WO2000046384A1 - Nucleic acid uptake and release vehicle - Google Patents

Abstract

A nucleic acid uptake and release vehicle comprising a ligand for a receptor covalently bonded by an amide linkage to an influenza virus hemagglutinin endosomolytic peptide is disclosed, which optionnaly may also comprise a polylysylleucyl peptide so as to provide additional sites for nucleic acid attachment; as well as a nuclear localization signal peptide so as to enhance intranuclear localization of delivered nucleic acid. The vehicle can be utilized to enhance the intracellular delivery of nucleic acid molecules.

Description

NUCLEIC ACID UPTAKE AND RELEASE VEHICLE

CROSS-REFERENCE TO RELATED APPLICATION

This is a continuation- in-part of Application No. 09/008,390, filed January 16, 1998, which is a continuation of Application No. 60/054947, filed August 7, 1997.

FIELD OF THE INVENTION The present invention relates to a nucleic acid uptake and release vehicle comprising a fusion protein wherein a ligand and a hemagglutinin endosomolytic peptide are covalently bonded by an amide linkage to provide a vehicle which can be utilized to enhance the intracellular delivery of moieties consisting of nucleic acids oligonucleotides or modified nucleic acid or oligonucleotides.

A preferred embodiment comprises an adenovirus penton base protein; and an influenza virus hemagglutinin endosomolytic peptide covalently bonded by an amide linkage that can be utilized to enhance the intracellular delivery of nucleic acid molecules, e.g., DNA vectors encoding sense or anti- sense oligonucleotides, as well as a DNA construct encoding the same, and a method of use of the same for intracellular delivery of nucleic acid molecules. Other suitable preferred ligands include epidermal growth factor. In a preferred embodiment, the vehicle can also contain a polylysylleucyl peptide so as to provide additional nucleic acid attachment sites; as well as a nuclear localization signal peptide so as to enhance intranuclear localization of the nucleic acid.

The invention may be modified by the incorporation of amino acids or peptides to provide additional sites of nucleic acid attachment and to impart flexibility to the secondary structure

The invention may be further modified by the incorporation of a nuclear localization signal.

BACKGROUND OF THE INVENTION

Genes have been used to treat certain diseases. In addition, anti-sense oligonucleotides and modified oligonucleotides have been used to treat diseases, such as viral infections, cancer, and genetic defects, so as to inhibit genes that are known to be associated with these diseases. However, heretofore it has been difficult to introduce nucleic acid molecules into cells at the levels required for such therapy. Therefore, there has been a desire in the art for the development of effective delivery systems which allow for intracellular bioavailability of nucleic acid molecules both ex vivo and in vivo.

SUMMARY OF THE INVENTION An object of the present invention is to provide a nucleic acid uptake and release vehicle. Another object of the present invention is to provide a DNA construct encoding a nucleic acid uptake and release vehicle.

Still another object of the invention is to provide a method of intracellular delivery of therapeutic nucleic acid molecules using the nucleic acid uptake and release vehicle.

These and other objects of the present invention, which will be apparent from the detailed description of the invention provided hereinafter, have been met in one embodiment, by a nucleic acid uptake and release vehicle (hereinafter "UTARVE") comprising a fusion protein wherein a ligand for a receptor and a hemagglutinin endosomolytic peptide are covalently bonded by an amide linkage to provide a vehicle which can be utilized to enhance the intracellular delivery of moieties consisting of nucleic acids oligonucleotides or modified nucleic acid or oligonucleotides.

In a further embodiment, UTARVE may be modified by the incorporation of amino acids or peptides to provide additional sites of nucleic acid attachment and to impart flexibility to the secondary structure

In yet another embodiment UTARVE may also be modified by the incorporation of a nuclear localization signal.

These and other objects of the present invention, which will be apparent from the detailed description of the invention provided hereinafter, have been met in the following most preferred embodiment, by a nucleic acid uptake and release vehicle (hereinafter "UTARVE") comprising:

(i) an adenovirus penton base protein covalently bonded by an amide linkage to

(ii) an influenza virus hemagglutinin endosomolytic peptide (Gly_n-HA-Gly_n). In a preferred embodiment, the UTARVE also comprises:

(iii) a polylysylleucyl (KL)_m peptide covalently bonded to the vehicle by an amide linkage. In still a more preferred embodiment, the UTARVE also comprises:

(iv) a nuclear localization signal peptide covalently bonded to the vehicle by an amide linkage.

In another embodiment, the above-described objects of the present invention have been met by a DNA construct encoding UTARVE. In still another embodiment, the above-described objects of the present invention have been met by a method of intracellular delivery of therapeutic nucleic acid molecules comprising contacting cells with a UTARVE-nucleic acid complex.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows HSV- 1 growth inhibition in Vero cells, where the Vero cells were exposed to the UTARVE- 1E4.5SA complex (•) or the UTARVE-IE1TI complex (o), and infected with HSV-1. UTARVE-complexes in which the two central nucleotides were inverted 1 E4,5 SAmul (Δ) and IE1TI mul (A), served as controls. Figure 2 shows HSV- 1 growth inhibition in Vero cells, where the Vero cells were exposed to the

UTARVE- 1E4,5SA complex (open bar) or uncomplexed 1E4,5SA (solid bar), and infected with HSV- 1. DETAILED DESCRIPTION OF THE INVENTION The invention comprises a ligand for a receptor covalently bonded by an amide linkage to an influenza virus hamaggluntin endosomolytic peptide.

As discussed above, in one embodiment, the present invention relates to a UTARVE comprising: (i) an adenovirus (Ad) penton base protein covalently bonded by an amide linkage to

(ii) an influenza virus hemagglutinin endosomolytic peptide (Gly_n-HA-Gly_n). The UTARVE targets the cell surface, and causes endosomolysis and increases membrane permeability for the nucleic acid molecules.

The Ad penton base protein contains a receptor binding site motif (RGD) for attachment to the ubiquitous cell receptors - integrins (Bai et al, J. Virol, 67:5198-5205 (1993).

As used herein, the expression "penton base protein" refers to the entire Ad penton base protein or fragments thereof which include at least amino acids 1-354 which contain the receptor binding motif. The particular adenovirus from which the Ad penton base protein sequence is derived is not critical to the present invention. Examples of such adenoviruses include Ad2, Ad5 and Ad3. The amino acid/nucleic acid sequences encoding Ad penton base proteins are well-known in the art. For example, the amino acid/nucleic acid sequence of the Ad2 penton base protein is described by Roberts et al ( J. Biol. Chem., 259:13968-13975 (1984); the amino acid/nucleic acid sequence of the Ad5 penton base protein is described by Newmann et al, Gene, 69:153-157 (1988); and the amino acid/nucleic acid sequence of the Ad3 penton base protein is described by Cuzange et al, Gene, 146:257- 259 (1994).

Other suitable ligands or base proteins include transferrin, cholera toxin B subunit, and epidermal growth factor.

The influenza HA endosomolytic peptide provides the endosomolytic function necessary for intracellular release of the UTARVE after its receptor mediated endocytosis (Wagner et al, Proc. Natl. Acad. ScL, USA, 89:7934-7938, 1992). As used herein, the expression "influenza HA endosomolytic peptide" refers to the wild-type sequence, as well as mutants thereof which retain the endosomolytic function, e.g., a mutant wherein Gly⁴ is mutated to Glu⁴; and a mutant with, e.g., 2-5 additional amino acids at the amino or carboxy terminus.

The amino acid sequence encoding the wild-type HA endosomolytic peptide, as well as the mutant thereof wherein Gly⁴ is mutated to Glu⁴, are well-known in the art (Wharton et al, J. Gen. Virol,

69:1847-1857, 1988); Wagner et al, Proc. Natl. Acad. ScL, USA, 89:7934-7938, 1992; and Plank et al, J. Biol. Chem., 269:12918-12924, 1994).

The glycine (Gly) residues flanking the HA endosomolytic peptide permit formation of α-helices, and, e.g., impart flexibility to the secondary structure at the HA-penton base protein junction. The number of Gly residues flanking the endosomolytic HA peptide, i.e., the value of n, is not critical to the present invention, but generally ranges from between 0 to 12, preferably 1 to 8. As discussed above, in a preferred embodiment, the UTARVE also comprises:

(iii) a polylysylleucyl (KL)_m peptide.

The (KL)_m peptide provides additional sites for complexing with the nucleic acid molecule. In (KL)_m, the K residues (Lys) interact with the nucleic acid, while the L residues (Leu) decrease potential stearic hindrance resulting from adjacent charged Lys residues. In the (KL)_ra peptide, the value of m is not critical to the present invention, but generally represents from 1 to 300 alternating lysine (K) and leucine (L) residues, preferably from 3 to 100 alternating lysine (K) and leucine (L) residues. (KL)_m can be used as the basis for generating tandem repeats of (KL)_m in an intermediate vector prior to, e.g., insertion adjacent to the penton base protein coding sequence, e.g., the Ad2 penton base protein coding sequence in pETllaHA/PB described in Example 1 below.

In still a more preferred embodiment, the UTARVE also comprises:

(iv) a nuclear localization signal peptide.

The particular nuclear localization signal (NLS) peptide employed is not critical to the present invention. Examples of such NLS peptides include the NLS peptide of the SV40 large T antigen (Van Dermonne et al, TIBS, 21:59-64, 1996);and those described by Boulikas, Critical Reviews in Eukaryotic

Gene Expression, 3:193-227, 1993), the NLS peptides listed in said references are incorporated by reference herein in their entirety. NLS peptide coding sequences can be introduced into DNA encoding a UTARVE by, for example, preparing a dsDNA encoding the NLS peptide of SV40 large T antigen using the following sense and anti-sense primers, respectively. 5'-AGATCTCATCGGACGACG-3' (SEQ ID NO: 1 ); and

5 '-TGATCAGAGGATCCTCGAC-3' (SEQ ID NO:2), so as to PCR amplify a single-stranded 87 base DNA molecule which encodes for the SV40 large T antigen NLS amino acid sequence:

Ser Ser Asp Asp Glu Ala Thr Ala Ser Asp Gin His Ser Thr Pro Pro Lys Lys Lys Arg Lys Val Glu Asp Pro (SEQ ID NO:3), flanked by sequences for restriction endonuclease cleavage by Bgll (italic) and Bell (underlined).

5' --4G-47XTTCATCGGACGACGAAGCTACCGCCGACAGTCAACATTCAACGC

CTCCC AAGAAGAAGCGCAAGGTCGAGGATCCTl CTGATCA-3 ' (SEQ ID NO:4).

This sequence includes the phosphorylation site for casein kinase II (CKII) which is required for optimal NLS function (Gams et al, Oncogene, 9:2961-2968, 1994).

The double-stranded 87 nucleotide DNA molecule can be inserted into a plasmid encoding (KL)_m, i.e., p(KL)_m, e.g., p(KL)₁₀. In particular, such a sequence can be inserted downstream of KL at the Bell site or upstream at the Bglϊl site or both. This results in one or two NLS sequences per UTARVE molecule. The orientation of the elements in UTARVE is not critical to the present invention. For example,

UTARVE can contain the elements covalently linked via peptide bonds in the following N-terminal to C-terminal (5' to 3' in the DNA) orientation shown in Table 1 below:

TABLE 1

Ad penton base protein: (Gly_n-HA-Gly_n) peptide; Gly_n-HA-Gly_n peptide:Ad penton base protein;

Ad penton base protein: Gly„-HA-Gly_n peptide :(KL)_m peptide;

Ad penton base protein: (KL)_m peptide: Gly_n-HA-Gly_n peptide;

Gly_n-HA-Gly_n peptide: Ad penton base protein: (KL)_m peptide;

Gly_n-HA-Gly_n peptide: (KL)_m peptide: Ad penton base protein; (KL)_m peptide: Ad penton base protein Gly _n-HA-Gly_n peptide;

(KL)_m peptide :Glyn-HA-Gly_n peptide: Ad penton base protein

Ad penton base protein: Gly„-HA-Gly_n peptide: NLS peptide;

Ad penton base protein: NLS peptide: Gly_n-HA-Gly_π;

Gly_n-HA-Gly_n peptide: Ad penton base protein: NLS peptide; Gly _n-HA-Gly_n peptide: NLS peptide: Ad penton base protein;

NLS peptide: Ad penton base protein: Gly_n-HA-Gly_n peptide;

NLS peptide :Gly_n-HA-Gly_n peptide: Ad penton base protein;

Ad penton base protein: Gly_n-HA-Gly_n peptide:(KL)_m peptide: NLS peptide;

Ad penton base protein: Gly_n-HA-Gly_n peptide: NLS peptide: (KL)_m peptide; Ad penton base protein: NLS peptide: Gly_n-HA-Gly_n peptide: (KL)_m peptide;

Ad penton base protein: (KL)_m peptide: Gly_n-HA-Gly_n peptide: NLS peptide;

Ad penton base protein: (KL)_m peptide: NLS peptide: Gly_n-HA-Gly_n peptide;

Ad penton base protein: NLS peptide:(KL)_m peptide: Gly_n-HA-Gly_n peptide;

Gly_n-HA-Gly_n peptide: Ad penton base protein: (KL)_m peptide: NLS peptide; Gly_n-HA-Gly_n peptide: Ad penton base protein: NLS peptide: (KL)_m peptide;

Gly_n-HA-Gly_n peptide: NLS peptide: Ad penton base protein: (KL)_m peptide;

Gly_n-HA-Gly_n peptide: (KL)_m peptide: Ad penton base protein: NLS peptide;

Gly_n-HA-Gly_n peptide:(KL)_m peptide: NLS peptide: Ad penton base protein;

Gly_n-HA-Gly_n peptide: NLS peptide :(KL)_m peptide: Ad penton base protein; (KL)_m peptide: Ad penton base protein: Gly_n-HA-Gly_n peptide: NLS peptide

(KL)_m peptide: Ad penton base protein: NLS peptide: Gly_n-HA-Gly_n peptide;

(KL)_m peptide: NLS peptide: Ad penton base peptide: Gly_n-HA-Gly„ peptide;

(KL)_m peptide: Gly_n-HA-Gly_n peptide: Ad penton base protein: NLS peptide;

(KL)_m peptide: Gly_n-HA-Gly_n peptide: NLS peptide : Ad penton base protein; (KL)_m peptide: NLS peptide: Gly_n-HA-Gly_n peptide: Ad penton base protein;

NLS peptide: Ad penton base protein: Gly_n-HA-Gly_n peptide: (KL)_m peptide; NLS peptide: Ad penton base protein: (KL)_m peptide: Gly_n -HA-Gly_n peptide; NLS peptide: Gly_n-HA-Gly_n peptide: Ad penton base protein: (KL)_m peptide; NLS peptide: Gly_n-HA-Gly_π peptide:(KL)_m peptide: Ad penton base peptide; NLS peptide:(KL)_m peptide: Ad penton base protein: Gly_n-HA-Gly_n peptide; and NLS peptide: (KL)_m peptide: Gly_n-HA-Gly_n peptide: Ad penton base protein.

The mechanism of action of UTARVE involves increased intracellular delivery (uptake) of the nucleic acid molecules, and increased release of the nucleic acid molecules from endocytic-like vesicles in which they are entrapped during intracellular delivery. Thus, the UTARVE is useful for enhancing the intracellular delivery of nucleic acid molecules, such as single or double stranded anti-sense or sense oligonucleotides for inhibition of gene expression, or single or double stranded sense oligonucleotides for gene delivery in gene therapy.

The particular nucleic acid molecule is not critical to the present invention. Examples of such include eukaryotic or viral gene expression vectors under complete, partial or no eukaryotic or non- eukaryotic regulatory control, which express RNA transcripts that are anti- sense to viral or cellular RNA transcripts. Such anti-sense oligonucleotides are useful to inhibit, reduce or alter expression from the viral or cellular transcripts associated with or critical to the infection/disease state and/or process. For example, pMK53, which expresses anti-sense HSV-2 ICP10 mRNA under the control of a viral gene promoter, such as the Cytomegalovirus immediate early (CMV, IE) promoter, has been used to inhibit ICP10 expression and induce apoptosis in cells which express both HSV-2 ICP10, as well as the cellular homolog of ICP10 (Smith et al, Virus Genes, 5:215-226, 1991).

The particular anti-sense oligonucleotide employed is not critical to the present invention. Examples of such anti-sense oligonucleotides include:

(1) Antisense AKT₂, which has been used to inhibit AKT₂ expression and tumorigenicity in PANCI cells (Chang et al, Proc. Acad. Natl. ScL, USA, 93:3636-3641, 1996);

(2) Antisense p53, which has been used to inhibit p53 expression and induce phenotypic changes(Mukhopadhyay et al, Anticancer Res., 16:1683-1689, 1996); and

(3) Antisense IE45SA, IE1T1, IE3T1, IE4T1, which have been used to inhibit expression of HSV-1 immediate early genes IE4, IE1, IE3 and IE4, respectively, as well as HSV-1 growth (Kulka et al, Proc. Natl. Acad. ScL, USA, 86:6868-6872, 1989; and Kulka et al,

Antimicrob. Agents Chemother., 38:675-680, 1994).

Additional examples of nucleic acid molecules which can be employed in the present invention include eukaryotic or viral gene expression vectors under complete, partial or no eukaryotic or non- eukaryotic regulatory control, which express enzymatic or non-enzymatic, regulatory, structural, immunogenic, co-factorial and/or mimetic DNA, RNA, proteins and/or peptides. These sense oligonucleotides are useful in order to inhibit, reduce or alter protein-protein interactions or protein- nucleic acid interactions or nucleic acid-nucleic acid interactions, or expression from the viral or cellular transcripts or gene promoters associated with or critical to the infection/disease state and/or process.

The particular sense oligonucleotide employed is not critical to the present invention. Examples of such sense oligonucleotides include: (1) Sense tissue factor, which has been used to overexpress tissue factor in order to determine the role of angiogenesis in tumor growth (Zhang, J. Clin. Invest., 92: 1320- 1327, 1994); and (2) Sense HIV- 1 protease, which has been used in cultured cells to induce cleavage of BCL- 2, and thereby cause apoptosis (Stack, Proc. Natl. Acad. ScL, USA, 93:9571-9576, 1996).

In addition, eukaryotic or viral gene expression vectors under complete, partial or no eukaryotic or non-eukaryotic regulatory control, which express catalytic RNAs or ribozymes which cleave covalent bonds in target RNA are useful in order to inhibit, reduce or alter expression from the viral or cellular transcripts associated with or critical to the infection/disease state and/or process, can be employed in the present invention as the nucleic acid molecule. Examples of such include:

(1 ) Anti-HPVl 8 ribozyme, which has been used to alter HPV18 mRNA expression through cleavage (of HPV mRNA), and thereby alter the growth rates and properties of selected cell lines (Chen et al, Cancer Gene Ther., 3:18-23, 1996); and

(2) Anti-ras ribozyme, which has been used to cleave the activated H-ras oncogene and thereby alter the malignant phenotype of an invasive human bladder cancer cell line

(Eastham et al, J.Urol., 156:1186-1188, 1996). In a preferred embodiment, the UTARVE-nucleic acid complexes are used as anti-viral agents, e.g., for the treatment of herpesvirus infections, as well as other human virus infections which can cause disease including the following virus families: Picornaviridae, e.g., Coxsackie virus; Togaviridae, e.g., Eastern equine encephalitis virus; Flaviviridae, e.g., St. Louis encephalitis virus; Coronaviridae, e.g.,

Coronavirus; Rhabdoviridae, e.g., Rabies virus; Filoviridae, e.g., Ebola virus; Paramyxoviridae, e.g., Respiratory syncytial virus; orthomyxoviridae, e.g., Influenza virus; Bunyaviridae, e.g., Rift valley fever virus; Arenaviridae, e.g., Lassa fever virus; Reoviridae, e.g., Human rotavirus; Calciviridae, e.g.,Norwalk virus; Retroviridae, e.g., HIV; Hepadnaviridae, e.g., Hepatitis B virus; Parvoviridae, e.g., Human parvovirusB-19; Papovaviridae, e.g., Human papillomavirus; Adenoviridae, e.g., Human adenovirus; and

Poxviridae, e.g., Vaccinia virus.

In another preferred embodiment, the UTARVE complexes are used as anti-viral agents for the treatment of animal virus infections that can cause disease including the following virus families: Picornaviridae, e.g., Foot-and-mouth disease virus; Togaviridae, e.g., Eastern equine encephalitis virus; Flaviviridae, e.g., Tick-borne encephalitis virus; Coronaviridae, e.g., Feline infectious peritonitis virus;

Rhabdoviridae, e.g., Rabies virus; Filoviridae, e.g., Ebola virus; Paramyxoviridae, e.g., Canine distemper virus; orthomyxoviridae, e.g., Influenza viruses swine, horse and fowl; Bunyaviridae, e.g., Rift valley fever virus; Arenaviridae, e.g., LCM virus; Reoviridae, e.g., African horse sickness virus; Calciviridae, e.g.., Feline calciviruses; Retroviridae, e.g., Avian sarcoma and leukosis viruses; Hepadnaviridae, e.g., Hepatitis Belike virus; Parvoviridae, e.g., Canine parvovirus B-19; Papovaviridae, e.g., Bovine papillomavirus; Adenoviridae, e.g., Equine adenovirus; and Poxviridae, e.g., Cowpox virus, Animal herpes viruses, e.g., equine.

In still another preferred embodiment, the UTARVE complexes can be used in the treatment of genetic diseases, such as hemophilia, adenosine deaminase deficiency, beta-thalassemia, diabetes and cystic fibrosis. Additionally, the UTARVE complexes can be used in the treatment of cancers, such as breast cancer, cervical cancer, lung cancer, bladder cancer, prostate cancer and acute lymphocytic leukemia (ALL) for which potentially involved genes have been identified.

The UTARVE complexes may also contain oligonucleotides, DNA, or expression vectors designed to inhibit, complement or replace the defective, aberrantly regulated and/or dysfunctional gene(s) associated with or critical to the targeted genetic disorder, or to the development and/or maintenance of the neoplastic state.

The interaction of the UTARVE complexes with non-integrin cell receptors, the endosomolysis following UTARVE-complex uptake and the efficient release of nucleic acid molecules which are covalently bound to the lysine residues in the Ad penton base protein or in the (KL)_m peptide, represent important issues/mechanisms relevant to the successful delivery of nucleic acid molecules to target cell/tissues. UTARVE design and versatility enables variation of its construction in order to address these important issues. This includes:

(i) increasing the flexibility of the molecule by varying the number of Gly residues in the

Gly_n-HA-Gly_n hinge region of the UTARVE in order to affect changes in its endosomolytic activity,

(ii) cross-linking of specific ligands to the UTARVE-nucleic acid complex in order to direct complex binding to a specific target (cell) receptor, and (iii) using various methods of nucleic acid complexation to UTARVE in order to achieve optimal intracellular release of the nucleic acid molecule from the UTARVE-nucleic acid complex.

Thus, in another embodiment, protein ligands or sugars or other moieties which can bind specifically to cell surface receptors may be cross-linked to UTARVE so as to alter/increase the binding specificity and uptake of the UTARVE-nucleic acid complex with the target cell/tissue. Cross-linking can be effected by any one of a variety of commercially available cross-linking/derivatizing agents and procedures, e.g.:

( 1 ) BMH, which is reactive with sulfhydryl groups (Chen et al, J.Biol Chem., 266: 18237- 18293, 1991);

(2) SPDP, which is reactive with amino and sulfhydryl groups (Carlsson et al, Biochein. J., 173:223-737, 1978); and

(3) MPBH, which is reactive with carbohydrate and sulfhydryl groups (Chamow et al, J. Biol. Chem.. 267:15916-15922. 1992).

The particular protein ligand employed is not critical to the present invention. Examples of such protein ligands include transferrin, cholera toxin B subunit, penton base protein, and epidermal growth factor. The particular sugar employed is not critical to the present invention. Examples of such sugars include complex sugars containing mannose and/or galactose, and their derivatives. Complexation of nucleic acid molecules to UTARVEs is accomplished so that:

(i) levels of intracellular UTARVE and nucleic acid molecules is optimized/maximized,

(ii) release of the nucleic acid molecules after intracellular penetration is optimized, (iii) the effect on the targeted gene in the absence of cellular toxicity is maximized, and (iv) biological activity resulting from alteration of the targeted gene is maximized. The complex can also be obtained by ionic bonding (for DNA vectors) or covalent amide bonding (for oligonucleotides) to the lysine residues in the (KL)_m peptide or in the Ad penton base protein in the UTARVE. For example, the nucleic acid molecule of interest is reacted with purified UTARVE in PBS over a range of molarity calculated to ionically bind, i.e., neutralize from 10%-50% of the negative charges on the nucleic acid molecule. A quantity of poly-L-lysine sufficient to neutralize the remainder of the charges can be added to achieve electroneutrality of the input nucleic acid molecule.

This technique is based on the procedure described by Wagner et al, Proc. Natl. Acad. Sci, USA, 89:6099-6103 (1992).

Also, the oligonucleotides can be derivatized such that they have a thiol (SH) group positioned on the alkyl side chain located on the 5' terminal nucleotide, and then conjugated to NH₂groups of lysine in the (KL)_m peptide or in the Ad penton base protein as described by Orgel et al, Nucl Acids Res.,

16:3671 (1988); and King et al, Biochem., 17:1499-1506 (1978), resulting in the attachment of the oligonucleotides to the UTARVE through disulfide, -S-S-, bridges. In this case, the primary means for intracellular release of oligonucleotides from the UTARVE is provided by the phosphodiester bond located between the nucleotides of the oligonucleotide that are susceptible to hydrolysis by endonucleases inside of the cell (Levis, J, Ph.D. Dissertation, The Johns Hopkins University, 1995).

Cleavage of the disulfide bond occurs after internalization and acidification of the UTARVE-S-S- oligonucleotide containing endosomal vesicle, thereby providing an alternative means for intracellular release of the oligonucleotide from the carrier.

In cases where the -S-S- bridge is unstable under culture and/or ligand linkage conditions, alternative methods for complexing oligonucleotides to UTARVE can be used, e.g., linkage chemistry based on an oligonucleotide derivative that has a carboxyl group attached to an alkyl linker arm on the 5' terminal nucleotide. The carboxyl moiety can be covalently attached to the NH₂ groups of lysine as described by Salter et al, FEBS Letters, 20:302-306 (1972). In this scheme, the conjugation of oligonucleotides to lysine residues requires the use of the activating agent EDAC [l-ethyl-3- (3- dimethylaminopropyl) carbodiimide] that reacts with the carboxyl group on the oligonucleotide derivative to form an urea-type intermediate which is then susceptible to nucleophilic attack by the NH₂ groups of pL. The end product of this conjugation reaction is the attachment of the oligonucleotide to UTARVE through an amide, -NH-CO-, i.e., peptide, linkage. This amide bond is not susceptible to cleavage under conditions that might reduce the -S-S- bond. Intracellular release of the oligonucleotide will rely on the cleavage of phosphodiester bonds in the nucleic acid molecule. Another conjugation method that can be employed involves making use of aminoxy functional groups positioned on the 5' terminal nucleotide of the oligonucleotide. This involves reaction of the NH2 groups of lysine with glyoxylic acid or the n-hydroxyphthalimide ester of glyoxylic acid to derivatize the lysine. The derivatized moieties are then conjugated to the oligonucleotide derivative through formation of an oxime bond, as described by Rose et al, J.Am. Chem. Soc, 116:30-33 (1994). The oxime linkage is stable over the pH range of 2-7, and the pH in the endosomal vesicles is in the acidic range. Thus, release of the oligonucleotide from the complex will rely on the cleavage of the phosphodiester bond located between the first and second nucleotide of the oligonucleotide (Levis, supra).

To inhibit, reduce or alter expression of the viral or cellular nucleic acids or proteins associated with or critical to the infection/disease state and/or process, the native backbone, phosphodiester, of the oligonucleotides can be replaced with a synthetic backbone, e.g., methylphosphonate, phosphorothioate, phosphodithioate, phosphoramidate; a mixed backbone, e.g., native and/or synthetic; oligoribonucleotides and their 2' modified derivatives, e.g., 2'-0-methylriboside phosphodiesters, 2 '-O-methylriboside methylphosphonates, or alternate 2 '-O-methylriboside methylphosphonates/phosphodiesters (alt-mr- OMP) which are 5', 3' and/or internally derivatized to contain a photofluor, e.g., BODIPY, and/or any cross-linking moieties, e.g., psoralen; and/or combinations of all of the above, which are known to function as viral or cellular anti-gene (target DNA), anti-sense (targets RNA) or aptamer (target proteins) (Stull et al, Pharmaceutical Res., 12:465-483, 1995).

Preferably, the nucleic acid molecule is modified such that the phosphodiester backbone is replaced with alt-mr-OMP or phosphorothioate backbones. In this case, the oligonucleotides are resistant to intracellular nucleases, but one phosphodiester bond remains located between the first and second nucleotide that is susceptible to hydrolysis by endonucleases inside the cell (Levis, supra)

As discussed above, in another embodiment, the above-described objects of the present invention have been met by a DNA construct encoding a UTARVE.

In still another embodiment, the above-described objects of the present invention have been met by a method of intracellular delivery of therapeutic nucleic acid molecules comprising contacting cells with a UTARVE-nucleic acid complex. Initially, the complexation and cellular delivery of nucleic acid molecules by UTARVE to target cells can be conducted using a β-galactosidase expression DNA construct (pCMVβ: CMVIE promoter/E. coli β-galactosidase gene cassette, Clontech, Palo Alto, CA), and evaluated on the basis of marker gene expression, β-galactosidase expression provides the enzymatic marker necessary to determine the efficiency of DNA transfection efficiency by UTARVE. Cells are incubated with UTARVE-β- galactosidase nucleic acid complexes under conditions which examine a range of UTARVE concentrations and times of incubation. At 24-48 hrs post-transfection, cells are screened for β- galactosidase expression by 5-bromo-4-chloro-3-indolyl-βD-galactoside (X-GAL) as described by Nielsen et al, Proc. Natl. Acad. Sci, USA, 80:5198-5202 (1983). Cell staining is observed by phase contrast light microscopy, and the transfection efficiency calculated as the percentage X-GAL stained cells.

To define the efficiency of uptake of the UTARVE-nucleic acid complexes, limiting dilutions of the complexes can be evaluated for the capacity to transfer detectable levels of fluorescence using UTARVE-nucleic acid molecule-BODIPY, or other photofluors, e.g., fluorescein. For example, logarithmic dilutions of the complexes in 2.0% (v/v) fetal calf serum/DMEM are applied to 5.0 x 10⁵ cells

(e.g., HeLa, Vero, or MRC5 cells) selected for the presence of different levels of receptors for the UTARVE on the cell surface. After 1 hr incubation at 37°C, fresh medium is added, and the cells are incubated for 1 , 2, 4, 8, 16 and 24 hrs. Cell-associated fluorescence is then quantitated at these times by FACS analysis (Nielsen et al, Proc. Natl. Acad. ScL, USA, 80:5198-5202, 1983). This will estimate the number of nucleic acid molecules that must be delivered/cell in a UTARVE-nucleic acid complex in order to detect the nucleic acid molecules in the cell.

To determine the optimum ratio of complex components, various molar ratios of complex components are evaluated for the capacity to mediate transfer of nucleic acid molecules. Intracellular localization can be determined by double immunofluorescence. Staining with fluorescein-labeled antibodies to coated vesicle proteins (viz. clathrin) can be carried out to determine if co-localization occurs. This can be carried out immediately after cell treatment, i.e., 1, 5, 10, 30, 60 mins, since endosomolysis mediated by viral proteins occurs almost upon delivery. If it occurs, fluorescence, e.g., red in cells given only UTARVE-nucleic acid complex and yellow in cells given the complex and stained with fluorescein-labeled anti-clathrin antibody, will be originally punctate, and become diffusely distributed throughout the cell at later times post infection.

Staining with fluorescein-conjugated HA antibody can also be carried out to:

(i) establish whether green fluorescence (UTARVE) is dissociated from red fluorescence

(nucleic acid molecules), and (ii) define the kinetics of dissociation. The ability of the nucleic acid molecules, delivered as a UTARVE-nucleic acid complex, to inhibit expression of the target gene, and the biological activity thereof, can be determined with unlabeled nucleic acid molecules. The results are then compared to recombinant adenovirus co-exposure in order to determine whether UTARVE is a superior delivery method. That is, cells are treated with the UTARVE-nucleic acid complex and transfected with the respective transcription unit, e.g., IE110 DNA or infected with HSV- 1. The parameters (time, complex concentration) are those which achieve optimal intracellular bioavailability. Gene expression and function alteration is determined 16-72 hrs later.

Activity is expressed as the IC₅₀and IC₉₀ of the complexed nucleic acid molecules.

The particular means of administration of the UTARVE-nucleic acid complexes to effect contacting of the cells is not critical to the present invention. For example, the UTARVE-nucleic acid complex can be administered topically either proximal and/or distal to the site of disease, to skin, mucous membranes and/or eye in the absence or presence of creams/ointments/lipid carriers, e.g., polyethylene glycol or liposomes, designed to facilitate complex uptake and/or stability at the site of topical application and/or disease.

The UTARVE-nucleic acid complex can also be administered intradermally either proximal and/or distal to the site of disease in the presence of buffered physiologic saline and/or other solutions containing (or not) lipid carriers designed to facilitate complex uptake and/or stability at the site of injection and/or disease.

The UTARVE-nucleic acid complex can also be administered subcutaneously either proximal and/or distal to the site of disease in the presence of buffered physiologic saline and/or other solutions containing (or not) lipid carriers designed to facilitate complex uptake and/or stability at the site of injection and/or disease.

The UTARVE-nucleic acid complex can also be administered intramuscularly either proximal and/or distal to the site of disease in the presence of buffered physiologic saline and/or other solutions containing (or not) lipid carriers designed to facilitate complex uptake and/or stability at the site of injection and/or disease. The UTARVE-nucleic acid complex can also be administered intravenously by injection and/or infused intravenously either proximal and/or distal to the site of disease in the presence of buffered physiologic saline and/or other solutions containing (or not) lipid carriers designed to facilitate complex uptake and/or stability at the site of injection and/or disease.

The UTARVE-nucleic acid complex can also be administered nasally or orally by inhalation and/or ingestion either proximal or distal to the site of disease either contained within (or not) biodegradable capsules in the presence of buffered physiologic saline and/or other solutions containing (or not) lipid carriers designed to facilitate complex uptake and/or stability at the tissue/organ of administration and/or disease.

The UTARVE-nucleic acid complex can also be administered ex vivo to cell cultures and/or tissues by incubation in cultures/growth media containing (or not) lipid carriers, e.g., liposomes, designed to facilitate complex uptake and/or stability at the site of injection and/or disease. The amount of UTARVE-nucleic acid complex to be administered will vary depending upon the age, weight, sex, and species of the subject (cells), as well as the disease to be treated and the nucleic acid molecule to be used. However, typically, the UTARVE-nucleic acid complex will be administered in an amount of from about 0.1 nM to 100 μM, preferably from about 0.1 nM to 10 μM. For example, the UTARVE-anti-sense oligonucleotide IE4,5SA complex described in Example 3 below, which has an alt- mr-OMP backbone, was administered in an amount of from about 0.1 to 50 nM.

The following examples are provided for illustrative purposes only and are in no way intended to limit the scope of the present invention.

EXAMPLE 1

Construction of Prototype UTARVEs

A prototype UTARVE was constructed by assembly of DNA sequences encoding: (i) the adenovirus type 2 penton base protein, which contains the RGD motif that binds the ubiquitous cell surface receptors-integrin, and is involved in receptor mediated uptake and endosomolysis, and (ii) an endosomolytic peptide derived from influenza virus HA protein. The HA peptide was used to increase the endosomolytic activity of the purified penton base protein.

More specifically, a dsDNA encoding the HA endosomolytic peptide (20 amino acids) plus 5' and 3' flanking sequences (Gly) were generated by PCR amplification of a primary single-stranded 70 nucleotide (nt) DNA sequence. The Gly residues permit formation of a-helices flanking the HA peptide

(Gly₂-HA-Gly₂), thereby imparting flexibility to the secondary structure at the HA-penton base junction. The sequence for Gly₂-HA-Gly₂was cloned into pETl laPB, a bacterial expression vector which contains the entire Ad2 penton base protein under the control of a T7 phage promoter, by blunt-end ligation in the Nhel site (mung bean nuclease treated) located upstream of the penton base protein initiator Met. The resulting plasmid, PETllaHA/PB, codes for a chimeric protein that begins with Gly₂-HA-Gly₂ (initiator

Met is encoded by vector), and is followed by and covalently bonded by an amide linkage to the penton base protein coding sequence. Positive clones were identified by sequencing.

To generate UTARVEΔHA, a dsDΝA sequence (top strand: 5'-GAGGTGGTGGTGG-3' (SEQ ID ΝO:5) encoding Gly₂-Gly₂ (no HA) was cloned into pETllaPB by blunt-end ligation in the Nhel site (mung bean nuclease treated) located upstream of the penton base protein initiator Met. The resulting plasmid pETl laΔHA/PB codes for a chimeric protein that begins with Gly₂-Gly₂ (initiator Met is encoded by vector), followed by the penton base protein coding sequence. Positive clones were identified by sequencing.

More specific details as to the construction of the above prototype UTARVE are set forth below. A. The Ad2 Penton Base Protein pETl laPB was constructed as described by Bai etal, J. Virol, 67:5198-5205 (1993). pETllaPB contains the entire Ad2 penton base protein under the control of the T7 phage promoter. The Ad2 penton base protein reading frame begins pETl 1 aPB, which also encodes a four-amino acid extension (Met-Ala- Ser-Thr) at the N-terminal of the Ad2 penton base protein, and continues with the first amino acid (Met) of the Ad2 penton base protein sequence. All of the signals for transcriptional and translational regulation of the Ad2 penton base protein gene are present within pET 1 laPB . This plasmid also contains an ampicillin resistance gene for positive (bacterial transformation) selection, as well as sequences important for its growth and replication in E. coli. only E. coli strains, such as BL21(DE3) (Studier et al, Methods En∑ymol, 185:60-89, 1990), which contain integrated copies of T7 RNA polymerase, are suitable for expression, isolation and purification of the Ad2 penton base protein after transformation with PETllaPB.

B. The HA Endosomolytic Peptide

A DNA molecule encoding the influenza HA endosomolytic peptide was first produced, and then cloned into PETllaPB.

More specifically, a dsDNA molecule was prepared which encodes the 20 amino acid HA endosomolytic peptide plus, 5' and 3' flanking sequences that provide the nucleotides suitable for cloning into pETl laPB, as well as encoding for glycine residues that flank HA (Gly₂-HA-Gly₂). The dsDNA was generated by PCR amplification of the following primary single-stranded 70 bp DNA sequence: 5' -GAGGTGGACTCTTCGAAGCAATTGCAGGTTTAATCGAAAACGGCTGGGA AGGCATGATCGACGGTGGTGG-3' (SEQ IDNO:6), using the following sense and anti-sense primers, respectively: 5'-GAGGTGGACTCTTCGAAGCA-3' (SEQ ID NO:7); and 5' -

CCACCACCGTCGATCATGCC-3' (SEQ ID NO: 8)

These primers each contain five 5' terminal nucleotides designed to encode for 2 glycines upon ligation into the Nhel site of pETllaPB.

PETllaPB was then digested with Nbel to cleave at a site located between the second and third amino acids (Ala and Ser) of the Ad2 penton base protein 4 amino acid extension (Met-Ala-Ser-Thr) in pETllaPB and blunt ended with mung bean nuclease. Next, the PCR amplified 70 bp fragment was ligated to the digested pETllaPB, to give rise to plasmid pETllaHA/PB. pETllaHA/PB encodes for a fusion protein that begins with a start Met followed by Gly₂-HA-Gly₂, a Thr residue and the Ad2 penton base protein coding sequence. E. coli strain DH5α was transformed with PETllaHA/PB, and growth selected on Amp containing culture media. Plasmid DNA was extracted from isolated colonies, and subjected to initial screening (for positive recombinants) using Munl(an isoschizomer of Mfel) restriction enzyme analysis. Plasmids containing both the Gly₂-HA-Gly₂ and Ad2 penton base protein sequence gain an additional Muni restriction site (present in the HA sequence). Positive clones were subjected to double-stranded sequencing using the PCR primers described- above to confirm in-frame ligation of Gly ₂-HA-Gly₂ sequence to the Ad2 penton base protein sequence 15 in pETllaHA/PB.

As discussed above, the number of flanking glycine residues may be varied according to the length of the primary nucleotide sequences which flank the HA encoding region. Alternatively the Gly₂- HA-Gly₂ sequence may be extended at the 5' and 3' ends by PCR amplification using sense and anti-sense primers that contain terminal sequences which encode for additional glycines. The strategy used to clone

Gly_n-HA-Gly_n sequences (where n > 2) into pETllaPB, and confirm the success of the cloning is similar to that used for Gly₂-HA-Gly₂.

Another prototype UTARVE was constructed to include 2 (KL)₁₀ units. To generate plasmids which contain 2 copies of (KL),₀, the PCR fragment generated from Seq ID 9- 1 1 were digested with Bglll and Bell and ligated into p(KL)₁₀that had been digested with Bell, which cuts only once in p(KL)₁₀. BgU and Bell digestion creates compatible cohesive ends which may be ligated to each other, but which cannot be digested with either enzyme after their ligation. This is important since recombinants which contain 2 copies of (KL)₁₀ as tandem repeats (required for in-frame expression) can be identified by restriction digestion analysis using Bglll and Bell, whereby a 146 bp band is observed after gel electrophoresis. Any other orientation or copy number (greater than two) does not produce a 146 bp band.

The strategy and methods described above to create 2 tandem copies of (KL)₁₀in pUC18 may be repeated in a sequential fashion to generate an increasingly larger number of tandem (KL)₁₀ repeats in PUC 18. The overall strategy of this method is to permit generation of (KL)₁₀repeats the polylysylleucyl codons of which are "in- frame" from one repeat to the next.

More specific details as to the construction of this prototype UTARVE are set forth below. C. The Polylsylleucyl Peptide

In order to generate additional attachment sites for nucleic acid molecules, 2 or more tandem repeats of polylysylleucyl peptides (KL)₁₀can be constructed that are suitable for cloning in-frame into either the BamHI site or an Fspl site (engineered by site-directed mutagenesis) in the Ad2 penton base protein coding sequence of PETl laHA/PB.

More specifically, a DNA molecule encoding (KL)₁₀ was prepared using the following sense and anti-sense primers, respectively:

5'-GGACGAATTCAAGATCTC-3' (SEQ ID NO:9); and 5'-CGATGAATTCTGATCAGA-3' (SEQ ID NO: 10) so as to PCR amplify a single-stranded 95 base DNA molecule:

5'-GGACGAATTC4G-47/C7€C [AAGCTT] ₁₀CTGATCAGAATTCATCG-3' (SEQIDNO:11), which encodes for 10 polylysylleucyl repeat sequences (identified in brackets) flanked by sequences for restriction endonuclease cleavage by EcoBJ (bold), Bqlll (italics) and Bell (underlined). After PCR amplification, the dsDNA was digested with EcoRl, and ligated into the EcoRl site of ϋcoRI-digested pUC18 DNA. The resulting recombinant clones were screened by restriction analysis to identify those that contain a single copy of (KL)₁₀ inserted into pUC18. This screening is based on the fact that the parental vector PUC18 does not contain restriction sites for either BgUl or Bell, and therefore confirmation of single copy insertion is accomplished by i?g/II/.9c/I-digestion of (KL)₁₀with subsequent gel electrophoresis. The presence of a 73 bp band (in addition to the vector band) indicates single copy (KL)₁₀ insertion, whereas band of 73 bp and 226 bp indicate the insertion of multiples of (KL)₁₀ inserts.

The single repeat construct was designated p(KL)₁₀.

Next, pETllaHA/PB was modified by site-directed mutagenesis to create a Fspl site at the end of the Ad2 penton base protein coding sequence (following amino acid 571). This new Fspl site allows for the cloning of the polylysylleucyl peptide encoding sequences in-frame into the HA/penton base protein truncated (BamHl site, Ad2 penton base protein amino acid 419) or full-length (Fspl site, Ad2 penton base protein amino acid 571).

More specifically, pETl laHA/PB was digested with BamHl and Hindlll, and the resulting 1.4 kb BamHllHindlll fragment, which includes the carboxyl half of the Ad2 penton base protein, was cloned directionally into M13mpl8, and subjected to site-directed mutagenesis using the following primer: 5' -AGGATGGACTATTATGCGCAAAA-3 ' (SEQ ID NO: 12)

(the mutated bases are underlined), and the Muta-gene™ M13 in vitro mutagenesis system/kit (BioRad, Hercules, CA) according to the manufacturer's instructions.

The resulting mutated 1.4 kb BamHllHindlll fragment was religated into BamHUHindlϊl-άigested pETl laHA/PB so as to replace the wild-type BamHllHindlll sequence with the mutated sequence. The resulting construct is designated pETllaHA/PBmut. As a result of this mutagenesis:

(i) an Fspl site was generated at the carboxyl-terminus of the Ad2 penton base protein coding sequence, and (ii) two novel stop codons in-frame with the Ad2 penton base protein sequence were generated immediately after the engineered Fspl site. With the exception of the 2 newly generated in-frame stop codons, all of the other signal sequences for transcription/translation initiation and termination of the HA/Ad2 penton base protein (with or without in- frame expression of polylysylleucyl sequences) remained functional and intact as described for pETllaPB.

The (KL)₁₀ repeat sequences were cloned into pETllaHA/PBmut at a position 3' adjacent to amino acid 419 (truncated) of the Ad2 penton base protein (BamHl site).

More specifically, pETllaHA/PBmut was subjected to Fspl digestion followed by BamHl digestion and electrophoretic gel purification to remove the BamHIIFspl Ad2 penton base protein fragment encoding amino acids 420 to 571.

A plasmid containing the 2 (KL)₁₀repeat sequence was digested with Bell, treated with Klenow I so as to blunt-end this site, subjected to a second digestion with Bglll and the 2 (KL)₁₀ sequences isolated, and purified by electrophoretic gel purification. The 2 (KL)₁₀ repeat encoding fragment was then cloned in-frame with the Ad2 penton base protein coding sequence by directional ligation into the Ti wHI/ESTJl-digested pETllaHA/PBmut. Directional ligation occurs because BamHl and Bglll are compatible cohesive ends and as such, ensure the correct orientation of the 2 (KL),₀ sequences upon ligation into pETl laHA/PBmut. These (KL)₁₀ repeats could also be cloned into pETl laHA/PBmut at a position 3' adjacent to amino acid 571 (full length) of the Ad2 penton base protein (Fspl site).

To clone the (KL)₁₀ sequences into the Fspl site of pETl laHA/PBmut, the plasmid containing the 2 (KL)₁₀ sequences is digested with Bglll and BcR treated with Klenow I so as to blunt-end these sites, and the (KL)₁₀ encoding fragment isolated and purified. pETl laHA/PBmut is then digested with Fspl, and ligated to the blunt-ended BglϊllBcll (KL)₁₀ encoding fragment.

(KL)₁₀ sequences cloned into PETllaHA/PBmut in an inverse (incorrect) orientation will regenerate a Bel site such that subsequent digestion with BamHl (site in Ad2 penton base protein coding sequence) and BcR will generate a 460 bp band as identified by gel electrophoresis. (KL)₁₀ sequences cloned in the correct (in-frame) orientation will not generate the 460 bp band after BamHllBcR digestion. D. Isolation and Purification of UTARVE

The UTARVEs were transformed in E. coli strain BL21 (DE3 ), which contains integrated copies of T7 RNA polymerase, and the UTARVE protein produced, isolated and purified as described by Bai et al, J. Virol, 67:5198-5205 (1993).

More specifically, expression was induced in late log phase cultures (the optical density at 600 nm was 0.8) in LB broth at 37°C by adjusting the medium to 0.4 mM isopropylthiogalactopyranoside

(IPTG). After 3 hrs shaking, the cells were collected by centrifugation, washed in STE buffer comprising

10 mM Tri-HCl (pH 8.0), 1.0 mM EDTA and 100 mM NaCl, lysed by freezing and thawing in the presence of 0.1 mg/ml lysozyme and sonicated (2 x 30 sec each) at 40% maximum output in a Branson

Sonifier. The insoluble fraction containing UTARVE was collected by centrifugation and resuspended in buffer comprising 20 mM Tris-HCl (pH 7.5), 1.0 mM EDTA and 0.1 % (w/v) Nonidet P-40. After three additional cycles of washing, the final pellet was resuspended in a small volume of 6.0 M urea, diluted with 9 volumes of buffer comprising 50 mM K₂P0₄(pH 10.7), 50 mM NaCl, 1.0 mM EDTA, and dialyzed against phosphate buffered saline (PBS) and then 50 mM phosphate buffer (pH 7.5). The resulting dialyzate was sterilized by filtration and stored at 4°C or frozen at -70°C. The UTARVE vectors are stable for at least 1 month at 4°C.

The purity of the UTARVE protein was confirmed by SDS-PAGE and Coomassie staining. Only one protein band was observed, representing the purified UTARVE. A similar band was seen for UTARVEΔHA.

EXAMPLE 2 Interaction of UTARVE and Nucleic Acid Molecules

The prototype UTARVE lacking the (KL)_m, but having the HA endosomolytic peptide obtained 18 in Example 1 was mixed with non-ionic oligonucleotides that have a deoxy-methylphosphonate backbone, and:

(i) are complementary to the translation initiation site of the immediate early (IE) gene 1

(IE ITT) of Herpes simplex virus- 1 (HSV-1); (ii) complementary to the splice acceptor junction of HSV-1 IE pre-mRNAs 4 and 5

(IE4,5SA); (iii) complementary to the translation initiation site of IE1TI in which the two central nucleotides were inverted (lElTlmul); (iv) complementary to the splice acceptor junction of HSV-1 IE pre-mRNAs 4 and 5 in which the two central nucleotides were inverted (lE4,5SAmul); or have an alternate 2'-0-methyl riboside methylphosphonate/phosphodiester backbone, and are:

(i) complementary to the splice acceptor junction of HSV-1 IE pre-mRNAs 4 and 5 (alt-mr-

IE4,5SA); or (ii) complementary to the splice acceptor junction of HSV-1 IE pre-mRNAs 4 and 5 in which two central nucleotides were inverted (alt-mr-iE4,5SA»7wl).

IE1T1 is complementary to the translation initiation site for the HSV-1 IE1 gene that codes for a major trans-activating protein designated IE110 or ICPO. IE1 TI has been shown to inhibit expression of IE110 and significantly reduce HSV- 1 growth. Its inhibitory activity for virus growth synergizes with that ofIE4,5SA (Kulka et al, Antimicrobial Agents Chemother., 38:675-680, 1993; Kulka etal, Antiviral Res., 20:115-130, 1994).

IE4,5SA has been shown to inhibit splicing of the target mRNA, as well as HSV-1 protein and

DNA synthesis, and it significantly reduces viral growth in vitro, and in infected animals (Kulka et al,

Proc. Natl.Acad.ScL, CΛS4,86:6868-6872, 1989; Kulka etal, Antimicro. Agents Chemother., 38:675-680,

1993); Kulka et aL Λwtmra/ Res., 20:115-130, 994; and Smith et al, Proc. Natl. Acad. ScL, USA, 83:2787-2791, 1986). Inhibition appears to involve faulty processing oflE mRNA 4 (IE4), or decreased synthesis of the encoded protein (IE68), since similar inhibitory effects have also been shown with oligonucleotides that target the translation initiation site of IE4, but not with oligonucleotides that target the translation initiation site of IES mRNA (Kulka et al, Antimicro. Agents Chemother., 38:675-680.

1993). lElTlmul is complementary to the translation initiation site of HSV-1 gene IE1, but has two inverted central nucleotides. It does not inhibit IE110 synthesis or HSV-1 growth (Kulka et al,

Antimicrobial Agents Chemother., 38:675-680, 1993 ; and Kulka et al, Antiviral Res. 20:115-130, 1994).

IE4,5SAmwl does not inhibit splicing of IE pre-mRNA 4,5 splicing or HSV- 1 growth (Kulka et al, Proc. Natl. Acad. ScL, USA, 86:6868-6872, 1989; Kulka et al, Antimicro. Agents Chemother., 38:675-680, 1993; and Kulka et al, Antiviral Res., 20:115-130, 1994). alt-mr-IE4,5SA has the same sequence as IE4,5SA, but the backbone consists of alternate 2 '-0- methylriboside methylphosphonate and phosphodiester groups. alt-mr-IE4,5SAλWMl has the same sequence as LE4,5SA»2wl, butthe backbone consists of alternate 2' -O-methylriboside methylphosphonate and phosphodiester groups.

The sequence of IE1TI is as follows: 5'-GCGGGGCTCCAT-3' (SEQ ID NO: 13). The sequence of IE4,5SA and alt-mr-IE4,5SA is as follows: 5'-TTCCTCCTGCGG-3' (SEQ ID

NO: 14).

The sequence of lElTlmul is as follows: 5'-GCGGGCGTCCAT-3' (SEQ ID NO: 15). The sequence of IE4,5SArøwl and alt-mr-IE4,5SAw«l is as follows: 5'-TTCCCTCTGCGG-3' (SEQ ID NO: 16). Methylphosphonate non-ionic oligonucleotides are synthesized so as to contain 3'5'-deoxy- methylphosphonate groups (d-MP) in place of all of the negatively-charged phosphodiester groups (Ts'o et al, Ann. N Y. Acad. ScL, 660:159-177, 1992), and synthesis was carried out by solid phase techniques, as described by Miller et al, In: Oligonucleotides and their analogues, Ed. Eckstein, Oxford University Press, Oxford pages 137-154, 1991). These oligonucleotides (OMPs) are known to penetrate mammalian cells in culture, are resistant to nuclease hydrolysis (Miller et al, 1991, supra), and specifically inhibit the expression of several target genes (Argis et al, Biochem., 25:6268-6275, 1986; Blake et al, -9røcΛewz., 24:6139-6145, 1985; Kulka et al, Proc. Natl. Acad. ScL, USA, 86:6868-6872, 1989; Kulka et al, Antimicro. Agents Chemother., 38:675-680, 1993; Kulka et al, Antiviral Res., 20: 115- 130, 1994; and Smith et al, Proc. Natl. Acad. ScL, USA, 83:2787-2791, 1986). Ionic alt-mr-OMP oligonucleotides were synthesized so as to contain alternating phosphodiester and 2'-0-methlyriboside methylphosphonate groups using solid phase techniques as described by Miller et al (1991), supra.

Next, the OMPs were cystamine derivatized, as described by Miller, In: Immun. Method of Enzymology, Ed. Lalley et al, 21 54-64 (1992), so as to have a thiol (SH) group positioned on the alkyl side chain located on the 5'-terminal nucleotide of the OMPs.

The resulting OMPs were conjugated as described by Orgel et al, Nucleic Acids Res., 16:3671 (1988); and King et al, Biochem., J : 1499- 1506 (1978), resulting in the attachment of the OMPs to the prototype UTARVE through disulfide, -S-S-, bridges. In this case, the primary means for intracellular release of OMPs from the prototype UTARVE is provided by the phosphodiester bond located between the first and second nucleotides of the OMP that is susceptible to hydrolysis by endonucleases inside the cell (Levis, supra).

In the second prototype UTARVE described in Example 1 , there are 2 (KL)₁₀ units. Since each

(KL)₁₀ unit contains 10 lysine residues, a total of 20 lysine residues are available for complexation to the

OMP. Complexation of the OMPs to the 2 (KL)₁₀ lysine residues to saturation yields 20 OMP per UTARVE. Based on these values, one may convert a given molar concentration of OMP to an approximation of either the number of UTARVE-OMP complexes/cell or the number of OMPs/cell within any assay. For example, for UTARVE-IEITI and UTARVE-IE4,5SA, in an assay volume of 50 μl, containing approximately 2.0 x 10⁴ cells, a 1.0 nM concentration of the OMP achieved an IC₅₀, which is equivalent to 7.5 x 10⁶ UTARVE-OMP complexes/cell or 1.5 x 10⁶ OMP molecules/cell. On the other hand, a 100 nM concentration of OMPs in the prototype complex achieved an IC₅₀, which is equivalent to approximately 7.5 x 10⁶ UTARVE-OMP complexes/cell or 1.5 x 10⁸ OMP molecules/cell.

EXAMPLE 3 Intracellular Bioavailabilitv and Activity of UTARVE Delivered Nucleic Acid Molecules Prototype UTARVEs lacking the (KL)_m,but having the HA endosomolytic peptide and complexed with IE1TI or IE4,5SA as described in Example 2 above, as well as their respective controls, lElTlmul, and IE4,5SArøwl, were analyzed for uptake, intracellular localization and activity as determined by inhibition of the target gene and virus growth. IE1TI and IE4,5SA are complementary to the translation initiation site of the IE1 gene of HSV-1 and the acceptor splice junction of the HSV-1 genes IE4 and IE5. Complexation was performed to saturate the lysine sites in the prototype UTARVE. The concentration of the complexes was estimated based on an OD₅₄₂.

More specifically, 10⁷ Vero cells were exposed to estimated concentrations of 1 - 1000 nM of the UTARVE-IEITI complex or the UTARVE-IE4,5SA complex, or their respective controls, UTARVE- lElTlmul, and UTARVE-IE4,5SAmwl, for 20 min, and then infected with 10 pfu HSV-1 (strain F) per cell. Virus titers were determined 24 hrs later by plaque assay as described by Aurelian et al, In: Clinical Virology Manual, 2nd Ed. Specter, Lancz Eds., Elsevier, N.Y., pages 73-99 (1992); and Kulka et al,

Proc. Natl. Acad. ScL, USA, 86:6868-6872, 1989). The results, which are expressed as % inhibition = virus titers in treated cells/virus titers in untreated cells x 100, are shown in Figure 1.

As shown in Figure 1, both UTARVE-IEITI (•) and UTARVE-IE4,5SA (o) complexes inhibit HSV-1 growth, as determined by plaque reduction relative to untreated cells, with IC₅₀ values of approximately 1.0 nM and IC₉₀ values of approximately 100 nM. The complexed controls UTARVE-

IE4,5SA«TO1 (Δ) and \E\Tlmu\ (A), did not inhibit virus growth. Significantly, IE1TI functions in the cytoplasm, at translation, while IE4,5SA functions in the nucleus, at splicing, (Kulka et al, Proc. Natl. Acad. ScL, USA, 86:6868-6872, 1989; Kulka et al, Antimicrob. Agents Chemother., 38:675-680, 1994). The finding that similar results are obtained with both OMPs indicates that the UTARVE complexed OMP can reach the intracellular site of its target, be it cytoplasmic or intranuclear.

In a second series of experiments, the activity of the UTARVE-IE4,5SA complex was compared to that of the uncomplexed IE4,5SA.

More specifically, 10⁷ Vero cells were exposed for 20 min to 100 nM, 10 nM and 1.0 nM of UTARVE-IE4,5SA complex or uncomplexed IE4,5SA, and infected with 10 pfu of HSV-1 (strain F) per cell. Virus titers were determined 24 hrs later. The results, which are expressed as % inhibition = virus titers in treated cells/virus titers in untreated cells x 100, are shown in Figure 2. As shown in Figure 2, the UTARVE-IE4,5SA complex (open bar) inhibited virus growth, as determined by plaque reduction relative to untreated cells, at all three concentrations. At these concentrations uncomplexed IE4,5SA (solid bar) did not inhibit virus growth. The IC₅₀ of the uncomplexed IE4,5SA was 25 μM, i.e., 2500-fold higher. These findings indicate that UTARVE complexation greatly increased the intracellular bioavailability of the OMP, as compared to the uncomplexed OMP.

In another series of experiments, IE4,5SA was bound to the photofluor BODIPY, and the resulting product was complexed to the prototype UTARVE to saturate the lysine sites as described above. 10⁷ Vero cells were then exposed to 50 μM of the resulting complex for 24 hrs, fixed with acetone and examined for intracellular fluorescence. It was compared to the fluorescence of similarly treated cells exposed to BODIPY-IE4,5SA that was not complexed to the prototype UTARVE.

The cells treated with the UTARVE-IE4,5SA-BODIPY complex showed a high level of cellular fluorescence distributed throughout the cytoplasm and the nuclei. Approximately 90% of the cells were positive. This compares to the relatively low levels of nuclear and small punctate fluorescence seen in approximately 40%> of the cells exposed to IE4,5SA-BODIPY that was not complexed to the prototype

UTARVE. A single large/dense endocytic-like vesicle evidenced good levels of fluorescence. These findings confirm the utility of UTARVE for improved intracellular delivery and bioavailability of nucleic acid molecules.

Thereafter, 10⁶ Vero cells, which have only an intermediate level of penton base protein receptors, wereexposedto 0.1-50nMalt-mr-IE4,5SA,UTARVEΔHA-alt-mr-IE4.5SA, orUTARVE-alt- mr-IE4. 5SA for 2 hrs, and infected with 5.0 pfu/cell of HSV-1 (strain F). HSV-1 titers were determined

24 hrs later by plaque assay. The IC₅₀for UTARVE-alt-mr-IE4,5SA was 1.0 nM, as compared to 100 nM for uncomplexed alt-mr-IE4,55A and 25 nM for UTARVEΔHA-alt-mr-IE4,5SA.

Next, 10⁷ Vero cells were treated for 24 hrs (saturation plateau) with 1.0 μM of [³²P]-alt-mr- OMP. Assuming that the volume of 10⁶ cells is 1.0 μl, the intracellular concentration was 1.6 μM. In cells treated for 2 hrs, intracellular concentration was 1.0 μM. In cells treated 24 hours with 1.0 μM of UTARVE-alt-mr-OMP, the intracellular concentration was 31 μM.

The data indicate that the improved antiviral activity of UTARVE-alt-mr-IE4,5SA, relative to the uncomplexed oligonucleotide, is due to increased intracellular levels when delivered as an UTARVE complex. Similar results were obtained in diploid cells (MRC5).

The above data indicate that: (i) UTARVE delivery increases intracellular oligonucleotide levels and antiviral activity, and (ii) the HA endosomolytic peptide component of UTARVE contributes significantly to the antiviral activity presumably because it enhances the endosomolytic activity of the penton base protein. The antiviral activity seen with UTARVE complexed OMP is well within the therapeutically significant doses (also in diploid cells). Indeed, IC₅₀ values for acyclovir are HSV strain dependent and range between 0.3-3.0 μM. The UTARVE-OMP complex is not toxic as evidenced by the finding that actin expression was not inhibited as determined by immunoblotting, and toxicity was not observed in dye release assays.

While the prototype UTARVE-OMP complexes were successful at inhibiting target virus replication, the inhibition was achieved at concentrations which probably saturate the number of cell surface receptors (average 1.0 x 10⁵ to 5.0 x 10⁵ per cell) involved in UTARVE cellular binding and uptake. It is believed that a preferred application of UTARVE in the delivery of OMPs or any other nucleic acid moieties as a complex may require lower molar concentrations than those giving 100%> receptor saturation. The flexibility of UTARVE design allows one to achieve this goal. For example, a length of 20 (KL)₁₀ units in UTARVE will increase the number of OMP molecules UTARVE particle to 200, as compared to 20 for the second prototype UTARVE. Using the results achieved with the prototype UTARVE as an example, this increase enables one to attain similar inhibitory OMP concentrations with 90%> less UTARVE particles than those used in the prototype experiment. In turn, this will reduce the saturation of the cell surface receptors by, 90%.

Thus, in another experiment, the second prototype UTARVE was complexed to an antisense expression vector for cyclin E, a protein which is involved in the regulation of the cell cycle. Coverslip cultures of breast cancer cells (MD-MBA- 157) were treated with unconjugated or UTARVE conjugated antisense expression vector for cyclin E (48 hrs) and pulsed (1 hr) with 100 nM bromodeoxy uridine (BudR) and processed for immunofluorescent detection of incorporated BudR. The percent of DNA synthesizing nuclei was scored against the total nuclei counter-stained with Hoechst 33258 (Molecular Probes Inc., Eugene, OR).

Inhibition of cyclin E in breast cancer cells with UTARVE delivered anti-sense expression vector for cyclin E caused a significant reduction in the proportion of replicating cells (in S phase). The S phase inhibition was respectively 42 and 99% for the unconjugated and UTARVE conjugated antisense cyclin E DNA.

EXAMPLE 4 Construction of a Penton Base Protein-Containing UTARVE A UTARVE was constructed by assembly of DNA sequences encoding: (i) the adenovirus type 2 penton base protein (PBP), which binds the cell surface receptors- integrin and (ii) an endosomolytic peptide derived from influenza virus HA.

More specifically, a dsDNA encoding the HA endosomolytic peptide (20 amino acids) plus 5' and 3' flanking sequences (Gly) were generated by PCR amplification of a primary single-stranded 70 nucleotide (nt) DNA sequence. The Gly residues permit formation of alpha-helices flanking the HA peptide (Gly₂-HA-Gly₂), thereby imparting flexibility to the secondary structure at the HA-penton base junction. The sequence for Gly₂-HA-Gly₂was first cloned into pGEMT (Promega, Madison, WI), a PCR cloning vector. The sequence for Gly₂-HA-Gly₂was removed from the pGEMT vector and cloned into pETl laPB (Bai M, Harfe B, Freimuth P, Mutations that alter an Arg-Gly-Asp (RGD) sequence in the adenovirus type 2 penton base protein abolish its cell-rounding activity and delay virus reproduction in flat cells. J. Virology 1993 ;67(9):5198-205 ), a bacterial expression vector which contains the entire Ad2 penton base protein under the control of a T7 phage promoter, by ligation in the Ndel site located upstream of the penton base protein initiator Met. The resulting plasmid, PETllaHA/PB, codes for a chimeric protein that begins with Gly₂-HA-Gly₂ (initiator Met is encoded by vector), followed by expression vector and penton base protein coding sequences. Positive clones were identified by sequencing.

To generate UTARVEΔHA, pETl laHA/PB was subjected to collapse ligation at the Ndel located upstream of the penton base protein initiator Met. The resulting plasmid pETl 1 aΔHA/PB codes for a chimeric protein that begins with an initiator Met, Ala, Ser and Thr all encoded by vector followed by the penton base protein coding sequence. Positive clones were identified by sequencing.

More specific details as to the construction of the above UTARVE are set forth below.

A. The Ad2 Penton Base Protein The Ad2 penton base protein reading frame begins pETl laPB, which also encodes a four-amino acid extension (Met-Ala-Ser-Thr) at the N-terminal of the Ad2 penton base protein, and continues with the first amino acid (Met) of the Ad2 penton base protein sequence. All of the signals for transcriptional and translational regulation of the Ad2 penton base protein gene are present within pETllaPB. This plasmid also contains an ampicillin resistance gene for positive (bacterial transformation) selection, as well as sequences important for its growth and replication in E. coli.

B. The HA Peptide

A DNA molecule encoding the influenza HA peptide was synthesized and then cloned into pETllaPB.

This was accomplished by preparing a dsDNA molecule which encodes the 20 amino acid HA plus, 5' and 3' flanking sequences that provide the nucleotides suitable for cloning into either pETl 1 a or pETllaPB, as well as encoding for HA flanking glycine residues (Gly₂-HA-Gly₂). The dsDNA was generated by PCR amplification of the following primary single-stranded 70 bp DNA sequence for HA: 5' -GAGGTGGACTCTTCGAAGCAATTGCAGGTTTAATCGAAAACGGCTGGGA AGGCATGATCGACGGTGGTGG-3' (SEQ ID NO: 17), using the following sense and anti-sense primers, respectively: 5'-CATATGGGAGGTGGACTCTTCGAAGCA-3^* (SEQ ID NO: 18); and 5'

-CATATGGCCACCACCGTCGATCATGCC-3' (SEQ ID NO: 19)

These primers each contain seven 5' terminal nucleotides designed to encode for 2 glycines upon ligation into the Ndel site of pETllaPB. The dsDNA was ligated into pGEMT according to manufacturer's (Promega, Madison, WI) recommended procedures and transformed into E. coli DH5α (Life Technologies) followed by growth selected on culture media containing ampicillin and 5-bromo-4- chloro-3-indoyl-β-D-galactose (X-GAL). Growth selection of pGEMT transformed DH5 on this culture media yields blue colonies. Disruption of lacZ expression in pGEMT via insertion/ligation of a PCR product can result in the formation of white colonies on this media. Plasmid DNA was extracted from white colonies (putative positive recombinants) and subjected to secondary screening using Ndel and Mfel restriction enzyme analysis. Positive clones (pGEMT/HA) were subjected to ds DNA sequencing using universal primer which targets vector sequences upstream of the insert.

The Gly₂-HA-Gly₂ encoding ds DNA sequence was removed from pGEMT by Ndel restriction digestion and cloned into Ndel digested pETl laPB to give plasmid pETl laHA/PB. The Ndel cleavage site in pETl laPB overlaps the vector encoded initiation (Met) codon. pETl laHA/PB encodes for a fusion protein that begins with a start Met followed by I) Gly₂-HA-Gly₂, ii) His, Met, Ala, Ser and Thr residues encoded by the expression vector and iii) the Ad2 penton base protein coding sequence.

E. coli DH5α was transformed with PETl laHA/PB, and growth selected on Amp containing culture media. Plasmid DNA was extracted from isolated colonies, and subjected to initial screening (for positive recombinants) using Ndel restriction enzyme analysis. Positive recombinants were subjected to a secondary screening using Mfel and Nhel restriction analysis in order to confirm the correct orientation of Gly₂-HA-Gly₂ sequence within pETl laHA/PB. The Mfel restriction site is present in the

HA encoding sequence and the Nhel site is located 5' adjacent to the AD2 PBP sequence. Positive clones were subjected to double-stranded sequencing using the universal forward primers described-above to confirm in-frame ligation of Gly ₂-HA-Gly₂ sequence to the Ad2 penton base protein sequence in pETllaHA/PB. Another UTARVE was constructed to contain a DNA molecule encoding Gly₂-HA-Gly₂ followed by a multiple cloning site (MCS) cloned into pETl la. The Ndel fragment encoding Gly₂-HA-Gly₂ from pGEMT/HA was cloned into the Ndel site of pETl la to give plasmid pETl laHA. This plasmid was further modified through restriction digestion with BamHl and Nhel, which cut downstream of the Gly₂-HA-Gly₂ encoding sequence, both sites made blunt-ended followed by a collapse ligation of these sites. The unique Hindlll site in this plasmid was deleted through a subsequent modification involving digestion with Hindlll followed by mung bean nuclease treatment and blunt-end ligation of this site to generate plasmid pETl laHA+. This cloning strategy creates a single site (BamHl) downstream of Gly₂-HA-Gly₂ for ligation/insertion of the MCS.

To generate the MCS, a ds DNA molecule was prepared which contains endonuclease cleavage sites for several enzymes including Bglll, Xhol, Ecll36III, Pmll, Ascl, Ncol, Sail, Bsrgl and BamHl as well as encoding for translational stop codons in all three reading frames. The ds DNA was generated by PCR amplification of the following primary single-stranded 54 bp DNA sequence 5' - CTCGAGCTCACGTGGCGCGCCATGGTCGACTGTACAGGATCCTAACTAGGTAAG-3' (SEQ ID 20) using the following sense and antisense primers, respectively 5'- A G A T C T C T C G A G C T C A C G T G G C G C G C 3 ' ( S E Q I D 2 1 ) a n d 5 ' -

AGATCTCTTACCTAGTTAGGATCCTG-3' (SEQ ID 22). The dsDNA was ligated into pGEMT according to manufacturer's recommended procedures and protocols and transformed into DH5 and growth selected on culture media containing ampicillin and 5-Bromo-4-chloro-3-indoyl-β-D-galactose (X-GAL). Growth selection of pGEMT transformed DH5α on this culture media yields blue colonies. Disruption of lacZ expression in pGEMT via insertion/ligation of a PCR product can result in the formation of white colonies on this media. Plasmid DNA was extracted from white colonies (putative positive recombinants) and subjected to secondary screening using BamHl restriction enzyme analysis which cuts only once (in the MCS) in a positive recombinant. Positive clones (pGEMT/MCS) were subjected to ds DNA sequencing using universal primer which targets vector sequences upstream of the insert. The MCS was removed from pGEMT/MCS and cloned into the unique BamHl site of pETl laHA+ to create plasmid pETl laHAmcs.

As discussed above, the number of flanking glycine residues may be varied according to the length of the primary nucleotide sequences which flank the HA encoding region. Alternatively the Gly₂-HA-Gly₂ sequence may be extended at the 5' and 3' ends by PCR amplification using sense and anti-sense primers that contain terminal sequences which encode for additional glycines. The strategy used to clone Gly_n-HA-Gly_n sequences (where n > 2) into pGEMT, pETllaPB and pETl laHAmcs and confirm the success of the cloning is similar to that used for Gly₂-HA-Gly₂.

Another UTARVE was constructed to include 1 (KL)₁₀ unit for which details are set forth below. C. The Polylsylleucyl Peptide

In order to generate additional attachment sites for drug molecules, a polylysylleucyl (KL),₀was constructed that was suitable for cloning in-frame with the Ad2 penton base protein coding sequence of

PETl laHA/PB.

More specifically, a DNA molecule encoding (KL)₁₀ was prepared using the following sense and anti-sense primers, respectively:

5'-GTCGACGGTACCGGATCCAAG-3' (SEQ ID NO:23); and 5'-GAATTCAGATCTGAGCTTAAG-3 ' (SEQ ID NO:24) so as to PCR amplify a single-stranded

90 base DNA molecule:

5'-GTCGACGGTACCGG^ΓCCΓAAGCTCAAGCTTAAACTCAAGCTTAAGCTCAAACTGAAG

CTTAAGCTCAAGCTTAAGCTC1AGATCTGAATTC-3' (SEQ ID NO:25), which encodes for 10 polylysylleucyl repeat sequences (identified in brackets) flanked by sequences for restriction endonuclease cleavage by Sail (bold), BamHl (italics), Bglll (underlined) and Hindlll (underlined within brackets). After PCR amplification, the dsDNA was ligated into pGEMT, transformed and grown as above. Growth selection of pGEMT (no insert) transformed DH5 on this culture media yields blue colonies. Disruption of lacZ expression in pGEMT via insertion/ligation of a PCR product can result in the formation of white colonies on this media. Plasmid DNA was extracted from white colonies and subjected to secondary screening using Hindlll restriction enzyme analysis. Positive clones

(pGEMT/KL) were subjected to ds DNA sequencing using universal primer which targets vector sequences upstream of the insert.

The KL₁₀ encoding fragment was removed from pGEMT/KL by Sall/Bglll digestion and cloned into the Sall/BamHI sites in the MCS of pETl laHAmcs to give plasmid pETl laHA/KL. Transformation into DH5α and growth selection on Amp containing media followed by Hindlll restriction analysis was used to identify positive clones. Final confirmation through ds DNA sequencing provided nucleotide sequence confirmation. Cloning the Sall/Bglll fragment into pETl laHAmcs affects a BsrGI to Acc65I change downstream of the Sail site in the MCS. Alternatively, the KL₁₀ encoding fragment may be cloned into pETl 1 aHAmcs as a BamHI/Bglll fragment thereby retaining the Bsrgl site within the MCS. The utility of this cloning strategy is to provide the potential for i) establishing an alternate cloning site within the MCS and ii) generating translational frame-dependent expression of a Cys codon (present in a BerGI but not Acc65I recognition sequence) upstream of the translational stop codons. A Cys amino acid can provide an alternative to Lys as a site for derivatization/complexation of drug moieties.

Another UTARVE, pETl laHA/PBP/KL was created to contain DNA sequences encoding in frame for Gly₂-HA-Gly₂ , PBP and KL,₀. This DNA sequence was constructed by cloning the PBP sequence from pETl laPBP in pETl laHA/KL using a two-step strategy. In the first step the 1.6kb

Mfel/Ascl fragment of pETl laPB, which encodes the carboxyl region of Gly₂-HA-Gly₂ and the coding sequence of PBP from amino acids 1 through 507, was cloned directionally into Mfel/Ascl digested pETl laHA/KL to create plasmid pETl laHA/PBpart/KL. Recombinants were growth selected on Amp containing media and positive recombinants identified by restriction analysis with BamHl wherein positives recombinant plasmids are cut twice with this enzyme. In the second step, a ds DNA PBP sequence encoding amino acids 495 to 571 was amplified by PCR using the following sense and antisense primers 5'-CGTGTTCAATCGCTTTCCCGAGAA-3' (SEQ ID26) and 5'- GTCGACAAAAGTGCGGCTCGATAGGACG-3' ( SEQ ID 27), respectively. The antisense primer contains six 5' terminal nucleotides designed to create a Sail restriction site 3' adjacent to the codon for PBP amino acid 571 with concomitant elimination of the translational stop codon. The dsDNA was ligated into pGEMT according to manufacturer's recommended procedures and protocols and transformed into DH5α and growth selected on culture media containing ampicillin and 5-Bromo-4- chloro-3-indoyl-β-D-galactose (X-GAL). Growth selection of pGEMT transformed DH5α on this culture media yields blue colonies. Disruption of lacZ expression in pGEMT via insertion/ligation of a PCR product can result in the formation of white colonies on this media. Plasmid DNA was extracted from white colonies (putative positive recombinants) and subjected to secondary screening using Ascl restriction enzyme analysis. Positive clones (pGEMT/PBterm) were subjected to ds DNA sequencing using universal primer which targets vector sequences upstream of the insert. PGEMT/PBterm was digested with Ascl and Sail and the 204bp fragment encoding PBP amino acids 507 to 571 cloned into the Ascl/Sall sites of pETl laHA/PBpart/KL to generate pETl laHA/PB/KL. E. Coli DH5α was transformed with this recombinant and growth selected on Amp containing culture media. Plasmid DNA is extracted from isolated colonies, and subjected to screening (for positive recombinants) using BamHl restriction enzyme analysis. Positive clones are subjected to ds DNA sequencing using the universal forward primer to confirm nucleotide sequence. pETl laHA/PB/KL encodes for a fusion protein that begins with a Met followed by I)Gly₂-HA-Gly₂, ii) His, Met, Ala, Ser, and Thr residues encoded by vector sequences, iii) the Ad2 penton base protein, iv) Val, Asp, Gly, Asn, Gly and Ser encoded by the

MCS, v) KL₁₀ and vi) end terminal amino acids Arg and Ser. C2. The EGF Protein

A DNA molecule encoding the mature EGF protein plus an additional amino acid, Ala, was first produced as two individual fragments and then joined to one complete molecule. The complete molecule was then cloned into pETl 1 aHAmcs and pETl 1 aHA/KL. While encoding for the mature EGF sequence, the complete DNA sequence was designed, based on codon degeneracy, to have a primary sequence different from that of the native sequence.

More specifically, a dsDNA molecule was prepared which encodes for Ala followed by the first 25 amino acids of EGF plus 5' flanking sequences that provide the nucleotides suitable for cloning into the appropriate site in vectors such as pETl laHAmcs or pETHA/KL and 3' flanking sequences that provide for cloning (ligation) adjacent with the second EGF fragment (below). The dsDNA was generated by PCR amplification of the following primary single-stranded 73 bp DNA sequence: 5'-GCCAACTCAGATTCAGAATGTCCACTGTCACACGATGGCTACTGCCTCCATGACGGAG TGTGCATGTATATCG-3' (SEQ ID NO:28), using the following sense and annti-sense primers, respectively: 5'-CTCGAGGCCAACTCAGATTCAGAATG-3' (SEQ ID NO:29); and

5'-AGGCCTCGATATACATGCACACTCCG-3' (SEQ ID NO:30).

The dsDNA was ligated into pGEMT according to manufacturer's recommended procedures and protocols and transformed into DH5α and growth selected on culture media containing ampicillin and 5-Bromo-4-chloro-3-indoyl-β-D-galactose (X-GAL). Growth selection of pGEMT transformed DH5α on this culture media yields blue colonies. Disruption of lacZ expression in pGEMT via insertion/ligation of a PCR product can result in the formation of white colonies on this media. Plasmid DNA was extracted from white colonies (putative positive recombinants) and subjected to secondary screening using Xhol and Pstl restriction enzyme analysis. Positive clones (pGEMT/EGFl) were subjected to ds DNA sequencing using universal primer which targetsvector sequences upstream of the insert.

A dsDNA molecule was prepared which encodes for amino acids 25 to 53 of EGF plus 5' flanking sequences that provide the nucleotides suitable for cloning (ligation) adjacent to the first EGF fragment (above) and 3' flanking sequences that provide for cloning inot the appropriate site in vectors such as pETl laHAmcs or pETl laHA/KL. The dsDNA was generated by PCR amplification of the following primary single-stranded 83 bp DNA sequence:

5'-TGGACAAATACGCATGCAACTGTGTTGTTGGATACATCGGCGAACGATGTCAATACCG CGATCTGAAATGGTGGGAACTGCGA-3 ' (SEQ ID NO:31 ), using the following sense and annti-sense primers, respectively: 5'-AGGCCTTGGACAAATACGCATGCAAC-3' (SEQ ID NO:32); and 5'-GTCGACTCGCAGTTCCCACCATTTCA-3* (SEQ ID NO: 33).

The dsDNA was ligated into pGEMT according to manufacturer's recommended procedures and protocols and transformed into DH5α and growth selected on culture media containing ampicillin and

5-Bromo-4-chloro-3-indoyl-β-D-galactose (X-GAL). Growth selection of pGEMT transformed DH5α on this culture media yields blue colonies. Disruption of lacZ expression in pGEMT via insertion/ligation of a PCR product can result in the formation of white colonies on this media. Plasmid DNA was extracted from white colonies (putative positive recombinants) and subjected to secondary screening using Sail restriction enzyme analysis. Positive clones (pGEMT/EGF2) were subjected to ds

DNA sequencing using universal primer which targetsvector sequences upstream of the insert.

The ds DNA sequence the second EGF fragment was removed from pGEMT/EGF2 by Stul/Sall restriction digestion and cloned into the Stul/Sall sites of pGEMT/EGF 1. The resulting recombinant plasmid pGEMT/EGF contains a novel DNA sequence which encodes for a native EGF protein plus a 5' Ala amino acid.

E. coli DH5α was transformed with pGEMT/EGF, and growth selected on Amp containing culture media. Plasmid DNA was extracted from isolated colonies, and subjected to initial screening (for positive recombinants) using Xhol/Ndel restriction enzyme analysis. Positive recombinants were subjected to a secondary screening using Xhol/Stul and Xhol/Sall restricition analysis in order to screen for the presence of one complete copy of EGF encoding DNA sequence within the vector. Positive clones were subjected to double-stranded sequencing using the universal forward primers described-above to confirm the nucleotide sequence of the EGF encoding DNA.

Another protype UTARVE, pETl laHA/EGF was created to contain a DNA sequence which encodes inframe for Gly₂-HA-Gly₂ and EGF. The Ala-EGF encoding dsDNA sequence was removed from pGEMT/EGF by Xhol/Sall restriction digestion and ligated into the Xhol/Sall sites of pETl laHAmcs to give pETl laHA/EGF. E. coli DH1 was transformed with putative recombinant, pETl laHA/EGF, and growth selected on Amp containing culture media. Plasmid DNA was extracted from isolated colonies, and subjected to initial screening (for positive recombinants) using Xhol/Sall restriction enzyme analysis. Positive clones were subjected to double- stranded sequencing using the universal forward primers described-above to confirm nucleotide sequence. pETl laHA/EGF encodea for a fusion protein that gegins with Met followed by i) Gly₂-HA-Gly₂, (ii) His, Met, Gly, Leu, Asp and Glu encoded by vector and MCS sequences, iii) Ala followed the EGF protein and iv) Val, Asp, Cys, Thr, Glu and Ser encoded by MCS.

Another protype UTARVE, pETl laHA/EGF/KL was created to contain DNA sequences encoding for Gly₂-HA-Gly₂ inframe with EGF followed by L₁₀. The L₁₀ encoding fragment was removed from pGEMT/KL by Sall/Bglll digestion and cloned into the Sall/BamHI sites in the MCS of pETl laHA/EGF to give pETl laHA/EGF/KL. Alternatively, the Ala-EGF containing Xhol/Sall fragment from pGEMT/EGF could be cloned into Xhol/Sall sites of pET 1 1 aHA/KL to generate the same recombinant plasmid. E. coli DH5α was transformed with putative recombinant pETl laHA/EGF/KL and growth selected on Amp containing culture media. Plasmid DNA was extracted from isolated colonies ans subjected to screening (for positive recombinants) using Hindlll restricition analysis.

Positive clones were subjected to ds DNA sequenceing using universal forward primer to confirm nucleotide sequence. pETl laHA/EGF/KL encodes for a fusion protein that begins with Met followed by i) Gly₂-HA-Gly₂ ii) His, Met, Gly, Leu, Asp, and Glu encoded by vector and MCS sequences, iii) Ala followed by EGF protein, iv) Val, Asp, Gly, Asn, Gly and Ser, v) KL₁₀ and vi) end terminal amino acids Arg and Ser.

D. Isolation and Purification of UTARVE

The UTARVEs were transformed in E. coil strain BL21(DE3 and the UTARVE protein produced, isolated and purified as described by Bai et al, J. Virol, 67:5198-5205 (1993).

More specifically, expression was induced in late log phase cultures (the optical density at 600 nm was 0.8) in LB broth at 37°C by adjusting the medium to 0.4 mM isopropylthiogalactopyranoside

(IPTG). After 3 hrs shaking, the cells were collected by centrifugation, washed in STE buffer comprising 10 mM Tri-HCl (pH 8.0), 1.0 mM EDTA and 100 mM NaCl, lysed by freezing and thawing in the presence of 0.1 mg/ml lysozyme and sonicated (2 x 30 sec each) at 40% maximum output in a Branson Sonifier. The insoluble fraction containing UTARVE was collected by centrifugation and resuspended in buffer comprising 20 mM Tris-HCl (pH 7.5), 1.0 mM EDTA and 0.1 % (w/v) Nonidet P-40. After three additional cycles of washing, the final pellet was resuspended in a small volume of 6.0 M urea, diluted with 9 volumes of buffer comprising 50 mM K₂P0₄(pH 10.7), 50 mM NaCl, 1.0 mM EDTA, and dialyzed against phosphate buffered saline (PBS) and then 50 mM phosphate buffer (pH 7.5). The resulting dialyzate was sterilized by filtration and stored at 4°C or frozen at -70°C. The purity of the UTARVE protein was confirmed by SDS-PAGE and Coomassie staining. Only one protein band was observed, representing the purified UTARVE. A similar band was seen for UTARVE-ΔHA.

While the invention has been described in detail, and with reference to specific embodiments thereof, it will be apparent to one with ordinary skill in the art that various changes and modifications can be made therein without departing from the spirit and scope thereof.

Claims

WHAT IS CLAIMED:

Claim 1. A nucleic acid uptake and release vehicle comprising a fusion protein wherein a ligand and influenza virus hemagglutinin endosomolytic peptide are covalently bonded by an amide linkage to provide a vehicle for the intracellular delivery of moieties consisting of nucleic acids, oligonucleotides or modified nucleic acids or oligonucleotides.

Claim 2. The vehicle of claim 1 wherein the ligand is selected from the group of ligands consisting of adenovirus penton base protein, transferrin, cholera toxin B subunit, and epidermal growth factor.

Claim 3. The vehicle of claim 1 wherein the ligand is adenovirus penton base protein.

Claim 4. The vehicle of claim 1 wherein the ligand is epidermal growth factor.

Claim 5. The vehicle of claim 1 wherein said protein is bonded to amino acids or peptides which provide additional sites of nucleic acid attachment.

Claim 6. The vehicle of claim 1 wherein said protein is bonded to a nuclear localization signal.

Claim 7. The vehicle of claim 1 wherein said protein is bonded to an amino acid or peptide which imparts flexibility to the secondary structure

Claim 8. The vehicle of claim 1 further comprising glycine residues flanking and covalently bonded by an amide linkages to the hemagglutinin endosomolytic peptide.

Claim 9. The vehicle of claim 5 where said amino acids or peptides are polylysylleucyl. Gam

10. The vehicle of claim 6 where said nuclear localization signal is the SV40 large T antigen nuclear localization signal peptide.

Claim 11. The vehicle of claim 5 wherein the amino acids or peptides which provide additional sites of nucleic acid attachment are bonded to the nucleic acid by ionic bonding.

Claim 12. The vehicle of claim 5 wherein the amino acids or peptides which provide additional sites of nucleic acid attachment are bonded to the nucleic acid by covalent bonding by an amide linkage.

Claim 13. The vehicle of claim 5 wherein the amino acids or peptides which provide additional sites of nucleic acid attachment are bonded to nucleic acid which has a thiol group positioned on the alkyl side chain by a thioether linkage.

Claim 14. The vehicle of claim 5 wherein the amino acids or peptides which provide additional sites of nucleic acid attachment are bonded to nucleic acid which has a carboxyl group attached to an alkyl linker arm by covalent bonding at an amide linkage.

Claim 15. The vehicle of claim 5 wherein the amino acids or peptides which provide additional sites of nucleic acid attachment are derivatized with glyoxylic acid or the n-hydrozyphthalimide ester of glyoxylic acid and are bonded to nucleic acid through formation of an oxime bond.

Claim 16. The DNA construct encoding the vehicle of claim 1.

Claim 17. The vehicle of claim 1 wherein the nucleic acids, oligonucleotides or modified nucleic acids or oligonucleotides are selected from the group of nucleic acids, oligonucleotides or modified nucleic acids or oligonucleotides consisting of single-stranded anti-sense oligonucleotides, double- standed anti-sense oligonucleotides, single-stranded sense oligonucleotides, and double-stranded sense oligonucleotides.

Claim 18. The vehicle of claim 1 wherein the nucleic acids, oligonucleotides or modified nucleic acids or oligonucleotides are anti-viral agents.

Claim 19. The process of treating a viral infection in a mammal in need of such treatment comprising the step: administering to the mammal in need of such treatment a suitable amount of the vehicle of claim 18.

Claim 20. The vehicle of claim 1 wherein the nucleic acids, oligonucleotides or modified nucleic acids or oligonucleotides are useful in treatment of genetic diseases.

Claim 21. The process of treating a genetic disease in a mammal in need of such treatment comprising the step: administering to the mammal in need of such treatment a suitable amount of the vehicle of claim 20.

Claim 22. The vehicle of claim 1 wherein the nucleic acids, oligonucleotides or modified nucleic acids or oligonucleotides are useful in treatment of cancer.

Claim 23. The process of treating cancer in a mammal in need of such treatment comprising the step: administering to the mammal in need of such treatment a suitable amount of the vehicle of claim 22.