CA2198861A1 - Recombinant adenovirus comprising a chimeric penton base protein - Google Patents

Abstract

A recombinant adenovirus comprising a chimeric penton base protein, which includes a nonpenton base sequence, and a therapeutic gene, a method of gene therapy involving the use of such adenovirus, and adenoviral transfer vectors for the generation of such recombinant adenovirus are provided.

Description

w o 96/07734 ~ PcTrusg5/o99o4 RECOMBINANT ADENOVIRUS COMPRISING A CHIMERIC PENTON BASE PROTEIN

TECHNICAL FIELD OF THE INVENTION
The present invention relates to a recombinant adenovirus comprising a chimeric adenoviral penton base protein and the use of a recombinant adenovirus comprising a chimeric adenoviral penton base protein in gene therapy.
BACKGROUND OF TH~ INVENTION
Adenoviruses belong to the family Adenoviridae, which is divided into two genera, namely Mastadenovirus and Aviadenovirus. Adenoviruses are nonenveloped, regular icosahedrons 65-80 nm in diameter (Horne et al., J. Mol. Biol., 1, 84-86 (1959)). The capsid is composed of 252 capsomeres of which 240 are hexons and 12 are pentons (Ginsberg et al., Virology, 28, 782-783 (1966)).
The hexons and pentons are derived from three different viral polypeptides (Maizel et al., Virology, 36, 115-125 (1968); Weber et al., ViroloqY, 76, 709-724 (1977)). The hexon comprises three identical polypeptides of 967 amino acids each, namely polypeptide II (Roberts et al., Science, 232, 1148-1151 (1986)). The penton contains a z5 penton base, which is bound to tne capsid, and a flber, which is noncovalently bound to and projects from the penton base. The fiber protein comprises three identical polypeptides of 582 amino acids each, namely polypeptide IV. The Ad2 penton base protein is an 8 X 9 nm ring-shaped complex composed of five identical protein subunits of 571 amino acids each, namely polypeptide III
(Boudin et al., Virology, 92, 125-138 (1979)). Proteins IX, VI, and IIIa are also present in the adenovirus coat and are thought to stabilize the viral capsid (Stewart et al., Cell, 67, 145-154 (1991); Stewart et al., EMBO J., 12(7), 2589-2599 (1993)).
Once an adenovirus attaches to a cell, it undergoes receptor-mediated internalization into clathrin-coated W096/07734 ~ ~ ~ 8 8 ~ ~ PCT~S95/09904 endocytic vesicles of the cell (Svensson et al., J.
Virol., 51, 687-694 (1984); Chardonnet et al., Virology, 40, 462-477 (1970)). Virions entering the cell undergo a stepwise disassembly in which many of the viral structural proteins are shed (Greber et al., Cell, 75, 477-486 (1993)). During the uncoating process, the viral particles cause disruption of the cell endosome by a pH-dependent mechanism (Fitzgerald et al., Cell, 32, 607-617 (1983)), which is still poorly understood. The viral particles are then transported to the nuclear pore complex of the cell (Dales et al., Virolo~y, 56, 465-483 (1973)), where the viral genome enters the nucleus, thus initiating infection.
An adenovirus uses two separate cellular receptors, both of which must be present, to attach to and efficiently infect a cell (Wickham et al., Cell, 73, 309-319 (1993)). First, the fiber protein attaches the virus to a cell by binding to an, as yet, unidentified receptor. Then, the penton base binds to ~ integrins.
Integrins are a family of heterodimeric cell surface receptors that mediate cellular adhesion to the extracellular matrix molecules fibronectin, vitronectin, m; ni n, and collagen, ag well as other molecules (Hynes, Cell, 69, 11-25 (1992)). Integrins are known to play important roles in cell signaling processes, including calcium mobilization, protein phosphorylation, and cytoskeletal interactions (Hynes, supra). The specificity with which an integrin binds to a particular ligand, such as those associated with an adenovirus, is a function of the paired ~ and ~ subunits of the integrin.
For example, integrin ~6~1 binds to lAm- n~ n, integrin ~2~1 binds to collagen and lAm; n; n, and integrin ~3~1 binds to collagen, lam;n;n, and fibronectin. Furthermore, different tissue types may have different complements of integrin subunits, thereby providing a mode of spatial control over integrin-ligand signal transfer or over ligand internalization. Accordingly, some integrins, W096/07734 ~ ~ ~ 8 8~1 PCT~S95/09904 such as those that include subunit ~v~ are broadly expressed on numerous cell types, whereas other integrins have a much more narrow tissue distribution. For - example, the integrins that include subunit ~2 are expressed only on leukocytes, such as neutrophils and macrophages, integrins including subunit ~4 are expressed only on lymphocytes and fibroblasts, and the integrin defined by the subunits a~3 is expressed only on platelets and megakaryocytes.
The specificity of integrin subunit complement also extends to the infectability of cells by different serotypes of adenovirus, because the particular ~ and ~
subunits dictate whether a virus can enter a cell. For example, the penton base of the adenovirus serotype Ad2 binds to integrins ~v~3 and ~v~S (Wickham et al. (1993), supra). Given that both receptors utilized by an adenovirus are expressed on most human cells, nearly all cells in a human body are susceptible to adenoviral infection.
A majority of integrins have been found to recognize short linear stretches of amino acids in binding to a specific ligand. The tripeptide motif arg-gly-asp (RGD), which is found in scores of matrix ligands, including laminin, fibronectin, collagen, vitronectin, and fibrinogen, has been implicated in the binding of ~3~1 ~5~ IIb~3~ ~m~2/ and most, if not all, of the five ~v~
containing integrins. The conformation of the RGD
sequence within a matrix ligand is thought to be a primary factor in integrin specificity (Pierschbacher et al., J. Biol. Chem., 262, 17294-17298 (1987)). Sequences that directly flank the RGD sequence have been shown to influence integrin specificity, presumably because of their effect on RGD conformation (Smith et al., Proc.
Natl. Acad. Sci. USA, 90, 10003-10007 (1993)). However, sequences distant from the RGD may also function in integrin specificity as has been shown for the binding of ~5~1 to fibronectin (Obara et al., Cell, 53, 649-657 RECTiFlED SHEET (RU' E 91 ISAIEP

8 ~ ~
.

(199~). Other int~grlns, which do not utili~e RGD, have been ~ound to bin~ similar 9hort llnear ~tretches ~f a~i~o acids ~ithln t~eir specif c l~gands. For example.
i~tegrin ~3 bind~ via the amino acid 3equence ly3-gln-ala-~ly-asp (KQA~) [SEQ ID NO:l~ in ~ibrinogen ~Kloczewiak et al., Biochemi9trY, 23, 1767-1774 (19~4)), while ~l bind~ via the core s~q~ence glu-ile-leu-asp-val (~ILDV~ ~SEQ ID NO:2] in fibronectin (Komorlya ee al., ~.
Biol. ~hem., 266, 15075-15079 (l~gl) ) . A structural motif (i.e., asn-pro-xaa-tyr (NPXY) ) present in the subunits of ~V-containing i~tegrins has been ~hown to be impo~tant fo~ internalization tSuzuki et al., Proc. ~a.tl.
Acad. Sci. USA, 87, 5354 (1990)).
The penton base se~uence i~ highly conser~ed amons serotypes of adenoviru8 and contains five copies of the R~D tripe~tide ~otlf (~eu~ann e~ al., Gene, 6~, 15~-157 (19~ . The RGD tr peptide i~ belie~ed to mediate binding to a~ inte~rins because exosenously added RG~
peptides ~an block pentcn ~a~e ~inding and adenoviral

2~ infec~icn (Wickha~ et al. (1993), su~ra), and adenovlrus~3 that have point ~uta~ion~ in the RGD
~equence of the penton base are restricted in t~eir ability to infect cells (Bai et al., J.. ~i~ol., 67, 51~8-5205 ~19g3)).
The pen~on ba~e gene~ from Ad2, Ad5, Adl2~ and Ad40 serotyge~ of adenovirus have been sequenced. Alignment of the ~equences reveals a high degree of con~er~a~ion over the entire sequence, except for the N-terminu~, and, in Ad2, Ad~, and Adl2, a hyperva~iable region that

3 0 includes the RGD ~equence. Only Ad40, cne of two enteric adenoviral serotype6, doe~ not ha~e an RGD sequence. Ad2 and ~d5 are identical in t~e hypervariable region and contain a large insert of amino acids flanking either ~ide of the RGD sequence. Secondary structural analysis o~ the hypervarlable re~icn~ of the three RG~-c~ntainlng penton ~ases predicts that, in each ca~e, the RGD i9 flanked by ~-helice6. Such 3trUcture~ are belieYed to AMEN~ED S~IFET

W096/07734 pcT~s9slo99n4 form the spikes seen in cryo-electron micrographic (cryo-EM) images of Ad2 penton bases (Stewart et al. (1993), supra).
Recombinant adenoviral vectors have been used for the cell-targeted transfer of one or more recombinant genes to diseased cells or tissue in need of treatment.
Such vectors are characterized by the further advantage of not requiring host cell proliferation for expression of adenoviral proteins (Horwitz et al., In Viroloqy, Raven Press, New York, vol. 2, pp. 1679-1721 (1990); and Berkner, BioTechniques, 6, 616 (1988)), and, if the targeted tissue for somatic gene therapy is the lung, these vectors have the added advantage of being normally trophic for the respiratory epithelium (Straus, In Adenoviruses, Plenan Press, New York, pp. 451-496 (1984)).
Other advantages of adenoviruses as potential vectors for human gene therapy are as follows:
(i) recombination is rare; (ii) there are no known associations of human malignancies with adenoviral infections despite common human infection with adenoviruses; (iii) the adenoviral genome (which is a linear, double-stranded DNA) can be manipulated to accommodate foreign genes that range in size; (iv) an adenoviral vector does not insert its DNA into the chromosome of a cell, so its effect is impermanent and unlikely to interfere with the cell's normal function;
(v) the adenovirus can infect non-dividing or term1 n~ 1 ly differentiated cells, such as cells in the brain and lungs; and (vi) live adenovirus, having as an essential characteristic the ability to replicate, has been safely used as a human vaccine (Horwitz et al. (1990), supra;
Berkner et al. (1988), supra; Straus et al. (1984), supra; Chanock et al., JAMA, 195, 151 (1966); Haj-Ahmad et al., J. Virol., 57, 267 (1986); and Ballay et al., EMBO, 4, 3861 (1985)).

w096/07734 PCT~S95/09904 The problem of using a recombinant adenovirus in gene therapy i9 that all cells that express the aforementioned two receptors used by the adenovirus to attach and infect a cell will internalize the gene(s) being administered - not just the cells in need of therapeutic treatment. Likewise, certain cells, such as lymphocytes, which lack the ~v integrin adenoviral receptors, will be severely impaired in the uptake of an adenovirus and will not be easily ~m~n~hle to adenovirus-mediated gene delivery. Accordingly, limiting adenoviralentry to specific cells and/or e~p~n~;~g the repertoire of cells ~m~nAhle to adenovirus-mediated gene therapy would be a significant improvement over the current technology. Targeted adenoviral gene delivery should expand the cells ~men~hle to gene therapy, reduce the amount of adenoviral vector that is necessary to obtain gene expression in the targeted cells, as well as reduce side effects and complications associated with increasing doses of an adenovirus, such as inflammation and the transfection of normal, healthy cells.
Attempts have been made to target a virus to specific cells by sterically blocking adenoviral fiber protein with antibodies and chemically linking tissue-specific antibodies to the viral particle (Cotten et al., Proc. Natl. Acad. Sci. USA, ~9, 6094-6098 (1992)).
Although this approach has demonctrated the potential of targeted gene delivery, the complexity and reproducibility of this approach present major hurdles blocking its application in clinical trials. The difficulties thus far encountered in targeting the virus by these methods involve the method o~ synthesis required, which is to make major alterations in the viral particles following their purification. These alterations involve additional steps that covalently link large molecules, such as polylysine, receptor ligands and antibodies, to the virus. The targeted particle complexes are not homogeneous in structure and their W096/07734 ~ 8 ~ 1 PCT~S95/09904 ._ efficiency is sensitive to the relative ratios of viral particles, linking molecules, and targeting molecules used.
The present invention seeks to overcome the problem 5 of lack of cell and tissue specificity of recombinant adenoviral gene therapy. It is an object of the present invention to provide a receptor-specific, preferably cell receptor-specific/tissue receptor-specific, recombinant adenovirus. A further object of the present invention is to provide means for generating such a recombinant adenovirus at the level of gene expression, thereby enabling purification of recombinant adenoviral particle~
by conventional techniques. Another object of the present invention is to provide a method of gene therapy involving the use of such a homogeneous adenovirus, without the need for additional components or further modification. These and other objects and advantages of the present invention, as well as additional inventive features, will be apparent from the following detailed description.

~RIEF SUMMARY OF THE lNv~NllON
The present invention provides a recombinant adenovirus comprising a chimeric penton base protein that 2S selectively binds to a given receptor, preferably a cell-specific/tissue-specific receptor, an adenoviral transfer vector comprising a recombinant penton base gene for the generation of a chimeric penton base protein, and a method of using a receptor-specific recombinant adenovirus comprising a therapeutic gene in gene therapy.

~RIEF DESCRIPTION OF THE FIGURES
Figure 1 is a partial restriction map of an - adenoviral transfer vector (pAT).
Figure 2 is a partial restriction map of the vector pRcP~5 (~RGD).

w096/07734 ~ PCT~S95/09904 DETAILED DESCRIPTION OF THE INVENTION
The present invention provides, among other things, a recombinant adenovirus comprising a chimeric penton base protein. The chimeric penton base protein comprises a nonpenton base amino acid sequence, which is specific for binding to a receptor, a receptor-specific antibody domain or epitope, in addition to or in place of a wild-type penton base amino acid sequence, preferably an amino acid sequence that is specific for binding to a receptor, most preferably an RGD amino acid sequence.
By "nonpenton base amino acid sequence" is meant any amino acid sequence that is not found in a wild-type penton base. Preferably, the nonpenton base sequence is less than ten amino acids, more preferably less than five amino acids, and most preferably about three amino acids.
By "RGD amino acid sequence" is meant the RGD amino acid sequence and the RGD amino acid sequence along with up to or including three amino acids flanking either or both sides of the RGD amino acid sequence. The nonpenton base amino acid sequence renders the adenovirus specific for a given receptor, preferably a cell-specific or tissue-specific receptor, or specific for a given receptor-specific antibody domain or epitope. The receptor is preferably one that the wild-type adenovirus does not bind or one that the wild-type adenovirus binds but with greater specificity upon the introduction of the nonpenton base amino acid sequence into the chimeric penton base protein.
Preferably, the RGD amino acid sequence of the penton base protein has been replaced at the DNA level with an amino acid binding sequence for a given receptor, receptor-specific antibody ~omA;n or epitope.
Alternatively, the RGD amino acid sequence has been rendered inactive at the DNA level by mutation of the RGD
amino acid sequence, such as by insertional mutagenesis, for example, or rendered conformationally inaccessible in the penton base protein, such as by insertion of a DNA

W096/07734 ~ 8 ~ 1 PCT~S95/09904 sequence into or adjacent to the adenoviral penton base gene sequence, wherein "gene sequence" refers to the complete penton base gene sequence as well as any lesser gene sequences that are capable of being expressed as functional penton base protein. Preferably, the DNA
sequence is inserted near the gene sequence encoding the RGD amino acid sequence, 90 as to move the gene sequence encoding the RGD amino acid sequence within the penton base gene sequence such that in the ch;meric penton base protein the RGD amino acid sequence is conformationally inaccessible for binding to a receptor. In the latter case, the inserted nonpenton base gene sequence that causes the conformational inaccessibility of the RGD
amino acid sequence in the penton base protein is preferably one that encodes an amino acid sequence that is specific for binding to a receptor, a receptor-specific antibody ~m~l n or epitope. Such a recombinant adenovirus can be used, for example, to study receptor binding, adenoviral attachment, and adenoviral infection in vitro or in vivo.
In a preferred embodiment of the present invention, the above-described recombinant adenovirus additionally comprises a gene capable of being expressed in a cell to which the virus has attached or by which the virus has been internalized and preferably is one having therapeutic utility. In yet another preferred embodiment of the present invention, the recombinant adenovirus is fiberless or further comprises a chimeric coat protein, such as a fiber or hexon, that includes an amino acid sequence that is specific for a receptor, a receptor-~pecific antibody ~om~in or epitope, preferably one that is specific for the same receptor, antibody ~om~ln or epitope as the nonpenton base sequence. Such recombinant - adenovirus can be used, for example, to study the effects of expression of the gene in a given cell or tissue ln vitro or 1n vivo. Alternatively, the recombinant adenovirus can be used for gene therapy.

W096/07734 PCT~S95/09904 r The recombinant adenovirus comprising a chimeric penton base protein and the recombinant adenovirus that additionally comprises a gene-capable of being expressed in a particular cell can be generated by use of a viral transfer vector, preferably an adenoviral transfer (pAT) vector, in accordance with the present invention. The viral transfer vector, preferably the pAT vector, comprises a chimeric adenoviral penton base gene sequence. The chimeric penton base gene sequence comprises a nonpenton base sequence in place of the RGD
amino acid sequence, which has been deleted, or in addition to the RGD amino acid sequence, which has been mutated or rendered conformationally inaccessible in the expressed chimeric penton base protein as described above. The nonpenton base sequence renders the adenovirus specific for binding to a receptor, receptor-specific antibody domain or epitope also as described above. Given that the penton base gene has been shown to be ~90~ conserved among 5 of the approximately 41 serotypes of adenovirus, it is expected that any one of the serotypes of adenovirus may be used as the source of the penton base gene. It is preferred, however, that one of the serotypes for which the penton base gene has been sequenced is used.
Restriction sites that are unique with respect to the adenoviral genome are introduced into the penton base gene sequence; preferably, such restriction sites are introduced into or flanking the RGD region of the penton base gene sequence by a suitable method, such as PCR
mutagenesis. These unique restriction sites may be any restriction site not already present in the penton base gene, and are preferably Eco RI and/or Spe I. Such sites facilitate the removal, inactivation, such as by sequence alteration, of the DNA sequence encoding the RGD amino acid sequence in a given adenoviral genome, such as a wild-type adenovirus, or the rendering of the RGD amino acid sequence conformationally inaccessible, thereby W096/07734 ~ 6 ~ PCT~SsS/o99o4 ._ altering or eliminating the ability of the penton base molecule to bind an ~v integrin receptor. A deleted RGD
amino acid sequence can be replaced with, or a mutated or conformationally inaccessible RGD sequence can be accompanied by, a different DNA sequence, preferably a DNA sequence encoding specificity for binding to a receptor, preferably a cell-specific or tissue-specific cell-surface receptor, or to a receptor-specific antibody domain or epitope, for example.
Preferably, the pAT vector is one into which any suitable receptor-specific sequence can be rapidly inserted. For example, a unique Spe I restriction site can be used to remove the RGD amino acid sequence.
Alternatively, sequences also can be inserted into the penton base gene sequence without the need for unique restriction sites through PCR. Becau~e a recombinant adenovirus can be created via ligation of recombinant sequences with viral DNA or via homologous recombination, the pAT vector preferably has either (1) unique restriction sites that allow ligation of a vector fragment with the complementing fragments of the re~lning viral genomes, as described in Example 4, or (2) adequate lengths of DNA on either side of the receptor-specific or antibody domain- or epitope-specific sequence that allow efficient homologous recombination with viral DNA, as described in Example 5. A preferred pAT vector is shown in Figure 1, which is a partial restriction map of such a vector. The pAT of Figure 1 was generated as described in Example 1.
DNA encoding short peptide sequences or protein dom~nc capable of binding to a given receptor, preferably a specific cell or tissue receptor, and capable of being internalized by the receptor, such as the receptor that is resident on leukocytes, is preferred for insertion into the penton base gene sequence in which the RGD amino acid sequence has been deleted, mutated, or rendered conformationally inaccessible. However, other W096l07734 PCT~S95/09904 8 ~ 61 DNA sequences, such as those that encode receptor-specific antibody dom~n~ and sequences that encode antigenic epitopes recognized by specific antibodies also may be used to replace the RGD amino acid sequence.
The size of the DNA used to replace the RGD amino acid sequence may be constrained, for example, by impeded folding of the penton base and improper assembly of the penton base/fiber complex. However, cryo-EM analysis of the penton base indicates that, if the RGD amino acid sequence exists at the spikes of the penton bases, there may be considerable room to accommodate large inserts without folding or assembly being affected adversely.
Also, the large differences in size between the hypervariable region of the Ad5 penton base and the Adl2 penton base indicates that the hypervariable region may be able to accommodate inserts of various sizes. Of course, assembly of the penton complex probably will not be an issue with respect to viruses that are constructed without a fiber molecule.
The target receptor can be any receptor that is internalized by a cell, is optimally cell-specific or tissue-specific, and desirably is expressed only on those cells or tissue to be treated. Tissue-specific integrins are preferred target receptors for chimeric penton base molecules for four reasons. First, many integrins bind ligands via short linear stretches of amino acids, and these stretches have been identified for many integrins.
Second, certain integrins are tissue-specific for cells ~mPn~hle to targeted gene therapy. For example, ~2-cont~;n~ng integrins are expressed only in leukocytes,thus allowing targeted delivery to normally non-susceptible target cells while also preventing delivery to normally susceptible non-target cells. Third, the wild-type penton base uses ~v integrins to enter cells so that the use of a related integrin molecule is more apt to be successful than the use of an unrelated receptor.
Fourth, integrins are efficiently internalized into the W096/07734 ~ ~ ~ 8 ~ 6 1 PCT~S95/09904 cell (Bretscher, EMB0, 11, 405-410 (1992)), which is required for viral infection of cells.
Changing the penton base protein such that it recognizes a receptor, such as a cell-specific or tis~ue-specific cell-surface receptor, typically will not alter viral particle attachment to cells because cell attachment is mediated by the adenoviral fiber protein.
Infection of the cells to which the chlmeric penton base protein has attached, however, will be generally impeded because the virus requires the presence of an ~v integrin to enter the cell. Therefore, an adenovirus with altered penton bases will bind to most cells but will enter only those cells expressing the receptor (e.g., the integrin) for the chimeric penton base. Those cells unable to express the appropriate receptor nonetheless will have adenoviral particles tethered on the cell membrane because the viral particles will be unable to enter the cell. Consequently, tethered adenoviral particles will be exposed to the immune system and disposed of accordingly.
Following fiber-mediated viral attachment to cells, it is possible that some virus may manage to non-specifically enter cells not expressing the cell- or tissue-specific receptor recognized by the chimeric penton base. Therefore, some leakiness in cell-specific or tissue-specific expression may be observed. However, if this is the case, fiber-deficient viral particles can be constructed that use the penton base as the primary cell attachment protein. It is known that fiber protein is nonessential for virion assembly or infectivity of cells (Falgout et al., J. Viroloqy, 62, 622-625 (1988)).
Such mutant adenoviral particles lacking fiber protein adsorb onto cells at a ten-fold lower rate. The fact that the mutant viruses do adsorb and at a lower rate than wild-type particles indicates that the adsorption is mediated by the penton base/Q~ integrin interaction, which has a 30-fold lower affinity than the fiber/fiber-W096/0773~ 8 ~ ~ PCTNS9~5~1 receptor interaction (Wickham et al. (1993), ~upra).
Accordingly, both attachment and internalization can be restricted only to those cells expressing the desired tissue-specific receptor.
By replacing the low affinity RGD amino acid sequence in a wild-type penton base with an amino acid sequence with a high affinity for a given receptor, such as a cell-specific or tissue-specific receptor, it is possible to generate a fiberless viral particle with as high, or higher, affinity for targeted cells than the wild-type fiber-expressing virus. Fiberless viral particles containing a chimeric penton base, compared to fiber-cont~in'ng viral particles, can be targeted more effectively to tissues because the altered virus is able to bind only to the cells expressing the cell-specific or tissue-specific receptor recognized by the chimeric penton base. The desired ligand and receptor characteristics are as described above.
A recombinant ~hlmeric penton base gene sequence can be moved from a pAT vector into baculovirus or a suitable prokaryotic or eukaryotic expression vector for expression and evaluation of receptor specificity, avidity, and other biochemical characteristics.
Accordingly, the present invention also provides recombinant baculoviral and prokaryotic and eukaryotic expression vectors comprising a chlm~ric adenoviral penton base gene sequence. The chimeric penton base gene sequence includes a nonpenton base sequence in addition to or in place of a wild-type penton base amino acid sequence, such as an RGD amino acid sequence, which is specific for binding to a receptor. The wild-type penton base amino acid sequence may be deleted, mutated, or rendered conformationally inaccessible as described above with respect to the recombinant adenovirus comprising a chimeric penton base protein. By moving the chimeric gene from pAT to baculovirus or a prokaryotic or eukaryotic expression vector, high protein expression is W096/07734 ~ 8 PCT~S95/09904 . ~;'1 achievable, resulting in approximately 50~ of .he total protein of the host being the chimeric penton base.
Accordingly, the present invention also provides a recombinant baculovirus comprising a chimeric penton base protein and a chimeric adenoviral penton base protein comprising a nonpenton base amino acid sequence in addition to or in place of a wild-type penton base amino acid sequence, such as an RGD amino acid sequence, which is specific for binding to a receptor. The nonpenton base amino acid sequence is specific for binding to a receptor or a receptor-specific antibody domain or epitope as described above. For protein characterization studies, the recombinant chimeric penton base protein (rcPB protein, such as rcPB5) can be purified using any suitable methods, such as those described by Wickham et al. (1993), supra.
Various characteristic parameters of the penton base protein of interest can be assessed as follows: Adhesion assays are used to evaluate the specificity of the interaction of the rcPB protein with its designated receptor, using the method of Wickham et al. (1993), supra, for example. Specificity and affinity of the receptor/rcPB interaction is assessed by Scatchard analysis as shown previously by Wickham et al. (1993), supra, for wild-type penton base protein. Receptor specificity is further assessed by using antibodies and peptides specific for the targeted receptor to block rcPB5 binding to cells, using conventional methods.
Internalization is assayed with rcPB protein and recombinant adenoviruses, e.g., Ad5 rcPB, as described previously by Wickham et al. (1993), supra, for example.
- rcPB binding to fiber protein is assessed by its ability to precipitate radiolabeled fiber molecules when coupled - to protein A-coated beads via an antibody to the penton base molecule. Alternatively, fiber binding can be assessed in an ELISA-based assay in which rcPB is coated onto ELISA plates followed by incubation of immobilized W096/07734 ~ 6 ~ ~CT~S95/09904 rcPB with soluble fiber molecules, using conver,tional methods. Fiber binding to the rcPB protein is then assessed by a further incubation with antibody specific for the fiber molecule, via conventional techniques.
Virus entry and gene expression are evaluated initially by using the pAT vector cont~;n;ng the insert of interest to generate recombinant virus expressing the chimeric penton base protein and a marker gene, such as ~-glucuronidase. ~-glucuronidase expression in cells infected with adenovirus cont~;n;ng the ~-glucuronidase gene (Ad-Gluc) can be detected as early as two hours after adding Ad-Gluc to cells. This procedure provides a quick and efficient analysis of cell entry of the recombinant virus and gene expression, and is implemented readily by an artisan of ordinary skill using conventional techniques.
A recombinant virus, which lacks a wild-type receptor binding sequence, such as the RGD amino acid sequence, in the penton base protein, can be produced in human embryonic cell line 293 (HEK 293), which has been shown to be able to replicate recombinant adenoviral particles lacking the RGD sequence, using previously described techniques (Bai et al. (1993), supra). A
recombinant virus, which lacks functional fiber proteins and in which the penton base acts as both the viral attachment and internalization protein, may be produced in cell lines which express the receptor to which the rcPB5 protein is targeted. For example, Ad5 rcPB5 cont~;n;ng an insert specific for ~3 can be produced in HeLa cells that express ~3 or 293 cells transfected with the gene for ~B3 .
The chimeric penton base protein may be present in the viral particle with the fiber, without the fiber (in which case the rcPB protein functions as the virus attachment and internalization protein), with a recombinant fiber molecule that is specific for a receptor, such as a cell-specific or tissue-specific W096/07734 ~ PCT~S9S/09904 receptor, or with a recombinant coat protein that is specific for a receptor, such as a cell-specific or tissue-specific receptor. See Examples 4-9 for a further description of these embodiments of the invention.
Recombinant adenoviruses of the present invention can be used to treat any one of a number of diseases by delivering to targeted cells corrective DNA, i.e., DNA
encoding a function that i8 either absent or impaired, or a discrete killing agent, e.g., DNA encoding a cytotoxin that, for example, is active only intracellularly, or DNA
encoding ribozymes or antisense molecules, for example.
Accordingly, use of the term "therapeutic gene" is intended to encompass these and other embodiments of that which is more commonly referred to as gene therapy and is known to those of skill in the art. Diseases that are candidates for such treatment include, for example, cancer, e.g., melanoma or glioma, cystic fibrosis, genetic disorders, and pathogenic infections, including HIV infection. For example, a recomhinant adenovirus having a penton base molecule recognized by ~ 3 receptors can be used to treat melanoma or glioma, and a recom.binant adenovirus recognized by ~3~1 receptors and expressing the cystic fibrosis tr~ncm~mhrane regulator (CFTR) gene can be used to treat cystic fibrosis by delivery to the epithelial cells of the lungs.
Furthermore, various blood-related diseases can be treated by using a recombinant adenovirus recognized by Qm~2 receptors to target neutrophils and macrophages, a recombinant adenovirus recognized by ~4~1 receptors to target lymphocytes, a recombinant adenovirus recognized by ~IIb~3 receptors to target plateletc and - megakaryocytes, and a recombinant adenovirus recognized by ~3 integrins to target endothelial cells undergoing anglogenesls .
One skilled in the art will appreciate that suitable methods of ~m~ni stering a recombinant adenovirus of the present invention to an ~nlm~l for purposes of gene W096107~34 ~ PCT~S95/09904 therapy (see, for example, Rosenfeld et al., Science, 252, 431-434 (1991); Jaffe et al., Clin. Res., 39(2), 302A (1991~; Rosenfeld et al., Clin. Res., 39(2~, 311A
(1991); Berkner, BioTechniques, 6, 616-629 (1988)), chemotherapy, and vaccination are available, and, although more than one route can be used to ~m;n; ster such a recombinant adenovirus, a particular route can provide a more immediate and more effective reaction than another route. Pharmaceutically acceptable excipients are also well-known to those who are skilled in the art, and are readily available. The choice of excipient will be determined in part by the particular method used to ~mln; gter the recombinant adenovirus. Accordingly, there is a wide variety of suitable formulations for use in the context of the present invention. The following methods and excipients are merely exemplary and are in no way limiting.
Formulations suitable for oral ~m~n;stration can consist of (a) liquid solutions, such as an effective amount of the compound dissolved in diluents, such as water, saline, or orange juice; (b) capsules, sachets or tablets, each containing a predetermined amount of the active ingredient, as solids or granules; (c) suspensions in an appropriate liquid; and (d) suitable emulsions.
Tablet forms can include one or more of lactose, mannitol, corn starch, potato starch, microcrystalline cellulose, acacia, gelatin, colloidal -~ilicon dioxide, croscarmellose sodium, talc, magnesium stearate, stearic acid, and other excipients, colorants, diluents, buffering agents, moistening agents, preservatives, flavoring agents, and ph~rm~cologically compatible excipients. Lozenge forms can comprise the active ingredient in a flavor, usually sucrose and acacia or tragacanth, as well as pastilles comprising the active ingredient in an inert base, such as gelatin and glycerin, or sucrose and acacia, emulsions, gels, and the W096/07734 ~ PCT~S95/09904 like containing, in addition to the active ingredient, such excipients as are known in the art.
The recombinant adenovirus of the present invention, alone or in combination with other suitable components, can be made into aerosol formulations to be ~m; n; stered via inhalation. These aerosol formulations can be placed into pressurized acceptable propellants, such as dichlorodifluoromethane, propane, nitrogen, and the like.
They may also be formulated as pharmaceuticals for non-pressured preparations such as in a nebulizer or an atomizer.
Formulations suitable for parenteral ~m; n; stration include aqueous and non-aqueous, isotonic sterile injection solutions, which can contain anti-oxidants, buffers, bacteriostats, and solutes that render the formulation isotonic with the blood of the intended recipient, and aqueous and non-aqueous sterile suspensions that can include suspending agents, solubilizers, thickening agents, stabilizers, and preservatives. The formulations can be presented in unit-dose or multi-dose sealed cont~;ners~ such as ampules and vials, and can be stored in a freeze-dried (lyophilized) condition requiring only the addition of the sterile liquid excipient, for example, water, for injections, immediately prior to use. Extemporaneous injection solutions and suspensions can be prepared from sterile powders, granules, and tablets of the kind previously described.
Additionally, the recombinant adenovirus of the present invention may be made into suppositories by mixing with a variety of bases such as emulsifying bases - or water-soluble bases.
Formulations suitable for vaginal ~m; n; stration may - be presented as pessaries, tampons, creams, gels, pastes, foams, or spray formulas cont~;n;ng~ in addition to the active ingredient, such carriers as are known in the art to be appropriate.

W096/0773~ 2 ~ PCT~S551~3~1 with the gene of interest, the composition employed, the method of administration, and the particular site and organism being treated. However, the dose should be sufficient to effect a therapeutic response.
In addition to the recombinant adenovirus of the present invention, the adenoviraI transfer vector also has utility in vitro. It can be used as a research tool in the study of adenoviral attachment and infection of cells and in a method of assaying receptor-ligand interaction. Similarly, the recombinant penton base protein comprising a nonpenton base amino acid sequence in addition to or in place of a wild-type receptor binding sequence, preferably the RGD sequence, can be used in receptor-ligand assays and as adhesion proteins in vitro or in vivo, for example.
The following examples further illustrate the present invention and, of course, should not be construed as in any way limiting its scope.

Exam~le 1 This example describes the construction of the adenoviral transfer vector (pAT) for making chimeric penton base molecules.
pAT, a partial restriction map of which is shown in Figure 1, was created by cloning the unique Bam HI/Pme I
fragment (13259-21561) from the Ad5 genome into pNEB 193 Rl-, a minor derivative of pNEB 193 (New England Biolabs, Beverly, MA), from which the unique Eco ~I restriction site was removed (SEQ ID NO:5~. The resulting vector was called pNEB 193 Rl- [Ad5 (Bam HI/Pme I)].
Two pairs of PCR primers were synthesized and used to amplify a region upstream (left side) and a region downstream (right side) from the RGD sequence. A unique Spe I site was inserte~ in~ the antisense primer (A5a(15147)S, SEQ I~ NO:14) u~ed to amplify the upstream region between the R~ sequence and a Bst XI site (15017). The Bst XI site is 35 bp upstream from the RECTIFIED SHEET (RULE 91) ISA/EP

W096/07734 ~ 8 ~ ~ PCT~S95/09904 penton base start codon (15052). A unique Spe I site was alsQ ~serted into the sense primer (A5a(15204)S, SEQ ID
~0:13~ to amplify the right side of the gene between the ~G~ amino acid sequence and an Asc I site (15671). The antisense primer (A5a(15659)) used ta ~plify the right side of the gene is shown as SEQ I~ NO:1 The sense primer (A5s(14153)E; SEQ ID NO:11) for the left side of the gene includes the naturaily occurring Bst XI site (15017) and also contains a unique Eco RI site 13 bp upstream from the penton base start codon, which is not in the original sequence. Using these primers, the left and right sides of the penton base gene were amplified and then cloned into pNEB 193 R1' [Ad5 (Bam HI/Pme I)] to create the pAT vector.
The net results of this cloning were the generation of a unique Eco RI site just upstream of the penton base start codon (SEQ ID NO:3) to facilitate subcloning into a baculovirus transfer vector and to facilitate the generation of a recom~inant adenovirus using the pAT
vector; the deletion of an 8 amino acid region (SEQ ID
NO:9) within the wild-type penton base containing the ~v integrin binding domain; and the incorporation of a unique Spe I site encoding the amino acids threo~i~e and serine in place of the deleted region (S~Q ID NO:~ so that alternative DNA sequences encoding receptor-binding amino acid motifs or other sequences of amino acids can be inserted at that site.

ExamPle 2 This example describes the construction of the base vector pRcPB5 (~RGD) for all cloning used to make a recombinant baculovirus for the expression of a recombinant chimeric Ad5 penton base protein (rcPB5 protein) and for making rcPB5 genes for incorporation5 into pAT and other related vectors.
pRcPB5 (^RGD), a partial restriction map of which is shown in Figure 2, was created by cloning the Eco RI/Bql RECTIFIED SHEET (RULE 91) ISA/EP

~- ~ S . ~ J .... ~ ; L '. L .. ~ ........ L .~
~ F 2 ~ ~ 8 ~ ~ ~

II fragmen~ from pAT intc the baculovirus transfer vec~or pAcSG2 ~pharmin~en~ San Diego, CA). A new sequence encoding a receptor binding damain is ln~erted into the unique Spe I Slt~ within the pen~on ba3e gene in pRcP~5 (~RGD). The re~lting vector i3 the~ u3ed to generate recombina~ baculovi~us con~truc~i so that rcPB5 Frotein can be generated in large quantit~es and e~aluated~
Based on evaluation of the rec~m~inant protein, the Eco RI/B~ ragment frsm the pRcP~5 deri~a~i~re i6 then cloned into pAT tO make the recombinant adencvirls (Ad5 rcP~).

Exa~le ~
T~is example describe~ the clonlng of integrin se~uences into t~e S~e I 5 'te of the pRcPB5 (~RG~) vector of Example 2.
A sequeAce recog~ized by a specifi~ integri~
recepror ig claned into t~e penton base cassette as follows. Overlapping sense and antisense oligcnuclectide~ encoding a desired integrin-specir~c sequence are synthesized. For example, the cverlapping sen6e ~Be~a 3RGD6, S~Q ID NO:l71 and antisense ~3eta 3RGDa, SEQ I~ ~O:l~) oligonucleotides that encode sequences that are speciic for the in~e~ri~ ~3 (SEQ ID
2~ NO:;5) a~d the overlap~ing ~en~e (~S-ls, SEQ ID NO:21) and antise~se (CS-la, S~Q ID NO:22) oligonucleotide that encode sequence~ that are specific for the integrin ~4~1 are sy~tne~ized. The ~en~e and antisense se~uence3 for each are mixed a~d co~verted to full-length double ~tranded ~equences ~y ~illing in the unpaired nucleotida~
with Klenow fragment. The resulting dau~le-stranded aequence ia then inserted into the Spe I site of the pRcPB5 (~RGD) ~ector following digestion with ~he endonuclea~es Xba I and SDe I. The resulting vec~ors, pRcP~5~a~3) and pRcpss ~4B1)~ are used t~ create t~e recombinant baculovirus ~ectors, AcNPV rcPB5~a~3~ and AcNP~ rcP~5(a4~l). The rcPB5 protein produced i~ in~ec~

At~EN~ED SHEET

W096/07734 ~ 8 ~ 1 PCT~S95/09904 recombinant baculovirus vectors, AcNPV rcPB5 (~3) and AcNPV rcPB5 (~4~1) . The rcPB5 protein produced in insect cells using the aforementioned baculovirus vectors is evaluated as described above. Based on the analyses in the baculovirus expression system, the Eco RI/Bql II
fragment from each of the vectors then is cloned into a pAT vector to produce the adenoviral vectors, pAT rcPB5 (~3) and pAT rcPBS (~4~1) .

Example 4 This example illustrates the creation of recombinant adenovirus particles containing a wild-type fiber protein and a chimeric penton base specific for ~4 integrins (Ad5 rcPB5 (~4) :wtFiber).
A chimeric penton base specific for ~4 integrin (Ad5 rcPB5 (~4) :wtFiber) can be constructed through ligation of plasmid DNA with viral DNA, which chimeric construct then can be transfected into cells using st~n~rd methods. The pAT vector cont~;n;ng the desired chimeric form of penton base is digested with Xmn I to generate a 1 kb piece of DNA cont~;n;ng the receptor-specific coding sequence. Ad5 DNA is digested with the restriction endonuclease Xmn I, which cuts wild-type Ad5 at positions 14561 and 15710 within the Ad5 genome. The two larger fragments are purified away from the smaller 1 kb piece cont~;n'ng the sequence encoding RGD and then ligated with the Xmn I fragment from the pAT vector. The ligated fragments are then transfected into the appropriate cell line to produce recombinant virus. The virus produced from the cells can be plaque purified and ~erified to contain the chtm~ric penton base gene u~ing PCR. The penton base produced in the Ad5-infected cells can be also purified, and its receptor-specificity verified.

Example 5 This example illustrates the creation of recombinant adenovirus particles containing wild-type fiber protein W096/07734 ~ 8 ~ ~ PCT~S9510~30 and chimeric penton base specific for ~4 integrins (Ad5 rcPB5 (~4) :wtFiber).
Adenovirus particles having a chimeric penton base specific for ~4 integrins (Ad5 rcPB5 (~4) :wtFiber) can be constructed through transfection of plasmid DNA and viral DNA into cells and generation of recombinants through homologous recombination. The pAT vector cont~;n~ng the desired chimeric form of penton base is linearized by digestion with Bam HI. AdS DNA is digested with the restriction endonuclease Xmn I which cuts wild type Ad5 at positions 14561 and 15710 within the Ad5 genome.
These two unique sites within the Ad5 genome flank the S~e I site in the pAT vector. The linearized pAT vector and the Xmn I-digested Ad5 DNA are then transfected into the appropriate cell line, such as 293 cells, to produce a recombinant virus by homologous recombination. The virus produced from such cells can be plaque purified and verified to contain the ch'meric penton base gene using PCR. The penton base produced in Ad5-infected cells also can be purified, and its receptor-specificity verified.

Example 6 This example illustrates the creation of recombinant adenovirus particles lacking fiber protein and chimeric penton base specific for ~4 integrins (Ad5 rcPB5 (~4 ):
-Fiber).
The Ad5 rcPB5 (~4) :wtFiber viral DNA can be digested with Bam HI and the 0-59 map unit (m.u.) fragment purified. The fiber gene can be deleted from plasmid cont~;nlng the Ad5 Bam HI/Sal I fragment (genome map units 59-100). The resultant plasmid is then digested with Bam HI and Sal I, ligated to the 0-59 m.u. piece of Ad5 rcPB5 (~4) :wtFiber DNA, and transfected into cells expressing ~4 integrins. The ~4 integrins can serve as both the attachment and internalization receptors for the Ad5 rcPB5 (~4) :-Fiber virus.

Example 7 This example iIl~3trate~ the creaticn of recombin~nt adeno~ u~ particles lack~ng fiber protein and containi~g chimeric penton base with an epitope reccgnized ~y a mcnocl~nal antibody ~a~tibodY-SPeC~iC sequence).
~ ira~ DNA is i301ated fro~ an Ad5 viru~ lacking fi~er, a3 in Example 6, and dige~t2d with ~he endonuclease X~n r as ~n ~xa~ple 5. The pAT ~ec~or containing tne chimeric penton base gene contai~ing t~e lo sequence encoding the anti~ody-speci~ic epitope i9 prepared a~ in Example 5. The dige~ted viral and pAT DNA
i~ transfected intc 253 cell~ to produce ~irus ky homolosous recomblnati~n. ~e recombinant ~ir~5 iS
passaged and a~plified by incubatin~ cell ly~aees with a ~ispecific an~ibody to av in~egrin~ ~expressed and ~nternalized in 2~3 cells) and the antibody-~pecific epitope ~n the chlmeric penton base, thereby allowing the recom~inant virus to attach and infect the 293 cells.
The virus proge~y from infected cell~ lack the bi~pe~ific antib~dy 50 th~t ~allowing their puriflcation fro~
infecte~ celL~ they can be targeted to ~IV-infected cells by combinin~ them wich a bispeci~ic antibody recognizing the antibody-specific epitope in the chimeric penton base and the HI~ gpl20.

Example ~
Thi~ example illustrates the creation of recombinant adenovirus particle~ containing chimeric fiber protein ~pecif ic for sialyl hewis X an~igen and cbimeric penton base s~ecific for ~4 integrins ~Ad5 rcPB5 (a4):rcFi~er The pr~tein P-selectin binds to ~iaIyl Lewis X
antigen present on lymphacyte~. The bi n~i n~ domain from P-sele~tin i9 pre~en~ on the rcFiber (~LX) protein to mediate viru3 binding. The left ar~ (0-59 m.u.) of the Ad5 r~PB5 (~4~:wtFiber YirU~ can be prepared ~y 3a~ HI
digestion and then purified.

A~IE~D SHE~

W096/07734 ~ 6 1 PCT~S95/09904 The rcFiber (sLX) gene can be cloned into the plasmid containing the 59-100 m.u. fragment of the Ad5 genome. The rcFiber (sLX) gene then can replace the wtFiber gene. The 59-100 m.u. Ad5 fragment containing rcFiber (sLX) then can be ligated to the 0-59 m.u. Ad5 fragment containing rcPB5 (~4). The ligated fragment then can be transfected into cells expressing the receptors for attachment (sialyl Lewis X) and internalization (~4 integrins) for the chimeric virus Ad5 rcPB5 (~4) :rcFiber (sLX).

Example 9 This example illustrates the creation of recombinant adenovirus particles containing chimeric hexon protein specific for sialyl Lewis X antigen, chimeric penton base specific for ~4 integrins, and lacking fiber protein [Ad5 rcPB5 (~4): -Fiber:rcHexon (sLX)].
The protein P-selectin binds to sialyl Lewis X
antigen present on lymphocytes. The binding domain from P-selection is present within an exposed loop on wtHexon, and the resultant rcHexon (sLX) protein can mediate virus binding. Genes encoding rcPB5 (~4) and rcHexon (sLX) can be cloned into the pAT vector that contains the penton base gene and the region encoding the exposed loop of hexon protein. A ligation of three fragments can be performed to generate the recombinant virus. Ad5 rcPB5 (~4) :-Fiber virus may be digested with Eco RI and Bam HI.
The left end of the genome up to the Eco RI site and the right end of the genome down to the Bam HI site can be purified and ligated with the Eco RI/Bam HI fragment from the pAT vector cont~1n~ng rcpBs (~4) and rcHexon (sLX).
The ligated fragment then can be transfected into cells expressing the receptors for attachment (sialyl Lewis X) and internalization (~4 integrins) for the chimeric virus Ad5 rcPB5 (~4) :rcFiber (sLX).

W096/07734 ~ PCT~S~5/~5v1 Example 10 This example describes the expression of wild-type Ad5 and mutant rcPB5 (~RGD) penton base protein in baculovirus.
s Using PCR, the wild-type Ad5 penton base sequence was amplified with an Eco RI site inserted at the 5' end of the gene. This gene was then cloned into the pAT
vector and then subcloned into the pAcSG2 baculoviral transfer vector, using methods known in the art.
Recombinant baculovirus encoding the Ad5 penton base gene (AcNPV wtP35) produced levels of penton base protein in insect cells that constituted approximately 50~ of the total cellular protein. The purified wtPB5 protein migrated consistently with respect to its known molecular weight on SDS-PAGE and can mediate cell adhesion like Ad2 penton base (Wickham et al. (1993), supra).
The core baculoviral transfer vector, pRcPB5 (^RDG), for expression of rcPB5 protein in the baculoviral expression system, has been used to make recombinant baculovirus, which expres~es rcPB5 protein. This rcPB5 protein, rcPB5 (ARGD), contains an 8 amino acid deletion in the RGD region as shown in Figure 2, which is a partial restriction map of the vector pRcPB5 (~RGD).
This protein migrates consistently with respect to its known molecular weight on SDS-PAGE and, in contrast to the wild-type protein (wtPB5), does not mediate cell adhesion, which i8 consistent with its lack of the RGD
cell adhesion motif. Chtmpric recombinant penton base molecules rcPB5 (~3) and rcPB5 (~4), each cont~;ntng inserts ~pecific for ~V~3 and ~4 integrins, respectively, also have been produced via the core baculoviral transfer vector, pRcPB5 (~RGD). The rcPB5 (~3) protein mediates epithelial cell (~+, ~4- ), but not lymphocyte (ov-, ~4+), adhesion; in contrast, RcPB5 (~4) supports lymphocyte, but not epithelial cell, adhesion.

W096/07734 PCT~S95/09904 Example 11 This example illustrates the targeting of adenovirus to lymphocytes.
The transfer vectors pAT (~V~3) and pAT (~4) were created by cloning the Eco RI/Bgl II fragments from their respective pRcPB5 vectors into the pAT (~RGD) vector.
These pAT vectors were then linearized by Bam HI
digestion and transfected with Xmn I-digested viral DNA
from an Ad5 variant, which lacks the E1 and E3 regions and contains a ~-galactosidase reporter gene under control of a CMV promoter, to obtain recombinant Ad5 virus, namely Ad5 rcPB (~RGD), Ad5 rcPB ( ~ 3), and Ad5 rcPB (~4 ), through homologous recombination. The Xmn I
endonuclease cuts the wild-type Ad5 at positions 14561 and 15710 within the genome, while the Bam HI/Pme I
fragment contained in the pAT vector spans the Ad5 region of 13258 to 21562. Thus, the regions of homologous overlap between the left and right viral sequences with the Bam HI/Pme I fragment are 1303 bp and 5851 bp, respectively. The presence of the chimeric inserts in the Ad5 rcPB (~3) and Ad5 rcPB (~4) viruses are verified through PCR.
The recombinant adenovirus Ad5 rcPB (~4 ) is analyzed for its targeting potential in relation to wild-type Ad5 and recombinant Ad5 rcPB (~RGD) by measuring levels of the reporter gene expression in cells exposed to each of the viruses. Delivery of ~-galactosidase to the B-lymphocyte cell line, Ramos, is significantly higher using the Ad5 rcPB (~4) virus than the Ad5 rcPB (^RGD) and wild-type Ad5 viruses. Conversely, delivery of ~-galacto-qida~e to an epithelial-derived cell line i9 significantly higher than the wild-type Ad5 virus in comparison to the Ad5 rcPB (~4) and Ad5 rcPB (~RGD) viruses. These results ~mo~trate that changes in penton base protein that redirect the protein to interact with a new receptor can be utilized to target the virus w096/07734 ~ B ~ PCT~Sss/o99o4 _ to new cells and tissues while specifically avoiding delivery to other cells and tissues.

All publications cited herein are hereby incorporated by reference to the same extent as if each publication were individually and specifically indicated to be incorporated by reference and were set forth in its entirety herein.
While this invention has been described with emphasis upon preferred embodiments, it will be obvious to those of ordinary skill in the art that the preferred embodiments may be varied. It is intended that the invention may be practiced otherwise than as specifically described herein. Accordingly, this invention includes all modifications encompassed within the spirit and scope of the appended claims.

SEQU~NCE ~9~ING

~; ~ Ga;~1ER~L INFo~MATIoN
( i) A~PLICANT: GEt~EC, In~-TrT~E OF ~ ;h,ioN: P~NTON 3RgE PROT}3IN AND hETHO~S OF ~SItlG SA~E
NUMBER OF SEQtlENCES: ~2 iv) COMPU~ E~EADA~ls FO~M:
~A) ~SEDIU~ TYP!3: Floppy diaX
(B) C:OMPU~ER: IH~ ~C co~pa~ible (C) OPE~lATINt~ SYST~ Pc-Dosll~s-l:)cs ~D~ SO~ E: PatentIn Releaae 1.0, Verolcn 1.25 ~PO~
v~ CD~h. APP~ICATION DATA:
tA) APPE~QTroN NUHB~: Wo pcTtus95/ossa4 ~8~ FI~IYG DATE: 01-AUG-L99, vi) PRIOR P.PPL~ lON DATA:
~A) APPLIQTI~N NU~ER: US 3a3162 ( 8 ~ FI ING DATE: OJ-S~3P-1994 2 ) INl~ 5ATION FOR SE~ ID NO: 1.
i~ 9E52UENC~5 CaMACTEP~ISTICS:
A ) ~NG1~1: 5 a~in~ acids (~ TYPE: amino acld (C) ST~ANDEDN~SS: ~ gle D ~ TOPOI.OGY: L lneaL~
~r ~C~ PrOt-in (Xi) 9EQU15NC~5 D3Sr~TPSION S~Q ID NO~1 r Yg G1n A1a G1Y A-P

( 2 ) INFORI~ATION FOR S~Q ID NO: 2:
U~ C~UAC~ STICS 1 ~A~ LI5NGTR: 5 am1nO aC~d~
(3) SYP~ nO aC1t ( C ) STR~U~UNe~g ~ ~ing1 ~ D ) TOP~ ne~r MOLE;C~ ; P~Otein ) SEQUBNC~ DBSCR~PT10-~: S~ rD Na:2:
G1U I1Q Leu A~p Val F~!r~fn ~HF~T ~

^ ~ t . ~; I .1. Li L . _ I J . `_,, " .~ . L ~

( Z ) INFOR~ TION EOR SEQ ~D NO: 3:
i ) Sl!~Qt~ÇNC2~ C~ CTERISTICS:
~1 LENGT~: ~S ba30 palrs ~B~ TYP~: nucl~ic acid C) S'rRANo~uN~SS ~ double ~D) T~P~r.OqY: llnlrn.~,,n ii ) MODECU~E TYPE; DNA ~ g~

~x) FE~AT~
A ) NAM~ S
~ B ) LOCATI:ON: 51. . 65 (xi) Si5~2~NC15 DBSC:RIPT~ON: SEQ ID NO: 3:
CTCACCl~AGG C:CATGGC~CC GAGC~ AT LCCCC~T ATG C~ 56 !Set ~q ~rg Aia Ala ( 2 ) :~NFQEWA~ oN FOR SBQ ID ~C: 4:
L) SEQIJ15NC~S CaM~CTER~SSrCS:
(A) r.ENGTH; 5 amino acid~
~B) TYPB~ a ~cit D ) ~OPODOGY: 1 Lnear ~ii) M~-T~'CUT~ TYP~: protei~
(xi) S~:QUENCD DESCRIPTrON; SEQ ID NO:~:
Me~ Arg Ala Ala S

12) l~ru -l~ssoN FO~I SISQ ID NO:5:
~i~ s~:o~ cE C~ARACSlS~rS~ICS:
~A) T3NGSE~ 65 ba~a pairs ( 9 ) ~PE: ~u~lel~ acid C ) Sl~ANDED~:SS: dou~l~
(D ) TOPOLOGY ~ r ( i~ ) ~r ~C~r ~ DNA ~ genomic ) ( iX ) Fl;ATURE
~A) r~3/ltEy; CDS
9 ) OCASSCN I 51. . 65 D S~F~T ~

. ~

(xi1 SEQU~NC~ DEsc~I2TToN: SEQ ~D NO:5:
CSCACCAA~G OcA~GGCAC~ GAGC~-T~ AT L'~CC~LAGT ATG C:~ 56 ~et ~g Arg ALa Ala ~2~ ~NFOR~ATION ~R SEQ ID NO:6:
( 1) SEQUENCB cHlu-a~ RIsTIcs: --tA~ L~al~: ~ a~ino acidc 3 ) TY?3~; amino acid D ) ~O~O~O&Y: 1 inear ( ii) ~OL~:C~ ~PE: protoin ( xl ) ~ ~ DESC~lPTION: SEQ ~D NO: 6:
g ala Ala 2 ~ INFORMATIO~ YOR SE~Q I3 NO2 7:
;uu~ s C~R~CT~UT~C~:
~A) 3ENGT~: 30 ba~e pair~
B ) T~ s nucleic acid ~C~ 5~ KCS: double (D ~ TO~WGY; Lnl~ .A
O~CU~S TYP25: DNA ~g-no~

( iX ~ FlSATU~:
~A~ Nl~/X~Y: CDS
B ) ~ 1..30 ~Xi I gEQlJUlCa~ D~3SC~ ON: Sl~ ~D ~10: 7:
GAC A~G AAC GAS AC~r AGT GCC ACa CGG G~ 30 A~p ~e~ A~n A~y ~hr ser Ala ~hr Arq Al~

( 2 ) INFO~TION POR ~lsq ~ NO~
( i ) S t~" b ~ CHA~Cl~RI8'TI CS:
~A) L3NGT~: 10 ~lno ~cidc ~B) SY~B: amLno acLd ~D) TOPOrOGy; llnear 1~3 M~OUT~ TYPB: protein : AMENl!ED SHEET

(xi ) S~;QU3NC~ DESC~IPSION: S~Q ID NO: ~:
Asp Het A~n Aep Thr Ser Ala Thr ~rg Ala NFOR!~ATION FOR SEQ ID NO:9:
( i ) SEQUISNCE ~MACTE~IsrIcs:
t~ LENG~H: ~ ba~e pairs ~B) TYP~3: nucl~i~ acid ( C ) STPAND~DNESS; do~bl~
D ) TOPOr OGY: unlc~awn Li ~ ~OLECU~E TYP~:: DNA ~ genomic ) ix ) FEA'rU~E;
( A ) NAY.E ¦ J~3Y: CD S
(8) ~ATION: 1..48 (xi) SEgu~5N-,c DE8C~IPTION: SEQ ID NO~
GAC ATG AAC GAT C~ GCC A~T CGC GGC GAC AC~ ~T GCC ACa CGG GCT 43 A~p ~et Aan Aap ~iA ~la Ile A~g Gly A~p 'Shr Phe Ala Thr Arg ~la ( 2 ) Lc~rO~ATION FOR SEQ ID NO: la ~L) ~ u~r~ ~p~ TsT~cs!
~A) r-ENOT~: 16 a~Lno acLdo ~) TYP13: amino acid ( D ) TopoI~oGr ~
r ~r ~ ~YPE: ~rotein ~xi~ SBQm~CB D~C~IPSION: SEQ I~ ~O: lO:
Acp Met Aan A~p ~i- Ala rI~ A~g Gly ~}j~ rhr ~ho Ala Shr Arg Ala 2 ) LN~ Io~ SISQ lD NC~
( ~ ~ g~U~ C;~ D~ I8TrCSl ~A) ~EIiG~EIs 5a ba~- p~irY
~S) TYPE: nuc1et c a~ld ( C ) ss~ANDEI~Rr:s9: ~inglc 0Pot~oGy ~ un Irn~ ,a nE~G~ TYPE: ~A (g~rJa~ic ) (xt ) SEQUENCB DESC5~IPTIOI~: SEQ ID NO:ll:
AM~CTQCC l~ CCr~TGG r~CCA~CG~ TG~.~ ,.. GAA.. 'C~CC;. TA 52 -t~ ~ st~

._ (2) INF~1~ASIC~N FOR SEQ ID NO:12:
v~ CT~SR~STICS:
( A ) ~ENGTH: 2 5 ba~e pai~
~) TYPE~ nucl~ic acid ~C) STE~RDEDN~35S: ~ ngle ~ ~ ) TOPO~O~;Y: unk~ wn IOL}~CU~E rYPE; DNA ~ gencmic ) ~xl~ SEQUENOE D~srgrpTIoN S~Q I~ ~:0:12:
CTGGa;GGCG CGCCA~AATC T~GTT 2 5 ( 2 ) I~F~ ATIor~ Fa~ ~Q ~D NO: 13:
t i ) S~4u ~ ctCE C~ARA~13R~STIC9 NGT~; 26 ba3~ paLr3 ~B) TYPE: nucleic acld ~ C ) S'rRANDlQ~NESS: 9ingle ( D ) TOPoLoGr: unkno~n (Li) I~ TYP2:~ DNA ~g~nomic) txL) S~QUl3~CF DESCRlPTION: S~Q ~D NO: 13;
~AAACrAGTG C~'~CACGGGC 'rGAG&A 2 ( 2 ) IN~IAT~ON JOR SSQ ID No: 14:
~i) SEQUENC~t CEARACTi~ISTICS:
~A) ~NG'r~: 26 ba~ pair~
~EI) TYPIC: nucleic acid C ~ S~AlilDleDNE~: ~ingle roPol,ocY ~ u~
SOE~CU~ SYP~: DNA (~r.:. ic~
(xL~ SEQUENCZ D~S~aTPSrQ~: S~Q ID No: 14;
A~AAI-T~-T~ ~ SGT CCSCCA 2 6 Z ) IN~A~ON PO~ S15~. ID NO:15:
i) SEQUENC$ C~ S~
~A~ L13NG~H: 68 ba~ palro ) TYP~: nuc lelc acid c ) STRA?~D~D~BSS: dol~b 1 c D ~ SOPOWGY un~cnown 11 ) MQ~,~:C~L~ ~YP~: DNA ~ geno~ic ) ix ) P~SATUR~ s ~A) ~a~lR~: CDS
t D ) ~OCATlON: 3 . . 68 bt~ ~, xi~ S~Q~ O~scR-pT7-oN: SEQ ID NO:15:
~A AC'r AGT G~;A GG~ ~GC TC~ TTT GGC CGC ~;GC GAC ASS CGC AAC ~GC 4 Thr ~er Gly Gly CyY Ser P!~e Gly Ar~ GLy A3p ;le Ar~ A~n Cy3 li GGC GGC CTG CAG TC'r AGA AAA 68 Gly Gly ~eu Gln Ser Arg Lye 2 ) INFOR~tATION E'C~ SEQ ID NO . 16:
(i) S~u~n~l: C~CTE:}~ISTrCS:
(A~ ~ENG~H: 22 a~ino acLda ~ 8 ) TYP~: a~lno acid ( D ~ 'rOPOI.06Y ~ Rar OI.i3CU~ TYPE: prot~i~
(xi~ s~Q-J. .~; D~SSCR~PTION: SEQ ~D Na.16:
rhr Ser Gly Gly Cya Ser Phe ~ly Arg Gly Asp Ile ~rg A~n Cy~ Gly Gly ~eu Gln S~r A~g ~yY

2 ) INFOR~ATrON FOR SEQ rD N<~- 17:
L) S2QU~3NC~3 CaARAC~IS'rrCS;
~A~ L~ENGT~ 41 ~a~-~ pairJ
3 ~ 5YPE: nucl~lc acid ~C~ S'lRANDXDNlSSS: ~ingle ~D ) TOPODXY ~ C..
~ Li ~ MOL~CU~ TY~B: D~A ( g~n 1 ~ ) (xL) SBQUINC~ D2~ IpTIoll: S~5Q ID NOtl7:
~AACTAGTGG AMASGCSCS ~ _ ~;CC~ GCGA~TSCG C 11 (2) IN~_LASIO~I FOR S~3Q ID ~la~
) S3QU~ c~cTERlsT~c3:
~A) L2NG~H: 42 bas~ pai.r~
nucl~lc a~d C ) STP~ ntn7-~s: ~i~gle t D ~ TOPOSO~Y ~
( i i ) ~SOL~:C~; TYPE: DNA ~ ~7~ - - i C ) (xi) Sl:QUENC9 DEgCRIPT~o~: ~BQ ID ~rc: 18:
AGAC ~;CA~;C~ CGCAGT~O~ AA~GTCGCCG CG 42 p,~JlE~3 S~

% ~ ~ ~

2 ) IN FORHATION FOR S~Q ID NO: 19:
u~ C~RACTERIST~cS;
~A) LE~GT~l: 110 ba~e pairs TYP~S: nucl~lc ~cid C ) STRA~ UP~ CS: doubl~
~1:1 ) Top~L~Gy 1~ nl~
OLLCl~ TYPE: DNA (~enc~m~c ) ?12ATUR~:
~A) ~ 13/XISY: CDS
t B ) LO~ATIQN: 3 . . 110 s3QuE~rcE DE;SC~IP~ION: SEQ ~D NO: l9:
~A AC~ AG'r ~;A GGA GAC GhG C~C ccA ~AG crc GTG ACS cTC C~ C 47 Thr s~ Gly Gly A~p Glu r~u Pro Gln ~e~ ~al Thr Le~ Pro Kl3 CQ AAC C~C CAC G~:T CCA GAG ATT CTC t:AC GSG CCA TCT ACT GGC GGC 95 Pro A~rl Leu Hi~ Gly Pro Glu Ile ~eu A~p val P~o .S~r Shr Gly GLy L~u Gln Ser Arg r.y~

l 2 ) T~ 2fATlON FOR SLQ LD NO: 20:
SEQt~NC:~ C~ rSTICS:
~A) ~ 36 a~ino aaid4 ~) TY~: a~lno ac~d ( D ) SO~
ii ) HO~ ~CU~ protein ~xi~ S~ 5NCL ~ CTP~ION; S~2 I~ NO:20:
hr Ser Gly GLy a~p G~u L--u Pr~ Gln ~eu Val Thr Leu ~r~ Eli~ ~r~
~n te~ Hin Gly Pro Clu Ile Leu A~p Val Pro 9cr Thr Gly Gly I,eu ln Ser Arg Ly~

~2~ F:lA'rION FOR 8E9 ID NO:21: _ L 1 S~QtJr',~ U~Y~T~c~:
~A) L~GTH: 65 ba~- palr~
(B) ~ I: nuc1~lc aci~
~ C ~ ST~ND~DN~
~o ) ToPo~oGr, ~ ~Q
~il) ~OI,~SCUT~15 ~ DNA ~g~ MFNnEn SH~T

L L . _ ~ J ~ lr~ J

(xi~ S~ n~ 0~5SC~IPT~O~: S~Q ID NO:21:
A~AcTAGrGG A'~ P^~CG~G CTCCCACAGC TCGT~ACTCT CCCAC~CCC~ A~CCTCCACG 60 ~2) INFOR~AT~ON FOR SEQ ID NO:22;
( i) 513~UENC~: C~ACTERISTICS:
( A ) ~ENGT~ ~ 60 ba~e pair~
3 ) TYPE: nucleic acid C ~ STRAN~5Dr~e~S: ~ ~ ngl e ~ D ) TOPOI,OGY: unknown oL~C~B TYP~: D~ (genc~nlc~
~xi~ SEl~ DR5C~I~TION: SEQ ~D t:O;22:
,,AGAC T~CAGGCC~C CAGTAG~C6 CACGTCGAGA A ~ AC CGTGGAGGTT 60 AMEN~EO S~F~T

Claims (39)

WHAT IS CLAIMED IS:

1. A recombinant adenovirus comprising a chimeric penton base protein, wherein said chimeric penton base protein comprises a nonpenton base amino acid sequence, which is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, in addition to or in place of a wild-type penton base amino acid sequence.

2. The recombinant adenovirus of claim 1, where said wild-type penton base amino acid sequence is an RGD
amino acid sequence.

3. The recombinant adenovirus of claim 1, wherein said recombinant adenovirus is fiber-deficient.

4. The recombinant adenovirus of claim 2, which further comprises, in addition to the chimeric penton base protein, a chimeric coat protein which includes an amino acid sequence that is (i) pecific for binding to a receptor, (ii) an antibody domain or (iii) an epitope.

5. The recombinant adenovirus of claim 4, wherein said amino acid sequence in said chimeric coat protein, which is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, (i) is specific for binding to the same receptor, (ii) is the same antibody domain or (iii) is the same epitope as the nonpenton base amino acid sequence in said chimeric penton base protein.

6. The recombinant adenovirus of claim 4, wherein aid chimeric coat protein is selected from the group consisting of a chimeric fiber and a chimeric hexon.

7. The recombinant adenovirus of claim 6, wherein said amino acid sequence in said chimeric coat protein, which is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, (i) is specific for binding to the same receptor, (ii) is the same antibody domain or (iii) is the same epitope as the nonpenton base amino acid sequence in said chimeric penton base protein.

8. A recombinant baculovirus comprising a chimeric adenoviral penton base gene sequence encoding a chimeric penton base protein, which chimeric adenoviral penton base gene sequence comprises (a) a nonpenton base gene sequence encoding an amino acid sequence that is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, and (b) a wild-type penton base gene sequence.

9. The recombinant baculovirus of claim 8, wherein said wild-type penton base gene sequence encodes an RGD
amino acid sequence.

10. The recombinant baculovirus of claim 9, wherein the gene sequence encoding the RGD amino acid sequence has been mutated.

11. The recombinant baculovirus of claim 10, wherein the gene sequence encoding the RGD amino acid sequence has been mutated by insertion of the nonpenton base gene sequence into the gene sequence encoding the RGD amino acid sequence.

12. The recombinant baculovirus of claim 9, wherein the gene sequence encoding the RGD amino acid sequence has been moved within the penton base gene sequence such that in the chimeric penton base protein the RGD amino acid sequence is conformationally inaccessible for binding to a receptor.

13. The recombinant baculovirus of claim 12, wherein the gene sequence encoding the RGD amino acid sequence has been moved within the penton base gene sequence by insertion of the nonpenton base gene sequence near the gene sequence encoding the RGD amino acid.

14. A recombinant baculovirus comprising a chimeric adenoviral penton base gene sequence, wherein said chimeric adenoviral penton base gene sequence includes a nonpenton base gene sequence, which encodes an amino acid sequence that is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, in place of a wild-type penton base gene sequence.

15. The recombinant baculovirus of claim 14, wherein said wild-type penton base gene sequence encodes an RGD amino acid sequence.

16. A chimeric adenoviral penton base protein comprising a nonpenton base amino acid sequence, which is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, in addition to or in place of a wild-type penton base amino acid sequence.

17. The chimeric adenoviral penton base protein of claim 16, wherein said wild-type penton base amino acid sequence is an RGD amino acid sequence.

18, A viral transfer vector comprising a chimeric adenoviral penton base gene sequence encoding a chimeric penton base protein, which chimeric adenoviral penton base gene sequence comprises a nonpenton base gene sequence encoding an amino acid sequence that is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, in addition to or in place of a wild-type penton base gene sequence.

19. The viral transfer vector of claim 19, wherein said viral transfer vector is an adenoviral transfer vector.

20. The viral transfer vector of claim 19, wherein said wild-type penton base gene sequence encodes an RGD
amino acid sequence.

21. The viral transfer vector of claim 20, wherein the gene sequence encoding the RGD amino acid sequence has been mutated.

22. The viral transfer vector of claim 21, wherein the gene sequence encoding the RGD amino acid sequence has been mutated by insertion of the nonpenton base gene sequence into the gene sequence encoding the RGD amino acid sequence.

23. The viral transfer vector of claim 20, wherein the gene sequence encoding the RGD amino acid sequence has been moved within the penton base gene sequence such that in the chimeric penton base protein the RGD amino acid sequence is conformationally inaccessible for binding to a receptor.

24. The viral transfer vector of claim 23, wherein the gene sequence encoding the RGD amino acid sequence has been moved by insertion of the nonpenton base gene sequence near the gene sequence encoding the RGD amino acid sequence,

25. A prokaryotic or eukaryotic expression vector comprising a chimeric adenoviral penton base gene sequence encoding a chimeric penton base protein, which chimeric adenoviral penton base gene sequence comprises a nonpenton base gene sequence encoding an amino acid sequence that is (i) specific for binding to a receptor, (ii) an antibody domain or (iii) an epitope, in addition to or in place of a wild-type penton base gene sequence.

26. The prokaryotic or eukaryotic expression vector of claim 25, wherein said wild-type penton base gene sequence encodes an RGD amino acid sequence.

27. The prokaryotic or eukaryotic expression vector of claim 26, wherein the gene sequence encoding the RGD
amino acid sequence has been mutated.

28. The prokaryotic or eukaryotic expression vector of claim 27, wherein the gene sequence encoding the RGD
amino acid sequence has been mutated by insertion of the nonpenton base gene sequence into the gene sequence encoding the RGD amino acid sequence.

29. The prokaryotic or eukaryotic expression vector of claim 26, wherein the gene sequence encoding the RGD
amino acid sequence has been moved within the penton base gene sequence such that in the chimeric penton base protein the RGD amino acid sequence is conformationally inaccessible for binding to a receptor.

30. The prokaryotic or eukaryotic expression vector of claim 29, wherein the gene sequence encoding the RGD
amino acid sequence has been moved by insertion of the nonpentan base gene sequence near the gene sequence encoding the RGD amino acid sequence.

31. The recombinant adenovirus of any of claims 1-7, wherein said adenovirus further comprises a therapeutic gene that is capable of being expressed in a cell to which the recombinant adenovirus has attached or by which the recombinant adenovirus has been internalized.

32. The recombinant adenovirus of claim 2, wherein said wild-type penton base amino acid sequence is specific for binding to a receptor,

33, The recombinant baculovirus of claim 8, wherein said wild-type penton base gene sequence encodes an amino acid sequence that is specific for binding to a receptor.

34, The recombinant baculovirus of claim 14, wherein said wild-type penton base gene sequence encodes an amino acid sequence that is specific for binding to a receptor.

35. The chimeric adenoviral penton base protein of claim 16, wherein said wild-type penton base amino acid sequence is specific for binding to a receptor.

36. The viral transfer vector of claim 18, wherein said wild-type penton base gene sequence encodes an amino acid sequence that is specific for binding to a receptor.

37. The prokaryotic or eukaryotic expression vector of claim 25, wherein said wild-type penton base gene sequence encodes an amino acid sequence that is specific for binding to a receptor.

38. A method of gene therapy comprising administering to a patient in need of gene therapy a therapeutically effective amount of a recombinant adenovirus of any of claims 1-7, 31, and 32.

39. The recombinant adenovirus of any of claim 1- 7, 31, and 32 for use in preparing a medicament.