (B).
Neither of the antisense oligonucleotide sequences complementary to the PDGFR-~ subunit mRNA reduced PDGFR-a subunit expression 14 days after the vascular injury (C-D), magnification (400X).
Figure 7: Antisense regulation of PDGFR-,B and PDGFR-a subunit expression on cultured vascular smooth muscle cells: Quiescent confluent rat vSMC were stimulated with 10 ng/ml of PDGF-BB and total proteins collected in Laemmli buffer 0, 1, 3, 6, 12, 24 and 48 hr later.
One group of control cells was left without additional therapy (open squares), while an identical cohort treated with 20 ,uM AS1-PDGFR-~
oligonucleotide 48 hrs, 24 hrs and immediately before PDGF-BB
exposure (filled squares). Total protein (30 ,ug/lane) was applied on SDS-PAGE under reducing conditions, PDGFR-~ and a protein expression were revealed by Western blot electrophoresis and immunohistochemistry, and quantified by image densitometry.

EXAM PLES
Examlple 1:
Induction of intimal hyperplasia: Balloon denudation of common carotid arterial endothelium was performed in male Sprague-Dawley (350-4259) (Charles River Breeding Laboratories, Kingston, MA). The rats were anesthetized with intraperitoneal injections of ketamine HCI 75 mg/kg (Ketaset, Aveco Co, Fort Dodge, IA) and xylazine HCI 5 mg/kg (Xylaject, Phoenix Pharma., St. Joseph, MO). Following exposure of the left common external carotid artery, a 2 French Fogarty balloon catheter (American Edwards Laboratories, Santa Ana, CA) was inserted through an arteriotomy into the common carotid artery to the aortic arch, insuflated sufficiently with air to produce slight resistance and withdrawn three times. Upon removal of the catheter, the externalcaroticl artery was ligated, the wounds were closed, and the animals were returned to their cages. Animals were sacrificed at different periods of time (7, 14, and 28 days) after injury with an overdose of ketamine and xylazine, exsanguinated and perfused with 50 ml of Ringer's lactate solution. The treated segment of the common carotid artery was removed, cut in 2 equal segments and fixed in 5% formalin solution. The segments were embedded in paraffin and eight sections of 6 ~um were obtained by microtome along the length of the specimen. Sections were stained with Hemal:oxylin-Eosin and the areas of the intima, media and adventitia, the intima:media area ratio and the percent of luminal occlusion were calculated for each arterial segment using computerized digital planimetry with a dedicated video microscope and 10 custonnized software. The nature of specimen treatment was kept from investigators until alFter completion of the data analysis.
Antisense oligonucleotides therapy: To study the possible contribution of PDGFR-~
subunit in neointima formation antisense oligonucleotide sequences to the receptor subunit were applied directly to balloon catheter denuded carotid arteries. We 15 employed two different antisense oligonucleotide phosphorothioate backbone sequences to the murine PDGFR-,B mRNA subunit (antisense 1 [AS1-PDGFR-~ :TAT
CAC -rcc TGG MG CCC]; SEQ ID No: 1 nucleotides 4 to 21; and antisense 2 [AS2-F'DGFR-~: TCT GAG CAC TM AGC TGG]; SEQ ID No.2 nucleotides 22 to 39).
Neither sequence contained more than two consecutive guanosines. Two scrambled 20 phosplhorothioate sequences (scramble 1 [SCR1 GTG ATA GTA TGC CGA GCA];
SEQ ID No: 3 and scramble 2 [SCR2 CGT TAC GTA AGC CTA GGA]; SEQ ID No: 4) were used as controls. All sequences were synthesized at the Massachusetts Institute of Teclhnology Biopolymers Laboratory. The oligonucleotides were deprotected, dried down, resuspended in Tris-EDTA (10 mmol Tris pH 7.4 and 1 mmol EDTA pH 8.0), 25 and quantified by spectrophotometry. To sustain the release and insure the local adminiistration of the oligonucleotide sequences directly to the injured arteries the oligon-lers were embedded within ethylene-vinyl acetate copolymer (EVAc; DuPont Co., V\'ilmington, DE) matrix release devices as previously described (17,24-26). After the endothelial denudation of the left common carotid artery, the EVAc devices 30 containing 400 1~9 of the scrambled or antisense PDGFR-~ oligomers were placed adjacent to the injured carotid arteries. In 14 days approximately 65% of the compound was released with a zero-order kinetics, and it has been esli",ated that approximately 1 % of the released oligomer would be delivered to the blood vessel wall from these types of devices (17, 25).

ImmunohistochemistryofPDGFR-a and-~ subunitexpression: Expression of PDGFR
-a and -~ subunits was determined immunohistochemically. Arterial sections were deparaffinized in xylene and ethanol baths, endogenous peroxidase activity was quenched in a solution of methanol (200 ml) plus hydrogen peroxide (3%, 50 ml), and 5 nonspecific binding antibody binding prevented by preincubating the tissues with serum (1:10) from species other than those used to raise the primary antibody. Arterial sections were then exposed to the primary antibody, PDGFR-a IgG (Santa Cruz Biotech., Santa Cruz, CA) or rabbit polyclonal anti-human PDGFR-~ IgG (UBI, LakePlacid, NY) diluted (1 :100, 1 :200, 1 :500, 1 :1000), or rinsed with PBS, and incubated 10 with a biotinylated goat anti-rabbit IgG (1 :400) (Dako, Carpinteria, CA). A Dot-blot and Western blot analysis were performed to confirm the cross reactivity of both rabbit antibodies to rat proteins. Peroxidase labelling was achieved with an incubation using avidin/'peroxidase complex (Vector Labs Inc., Burling'ame, CA), and antibody visualization established after a 5 min exposure to 0.05% 3,3'-diaminobenzidine (Sigma Chem, St Louis, MO) in 0.05 M Tris-HCI at pH 7.6 with 0.003% hydrogen peroxide. The arteries were counterstained by a rapid immersion (10 seconds) in Gill's hematoxylin #3 solution, and rinsed in tap and distilled water.
Cell culture: Vascular smooth muscle cells (vSMC) of rat thoracic aorta were isolated by the explant technique (27). The cells were seeded in culture dishes (35 mm), grown to conlluence in Dulbecco's modified Eagle medium (DMEM) supplemented with 10%
fetal Ibovine serum (complement-heat inactivated), penicillin (50 U/ml) and streptomycin (50 mg/ml), and used between the 6th and 10th passage. At confluence, the me!dium was replaced with DMEM, 0.1% FBS and antibiotics, two groups of cells were treated either with AS1-PDGFR-~ or SCR1-PDGFR-~ (direct application not embeclded into EVAc matrices) at 0, 24 and 48 hrs, whereas a third group was untreated and served as control. PDGF-BB (10 ng/ml) was added and total proteinsfrom the cells were collected 0,1, 3, 6, 12, 24 and 48 hrs later.
Western blot analysis of PDGFR-a and -~ protein subunit: Total proteins were prepar~ed by washing the cells with ice cold PBS, and the addition of 100 ~ul of Laemmli buffer containing EDTA 1 mM, phenylmethylsulfonyl fluoride 1 mM, leupeptin 10 ug/ml and NaVO31 mM. The extracted cell proteins were boiled for 5 min, and a 30 ,ul aliquot (~30 ,~lg protein) of each sample was separated by 7.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions (MinigelApparatus, Bio-Rad) and transblotted onto 0.45 ,um polyvinylidene difluoride memblranes (PVDF, Millipore). The membranes were blocked in TBS-5% Blotto (Tris-HCI 10 mM, NaCI 150 mM pH 7.85; 5% non fat dry milk Bio-Rad) for 1 hr at room temperature with gentle agitation. Membranes were washed with TBS and TTBS (TBS;0.05% Tween 20 Bio-Rad), and incubated with rabbit polyclonal anti-human PDGFR-~i IgG antibodies (dilution 1 :200 in TTBS) for 2 hrs at room temperature. The membranes were washed with TTBS and incubated with alkaline-phosphatase goat anti-rabbit IgG
(1 :100) for 2 hrs at room temperature. Membranes were washed with TTBS and TBS
and alkaline phosphatase bound to secondary antibodies was revealed by chemiluminescence (Bio-Rad kit). Prestained molecular weight marker proteins 10 (Bio-Rad) were used as standards for SDS-PAGE. To probe the immunoblots with second antiserum, the PVDF membranes were stripped by incubation in 62.5 mM
Tris-HCI, pH 6.7, 2% SDS, and 100 mM 2-mercaptoethanol for 30 min at 50~C, gentle agitation. The blots were then washed twice with TBS, then washed at least 5 times to remove traces of 2-mercaptoethanol. Then, the blots were incubated with polyclonal 15 anti-human PDGFR-a antibodies (dilution 1 :200 in TTBS) and processed as described above.
Statistical analysis: Data are mean i SEM. Statistical comparisons were determined by variance analysis followed by an unpaired Student's t-test with Bonferroni's correction for multiple comparisons. Data were considered to be significantly different 20 if P < 0.05 was observed.
RESULTS
Neointimal hyperplasia: Effects of PDGFR-~ mRNA antisense oligonucleotides:
Neoinltimal formation determined 14 days after balloon deendothelialization of rat common carotid arteries served as controls for all subsequent experiments. At this 25 time an intima:media area ratio of 1.37 i 0.15 was observed (Fig. 1). Antisense sequences directed against the PDGFR-~ subunit mRNA were used to reduce receptorsubunit expression. The sustained release of antisense oligonucleotide sequencesAS1-PDGFR-~ or AS2-PDGFR-~ from EVAc matrices placed adjacent to the injured artery reduced the intima:media area ratio to 0.27 i 0.09 and 0.55 i 0.11, but neither 30 scrambled oligonucleotide sequence significantly affected neointimal thickening (SCR1 1.5 i 0.12 and SCR2 1.66 i 0.13, Fig. 1). Medial areas were no different in any treated or control groups (data not shown).

Protein expression of PDGFR-a and -~ subunit: In the absence of vascular injury basaN~xpression of PDGFR-,B subunit was observed on medial vSMC. 26.5 i 2.5% of these cells were immunohistochemically identified with an antibody that specifically recognizes the PDGFR-~ protein (Figs. 2, 3A). 14 days after a denuding injury, PDGFR-~ protein doubled on medial vSMC (51.2 i 5%, p < 0.001) and became evident on 74.5 i 2.5% of the intimal cells (Figs. 2, 3B). The perivascular sustained release of both antisense sequences significantly reduced PDGFR-~B subunit expression in both vascular compartments, yet the sequence closer to the 5' mRNAend, AS1-PDGFR-~, was more potent at reducing receptor subunit and neointima formal:ion. Two weeks after the treatment of vascular injured carotid arteries with AS1-PDGFR-~, only 4.4 i 1.8% of medial cells and 2.8 i 1.6% of intimal cells retained PDGFR-~ subunit expression (p < 0.001 compared with controls (Bl), Figs.2, 3C). The AS2-PDGFR-~ oligonucleotide reduced these values to 15.9 i 5.2% and 19.1 i 5.2%
respectively (p < 0.001 compared with controls (Bl), Figs. 2, 3D). Scrambled oligonucleotide sequences had no effect on receptor subunit expression (data notshown). The suppression of neointima with application of antisense PDGFR-~
oligomlers followed inhibition of PDGFR-~ subunit expression in an exponential fashion (Intima: Media area ratio = e(~)); where ~ was the percent of all cells expressing the PDGFR-~ subunit and T was defined as the exponential constant. Intimal thickening correklted with medial PDGFR-~ subunit expression with an exponential constant (T) of 17.64 (p < 0.01; r = 0.82, Fig. 4A), and with intimal receptor expression with an expon~ential constant (T) of 0.32 (p < 0.001; r = 0.96, Fig. 4B).
Specificity of the antisense oligonucleotide effect for PDGFR-~ mRNA was demonstrated through similar immunohistochemical identi~icalion of PDGFR-a subunit protein expression. In absence of vascular injury PDGFR-a subunit expression wasobserved on 32.8 i 4.6% of medial vSMC (Figs. 5, 6A). Fourteen days after denuding injury l'DGFR-a expression increased on medial vSMC (52.7 i 3.4%, p < 0.001) andwas noted on 57.3 i 4.2% of the intimal cells (Figs. 5, 6B). Despite their effects on PDGFR-~ subunit expression the sustained perivascular release of either antisense sequences for 14 days after a vascular injury did not affect the PDGFR-a subunitexpression. PDGFR-a protein expression in the media and intima of rat carotid treated with AS1-PDGFR-~ was 58.5 i 3.2% and 61.5 i 2.8% respectively (Figs. 5, 6C), and59.4 ~: 3.5% and 62.9 i 3.8% respectively for animals treated with AS2-PDGFR-~

CA 02228977 l998-02-03 (Figs. 5, 6D). Treatment with scrambled oiigonucleotide sequences did not alter the expression of PDGFR-a as compared to control animals (Bl) (data not shown).
Profein expression of PDGFR-~ subunit on cultured smooth muscle cells: vSMC weregrown to confluency on 35 mm Petri dishes, then kept quiescent in DMEM with 0.1 %
FBS, AS1-PDGFR-,B (20,uM) or SCR1-PDGFR-~ oligonucleotide (20,uM) were added at 0, 24 and 48 hrs, a third group of cells was untreated with oligonucleotide and servecl as control. Two days after the first oligonucleotide application, PDGF-BB (10 ng/ml) was added in each group. At 0, 1, 3, 6,12, 24 and 48 hrs after the addition of PDGF-BB, the cells were washed with cold PBS, Laemmli buffer (100 ,ul) was added, 10 total proteins collected, quantified by bioassay, and the expression of PDGFR-~ at each time point was determined by Western blot electrophoresis and quantified byimage densitometry. Significant basal PDGFR-~ protein expression was noted in vSMC (Fig. 7). These values decreased by 53% one hour after stimulation with PDGF-BB, and by an additional 32% 11 hrs after that, to be reexpressed near basal 15 levels 48 hours after initial stimulation. AS1 -PDGFR-~ suppressed protein expression by over 75% at baseline, and for the duration of the experiment (Fig. 7). These effects were <;pecific for the PDGFR-~ target gene as PDGFR-a protein expression was unaffected by the antisense PDGFR-~ oligonucleotide sequence. The SCR1-PDGFR-~ oligonucleotide sequence had no affect on the normal pattern of 20 PDGFR-~ protein expression seen in control vSMC at baseline and following stimulation with PDGF-BB (data not shown).
DISCUSSION
I n previous reports we and other investigators showed that antisense oligonucleotide sequences complementary to DNA binding proteins and cell-cycle 25 regulators genes such as c-myb, c-myc, cdc2, cdk2, nonmuscle myosin and PCNA
inhibited target protein expression, suppressed vSMC proliferation in vitro and in vivo and inhibited neointimal formation in injured arteries of different animal species (10-17, 28-31) . To date, targeted genes were principally those involved in cell cycle progression. However, these genes are not unique to vSMC, but are also expressed30 in other cell types and their use might induce side effects in tissues with high rates of proliferation. Growth factors play a central role in all phases of the vascular response to injuly, and yet no studies have yet to be reported on the consequences of antisense sequences directed against growth factor and/or their receptors. PDGF, for example, is critical to vSMC migration and intimal thickening (1, 7, 32) in a manner fairly selective for vSMC (7, 8, 32) and as a result became the focus of the present manuscript.
We employed two antisense oligonucleotide sequences selective for either positions 4-21 or 22-39 of the PDGFR-,B mRNA subunit. As PDGFR-~ subunit 5 expression is reexpressed after initial down-regulation following PDGF-BB stimulation in vitro (Fig. 7) and is manifest over the fu112 week period after in vivo injury (Figs. 2, 3), the oligonucleotides were embedded in EVAc matrices to provide a sustained releasr during the entire experimental procedure. Previous studies demonstrated the need to match the kinetics of oligonucleotide release to the kinetics of antisense target 10 gene cxpression. When gene expression is prolonged, as it is for c-myc, a more sustained oligonucleotide release device was required to demonstrate biologic effect (17). Sustained release of the two antisense oligonucleotide sequences compl~ -mentary to PDGFR-~ mRNA reduced arterial intimal thickening by 80 and 60%
respectively. In normal rat carotid arteries approximately 25% of the medial vSMC
15 stained positively for PDGFR-~ subunit protein. Two weeks after vascular injury this expression more than doubled in medial vSMC, and close to 75% of the cells forming the neointima stained positively as well. Interestingly, while both antisense sequences reduced PDGFR-~ subunit protein expression below the basal level (25%) observed in the media of uninjured rat carotid arteries, the oligonucleotide sequence closer to 20 the 5'-mRNA region was almost four times more potent at inhibiting PDGFR-~expression in medial and intimal vSMC. The variable response to these two sequences enable!d delineation of a correlation between PDGFR-~ levels and neointimal potential.
In arteries where PDGFR-~ subunit expression was reduced below basal levels, i.e., in fewer than ~25-30% of all cells, only minimal intimal thickening was observed. When 25 PDGFR-~ subunit expression exceeded basal levels, intimal proliferation rose expon,entially (Fig. 4).
Though the first antisense sequence (AS1) almost completely reduced PDGFR-,B subunit protein expression by day 14, it did not completely abolish intimal hyperplasia. This observation raises the possibility that while PDGFR-BB stimulation 30 may contribute up to 80% of the neointima formation, the secretion of other growth factors or peptides might contribute to the residual fraction (33-36). Alternatively, the lack ol complete inhibition of neointima may stem from the inability of the sustained antisense delivery system to fully suppress the immediate-early PDGF effect. TheEVAc matrices allow the release of their embedded contents over the entire course of the experiment, not as a large bolus at the time of injury. Upon vascular injury the almosl: immediate platelet adhesion to subendothelial connective tissue induces the releas~- of platelet PDGF-BB which stimulates its PDGFR-BB, and the interval of time betwel n balloon denudation and oligonucleotide release upon application may well 5 have allowed sufficient growth factor-receptor interaction to activate the intracellular eventci that lead to neointima formation. Indeed, our in vitro study revealed first, a compllex pattern of PDGFR-~ protein expression in response to stimulation with PDGF-BB, with initial suppression of heightened basal levels that returned within 48 hrs, and second, that pretreatment with AS1-PDGFR-~ oligonucleotide reduced 10 receptor subunit expression at baseline by 4 fold, and upon stimulation with PDGF-BB
for the duration of the experiment (Fig. 7). The administration of antisense PDGFR-~oligon-lers days before the surgical procedure might reduce the basal expression of PDGFR-,B subunit sufficiently to prevent its interaction with PDGF-BB or prevent the biological activity induction related to their interaction after the injury. Such studies 15 could also allow one to determine the impact of these early interactions to residual intimal thickening.
The use of antisense technology is beset by questions of specificity (37-39).
Recent reports have raised concern that the antiproliferative activity of antisense oligonucleotides to c-myb and c-myc, for example, was arose from aptameric rather 20 than a hybridization-dependent antisense mechanism (37, 38). It was hypothesized that oligonucleotides with four sequential guanosines might bind to serum proteins including growth factors such as bFGF, aFGF, PDGF and VEGF, reducing the interaction of these growth factors with their receptors, and the intracellular signal transduction leading to gene protein expression (such as c-myc and c-myb) involved 25 in cell cycle progression (39). Nonetheless, other studies have shown specific in vivo and/or in vitro effects with antisense oligonucleotides lacking multiple sequential guano~sines to these and other genes involved in cell cycle progression such as cdc2, cdk2, nonmuscle myosin and PCNA (11-14, 28). Neither antisense sequence used in the shldies we now report possessed more than two contiguous guanosines. To more30 definitely address this issue however, we examined the effects of the sequences on the a subunit. As antisense sequence can discriminate between oligonucleotide sequences that differ by one or two bases (15, 40, 41 ) we compared effects of AS1 on the PC)GFR-~ and -~ subunits. Quantitative analysis of protein expression on vSMC
in culture confirmed immunohistochemical identification of antigenicity in vivo. The antisense sequences directed against the -~ subunit inhibited only this targeted protein subunit without affecting the PDGFR-a protein subunit expression (Figs. 5-7).
Scrambled oligonucleotide sequences also failed to reduce neointima formation, or PDGFR-~ subunit protein expression in vitro or in vivo.
It is interesting to note that the antisense sequence closer to the 5' end of PDGFR-~ mRNA was more potent at inhibiting intimal thickening and PDGFR-~ protein expression than the AS2 sequence. This is in accordance with previous reports which have shown that the biologic effects of antisense oligomers are dictated in part by the location of the sense target sequence. Antisense oligonucleotides directed at or near 10 the 5' translation initiation site were most effective at inhibiting gene expression, and in some cases a shift of few base pairs in the targeted sequence was sufficient to induce drastic variation in target gene inhibition (42-45). This discrepant effects between similar sequences remains enigmatic. Possible explanations could be that the secondary structure of the mRNA close to the initiation codon might offer a more15 favourable hybridization site for the antisense sequence. Downstream regions of the mRNA might fold and reduce the hybridizing access for the antisense sequences.
Alternatively, antisense sequences complementary to or near the 5' mRNA region may be more potent at preventing mRNA translation (46-48). These and other issues will require further study before antisense technology can reach its full potential.
CONCLUSION
We observed that the sustained perivascular application of antisense oligonucleotide sequences complementary to PDGFR-~ mRNA not only prevented overexpression of PDGFR-,B protein in healing medial and intimal vSMC but did so in a manner commensurate with effects on intimal thickening. Almost complete abolition 25 of PD(,FR-~ protein expression was achieved with the antisense sequence closer to the 5' PDGFR-,B mRNA. The antisense PDGFR-~ effect was specific. The oligomers emplo'yed did not bear 4 contiguous guanosines eliminating concern for non-specific, aptameric binding, and only the antisense sequences suppressed protein expression, and only of the target PDGFR-,B, and not the PDGFR-a protein subunit.
As PDGFR-~ expression is specific to mesenchymal cells such as vSMC and fibroblasts, the regulation of this cell membrane receptor might provide an important advantage over the inhibition of cell cycle proliferative proteins which are expressed ubiquitously. Regulation of PDGFR-~ could contribute to the prevention of intimal thickening without affecting the proliferation of unrelated but critical cells. Further investiigations are needed to determine whether and how the neointima will respond with rlelease of PDGFR-~ protein expression inhibition after the removal or the degradation of the antisense oligomers. Finally, our results demonstrate again the value of antisense technology in helping elucidate the mechanisms involved in vascular healin!3, and as a possible approach to the prevention and progression of the accele!rated arteriopathies that follow vascular intervention.
It is readily apparent from the foregoing that antisense oligonucleotides to PDGFR-~ mRNA successfully prevented restenosis. Other antisense oligonucleotidesmay be designed from the sequences of the receptor PDGFR-,B subunit and used with 10 success. The antisense may be adjuncted with any other antisense oligonucleotides which also show inhibition of intimal thickening. Examples thereof are those already described in WO 93/08845 and USP 5,593,974 which hybridize with c-myb (SEQ. ID.
No. 5) NMMHC (SEQ. ID. No. 6) and/or PCNA (SEQ. ID No. 7) mRNAs.
Examlple 2:
It has been shown that, in vitro, antisense oligonucleotides to both c-myb (Seq.ID No 5) and NMMHC (Seq. ID No. 6) caused substantial suppression of cellular proliferation while the sense oligonucleotides had no effect and were similar to the results obtained using just Tris-EDTA buffer as a control.
Antisense c-myb oligonucleotide:
Sequence ID No. 5 GTGTCGGGGTCTCCGGGC
Antisense NMMHC oligonucleotide:
Sequence ID No. 6 CATGTCCTCCACCTTGGA
The inhibitory action of antisense phosphorothiolate oligonucleotides directed against NMNHC or c-myb was concentration-dependent (antisense NMMHC: 32% vs 65% suppression at 2 ,uM and 25 ,uM, respectively; antisense c-myb: 33% vs 50%
suppression at 2 ,uM and 25 ,uM respectively). Previous estimates of the relative abundance of these two messages indicated that c-myb mRNA occurs at extremely 30 low concentrations in exponentially growing SMC (less than 0.01% of poly A+ RNA), whereas NMMHC mRNA is present at significantly higher levels. The observed concentration dependence of the two antisense oligonucleotides with regard to growth inhibition was consistent with the relative abundance of the two mRNAs.

The antiproliferative effects of the antisense and sense phosphorothiolate oligonucleotides were also evaluated with the BC3H1 cell line as well as with primary rat and mouse aortic SMC. The data obtained showed that growth of the three celltypes is greatly suppressed with phosphorothiolate antisense but not sense NNMHC5 or c-myb oligonucleotides. The admixture of antisense c-myb oligonucleotides for 4 hr produc:ed an antiproliferative effect which is identical to that observed with continuous exposure for 72 hr (50% suppression of cell growth).
The treatment of SMC with antisense NMMHC oligonucleotides produced no growth inhibitory effect at either time point, whereas exposure to antisense c-myb oligonucleotides generated a 19% suppression of proliferation at 72 hr and a 40%suppression of proliferation at 120 hr.
Example 3: Release of oligonucleotides from polymeric matrices.
Release of Oligonucleotides from PluronicTM Gel Matrix Matrices made from a poly(ethylenoxide-propylene oxide) polymer containing c-myb and NMMHC antisense oligonucleotides (described in Materials and Methods) were prepared in order to test the rate of release of the oligonucleotides from the matric,es. The test samples were prepared by weighing 1.25 9 of UV sterilized PluronicTM 127 powder (BASF Corp., Parsippany, N.J.) in scintillation vials and adding 3.25 n~l of sterile water. Solubilization was achieved by cooling on ice while shaking.
To these solutions were added 500 1~1 of a sterile water solution containing theoligonucleotides (5.041 mg/500,ul). The final gels contained 25% (w/w) of the polymer and 1 mg/g oligonucleotides.
The release kinetics of the gels containing oligonucleotides were determined by placing the gels in PBS and measuring the absorption (OD) over time. The results for four test gels indicate that oligonucleotides are released from the gels in less than one hour.
Release of oligonucleotides from EVAc matrices The release of oligonucleotides from ethylene vinyl acetate (EVAc) matrices was demonstrated.
Matrices were constructed and release was determined as described by Murray et al. (1983), In Vitro., 19: 743-748. Ethylene-vinyl acetate (EVAC) copolymer (ELVAX
40P, C)uPont Chemicals, Wilmington, DB) was dissolved in dichloromethane to forma 10% weight by volume solution. Bovine serum albumin and the oligonucleotide were dissolved together at a ratio of 1000 - 2000:1 in deionized HO, frozen with liquid N end then Iyophilized to form a dry powder. The powder was pulverized to form a homogeneous distribution of particles less than 400 microns in diameter. A knownquantilty of the powder was combined with 4-10 ml of the 10% (w/v) EVAc copolymer solution in a 22 ml glass scintillation vial. The vial was vortexed for 10 seconds to form a homogeneous suspension of the drug particles in the polymer solution. This suspension was poured onto a glass mold which had been precooled on a slab of dry ice. After the mixture froze it was left in place for 10 minutes and then removed from the mold and placed into a -20~C freezer for 2 days on a wire screen. The slab was dried for an additional 2 days at 23~C under a 600 millitorr vacuum to remove residual 10 dichloromethane. After the drying was complete 5 mm X 0.8 mm circular slabs are excised with a #3 cork borer.
The results indicate that about 34% of the oligonucleotide was released within the first 48 hours.
Example 4: In vivo application of oligonucleotides to inhibit c-myb and NMNHC
15 in rats.
I. Animal Model.
Balloon stripping of the rat carotid artery is used as a model of restenosis in vivo. F'ats were anesthetized with Nembutal (50 mg/kg). A left carotid dissection was carriecl out and a 2F Fogarty catheter was introduced through the arteriotomy incision 20 in the internal carotid artery. The catheter was advanced to the aortic arch, the balloon was inflated and the catheter withdrawn back to the arteriotomy site. This was repeated two more times. Subsequently, the balloon being withdrawn, the internalcaroticl was tied off, hemostasis achieved, and the wound closed.
Il. Oligonucleotide Delivery.
The oligonucleotides were applied with a hydrogel and with an implantable ethylene vinyl acetate (EVAc) matrix. A polyethylene oxide-polypropylene oxide polym~er (PluronicTM 127, BASF, Parsippany, NJ) was used as a hydrogel. The PluronicTM gel matrices were prepared as described in Example 3. Briefly, sterile solutions of PluronicTM 127 were prepared by weighing 1.25 g of UV sterilized Pluronic 30 powder into a scintillation vial and adding 3.25 ml of sterile water. Solubilization was achieved by cooling on ice while shaking, forming a solution containing 27.7% byweight of the polymer. To these solutions were added 500,uL of a sterile water solution of the antisense c-myb (See Example 2) oligonucleotides (5.041 mg/500,uL). The final gels were 25% w/w of PluronicTM polymer and 1 mg/g oligonucleotide. Drug-free 25%

(w/w) gels were prepared as controls. The EVAc matrices were prepared as described in Example 3, and contained 40 ~ug of oligonucleotide.
Immediately after balloon injury, 200 ,ul of Pluronic/oligonucleotide solution (which contained 200 ,ug of the oligonucleotide) was applied to the adventitial surface 5 of the artery and gelling was allowed to occur. The antisense/EVAc matrix (which contained 40 ~9 of the oligonucleotide) and drug-free gels were applied in the same manner.
Ill. Quantification of Effect.
After 14 days, the animals were sacrificed and the carotid arteries were 10 perfused under pressure (120 mmHg with Ringer s Lactate. Both carotid arteries were excise!d and fixed in 3% formalin. Thin sections were then prepared for light microscopy in a standard manner. The slide was visualized and digitized using a dedica~ted computer system and by a hand held plenymeter and the area of neointimal proliferation calculated (in sq mm).
In control animals which received no treatment, or which were treated with the drug-hee gel, there was extensive restenosis, characterized by symmetric neointimal formation along the entire length of the injured artery, narrowing the lumen by about 60%, resulting in an intima ratio of 1.4.
In animals treated with antisense c-myb oligonucleotides, there was minimal 20 restenosis, minimal proliferative rim (less than 10% of the lumen) that was limited to the portion of the artery in direct contact with oligonucleotide, with an intima/media ratio of 0.09. This effect was most pronounced for animals treated with the antisense/Pluronic(g). The intima/media ratio obtained using EVAc/antisense was about 0.45. However, the EVAc matrix contained 40 Hg of oligonucleotide, compared with25 200 ,ug of oligonucleotide administered in the Pluronic gel, which may account for some of the difference.
Seven rats in each treatment group were subjected to balloon angioplasty, and the arterial walls treated as follows: with a drug free hydrogel (PluronicTM127 as described above), a hydrogel containing sense c-myb, a hydrogel containing antisense 30 c-myb, and no treatment at all. Similar high levels of neointimal proliferation occurred in all animals except those treated with antisense c-myb, where the levels of proliferation were dramatically lower.

Example 5: Inhibition of PCNA using antisense oligonucleotides.
Using the same methodology as in Example 2, antisense for PCNA having the sequence:
Sequence ID No. 7:
GAT CAG GCG TGC CTC AAA, was applied to SV-SMC cells in culture. Sense PCNA was used as a negative control;
NMMHC-B was used as a positive (inhibitory) control.
There was no suppression of smooth muscle cell proliferation in the negative control; there was 52% suppression using antisense NMMHC-B and 58% suppression with antisense PCNA.
Examlple 6: In vivo application of antisense oligonucleotides to inhibit smooth musclle cell proliferation in rabbits.
New Zealand white rabbits (1-1.5 Kg) were anesthetized with a mixture of ketamine and zylazine and carotid dissection was performed as described in Example 3. A 5F Swan-Ganz catheter was inserted and positioned in the descending aorta with fluoroscopic guidance. The Swan-Ganz catheter was exchanged over the wire for anangioplasty catheter with a 3.0 mm balloon. The common iliac artery was angioplastied 3 times at 100 PSI for 90 seconds each time. A Wolinsky catheter was introduced and loadecl with oligonucleotide solution in a total volume of 5cc normal saline. Saline was injected as a control in a counterlateral iliac artery. The oligonucleotides were a mixture of anti~sense mouse c-myb and human NMMHC (200 LUM of each), described above.
The n-lixture was injected under 5 atmospheres of pressure over 60 seconds. Two rabbits were treated with antisense oligonucleotide.
The animals were sacrificed 4 weeks later and the arteries were processed as described in Example 4 for rat arteries.
The results indicated a 50% reduction is neointimal proliferation in rabbit arteries treated with antisense compared to saline alone.
Examlple 7. Inhibition of proliferation of baboon smooth muscle cells using antisense oligonucleotides.
Using the same methodology as in Example 2, primary baboon smooth muscle cells (gift from Dr. Hawker, Emory University) were treated with antisense human myb and human NMNHC. The cells were allowed to grow for 72 hours after treatment with the oligonucleotides, then counted as described in Example 2. The results show that hNMMlHC caused 65.5% growth suppression and c-myb caused 59.77% growth suppression in the baboon cells.
Example 8: Compositions for use in the prevention of restenosis It will be appreciated from the above teachings that antisense oligonucleotides to PCNA, NMMHC, c-myb and PDGFR-~ and any mixture thereof can be made and used by themselves or in a suitable carrier. The above examples are therefore not restrictive.
Equivialents One skilled in the art will recognize several equivalents, modifications, variations of the present method from the foregoing detailed description. Such equivalents, modifications and variations are intended be encompassed by the appended claims.
REFERENCES
1. Clowes AW, Schwartz SM. Significance of quiescent smooth muscle migration in the injured rat carotid artery. Circ. Res. 1985; 56: 139-145.
15 2. Lindner V, Reidy MA. Proliferation of smooth muscle cells after vascular injury is inhibited by an antibody against basic fibroblast growth factor. Proc. Natl.
Acad. Sci USA. 1991; 88: 3739-3743.
3. Olson NE, Chao S, Lindner V, Reidy MA. Intimal smooth muscle cell proliferation after balloon catheter injury: the role of basic fibroblast growthfactor. Am. J. Pathol. 1992; 140: 1017-1023.
4. Schwartz SM, Deblois D, O'Brien ERM. The intima: Soil for atherosclerosis and restenosis. Circ. Res.1995; 77: 445-465.
5. Clowes AW, Clowes HM, Reidy MA. Kinetics of cellular proliferation after arterial injury lll. Endothelial and smooth muscle growth factor in chronically denuded vessels. Lab. Invest. 1986; 54: 295-303.
6. Koyama N., Hart CE, Clowes AW. Different functions of the platelet-derived growth factor-a and -,B receptors for the migration and proliferation of cultured baboons smooth muscle cells. Circ. Res. 1994; 75: 682-691.
7. Jawien A, Bowen-Pope DF, Lindner V, Schwartz SM, Clowes AW.
Platelet-derived growth factor promotes smooth muscle migration and intimal thickening in a rat model of balloon angioplasty. J. Clin. Invest. 1992; 89:
507-511.

8. Ferns GM, Raines EW, Sprugel KH, Motani AS, Reidy MA, Ross R. Inhibition of neointimal smooth muscle accumulation after angioplasty by antibody to PDGF. Science. 1991; 253: 1129-1132.
9. Lau K-W, Sigwart U. Restenosis an accelerated arteriopathy: pathophysiology, preventive strategies and research horizons. In Molecular interventions and local drug delivery. Edelman ER, Levy RJ editors. Saunders WB Company, Cambridge UK.1995; 1-28.
10. Simons M, Edelman ER, DeKeyser JL, Langer R, Rosenberg RD. Antisense c-myb oligonucleotides inhibit intimal arterial smooth muscle cell accumulation in vivo. Nature.1992; 359: 67-70.
11. Morishita R, Gibbons GH, Ellison KE, Nakajima M, Zhang L, Kaneda Y, Ogihara T, Dzau VJ. Single intraluminal delivery of antisense cdc2 kinase and proliferating-cell nuclear antigen oligonucleotides results in chronic inhibition of neointimal hyperplasia. Proc. Natl. Acad. Sci. USA. 1993; 90: 8474-8478.
15 12. Morishita R, Gibbons G, Ellison KE, Nakajima M, von der Leyen H, Zhang L, Kaneda Y, Ogihara T, Dzau V. Intimal hyperplasia after vascular injury is inhibited by antisense cdk2 kinase oligonucleotides. J. Clin. Invest. 1994; 93:
1458-1464.
13. Abe J-l, Zhou W, Taguchi J-l, Takuwa N, Miki K, Okazaki H, Kurokawa K, Kumada M, Takuwa Y. Suppression of neointimal smooth muscle cell accumulation in vivo by antisense cdc2 and cdk2 oligonucleotides in rat carotid artery. Biochem. Biophys. Res. Comm..1994; 198: 16-24.
14. Simons M, Edelman ER, Rosenberg RD. Antisense proliferating cell nuclear antigen oligonucleotides inhibit intimal hyperplasia in a rat carotid artery injury model. J. Clin. Invest.1994; 93: 2351-2356.
15. Bennett, MR, Anglin S, McEwan JR, Jagoe R, NewbyAC, Evan Gl. Inhibitionof vascular smooth muscle cells proliferation in vitro and in vivo by c-myc antisense oligodeoxynucleotides. J. Clin. Invest.1994; 93: 820-828.
16. Shi Y, Fard A, Galeo A, Hutchinson HG, Vermani P, Dodge GR, Hall DJ, Shaheen F, Zalewski A. Transcatheter delivery of c-myc antisense oligomers reduces neointimal formation in a porcine model of coronary artery balloon injury. Circulation. 1994; 90: 944-951.
17. Edelman ER, Simons M, Sirois MG, Rosenberg RD. C-myc in vasculoproliferative disease. Circ. Res. 1995; 76: 176-182.

18. Crooke R. In vitro toxicology and pharmacokinetics of antisense oligonucleotides. Anticancer Drug Des.1991; 6: 609-646.
19. Loke S, Stein C, Zhang X, Mori K, Nakanishi M, Subashinge C, Cohen J, Neckers L. Characterization of oligonucleotide transport into living cells. Proc.
Natl. Acad. Sci. USA.1989; 86: 3474-3478.
20. Yakubov LA, Deeva EA, Zarytova VF, Ivanova EM, Ryte AS, Yurchenko LV, Vlassov W. Mechanism of oligonucleotide uptake by cells: involvement of specific receptors? Proc. Natl. Acad. Sci. USA. 1989; 86: 6454-6458.
21. Wagner R, Nishikura K. Cell cycle expression of RNA duplex unwinding activity in cells. Mol. Cell Biol. 1988; 8: 770-777.
22. Raines EW, Bowen-Pope DF, Ross R. Platelet-derived growth factor. In Peptide growth factors and their receptors 1. M.B. Sporn and A.B. Roberts, editors. Springer-Verlag, NewYork. 1991; 173-262.
23. Majesky MW, Reidy MA, Bowen-Pope DF, Hart CE, Wilcox JN, Schwartz SM.
PDGF ligand and receptor gene expression during repair of arterial injury. J.
Cell Biol.1990; 111: 2149-2158.
24. Langer R, Brown L, Edelman ER. Controlled release and magnetically modulated release systems for macromolecules. Drug and enzyme targeting.
Methods Enzymol.1985; 112: 399-423.
20 25. Edelman ER, Adams DA, Karnovsky MJ. Effect of controlled adventitial heparin delivery on smooth muscle cell proliferation following endothelial injury. Proc.Natl. Acad. Sci. USA. 1990; 87: 3773-3777.
26. Rhine WD, Sukhatme S, Hsieh DST, Langer R. A new approach to achieve zero-order release kinetics from diffusion-controlled polymer matrix systems.
In Controlled Release of Bioactive Materials. R. Baker, editor. Academic Press, New York. 1980; 177-187.
27. Ross R. The smooth muscle cell: Il. Growth of smooth muscle cell in culture and formation of elastic fibers. J. Cell Biol. 1971; 50: 172-186.
28. Simons M, Rosenberg RD. Antisense nonmuscle myosin heavy chain and c-myb oligonucleotides suppress smooth muscle cell proliferation in vitro. Circ.Res. 1992; 70: 835-843.
29. Ebbecke M, Unterberg C, Buchwald A, Stohr S, Wiegand V. Anti-proliferativeeffects of a c-myc antisense oligonucleotide on human arterial smooth muscle cells. Basic Res. Cardiol. 1992; 87: 585-591.

30. Biro S, Fu YM, Yu ZX, Epstein SE. Inhibitory effects of antisense oligodeoxynucleotides targeting c-myc mRNA on smooth muscle cell proliferation and migration. Proc. Natl. Acad. Sci. USA. 1993; 90: 654-658.
31. Shi Y, Hutchinson HG, Hall DJ, Zalewski A. Downregulation of c-myc expression by antisense oligonucleotides inhibits proliferation of human smooth musclecells. Circulation. 1993; 88: 1190-1195.
32. Nabel EG, Yang Z, Liptay S, Sang H, Gordon D, Haudenschild CC, Nabel GJ.
Recombinant platelet-derived growth factor B gene expression in porcine arteries induces intimal hyperplasia in vivo. J. Clin. Invest. 1993; 91:
1822-1829.
33. Baumgartner HR, Platelet interaction with vascular structures. Thromb Diath.
Haemorrh. Suppl. 1972; 51: 161 -176.
34. Heldin CH, Watson A, Westermark B. Partial purification and characterization of platelet factors stimulating the multiplication of human glial cells. Exp. Cell Res. 1977; 109: 429-437.
35. Assoian RK, Grotendorst GR, Miller DM, Sporn MB. Cellular transformation by coordinate action of three peptide growth factors from human platelets. Nature.
1984; 309: 804-806.
36. Hwang DL, Latus LJ, Lev-Ran A. Effects of platelet-contained growth factors (PDGF, EGF, IGF-1, and TGF-~) on DNA synthesis in porcine aortic smooth muscle cells in culture. Exp. Cell Res.1992; 200: 358-360.
37. Burgess TL, Fisher EF, Ross SL, Bready JV, Qian Y-X, Bayewitch LA, Cohen AM, Herrara CJ, Hu SS-F, Kramer TB, Lott FD, Martin FH, Pierce GF, Simonet L, Farrell CL. The antiproliferative activity of c-myb and c-myc antisense oligonucleotides in smooth muscle cells is caused by a nonantisense mechanism. Proc. Natl. Acad. USA. 1995; 92: 4051-4055.
38. Guvakova MA, Yakubov LA, Vlodavsky 1, Tonkinson JL, Stein CA.
Phosphorothioate oligodeoxynucleotides bind to basic fibroblast growth factor, inhibit its binding to cell surface receptors, and remove it from low affinity binding sites on extracellular matrix. J. Biol. Chem.1995; 270: 2620-2627.
39. Stein CA. Does antisense exist? Nature Med. 1995; 1: 1119-1121.
40. Wang A, Creasy A, Lardner M, Lin L, Strickler J, Van Arsdell J, Yamamoto R, Mark D. Molecular cloning of the complementary DNA for human tumor necrosis factor. Science. 1985; 228: 149-154.

41. Holt JT, Redner RL, Nienhuis AW. An oligomer complementary to c-myc mRNA
inhibits proliferation of HL-60 promyelocytic cells and induces differentiation.Mol. Cell Biol.1988; 8: 963-973.
42. Paules RS, Buccione R, Moschel RC, Vande Woude GF, Eppig JJ. Mouse mos protooncogene product is present and functions during oogenesis. Proc. Natl.
Acad. Sci. USA.1989; 86: 5395-5399.
43. Daaka Y, Wickstrom E. Target dependence of antisense oligodeoxynucleotide inhibition of c-Ha-ras p21 expression and focus formation in T24-transformed NIH3T3 cells. Oncogene Res.1990; 5: 267-275.
10 44. Liebhaber S, Russell JE, Cash F, Eshelman SS. Intramolecular duplexes in eukaryotic mRNA suppress translation in a position-dependent manner. J. Cell.
Biochem. 1991; 15D: CD007 (Abstract).
45. Speir E, Epstein SE. Inhibition of smooth muscle cell proliferation by an antisense oligodeoxynucleotide targeting the messenger RNA encoding proliferating cell nuclear antigen. Circulation. 1992; 86: 538-547.
46. Wickstrom E, Simonet W, Medlock K, Ruiz-Robles 1. Complementary oligonucleotide probe of vesicular stomatitis virus matrix protein mRNA
translation. Biophys. J.1986; 49: 15-17.
47. Kozak M. Influences of mRNA secondary structure on initiation by eucaryotic ribosomes. Proc. Natl. Acad. Sci USA.1988; 85: 2850-2854.
48. Jaroszewski JW, Kaplan O, Syi JL, Sehested M, Faustino PJ, Cohen JS.
Concerning antisense inhibition of the multidrug resistance gene. Cancer Commun.1990; 2: 287-294.

Claims (38)

What is claimed is:

1. A method of inhibiting translation or transcription of a target nucleic acid sequence encoding a protein involved in smooth muscle migration and/or proliferation in vivo, the method comprising: directly depositing onto a surface or within the vascular smooth muscle within the body of a mammal suffering of vascular injury at least one oligonucleotide complementary to the target sequence, in an amount sufficient topenetrate cells of the tissue, to hybridize with said target nucleic acid, and to inhibit intracellular translation or transcription of said target sequence, said protein being platelet-derived growth factor .beta.-receptor subunit (PDGFR-.beta.).

2. The method of claim 1 which results in prevention of restenosis.

3. The method of claim 1 wherein the oligonucleotide is in a physiologically compatible solution and wherein it is applied by injection.

4. The method of claim 1 wherein the solution is applied to the tissue using an infusion pump, stent, or catheter.

5. The method of claim 1 wherein the oligonucleotide further comprises an antisense sequence complementary to the sequence of a gene selected from the group consisting of c-myb, NMMHC and PCNA.

6. The method of claim 1 wherein said oligonucleotide sequence comprises about 14 to 38 nucleotide bases.

7. The method of claim 1 wherein the oligonucleotide is treated to render it resistant to degradation or extension by intracellular enzymes.

8. The method of claim 7 wherein the treatment comprises substituting at least one backbone phosphodiester linkage of the oligonucleotide with a linkage selected from the group consisting of phosphorothioate, methylphosphonate, sulfone, sulfate, ketyl, phosphorodithioate, various phosphoramidate, phosphate ester, bridged phosphorothioate and bridged phosphoramidate linkages.

9. The method of claim 7 wherein the treatment comprises capping a 3'-nucleotidewith a structure resistant to addition of nucleotides.

10. The method of claim 1 wherein the oligonucleotide is delivered to the tissue in a concentration of between approximately 30 and 3000 µg oligonucleotide per square centimeter of tissue surface area.

11. The method of claim 1 wherein the target nucleic acid sequence comprises a mRNA.

12. The method of claim 11 wherein said oligonucleotide sequence inhibits translation of the mRNA.

13. The method of claim 1 wherein the oligonucleotide is incorporated into a carrier.

14. The method of claim 1 wherein the carrier comprises an implantable matrix.

15. The method of claim 1 wherein the carrier comprises a hydrogel.

16. The method of claim 1 wherein the hydrogen comprises a material which is liquid at a temperature below 37°C.

17. The method of claim 16 wherein the hydrogel material comprises a polyoxyethylene oxide polypropylene oxide copolymer.

18. The method of claim 17 wherein the polymer comprises from about 10 to about 80% by weight polyethylene oxide and from about 20 to about 90% polyethylene oxide.

19. The method of claim 18 wherein the polymer comprises about 70% by weight polyethylene oxide and about 30% by weight polypropylene oxide.

20. The method of claim 1 wherein the oligonucleotide is deposited extravascularly.

21. The method of claim 1 wherein said oligonucleotide is deposited onto or beneath an adventitial surface of the vascular system.

22. An oligonucleotide adapted for cell penetration and resistant to intracellular degradation comprising nucleotide bases complementary to mRNA encoding the protein PDGFR-.beta..

23. The oligonucleotide of claim 22 wherein said oligonucleotide comprises about14 to 38 nucleotide bases.

24. The oligonucleotide claim 22 wherein said oligonucleotide inhibits translation of the target sequence.

25. The oligonucleotide of claim 22 having the nucleotide sequence defined in SEQ.
ID. No: 1 or 2.

26. A composition of oligonucleotides comprising at least one oligonucleotide asdefined in any one of claims 22 to 25 and at least one oligonucleotide complementary to mRNA encoding the protein NMMHC, c-myb or PCNA.

27. The composition of claim 26 wherein said least one oligonucleotide complementary to the mRNA encoding the protein NMMHC, c-myb or PCNA has the nucleotide sequence selected from SEQ. ID. No. 5, 6 and 7.

28. An implant for local sustained delivery of antisense oligonucleotides comprising a carrier containing antisense oligonucleotides as defined in any one of claims 22 to 27.

29. The implant of claim 28 wherein the carrier comprises a polymer.

30. The implant of claim 29 wherein the polymer comprises a hydrogel material which is liquid at temperatures below about 37°C and gels at body temperature.

31. The implant of claim 30 wherein the hydrogel comprises a polyethylene oxide - polypropylene oxide copolymer.

32. The implant of claim 29 wherein the oligonucleotide is present at a concentration in said polymer sufficient to release into a surface on which it is deposited from about 30 to about 3000 µg oligonucleotide per square centimeter of tissue surface area.

33. The implant of claim 29 wherein the polymer is selected from the group consisting of polyanhydrides, polyorthoesters, polylactic acid, polyglycolic acid, poly (lactic acid-glycolic acid) copolymers, collagen, dextran, protein polymers, ethylenevinyl acetate, polyvinyl acetate, polyvinyl alcohol, polymers of propylene glycol and etlhylene oxide, and derivatives thereof.

34. The implant of Claim 28 which can prevent restenosis.

35. The implant of claim 28 which can prevent thrombosis.

36. The implant of claim 28 which can prevent development of a pulmonary disorder.

37. The implant of claim 28 which can prevent development of a cardiovascular disorder.

38. Use of an oligonucleotide as defined in any one of claims 22 to 27 for inhibiting translation or transcription of the target nucleotide sequence in the smooth muscle tissue.