CA2114800A1 - New conjugates for the introduction of nucleic acid into higher eukaryotic cells - Google Patents

Abstract

Abstract Conjugates in which a virus is bound via an antibody to a substance having an affinity for nucleic acid, for transporting gene constructs into higher eucaryotic cells. Complexes of the conjugates and nucleic acid are internalised in the cell, possibly with the co-operation of a cell-type-specific internalising factor, whilst the virus as part of the complex brings about the internalisation and the release of the contents of the endosomes, in which the complexes are located after entering the cell. Pharmaceutical preparations in which the nucleic acid is a therapeutically active gene construct, particularly for use in gene therapy, and transfection kits.

Description

21~800 S014-519.572 New Conjugates for Introducing Nucleic Acid Into Higher Eucaryotic Cells The invention relates to the introduction of nucleic acids into higher eucaryotic cells.
There is a need for an efficient system for S
introducing nucleic acid into live cells particularly in gene therapy. Genes are introduced into cells in order to achieve in ViVO synthesis of therapeutically effective genetic products, e.g. in order to replace the defective gene in the case of a genetic defect.
"Conventional" gene therapy is based on the principle of achieving a lasting cure by a single treatment. -~
However, there is also a need for methods of treatment in which the therapeutically effective DNA (or mRNA) is ~ ;~
administered like a drug ("gene therapeutic agent") once - D
or repeatedly as necessary. Examples of genetically ~
caused diseases in which gene therapy represents a ; -promising approach are hemophilia, beta-thalassaemia and "Severe Combined Immune Deficiency" (SCID~, a syndrome caused by the genetically induced absence of the enzyme adenosine deaminase. Other possible applications are in immune regulation, in which humoral or intracellular immunity is achieved by the administration of functional nucleic acid which codes for a secreted protein antigen or for a non-secreted protein antigen, by immunization.
Other examples of genetic defects in which a nucleic acid which codes for the defective gene can be administere~, e.g. in a form individually tailored to the particular requirement, include muscular dystrophy (dystrophin gene), cystic fibrosis (cystic fibrosis transmembrane conductance regulator gene), hypercholesterolemia (LDL receptor gene). Gene-therapy methods of treatment are also potentially of use when hormones, growth factors or proteins with a cytotoxic or :

~ ~ .

immune-modulating activity are to be synthesized in the body.
Gene therapy also appears promising for the treatment of cancer by administering so-called "cancer vaccines". In order to increase the immunogenicity of tumor cells, they are altered to render them either more antigenic or to make them produce certain im~une modulating substances such as cytokines in order to trigger an immune response. This is accomplished by transfecting the cells with DNA coding for a cytokine, e.g. IL-2, IL-4, IFN-gamma, TNF-alpha. To date, gene transfer into autologous tumor cells has chiefly been accomplished via retroviral vectors.
The mode of activity of antisense RNAs and DNAs as well as ribozymes enables them to be used as therapeutic agents for blocking the expression of certain genes (such as deregulated oncogenes or viral genes) in vivo.
It has already been shown that short antisense oligonucleotides can be imported into cells and exert their inhibiting effect therein (Zamecnik et al., 1986), even if their intracellular concentration is low, caused, inter alia, by their restricted uptake by the cell membrane as a result of the strong negative charge of the nucleic acids.
Various techniques are known for gene transfer into mammalian cells in vitro but their use in vivo is limited (these include the introduction of DNA by means of liposomes, electroporation, microinjection, cell fusion, DEAE-dextran or the calcium phosphate precipitation method).
In recent times, biological vectors have been developed to bring about the transfer of genes by using the efficient entry mechanisms of their parent viruses.
This strategy was used in the construction of ;
recombinant retroviral and adenoviral vectors in order to achieve a highly efficient gene transfer in vitro and in vivo (Berkner, 1988). For all their efficiency, ::.: :. , ' 8 ~ 0 these vectors are subj ect to restrictions in terms of the size and construction of the DNA which is transferred. Furthermore, these agents constitute safety risks in view of the co-transfer of viable viral gene elements of the original virus. Thus, for example, the use of retroviruses is problematic because it involves, at least to a small percentage, the danger of side effects such as infection with the virus (by recombination with endogenous viruses or contamination with helper viruses and possible subsequent mutation into the pathogenic form) or the formation of cancer.
Moreover, the stable transformation of the somatic cells of the patient, as achieved by means of retroviruses, is not desirable in each case because this can only make the treatment more difficult to re~erse, e.g. if side effects occur.
In order to circumvent these restrictions, alternative strategies for gene transfer have been developed, based on mechanisms which the cell uses for the transfer of macromolecules. One example of this i~
the transfer of genes into the cell via the extremely efficient route of receptor-mediated endocytosis (Wu and Wu, 1987, Wagner et al., 1990 and EP-Al 0388 758). This approach uses bifunctional molecular conjugates which have a DNA binding domain and a domain with specificity for a cell surface receptor (Wu and Wu, 1987, Wagner et al., 1990). If the recognition domain (hereinafter referred to as the "internalizing factor") is recognized by the cell surface receptor, the conjugate is internalized by the route of receptor-mediated endocytosis, in which the DNA bound to the conjugate is also transferred. Using this method, it was possible to achieve gene transfer rates at least as good as those achieved with the conventional methods (Zenke et al., 1990) .
Whereas this vector system is able to transport large quantities of DNA into cells having the suitable , ~

` 2~4~0 cell surface receptor, the corresponding gene expression very often does not accord with the transfer capacity (Cotten et al., 1990). It was assumed, inter alia, that the reason for this phenomenon is that the DNA conveyed into the cell by receptor-mediated endocytosis lands in lysosomes where it undergoes degradation (Zenke et al., 1990, Cotten et al., 1990~. Therefore, the fact that the DNA internalized in lysosomes does not have any specific mechanism for leaving the intracellular vesicle system constitutes a restriction which is inherent in this transport system.
The aim of the present invention was to reduce or eliminate these restrictions.
A plurality of viruses effect their entry into the eucaryotic host by means of mechanisms which correspond in principle to the mechanism of receptor-mediated endocytosis. Virus infection based on this mechanism generally begins with the binding of virus particles to receptors on the cell membrane. After this, the virus is internalized into the cell. This internalizing process follows a common route, corresponding to the entrance of physiological ligands or macromolecules into the cell: first of all, the receptors on the cell surface arrange themselves in groups, to form a so-called "coated pit", and the membrane is inverted inwardly and forms a vesicle surrounded by a coating.
After this vesicle has rid itself of its clathrin coat, acidification takes place inside it by means of a proton pump located in the membrane. This triggers the release of the virus from the endosome. Depending on whether the virus has a lipid coat or not, two types of virus release from the endosome were taken into account: in the case of so-called "naked" viruses (e.g. adenovirus, poliovirus, rhinovirus) it was suggested that the low pH
causes changes in conformation in virus proteins. This ~
exposes hydrophobic domains which are not accessible at ~ ~-the physiological pH. These domains thus acquire the 21~800 .. ..

ability to interact with the endosome membrane and thereby cause the release of the virus genome from the endosome into the cytoplasm. As for viruses with a coat (e.g. vesicular stomatitis virus, Semliki Forest virus, influenza virus) it is presumed that the low pH modifies the structure or conformation of some virus proteins, thereby promoting the fusion of the virus membxane with the endosome membrane. Viruses which penetrate into the cell by means of this mechanism have certain molecular peculiarities which enable them to break up the endosome membrane in order to gain entry into the cytoplasm.
Other viruses, e.g. the coated viruses Sendai, HIV
and some strains of Moloney leukaemia virus, or the uncoated viruses SV40 and polyoma, do not need a low pH-milieu for penetration into the cell; they can either bring about fusion with the membrane directly on the surface of the cell (Sendai virus, possibly HIV) or they are capable of triggering mechanisms for breaking up the cell membrane or passing through it. It is assumed that the viruses which are independent of pH are also capable of using the endocytosis route (McClure et al~, 1990).
In experiments which preceded the present invention it was established that gene transfer by means of nucleic acid complexes in which the nucleic acid is complexed with polycations, optionally coupled to an internalizing factor, e.g. with transferrin-polylysine conjugates, is significantly increased by treatment with adenoviruses, specific retroviruses or with virus fragments. This effect was achieved by making use of I the phenomenon that these viruses are taken up into the cells by endocytosis mechanisms and have a specific mechanism for escaping from the vesicle system by breaking open the endosomes, e.g. in the case of the adenoviruses (Pastan et al., 1986).
Starting from these observations, the problem of the invention was solved by developing a bioconjugate which contains the virus as an integral part of its functional construct.
The invention thus relates to a conjugate which has the ability to form complexes with nucleic acid and which comprises an internalizing factor and a substance having an affinity for nucleic acid, for introducing nucleic acid into higher eucaryotic cells. The conjugate is characterized in that the internalizing factor is a virus which is bound to the nucleic acid-binding substance via an antibody in such a way that it is capable per se of penetrating into the cell as part of the conjugate/nucleic acid complex and of releasing the contents of the endosomes, in which the complex is located after entering the cell, into the cytoplasm.
The invention in a further aspect relates to complexes in which the conjugates according to the invention are complexed with nucleic acid.
The ability of the virus to penetrate into the cell ~-and release the content of the endosomes, in which the conjugate/nucleic acid complex is located, into the cytoplasm, is hereinafter referred to as the "up take function".
The conjugates according to the invention combine the advantages of vector systems based on internalizing factor conjugates with the advantages which the viruses bring into these systems.
Compared with gene transfer by receptor-mediated endocytosis, the virus-polycation-DNA complexes according to the invention have the advantage that they circumvent the fundamental restriction inherent in the known molecular conjugate systems, in that, unlike the ~
known conjugates, they have a specific mechanism which ~--enables them to be released from the cell vesicle system. Compared with biological vectors, the vector ~
system according to the invention constitutes a -fundamental conceptual departure from the recombinant viral vectors, in that the foreign DNA which is to be transported is carried on the outside of the virion.

- 21~4800 Consequently, the conjugates according to the invention can transport very large gene constructs into the cell, with no restrictions of any kind as to the sequence.
Suitable viruses include, on the one hand, those which are able to penetrate into the cell by receptor-mediated endocytosis and to bring about their release -and hence the release of the nucleic acid - from the endosome into the cytoplasm. (The suitability of viruses within the scope of the present invention is further defined in that they retain this property even when they are a component of the nucleic acid complexes). Without wishing to be tied to this theory, this mechanism could benefit the nucleic acid complexes transferred into the cell in so far as the ability of the virus to release the contents of the endosomes prevents the fusion between the endosomes and lysosomes and consequently prevents the enzymatic decomposition which normally occurs in these cell organelles.
The higher eucaryotic cells are well known and do not include yeast. (Watson et al., 1987). Examples of higher eucaryotic cells capable of adenovirus infection are described by Fields and Knipe, 1990.
Viruses whose uptake function, occurring at the start of infection, occurs by receptor-mediated endocytosis and which are suitable as part of the conjugates according to the invention by virtue of this property, include on the one hand viruses without a lipid coat such as adenovirus, poliovirus, rhinovirus, and on the other hand the enveloped viruses vesicular stomatitis virus, Semliki Forest virus, influenza virus;
pH-dependent strains of Moloney virus are also suitable.
Particularly suitable viruses for use in the present invention are adenovirus subgroup C, type 5, Semilki Forest Virus, Vesicular Stomatitis Virus, Poliovirus, Rhinoviruses and Moloney Leukemia Virus. The use of RNA viruses for the present invention which have no reverse transcriptase has the advantage that 2~8~0 transfection in the presence of such a virus does not lead to the formation of viral DNA in the cell.
An important advantage derived from the present invention is that the DNA to be transferred is not integrated into the genome of the parent virus, as in the case with standard recombinant viral vectors (see Berkner, 1988; Eglitis and Anderson, 19~8). Thus, the present invention provides much greater flexibility as to the design of the foreign gene sequence to be expressed, as transcription is not dependent on promoters in the parent virus gene. In addition, this strategy allows a greatly increased size of DNA that can be transferred, as the packaging constraints of the -~
virus do not limit the amount of DNA that can be carried on the exterior. Over and above these practical and immediate advantages, important potential safety features derive from the design of the vector.
Conventional recombinant viral vectors mediate obligatory co-delivery of genome elements of the parent virus from which potential safety hazards derive (Ledley, 1989; Anderson, 1984). Since the conjugates according to the invention selectively exploit viral entry features, the viral genome is not an essential feature. This design allows the possibility of modifying the present system with a functionally and/or --structurally inactivated viral genome to minimize the safety hazards deriving from the transfer of viable genes from the parent virus.
Within the scope of the present invention, the term viruses - provided that they have uptake function as defined above - includes in addition to the wild types, mutants which have lost certain functions of the wild type, other than their uptake function, especially their ability to replicate, as a result of one or more mutations. However, mutants which have lost their uptake function can be employed in the practice of the invention so long as they are employed as part of a . J : ': ~ . ~ ` . r s ~ ~ ` - . , ! . ; , .; : ' ' , ~

8 ~ 0 -_ 9 "ternary co~plex~ as defined herein and the mutant virus has not lost its endosomolytic activity.
Mutants may be produced by conventional mutagenesis processes by mutations in virus-protein regions which are responsible for the replicative functions and the uptake function and which may be complemented by a -packaging line. These include, e.g. in the case of adenovirus, ts-mutants (temperature sensitive mutants), ElA- and ElB-mutants, mutants which exhibit mutations in MLP-driven genes (Berkner, 1988) and mutants which exhibit mutations in the regions of certain capsid proteins. virus strains which have corresponding natural mutations are also suitable. The ability of viruses to replicate can be investigated, for example, using plaque assays known from the literature, in which cell cultures are covered with suspensions of various virus concentrations and the number of lysed cells which is visible by means of plaques is recorded (Dulbecco, 1980).
Other viruses which may be suitable for use within the scope of the invention include so-called defective viruses, i.e. viruses which, in one or more genes, lack the function necessary for autonomous virus replication, for which they require helper viruses. Examples of this category are DI-particles (defective interfering particles) which are derived from the infectious standard virus, have the same structural proteins as the standard virus, have mutations and require the standard virus as a helper virus for replication (Huang, 1987;
Holland, 1990). Examples of this group also include the satellite viruses (Holland, 1990). Another grou~ is the class of parvoviruses called the adeno-associated virus (Berns, 1990).
Since the uptake cycles of many viruses into the cell have not yet been fully explained, it must be assumed that there are other viruses which have the endosomolytic activity required for their suitability for use in this invention.
Also suitable within the scope of this invention may be attenuated live vaccines (Ginsberg, 1980) or vaccination strains.
The term viruses within the scope of the present invention also includes inactivated viruses, e.g.
viruses inactivated by chemical treatment such as treatment with formaldehyde, by W -radiation, by chemical treatment combined with W -radiation, e.g.
psoralen/ W-radiation, by gamma-radiation or by neutron ~-bombardment, as well as parts of viruses, e.g. the protein content freed from nucleic acid (the empty virus ~ -capsid), provided that they have the uptake functions of ~--the intact virus. ~-Inactivated viruses that are also used for vaccines, for example, may be prepared by standard methods known from the literature (Davis and Dulbecco, 1980, Hearst and Thiry, 1977) and then tested to see whether they are suitable as components of the conjugates according to the invention.
The virus may possibly be a chimeric virus whi¢h has a foreign epitope in a region which is not essential for the uptake function. However, even when such chimeric viruses have lost their uptake function, they -may be employed within the scope of combination ~-complexes, so long as the virus has not lost its endosomolytic properties.
In order to select a virus, an inactivated virus or a virus component for the particular transfection which is to be carried out, the process used may be, for example, to investigate the virus first of all in preliminary tests to see whether it has an effect when the nucleic acid/polycation complexes are taken up into the target cell. Furthermore, its uptake functions may be tested by using it in transfection with bioconjugates, e.g. transferrin-polycation conjugates or another bioconjugate with specificity for the target 2~1~8~

cell to be transfected, and checking its ability to increase the gene transfer capacity by measuring the expression of a reporter gene.
When intact viruses are used, tests are carried out, preferably in parallel to the preliminary tests investigating the virus for its suitability for the proposed transfection, to see whether the virus is capable of replicating. The investigation for ability to replicate is carried out using plaque assays (see above) in the case of cytopathic viruses or in the case of viruses which significantly impair the growth of the host cells. For other viruses, detection methods specific to the virus in question are used, e.g. the hemagglutination test or chemico-physical methods (using an electron microscope).
Within the scope of this invention, the preferred viruses are those which can be produced in a high titre, which are stable, have low pathogenicity in their native state and in which a targeted elimination of the ability to replicate i5 possible, especially adenoviruses. If a specific cell population is to be transfected, viruses which specifically infect this cell population are I preferred. If the transfection is intended to attack I different cell types, viruses which are infectious for a wide range of cell types are used.
In any case, for therapeutic use of the invention vivo, only those viruses or virus components may be used in which the safety risks are minimized as far as possible, particularly the risk of replication of the virus in the cell and recombination of virus DNA with host DNA.
In preliminary tests, adenovirus preparations were inactivated using a conventional W sterilizing lamp or with formaldehyde and it was found, surprisingly, that the extent of inactivation of the viruses was substantially greater than the reduction in the gene transfer effect. This is a clear indication that ` - 2~ 1~800 ,. .

mechanisms connected with the normal infection mechanism in the active virus can be destroyed without eliminating ;
the effect which is essential for gene transfer.
Substances with an affinity for nucleic acid which may be used according to the invention include, for example, homologous polycations such as polylysine, polyarginine, polyornithine or heterologous polycations having two or more different positively charged amino acids, these polycations possibly having different chain lengths, and also non-peptidic synthetic polycations such as polyethyleneimine. Other substances with an affinity for nucleic acid which are suitable are natural DNA-binding proteins of a polycationic nature such as histones or protamines or analogues or fragments thereof.
~ he sensitivity of a given cell line to transformation by a virus which facilitates the entry of conjugates into the cell or constitutes a ligand for this type of cell depends on the presence and number of surface receptors for the virus on the target cell.
Methods of determining the number of adenovirus receptors on the cell surface are described for HeLa and KB cells by Svensson, 1985, and Defer 1990. It is assumed that the adenovirus receptor is expressed fairly ~
ubiquitously. -Therefore, many cell lines can be transformed with a vector system which contains an adenovirus or a part thereof. However, some higher eukaryotic cells have few or no viral receptors. If such cells are to be transformed, it may be necessary to use a second conjugate of an internalising factor which is bound to a substance having an affinity for nucleic acid, the internalising factor being specific for a surface receptor of the higher eukaryotic cell, the virus conjugate and the internalising factor conjugate being complexed with the nucleic acid. Such complexes can successfully be used to aid the transformation of higher :

- `- 2~1~800 eukaryotic cells, such as epithelial respiratory tract cells which have a relatively low cell surface population of adenovirus receptors (e.g. the cell line HREl).
In a preferred embodiment of the invention, the complexes may therefore optionally contain, in addition to the virus conjugate, another conjugate in which a substance having an affinity for nucleic acid, generally the same one as in the virus conjugate, is coupled with an internalizing factor having an affinity for the target cell. This embodiment of the invention is used particularly when the target cell has no or few receptors for the virus. In the presence of another internalizing factor-binding factor conjugate, these endosomolytic conjugates profit from the internalizing ability of the second conjugate, by being complexed to the nucleic acid together with the second conjugate and being taken up into the cell as part of the resulting complex, hereinafter referred to as a "combination complex" or "ternary complex".
Specifically, preliminary tests can determine whether the use of an (other) internalizing factor -permits or improves the uptake of nucleic acid complexes, by carrying out parallel transfections with nucleic acid complexes, first without any additional internalizing factor, i~e. with complexes consisting of nucleic acid and virus conjugate, and then with complexes in which the nucleic acid is conjugated with another conjugate containing an additional internalizing factor for which the target cells have a receptor. If an additional internalizing factor is used, it is defined particularly by the target cells, e.g. by specific surface antigens or receptors specific to a ~ `
cell type which thus permit the targeted transfer of nucleic acid into this type of cell.
The term "internalizing factor" for the purposes of the present invention refers to ligands or frag~ents ::: :~
:: ,~-.

211~8~0 .

thereof which, after binding to the cell are internalized by endocytosis, preferably receptor-mediated endocytosis, or factors the binding or internalizing of which is carried out by fusion with elements of the cell membrane.
Suitable internalizing factors include the ligands transferrin (Klausner et al., 1983), conalbumin (Sennett et al., 1981), asialoglycoproteins ~such as asialotransferrin, asialorosomucoid or asialofetuin) (Ashwell et al., 1982), lectins (Goldstein et al., 1980 and Shardon, 1987) or substances which contain galactose and are internalized by the asialoglycoprotein receptor, mannosylated glycoproteins tStahl et al., 1987), lysosomal enzymes (Sly et al., 1982), LDL (Goldstein et al., 1982), modified LDL (Goldstein et al., 1979), lipoproteins which are taken up into the cells via receptors (apo B100/LDL); viral proteins such as the HIV
protein gpl20; antibodies (Mellman et al., 1984; Kuhn et al., 1982), Abrahamson et al., 1982), or fragments thereof against cell surface antigens, e.g. anti-CD4, anti-CD7; cytokines such as interleukin-1 (Mizel et al., 1987), Interleukin-2 (Smith et al., 1985), TNF (Imamure et al, 19~7), interferons (Anderson et al., ~982~; CSF
(colony-stimulating factor), (Walker et al., 1987);
factors and growth factors such as insulin (Marshall, 1985), EGF tepidermal growth factor), (Carpenter, 1984);
PDGF (platelet-derived growth factor) (Heldin et al., 1982); TGFB (transforming growth factor B), (Massague et al., 1986), nerve growth factor (Hosang et al., 1987)r ;
insulin-like growth factor I (Schalch et al., 1986),~LH, FSH, (Ascoli et al., 1978), growth hormone (Hizuka et al., 1981), prolactin (Posner et al., 1982), glucagon (Asada-Xubota et al., 1983), thyroid hormones (Cheng et al., 1980); ~-2-macroglobulin protease (Kaplan et al., 1979); and "disarmed" toxins. Other examples are ~ ~-immunoglobulins or fragments thereof as ligands for the ~-;
Fc-receptor or anti-immunoglobulin antibodies which bind . , -:
' ', - ~`` 2il48~0 to SIgs (surface immunoglobulins). The ligands may be of natural or synthetic origin (see, Trends Pharmacol.
Sci. (1989), and the references cited therein).
The following are essential requirements for the suitability of such internalizing factors according to the present invention, a) that they can be internalized by the specific cell type into which the nucleic acid is to be introduced and their ability to be internalized is not affected or only slightly affected if they are conjugated with the binding factor, and b) that, within the scope of this property, they are capable of carrying nucleic acid "piggyback"
into the cell by the route they use.
Without being pinned down to this theory, the combination complexes are taken up by cells either by binding to the surface receptor which is specific to the internalizing factor or, if a virus or virus component is used, by binding to the virus receptor or by binding to both receptors by receptor-mediated endocytosis.
When the endosomolytic substance is released from the endosomes, the DNA contained in the complexes is also released into the cytoplasm and thereby escapes the lysosomal degradation.
The presence of viruses, virus components or non-viral endosomolytic agents as components of endosomolytic conjugates in the DNA complexes has the following advantages:
1) Wider applicability of the gene transfer technology with nucleic acid complexes, since the ¦ endosomolytic agents themselves, especially if a virus or virus component is used, may constitute the internalizing factor or may also be complexed -to the DNA in conjunction with another internalizing factor (e.g. transferrin or asialofetuin etc.). In this way it is possible to make use of the positive effect of the viruses even -` 211~8~0 for cells which do not have any receptor for the virus in question.
2) Improvement in the efficiency of gene transfer, since the binding of the endosomolytic conjugates to the DNA ensures that they are jointly taken up into the cells. The coordinated uptake and release of viruses and DNA also gives rise to the possibility of a reduction in the quantity of DNA
and viruses required for efficient gene transfer, which is of particular importance for use in vivo.
In the experiments carried out according to the invention, human transferrin was used as an additional internalizing factor; moreover, the performance of the conjugates according to the invention was demonstrated by means of complexes of DNA and polylysine-conjugated virus which contained no additional internalizing factor-binding factor conjugate.
The binding of the virus to the substance having an affinity for nucleic acid is achieved by covalent bonding of the substance with an affinity for nucleic acid to an antibody. It is preferable to use an antibody which binds to an epitope in a virus protein region not involved in the uptake function of the virus. --~
In the tests carried out within the scope of the ~ -~
invention, the binding between an adenovirus and a ~-polycation was achieved by covalently conjugating an antibody with specificity for the adenovirus capsid to a polylysine molecule. It is known that the adenovirus fibre and penton proteins are essential for the binding of the virus and its uptake into the cell, whereas the main capsid protein hexon is of lesser importance in these processes. Therefore, an antibody was used which brings about the binding of the adenovirus to polylysine r. :~
by recognition of an epitope on the hexon protein. This specific binding was achieved by using, on the one hand, a chimeric adenovirus which has a foreign epitope in the surface region of its hexon protein. On the other hand, a monoclonal antibody was used which is specific for the.
heterologous epitope. (This construction is diagrammatically shown in Fig. 1). This results in a binding of the adenovirus to polylysine without functionally destroying the capsid proteins.
The use of a special antibody for establishing the bond between the virus and the nucleic acid-binding substance is not critical. The prerequisite for the suitability of a particular antibody is that it should not neutralize, or should only partly neutralize, the uptake function of the virus.
Within the scope of the present invention, antibodies against epitopes in virus protein regions which are not essential for the uptake function are preferred. Examples of such virus regions are the hexon protein of the adenovirus mentioned above or influenza neuraminidase.
However, antibodies with specificity for virus proteins which are involved in the uptake function are also suitable, provided that it is ensured, by maintaining a suitable stoichiometric ratio, that the antibody occupies only part of the cell binding regions of the virus, so that there are still sufficient domains free for the binding of the virus to the cell. On the other hand, antibodies which block the uptake function of the virus may be used so long as the complex further comprises a second conjugate comprising an internalizing factor and a nucleic acid-binding substance, i.e. a combination complex.
The quantity of antibody suitable for the specific application can be determined by titration.
Monoclonal antibodies, possibly the Fab' fragments thereof, are preferred.
If the virus is a chimeric virus with a foreign epitope, the antibody is directed against this epitope.
Preferably, the virus is a chimeric adenovirus where the coding sequence for the hexon region has been modified 2 ~ O

to include a sequence coding for a heterologous protei~
against which an antibody can be raised. The hexon protein is composed of a hiyhly conserved base domain and three less conserved loops that are highly exposed on the surface of the viron (Roberts et al., 1986).
There are several short regions in these loops where the Ad2 and Ad5 amino acid seguences are dissimilar, with Ad5 showing changes as well as deletions compared to Ad2. These are potential sites for the insertion of the heterologous gene sequences coding for the heterologous protein which may be used to link the adenovirus immunologically to the substance having affinity for the nucleic acid. Preferably, the heterologous gene ~ -~
sequence is inserted in the Ad5 gene sequence at amino acid positions 161-165, 1~8-194, 269-281 and 436-438, referred to as sites I, II, III and IV, respectively (see Figure 8). At each potential site, a unique restriction site may be created by means of site~
directed mutagenesis of a subclone of the Ad5 hexon gene. Nucleotides coding for nonconserved amino acids may be deleted at the same time, leaving more space for insertion of the heterologous gene sequence. In general, because of the small numbers of amino acids which can be inserted at sites I, II, III and IV (up to about 65 amino acids), the heterologous gene sequence codes for only the amino acids corresponding to the epitope and a minimal number of flanking sequences.
(The possible insertion sites created in the AdS hexon gene sequence by directed mutagenesis are shown in Fig.
8; for the three-dimensional sketch of the hexon sub-unit the representation by Roberts et al., 1986 has been adapted.) The epitope specificity of a particular monoclonal antibody to a heterologous protein may be determined by peptide scanning, see Geysen, et al., 1984, 1985, 1986, 1987, and EP-A-392,369, the disclosures of which are fully incorporated by reference herein. According to ..

`-- 211~81~

this method, overlapping 8-amino acid long peptides of the heterologous protein are prepared by methods of solid phase synthesis. For example, peptide 1 consists of amino acids 1-8, peptide 2 of amino acids 2-9, and so on. The peptides remain bound to the solid carrier after synthesis. Hybridoma cell culture supernatants or purified monoclonal antibodies thereof are then tested for reactivity to the immobilized peptides by ELISA.
Once the epitopic region is identified, the gene sequence coding for the epitopic region may then be inserted into any one of the restriction sites of regions I, II, III or IV. There are many examples of proteins and antibodies which are specific for the protein. One of ordinary skill in the art can select a -~
heterologous protein-antibody combination which is operable in the present invention with no more than routine experimentation. For example, the known coding sequence for a protein, for which an antibody thereto is also known, may be inserted into the hexon region of an adenovirus. The resulting chimeric virus can then be tested for immunological binding, for example, to labelled antibody in a competition, ELISA, or other immunoassay format. Such immunoassay techniques are well known and are practised routinely by those of ordinary skill in the art.
I The antibody-polycation conjugates can be produced chemically by a method known per se for the coupling of peptides, preferably using the method described by Wagner et al., 1990, and in EP-Al 388 758.
If monoclonal antibodies have suitable carbohydrate side chains, particularly terminal sialic acids, in the constant region of the heavy chain, the conjugates may be prepared by binding the polycation to the carbohydrate side chain, using the method described by Wagner et al., l991bo Another aspect of the invention relates to complexes which are taken up into higher eucaryotic -~
, .~
. ~: :

- " 211~8~0 cells, containing nucleic acid and a conjugate of an internalizing factor and a substance having an affinity for nucleic acid. The complexes are characterized in that the internalizing factor is a virus which is bound to the substance having an affinity for nucleic acid via an antibody in such a way that it has the ability to penetrate into the cell as part of the conjugate/nucleic acid complex and release the contents of the endosomes, in which the co~plex is located after entering the cell, into the cytoplasm.
As for the qualitative composition of the nucleic acid complexes, generally the nucleic acid to be transferred into the cell is determined first. The nucleic acid is defined primarily by the biological effect which is to be achieved in the cell and, in the case of use for gene therapy, by the gene or gene section which is to be expressed, e.g. for the purpose of replacing a defective gene, or by the target sequence of a gene which is to be inhibited. The nucleic acids to be transported into the cell may be DNAs or RNAs, whilst there are no restrictions imposed on the nucleotide sequence.
If the invention is applied on tumour cells in order to use them as a cancer vaccine, the DNA to be introduced into the cell preferably codes for an immunomodulating substance, e.g. a cytokine, such as IL-2, IL-4, IFN-gamma, TNF-alpha. Combinations of cytokine encoding DNAs may be particularly useful, e.g. IL-2 and IFN-gamma. Another useful gene for insertion into tumour cells may be the multi drug resistance gene (mdr).
It is also possible to introduce two or more different nucleic acid sequences into the cell, e.g. a plasmid containing cDNAs coding for two different proteins under the control of suitable regulatory sequences or two different plasmid constructs containing -different cDNAs.

2~1~800 . ~

Therapeutically effective inhibiting nucleic acids for transfer into the cells in order to inhibit specific gene sequences include gene constructs from which antisense-RNA or ribozymes are transcribed. Furthermore, it is also possible to introduce oligonucleotides, e.g.
antisense oligonucleotides, into the cell. Antisense oligonucleotides comprise preferably 15 nucleotides or more. Optionally, the oligonucleotides may be multimerized. When ribozymes are to be introduced into the cell, they are preferably introduced as part Oc a gene construct which comprises stabilizing gene elements, e.g. tRNA gene elements. Gene constructs o~
this type are disclosed in EP A O 387 775.
Apart from nucleic acid molecules which inhibit genes, e.g. viral genes, due to their complementarity, genes with a different mode of inhibitory action may be employed. Examples are genes coding for viral proteins which have so-called trans-dominant mutations (Herskowitz, 1987). Expression o~ the genes in the cell yields proteins which dominate the corresponding wildtype protein and thus protect the cell, which acquires "cellular immunity", by inhibiting viral replication. Suitable are trans-dominant mutations of viral proteins which are required for replication and expression, e.g. Gag-, Tat and Rev mutants which were shown to inhibit HIV replication (Trono et al., 1989;
Green et al., 1989; Malim et al., 1989).
Another mechanism of achieving intracellular immunity involves expression of RNA molecules containing the binding site for an essential viral protein, e.g.
so-called TAR decoys (Sullenger et al, 1990).
Examples of genes which can be used in somatic gene therapy and which can be transferred into cells as -~
components of gene constructs by means of the present invention include factor VIII (hemophilia A) (see, e.g.
Wood et al., 1984), factor IX (hemophilia B) (see, e.g.
Kurachi et al., (1982), adenosine deaminase (SCID) (see, . .

211~8~
`

e.g. Valerio et al., 1984), a-l antitrypsin (emphysema of the lungs) (see, e.g. Ciliberto et al., 1985) or the cystic fibrosis transmembrane conductance regulator gene (see, e.g. Riordan et al., 1989). These examples do not constitute a restriction of any kind.
As for the size of the nucleic acids, a wide range is possible; gene constructs of about 0.15 kb (in case of a tRNA gene containing a ribozyme gene) to about 50 kb or more may be transferred into the cells by means of the present invention; smaller nucleic acid molecules may be utilised as oligonucleotides.
It is obvious that, precisely because the present invention is not subject to any restrictions as to the gene sequence and even very large gene constructs can be transported with the aid of the invention, the possible applications are extremely wide.
When determining the molar ratio of antibody-polycation:nucleic acid it should be borne in mind that complexing of the nucleic acid(s) takes place. In the course of earlier inventions it had been established that the optimum transfer of nucleic acid into the cell can be achieved if the ratio of conjugate to nucleic acid is selected so that the internalizing factor-polycation/nucleic acid complexes are substantially electroneutral. It was found that the quantity of nucleic acid taken up into the cell is not reduced if some of the transferrin-polycation conjugate is replaced by non-covalently bound polycation; in certain cases there may even be a substantial increase in DNA uptake (Wagner et al., l991a). It had been observed that the DNA inside the complexes is present in a form condensed into toroidal structures with a diameter of 80 to 100 nm. The quantity of polycation is thus selected, with respect to the two parameters of electroneutrality and the achievement of a compact structure, whilst the quantity of polycation which results from the charging of the nucleic acid, with respect to achieving ;~ r~ r ~

...... . . . ~ .. .. .. . ... ...... . .... .. .. . .. .. .

~1~4~0~

electroneutrality of the complexes, as preferred according ~o the invention, generally also guarantees compacting of the DNA.
A suitable method of determining the ratio of components contained in the complexes according to the invention is first to define the gene construct which is to be transferred into the cells and, as described above, to determine a virus which is suitable for the particular transfection. ~hen an antibody which binds to the virus is conjugated with a polycation and complexed with the gene construct. Starting from a defined quantity of virus, titrations may then be carried out by treating the target cells with this (constant) quantity of virus and decreasing concentrations of DNA complex (or optionally vice versa). In this way the optimum ratio of DNA complex to virus is determined. In a second step the cells are treated with decreasing concentrations of the virus/DNA
complex mixture (at a constant ratio of virus to complex) and the optimum concentration is determined.
Preferably, the virus is an adenovirus and the molar ratio of adenovirus to substance having an affinity to the nucleic acid is about 1:1 to about 1:100.
The length of the polycation is not critical, so long as the complexes are substantially electroneutral, -with respect to the preferred embodiment. The preferred range of polylysine chain lengths is from about 20 to about 1000 lysine monomers. However, for a given length of DNA, there is no critical length of the polycation.
Where the DNA consists of 6,000 bp and 12,000 negative charges, the amount of polycation per mole DNA may be, e.g.:
60 molecules of polylysine 200 ~ ~
30 molecules of polylysine 400; or ~- -120 molecules of polylysine 100, etc. -~
One of ordinary skill in the art can select other combinations of polycation length and amount of : ::
'c ~ ~ ~

~1 148~

polycation with no more than routine experimentation.
The complexes according to the invention can be prepared by mixing the components nucleic acid and antibody-bound polycation, which are present in the form of dilute solutions. The DNA complexes can be prepared at physiological saline concentrations. Another possibility is to use high salt concentrations (about 2 M NaCl) and subsequently adjust to physiological conditions by slow dilution or dialysis. The best sequence for mixing the components nucleic acid, antibody-polycation conjugate and virus is determined by individual preliminary tests.
The invention relates in another aspect to a process for introducing nucleic acid into higher eucaryotic cells, in which the cells are brought into contact with the complexes according to the invention in such a way that the complexes are internalized and released from the endosomes.
The present invention relates in another aspect to pharmaceutical preparations containing as active component a complex consisting of therapeutically active nucleic acid, preferably as part of a gene construct, and an antibody coupled via a polycation. Preferably, this preparation is in the form of a lyophilisate or in a suitable buffer in the deep-frozen state and the virus preparation is mixed with the complex solution shortly before use. Possibly, the virus may already be contained in the pharmaceutical preparation, in which case it is in deep-frozen state. Any inert pharmaceutically acceptable carrier can be used, e.g.
saline solution or phosphate-buffered saline solution or any carrier in which the DNA complexes have suitable solubility properties to allow them to be used within the framework of the present invention. For methods of formulating pharmaceutical preparations reference is made to Reminington's Pharmaceutical Sciences, 1980.
Possibly, the components required for transfection, -- 211~0~
-namely DNA, virus preparation and antibody conjugate or the conjugation partners, possibly internalising factor conjugate and possibly free polycation, are kept separate in a suitable buffer or kept partly separate as components of a transfection kit, which is a further subject of the present invention. The transfection kit of the present invention comprises a carrier which contains one or more containers such as test tubes, ampoules or the like which contain the materials required for transfection of the higher eukaryotic cells according to the present invention. In such a transfection kit, a first container may contain on2 or more different DNAs. A second container may contain one or more different internalising factor conjugates, -allowing the transfection kit to be used as a modular system. Whether the components are present as a ready -~
to use preparation or are kept separately to be mixed together immediately before use will depend not only on the specific application but on the stability of the complexes, which can be investigated routinely using stability tests.
For therapeutic purposes the preparations may be administered systemically, preferably by intravenous route. The target organs for this type of administration may be, for example, the liver, spleen, -lungs, bone marrow and tumours.
Recently, the feasibility of using myoblasts (immature muscle cells) to carry genes into the muscle fibres of mice was shown. Since the myoblasts were shown to secrete the gene product into the blooa, this method may have a much wider application than treatment of -genetic defects of muscle cells like the defect involved in muscular dystrophy. Thus, engineered myoblasts may be used to deliver gene products which either act in the blood or are transported by the blood.
Examples for local application are the lung tissue (use of the complexes according to the invention as a -`- 21~8~

fluid for instillation or as an aerosol for inhalation), direct injection into the liver, muscle tissue or into a tumour or local administration in the gastro-intestinal tract. Another method of administering the pharmaceutical composition is via the bile duct system.
This method of adminstration penmits direct access to the hepatocyte membranes on the bile ducts, thereby avoiding interaction with constituents of the blood.
Therapeutic application may also be ex vivo, in which the treated cells, e.g. bone marrow cells, hepatocytes or myoblasts, are returned to the body (e.g.
Ponder et al., 1991, Dhawan et al., 1~91. Another ex vivo application of the present invention relates to so-called cancer vaccines. The principle of this therapeutic possibility is to isolate tumour cells from a patient and transfect the cells with a DNA coding for a cytokine. In a next step the cells are optionally inactivated, e.g. by radiation, so that they no longer replicate but still express the cytokine. Then the genetically altered cells are administered to the patient from whom they were taken in the form of a vaccine. In the area surrounding the vaccination site the cytokines secreted activate the immune system partly by activating cytotoxic T-cells. These activated cells are able to exert their activity in other parts of the body and attack even untreated tumour cells. In this way the ris~ of tumour recurrence and metastasis development is reduced. A suitable procedure for using cancer vaccines for gene therapy has been described by Rosenberg et al., 1992. Instead of the retroviral vectors proposed by Rosenberg, the gene transfer system according to the present invention can be used.
In order to determine the capacity for gene transfer of adenovirus-antibody-polycation/DNA
complexes, a plasmid containing the gene coding for Photinus Pyralis Luciferase (De Wet et al., 1987) as reporter gene was used as the DNA. HeLa cells were used 2 ~

as target cells for the complexes; these cells have a defined population of cell surface receptors for adenoviruses (Philipson et al., 1968). When the components of the conjugate according to the invention (virus, antibody-polylysine-conjugate, DNA) were used in conjunction, high values were obtained for the expression of the luciferase reporter gene (Fig. 3)~
comparative experiments showed that the adenovirus only slightly increased the transfer of non-complexed plasmid-DNA. It was also found that DNA which was complexed with the antibody-coupled polylysine (without binding to the virus) was not appreciably taken up in HeLa cells. In sharp contrast to this, high gene expression values were obtained with the complex if the DNA was able to interact with the adenovirus by binding via the antibody. This effect was stopped when the ~-virions were heat treated before the complexing. Since this treatment selectively removes the viral uptake functions without destroying the structural integrity of ~-the virus (Defer et al., 1990), it can be concluded from these experiments that it is the specific uptake functions of the adenovirus which constitute the crucial contribution to the success of gene transfer. It was also found that competition for the heterologous epitope on the surface of the chimeric adenovirus by a specific, non-polylysine-bound monoclonal antibody also brings about a reduction in the net gene expression. This effect did not occur when a non-specific antibody was used. It is therefore the specific interaction between the antibody-bound DNA and the corresponding adenovirus surface epitope which is essential to the achievement of functional gene transfer by means of the complex. It was also found that polylysine-complexed DNA was not appreciably transferred into the target cells by the adenovirus. This is an indication that the gene transfer capacity of the complexes is not based on the condensing of DNA but depends on the antibody-mediated - -`` 211~800 :

binding of the reporter gene to the virion~
In accordance with this, the use of a virus which did not have the epitope recognized by the polylysine-coupled antibody could not achieve the high gene expression values achieved by a virus which did have this epitope. However, this virus was able to increase the extent of gene transfer above the background level.
Since it is known that adenoviruses are capable of non-specifically augmenting the cellular uptake of macromolecules through the liquid phase (Defer et al., 1990), this result was not unexpected. The fact that this non-specific transport brought about a significantly lower expression of the reporter gene than the specific virus which was able to bind to the antibody-polylysine/DNA complex demonstrates the importance of specific binding of the components of the complex.
The interaction of plasmid DN~ with polylysine conjugates results in significant structural changes in the DNA molecule, which are most clearly characterized by striking condensation into a toroidal structure of 80 to 100 nm (Wagner et al., l991a). The diameter of the virus is of the order of 70 to 80 nm (Philipson, 1983).
It was therefore assumed, on the basis of steric considerations, that the optimum ratio of adenovirus to antibody-polylysine-complexed DNA should be no more than 1:1. Furthermore, the diameter of the coated pits by means of which the initial uptake step of receptor-mediated endocytosis is carried out, is about 100 nm (Darnell et al., 1975). On the basis of this fact it was assumed that multimers exceeding this size would be restricted in their uptake capacity. Within the scope of the present invention these correlations were analyzed, whilst the use of adenovirus in molar excess relative to the antibody-polylysine-complexed DNA showed that the maximum expression of reporter gene was achieved at a ratio of 1:1 (Fig. 4). The optimum , ~

4 8 ~

conjugate, within the scope of the experiments carried out, was therefore found to be one which consists of a single adenovirus internalizing domain in conjunction with a single antibody-polylysine/DNA binding domain.
Next, the gene transfer efficiency of adenovirus-antibody-polycation conjugates having this optimum ratio was investigated. If logarithmic dilutions of the complex were added to the target cells, there was a corresponding logarithmic reduction in expression of the reporter gene (Fig. 5), whilst it was noticeable that 107 ~-DNA molecules, applied to lo6 HeLa cells using this vector system, resulted in the detectable expression of the reporter gene. Surprisingly, therefore, efficient expression of a foreign gene was achieved with as few as 10 DNA molecules per cell in the form of adenovirus-polycation-DNA complexes.
Therefore, with regard to the magnitude of DNA
uptake, the conjugates according to the invention show clear superiority over the DNA gene transfer vectors, which are required in numbers of approximately 500,000 DNA molecules per cell (Felgner et al., 1987, Felgner et al., 1989, Maurer, 1989). Since these methods efficiently convey the majority of the DNA into the target area of the cells, namely the cytosol (Felgner et al., 1989, Malone et al., 1989, Loyter et al., 1982), the efficiency of the conjugates according to the invention may possibly not be based exclusively on the increase in release of the foreign DNA into the cytoplasm; other mechanisms on the route of gene transfer may also be enhanced.
In the configuration of the adenovirus-polylysine-DNA complexes, the adenovirus part acts both as an agent for breaking up endosomes and also as the ligand domain of the complex. Therefore, the gene transfer efficiency of the complexes for a givPn target cell should reflect the relative number of adenovirus cell surface receptors. Both the cell line HeLa and the cell line .

`- 21~48~

Ks, which have large amounts of adenovirus receptors (Philipson et al. 1968) exhibited a correspondingly high degree of accessibility for gene transfer by means of adenovirus-polylysine-DNA complexes (see Fig. 6). By contrast, the relatively low number of adenovirus receptors which characterises the cell lines MRC-5 (Precious and Russell 1985) and HBE1, is reflected in a low gene transfer level by means of the complexes into these cells.
Ternary complexes which contained another cell surface ligand (internalising factor) in addition to the adenovirus, showed significantly higher levels than the conjugates which had only transferrin or adenovirus ligand domains (Fig. 7A). The magnitude of this increase was clearly not based on an additive effect of transferrin-polylysine-DNA complexes plus adenovirus-polylysine-DNA complexes. Since the ternary complexes can be internalised by means of the adenovirus or transferrin uptake mechanism, this obvious interaction leads one to suppose that the adenovirus domain facilitates entry into the cell by both routes, presumably on the basis of endosomolysis brought about by adenovirus. In order to demonstrate the selective co-operation of the adenoviral domain of the ternary complex when endosomes are broken up, the combination complexes were applied to cell lines which have different degrees of receptiveness for adenovirus-polylysine-DNA complexes (Fig. 7B). The epithelial respiratory tract cell line exhibits very low gene transfer values, compared with HeLa cells, achieved by AdpL-DNA complexes (Fig. 6), reflecting the relatively ~ -low cell surface population of adenovirus receptors, characteristic of this cell line. By clear contrast, the use of ternary AdpL/TfpL complexes resulted in levels of gene expression which were comparable with those seen in HeLa cells. The receptiveness of this cell line for gene transfer by the ternary complexes - 2~8~ ~

agrees with the concept that the adenovirus domain is taken up by means of the transferrin mechanism, whilst intensifying the gene transfer by causing endosomes to break up. From this arises the possibility of using the endosomolytic property of adenovirus and other viruses in the construction of conjugates which are thereby enabled to escape from the cell vesicle system.
Within the scope of the present invention, the direct in vivo transfer of a gene into the respiratory tract epithelium by means of the complexes according to the invention was demonstrated in a rodent model. This ~-supports the possibility of using the present invention to achieve transient gene expression in the respiratory tract epithelium. The possibility of achieving genetic modification of respiratory tract cells in situ is a possible strategy of gene therapy for diseases of the respiratory tract epithelium. In the tests carried o~t, the transferrin-polylysine-DNA complexes yielded a low level of reporter gene expression. This agrees with the fact that this type of conjugate should be enclosed in the endosomes. The binary adenovirus polylysine complexes brought about a significantly higher gene expression. This was further increased by using a second internalising factor in the combination complex hTfpL/AdpL. To find out whether the net gene expression agrees with the transduction frequency, the proportion of cells which had been transduced with the various types of complex was determined. It was discovered that there is such an agreement; the respiratory tract epithelium modified with hTfpL in the primary culture showed less than 1% transduction frequency, the AdpL
-complexes showed frequencies in the range from 20 to 30 and the combination complex showed more than 50%
modified cells.
The experiments carried out in vivo on rodent models agreed with the results obtained in the primary -culture. The examination of histological lung sections . '~' 211~80~

from rats which had been treated with lacZ combination complexes showed uneven zones of ~-galactosidase activity which contained numerous positive cells. The positive regions were associated with the bronchiolar and distal regions of the respiratory tract.

Summary of the Fiaures Fig. 1: Diagrammatic representation of adenovirus-polycation-DNA complexes containing a foreign epitope on the adenovirus capsid.
Fig. 2: Preparation of the chimeric adenovirus Ad5-P202.
Fig. 3: Gene transfer to HeLa cells using adenovirus-polycation-DNA complexes.
Fig. 4: Determining the optimum ratio of adenovirus and polylysine-antibody-complexed DNA.
Fig. 5: Determin~ng the gene transfer achieved by means of adenovirus-polycation-DNA complexes.
Fig. 6: Gene transfer to various eukaryotic cell mediated by adenovirus-polycation-DNA complexes.
Fig. 7: Gene transfer mediated by combination complexes containing adenovirus and human transferrin conjugates.
A: Comparison of the efficiency of combination complexes with binary complexes in HeLa cells.
B: Comparison of HeLa cells and HBEl cells with regard to the efficiency of gene transfer by combination complexes.
Fig. 8: Three dimensional representation of the hexon sub-unit; possible insertions sites in the AD5 hexon gene sequence, obtained by site directed ~ -mutagenesis;
Fig. 9: Transfection of primary respiratory tract epithelial cell cultures. Relative level of net gene transfer;
Fig. 10: Transfection of primary respiratory tract epithelial cells cultures. Relative frequency of ' ~ ' '~ :
211~80~

transduction; ~ -Fig. 11: Gene transfer via the intra-tracheal route in vivo. Relative level of net gene transfer in vivo;
Fig. 12: Gene transfer through the intra-tracheal route in vivo. Localisation of heterologous gene expression in the respiratory tract epithelium;
Fig. 13: In vivo application of chimeric adenovirus-lectin-polylysine DN~ complexes.

2~14800 Examples The invention is illustrated by means of the following Examples:

Example 1 Preparation of antibody-polylysine coniuqates 1) Preparation of the chimeric adenovirus Ad-P202 In order to make changes in the Ad5 hexon gene it was first necessary to subclone the gene. The plasmid pEcoRIA-Ad5 (Berkner and Sharp, 1983) contains the left-hand part of the adenovirus genome of map unit (m.u.) 0 to 76. The hexon gene is between m.u.52 and m.u.60. A
2.3 kbp HindIII/SstI-fragment contains that part of the hexon gene in which the change is to be made. Since a plurality of HindIII and SstI sites are contained in pEcoRIA-Ad5 it was necessary to construct several intermediate plasmids in order to be able to assemble the altered hexon gene in the original plasmid. A
SalI/BamHI fragment (m.u. 46 to 60) contains the hexon gene without any additional HindIII or SstI sites. ~-First of all, the adenovirus DNA was recloned from m.u.
0 to 76 by using a vector designated pl42 (derived from -the commercially obtainable plasmid pIBI24 (IBI, Inc.) by restriction digestion with PvuII, followed by the insertion of an EcoRI linker) which contains no SstI or BamHI sites. Then the SalI sites at m.u. 26 were eliminated ~y deleting the XbaI fragment (m.u. 3.7 to 29); the resulting vector was designated pl41-12.
Finally, the desired HindIII/SstI-fragment was cloned in M13mpl8 and was therefore ready for mutagenesis. Site directed mutagenesis was carried out with one of the resulting clones using the method described by Kunkel, 1985. The codons 188 to 194 of the hexon gene were removed and at this position a unique PmlI-site occurring only once was introduced. The resulting clone ~ A~','!`'`'`' `~ ~ ~

2i~4800 (167-1) was then cut with PmlI and a double stranded oligonucleotide coding for the amino acids 914-928 of the mycoplasma pneumoniae Pl-protein was inserted (Inamine et al., 1988). The Pl-sequence contains an epitope which is recognized by the monoclonal antibody 301, the preparation of which is described hereinafter.
The modified HindIII/SstI-fragment was isolated ~rom pl67-1 and ligated back into the original plasmid pEcoRIA-Ad5. The preparation of Ad-P202 is shown in Fig. 2.

2) Preparation of a monoclonal antibody with specificity for the chimeric adenovirus (MP301) a) Immunization The monoclonal antibody was prepared by standard methods.
The Mycoplasma pneumoniae strain M-129 ~ATCC#29342) was used as the antigen. After cultivation in a culture flask (~u et al., 1977) it was washed 3 times with PBS, Mycoplasma pneumoniae was harvested and taken up in 0.5 ml of PBS. 10 ~g of the antigen were used for immunization: 3 female BALB/c mice about six weeks old were immunized in accordance with the following protocol:

1st immunization: about 10 ~g of antigen per mouse in complete Freund's adjuvant by intraperitoneal route.

2nd immunization: about 10 ~g of antigen per mouse in incomplete Freund's adjuvant by subcutaneous route, 3 weeks after the first immunization.

3rd immunization: about 10 ~g of antigen per mouse in incomplete Freund's adjuvant by intraperitoneal route, 2 weeks after the second immunization.

4 8 1~ 0 One week later, samples of serum were taken from the mice and the serum titres were measured. The mouse with the hi~hest titre was boos~ed by i.v. injection of 10 ~g antigen into the tail; the spleen cells of this mouse were taken out after 3 days for fusion with hybridoma cells.

b) Fusion:
About lo8 spleen cells were fused with about 1O8 myeloma cells of the line SP2/0 Agl4 (ATCC CRL-1581) in the presence of PEG 4000 (50% in serum-free culture medium) using the method of Kohler and Milstein, 1975.
Then the cells were grown for 2 weeks in HAT-selection medium, then for one week in HT-medium and finally in normal culture medium (DMEM plus 10% FCS plus penicillin, streptomycin). By means of radioimmuno-sorbent assay (RIA) screening was carried out for antibody-producing clones and specificity for the Mycoplasma pneumoniae Pl protein was determined using -~
Western blot. The "soft agar" method was used to obtain monoclones.

c) Investiaation of the monoclonal antibody MP301 for neu~tralizina effect o~ adenovirus Ad-P202 In order to determine whether the monoclonal antibody MP301 neutralizes the ability of the virus to infect cells, the titre of Ad-P202 was determined once with and once without the addition of antibody (7 ~g/ml), using HeLa-cells (approximately 50% confluent in 2% FCS/DMEM on 96-well plates) as the target cells.
Serial dilutions were prepared of Ad-P20~ which were applied to the HeLa-titre plates with or without antibody. The plates were incubated for 48 hours at 37~, 5% CO2, stained with crystal violet and investigated for IC 50 (inhibition concentration, about 50% cell lysis). The titre of 1:2048 was obtained with and without antibody.

2il~

d) Preparation of MP301-Dolylvsine coniuqates Coupling of the monoclonal antibody to polylysine was carried out using the method described by Wagner et al., 1990, and in EP-Al 388 758.
20.6 nmol (3.3 mg) of the monoclonal antibody MP301 in 1 ml of 200 mM HEPES pH 7.9 were treated with a 5 mM
ethanolic solution of SPDP (loo nmol). After 3 hours at ambient temperature the modified antibody was gel-filtered over a Sephadex G-25 column, thereby obtaining 19 nmol of antibody modified with 62 nmol of dithiopyridine linker. The modified antibody was allowed to react with 3-mercaptopropionate-modified polylysine (22 nmol, average chain length 300 lysine monomers, FITC-labelled, modified with 56 nmol mercapto-propionate linker~ in 100 mM HEPES pH 7.9 under an argon atmosphere. Conjugates were isolated by cation exchange chromatography on a Mono S HR5 column (Pharmacia).
(Gradient: 20 to 100% buffer. Buffer A: 50 mM HEPES pH
7.9; buffer B: buffer A plus 3 M sodium chloride. The product fraction eluted at a salt concentration of between 1.65 M and 2 M. Dialysis against HBS (20mM
HEPES pH 7.3, 150 mM NaCl) produced a conjugate consisting of 9.1 nmol MP301 and 9.8 nmol polylysine.

Example 2 Gene transfer bY means of adenovirus-polycation-DNA-complexes in EucarYotic cells In the course of the experiments carried out in this Example, various combinations of specific and non-specific complex components were examined for their ability to transport a reporter gene into HeLa and other cells.
Complexing of DNA with the antibody-coupled polylysine was carried out by diluting 6 ~g of purified pRSVL-DNA in HBS (150 mM NaCl, 20 mM HEPES, pH 7.3) to a total volume of 350 ~1 and purifying it with 9.5 ~g of 2~148~1D

MP301pL in 150 ~1 of total volume of the same buffer.
(pRSVL contains the Photinus pyralis luciferase gene under the control of the Rous Sarcoma virus LTR
enhancer/promoter (Uchida et al., 1977, De Wet et al., 1987), prepared by Triton X Lysis Standard Method (Maniatis), followed by CsCl/EtBr equilibrium density gradient centrifugation, decolorizing with butanol-l and dialysis against 10 mM tris/HCl pH 7.5, 1 mM EDTA in 350 ~1 HBS (150 mM NaCl, 20 mM HEPES, pH 7.3).) The quantity of antibody-coupled polylysine is based on a calculation of the guantity required to achieve electroneutrality of the imported DNA. The polylysine-antibody-complexed DNA was diluted in HBS to a final concentration of 2 x 1011 DNA molecules per ml. The adenovirus P202-Ad5 was diluted in ice~cold DMEM, supplemented with 2% FCS, to a final concentration of 2 x 10l1 virus particles per ml. Equal volumes of antibody-polylysine DNA and virus were combined and incubated for 30 minutes at ambient temperature. The target cells used for the gene transfer were HeLa cells which had been grown in DMEM medium supplemented with 5% ~-FCS, 100 I.U. penicillin/ml and 100 ~g streptomycin/ml, in 60 mm tissue culture dishes (300,000 cells~. For comparison to HeLa cells, the cell lines HBEl, KB (ATCC
No. CCL 173 and MRC-5 (ATCC No. CCL 171) were evaluated.
HBEl, a respiratory cell line, was grown in F12-7X
medium as described by Willumson et al., 19~9. KB and MRC-5 were grown in Eagle's minimal essential medium/10%
heat-inactivated FCS/penicillin at 100 international units per ml/streptomycin at 100 ~g per ml/10 mM
nonessential amino acids/2 mM glutamine.
Before application of the transfection medium, the ; -plates were cooled at 4C for 30 minutes, the medium was removed, 1 ml of transfection medium was added and the cells were incubated for 2 hours at 4C. This step was carried out in order to bring about binding of the DNA
complexes to the cells without them being internalized. -~
, . ~: -. .

.. . ~

211~
-After this binding step, the plates were washed three times with ice-cold 2% FCS/DMEM in order to eliminate any non-bound reaction components in the liquid phase.
After the addition of 2 ml of ice-cold 2% FCS/DMEM the plates were slowly heated. Then the plates were placed in an incubator for 16 hours (37C, 5% C02). In order to measure the expression of reporter gene, cell lysates -~
were prepared, standardized in terms of their total protein content and investigated for luciferase activity exactly as described by Zenke et al., 1990. (The luminometer was calibrated so that one picogram of luciferase yields 50,000 light units.) pRSVL reporter plasmid DNA was combined with adenovirus P202-Ad5 without having been previously complexed with the polylysine antibody conjugate ~DNA +
P202-Ad5). Furthermore, pRSVL-DNA, complexed with the antibody-coupled polylysine, was investigated in the absence of the specific virus (DNA + MP301pL) and these two reaction media were compared with a reaction medium containing the total combination of the complex components (DNA + MP301pL + P202Ad5). Analogously, the complexes were investigated for their ability to perform gene transfer by using a specific antibody which had been heat inactivated before complexing (50-C, 30 min) (DNA + M~301pL + P202-Ad5). Competition experiments were carried out with the specific adenovirus in the presence of the polylysine-coupled antibody MP301 plus a ten-fold molar excess of non-polylysine-coupled MP301 (DNA + MP301pL + MP301 + P202-Ad5) or in the presence of MP301pL and a ten-fold molar excess of non-coupled irrelevant monoclonal antibody, anti-rat-IgG (DNA +
MP301pL + anti-rat IgG + P202-AD5). Furthermore, before incubation with the specific virus, the reporter plasmid DNA was complexed with non-conjugated polylysine (4 ~g) in an amount equimolar to the antibody-coupled polylysine (DNA + pL + P202-Ad5). The complex forming reactions using the adenovirus WT300, which lacks the sa~
: ::

epitope recognized by MP301, were carried out exactly as for the specific virus P202-ADs. The experiments were carried out three times in all. The results are shown in Fig. 3; the data represent mean values + SEM. The dotted horizontal line shows the background signal oP
untreated HeLa cells. The results obtained with the cell lines HBEl, KB and MRC-5 compared with He~a cells are shown in Fig. 6.

Example 3 Determination of oPtimum ratio of adenoyirus and antibodv-polvlYsine/DNA for aene transfer In the experiments carried out, the results of which are given in Fig. 4, adenovirus-antibody-polylysine/DNA complexes with the complex components in various proportions were examined for their ability to permit gene transfer into HeLa cells. The complex forming reactions were carried out as given in Example 2, except that 2.5 x 101 DNA molecules complexed with the antibody-polylysine conjugate were used, with different amounts of the specific adenovirus P202-Ad5.
The cultivation of the cells, the application of the ;
complexes to the cells, incubation of the cells and measurement of the reporter gene expression were as in Example 2. The data shown represent mean + SEM from four different experiments.
:~,~" ', Example 4 The measurement of the qene transfer performance of adenovirus-~olvcation-DNA complexes Limiting dilutions of the complex, prepared exactly as in Example 2, were investigated to see how effective they are at enabling the detectable expression of the reporter gene in HeLa cells. After complex formation, logarithmic dilutions of the complex in 2% FCS/DMEM were . :
'~
: ::
,~
:', ~ :

~ ; } ~ i 2 ,:

21~8~0 prepared. 1 ml aliquots of the various dilutions were applied to 60 mm tissue culture dishes which contained 5 x 105 HeLa cells. After one hour incubation (37~, 5 Co2), 3 ml of 5% FCS/DMEM were added and the plates were incubated for a further 16 hours under the same conditions. The reporter gene expression was measured as in Example 2. The values for luciferase expre~sion given in Fig. 5 corxespond to the mean values + SEM from 3 or 4 experiments. The dotted horizontal line shows the background signal of untreated HeLa cells.

Example 5 Gene Transfer bY means of Combination Complexes Containing Adenovirus and Human Transferrin To prepare ternary complexes containing a combination of adenovirus and human transferrin domains, the epitope-tagged adenovirus P202-Ad5 (2.5 x 101 particles) was diluted in 750 ~1 2% FCS/DMEM and combined with polylysine monoclonal antibody MP301pLys (2 ~g) diluted in 250 ~1 HBS. Incubation was performed for 30 minutes at room temperature. Plasmid DNA pRSVL
(6 ~g~ diluted in 250 ~1 HBS was then added to the mixture and incubated for an additional 30 minutes at room temperature. The resulting adenovirus-polylysine-DNA complexes were predicted to possess incompletely condensed DNA based upon total polylysine content. To complete DNA condensation and contribute a human transferrin moiety to the complexes, human transferrin polylysine conjugates ~Wagner et al., 1990) (9 ~g) diluted in 250 ~1 HBS were added to the adenovirus-polylysine-DNA complexes. A final incubation of 30 minutes at room temperature was performed. The resulting combination complexes were incubated with tissue culture cells to achieve specific binding of the formed complexes (4~C, 2 hours). The plates were then washed three times with ice-cold 2% FCS/DMEM and r ~ ' A~

8 0 ~

returned to the in_ubator (37C, 5% co2) for 16 hours after the addition of 2 ml 2% FCS/DMEM. Evaluation of reporter gene expression was as before.
Fig. 7A shows the relative values for gene expression brought about by human transferrin-polylysine-DNA complexes (hTfpL), adenovirus-polylysine-DNA complexes (AdpL) and ternary complexes, in HeLa cells. Fig. 7B shows the relative accessibility for gene transfer by ternary complexes (AdpL/hTfpL) of HeLa and HBEl cells.

Example 6 Gene Transfer in Respiratory Tract Epithelial Cells usinq Binary and TernarY Adenovirus Com~lexes For these experiments the rat Sigmodon hispidus ("Cotton Rat") was used which has been found to be a suitable animal model for human adenoviral lung diseases (Pacini et al., 1984). In addition, the binary and -~
ternary complexes described in the preceding example were used.

a) Transfection of Primary ResPiratory Tract EPithelial Cell Cultures The primary cultures were prepared using known methods (Van Scott et al., 1986). The dissociated cells were harvested, washed three times with F12-7X medium and plated out into 3cm tissue culture dishes at a density of 5 x 105 cells per dish. The cells were kept in F12-7X medium and when 50 to 75% confluence was ~-achieved they were used for the gene transfer experiments, and this normally took 2-3 days. For the gene transfer experiments the complexes were applied directly to the cells and incubated for 24 hours. For these experiments, pCMV DNA was used. The plasmid pCMV
was prepared by removing the BAMHl insert of the plasmid pSTCX556 (Severne et al., 19~8), treating the plasmid 211~0 with klenow fragment and using the HindIII/SspI and klenow-treated fragment from the plasmid pRSVL, which contains the sequence coding for luciferase, or the sequence coding for ~-galactosidase (MacGregor and Caskey, 1989), and the resulting plasmids were designated pCMVL and pCMV~-gal. Complex formation was carried out analogously to pRSVL.

i) Relative Level of Net Gene Transfer For these experiments the reporter plasmid pCMVL
was used. The cells were investigated for luciferase gene expression after 24 hours; the results are given in Fig. 9. The bottom axis shows the measurement of the unmodified cells whilst the Y axis shows the luciferase gene expression as light units per 25 ~g of total protein, obtained from cell lysates. The experiments were each carried out 3-4 times and the results are mean + SEM.

ii) Relative Transduction Frequency In these experiments the plasmid pCMV~-gal was used as reporter DNA. The cells were transfected as described above and after 24 hours the reporter gene expression was determined by staining using the method described by MacGregor et al., 1989. The results are shown in Fig. 10 (magnification: 320 x). A: hTfpL, B:
AdpL, C: hTfpL/AdpL.

¦ b) Gene Transfer via the Intra-tracheal Route in vivo I The animals were anaesthetised with methoxyflurane.
After a vertical cut had been made in the ventral side of the neck the wind pipe was cut off squarely. The complexes (250-300 ~1: 3 ~g of plasmid DNA) were injected directly into the wind pipe in full view in the animals which had been positioned at an angle of 45.
The animals were killed with C02 and the wind pipe and lungs were harvested en bloc after in situ rinsing with s o n cold phosphate buffered saline solution (PBS). For the luciferase test the lung tissue was homogenised in extraction buffer, the lysates were standardised to a total protein content and the luciferase gene expression was measured as described.

i) Relative Level of Net Gene Transfer in vivo 24 hours after transfection the luciferase expression was measured. The results are shown in Fig.
11. The light units specified relate to 1250 ~g of total protein, obtained from the lung lysates. The ;~
experiments were each carried out 3-4 times and the results are given as mean values + SEM.

ii) Localisation of Heteroloqous Gene Expression in the Res~iratorv Tract E~ithelium For these tests, the plasmid pCMV~-gal was used as reporter DNA; hTfpL/AdpL combination complexes were used. 24 hours after the injection, 14 ~g thick frozen sections of the harvested lungs were investigated for expression of the reporter gene by staining with X-gal and counter-staining with Nuclear Fast Red. The stainings are shown in Fig. 12 (magnification: 600 x);
they show the results of transfection of rats treated with hTfpL/AdpL complexes, containing an irrelevant non-lacZ plasmid designated pRc/RSV or pCMV~-gal, containing ~ ~
the lacZ reporter plasmid. A: Example of a bronchiole ~ ~;
treated with complexes containing pRc/RSV; B: Example of a bronchus, treated with complexes containing pCMB~-gal;
C: Example of the distal respiratory tract region treated with complexes containing prC/RSV; D: Example of the distal respiratory tract region treated with complexes containing pCMV~-gal; E: Enlargement of the ~-galatosidase-positive region from lungs, treated with complexes containing pCMV~-gal (magnification: 1.000 x).

~148~Q

Example 7 Construction of Coniuqates with Specificity for Lunq Reaions conjugates were constructed with a view to selective binding to the ciliated section of the respiratory tract epithelium. The construction of such conjugates demands (generally as well as in this particular instance) confirmation of the binding properties of the ligand candidates in the conjugate configuration. SNA lectin was selected as a candidate.

a) Pre~aration of Ternarv Adenovirus-Polvlvsine/
Lectin-PolylYsine/DNA ComDlexes As in the preceding examples the chimeric adenovirus P202 (2 x 101 particles) was combined with the antibody polylysine conjugate MP301pL (1.2S ~g) in 250 ~1 HBS and incubated for 30 minutes at ambient temperature. Then the reporter plasmid pCMVL was added (6 ~g in 125 ~1 of HBS) and incubation was continued for a further 30 minutes at ambient temperature. A
commercially available biotinylated lectin SNA (E-Y Lab, San Mateo, CA: 2.8 ~g) in 62.5 ~1 of HBS was combined with streptavidin-polylysine (1.35 ~g in 62.5 ~1 HBS) and left to stand for 30 minutes at ambient temperature in order to form SNA-polylysine. The SNA-polylysine was combined with the above reaction mixture in order to form SNA-adenovirus-polylysine DNA complexes. As a comparison, complexes were prepared without the SNA
ligand as described in the preceding examples.

b) In vivo use of Lectin Com~lexes in the Lunqs The cell-specific tropism of lectin SNA for the ciliated human respiratory tract epithelium has its counterpart in the ferret, which was used as animal model in the experiments. Male animals weighing about 1.5 kg were used. For each animal the complexes . ~

prepared in a) were used in four-fold amounts. The animals were anaesthetised and the complexes were introduced into the central lobe of the right lung by means of a bronchus scope. After 24 hours the different lung regions were harvested, homogenised and ~-investigated for luciferase activity. The lung regions investigated included parts which were not in contact with the complexes when they were administered (left hand upper lung section, parenchyme of the left hand upper lobe, lower part of the wind pipe) and parts which were in contact with the complexes (right hand central lobe, parenchyme of the right hand central lobe). As can also be seen from Fig. 13, the regions which were in contact with the complexes exhibited luciferase ;
expression (right hand central lobe, parenchyme of the right hand central lobe; second and third bars), wh~reas the other regions (lower part of the wind pipe, first bar; parenchyme of the left hand upper lobe, fourth bar;
left hand upper lung section, fifth bar) showed no expression.

c) Investiaating the_Specificitv of the Liqand In parallel thereto, the specificity of binding of lectin conjugates was investigated. Since no anti~
lectin antibody was available, transferrin/lectin-polylysine/DNA complexes were prepared and binding was carried out with a primary anti~transferrin antibody, backed up by a secondary horse-raddish peroxidase-coupled anti-mouse antibody. The conjugates were detected in the apical part of the ciliated cell population, but the test design did not allow of any clear conclusions as to specific binding.

-- 21~8~

Bibliography:

Anderson, W.F., 1984, Science 226, 401-409.
Anderson, P. et al., 1982, J. Biol. Chem., 257, 11301-11304.
Abrahamson, D.R. et al., 1981, J. Cell Biol., 91, 270-280.
Asada-Kubota, M. et al., 1983, Exp. Pathol., 23, 95-101.
Ascoli, M. et al., 1978, J. Biol. Chem., 253, 7832-7838.
Ashwell, G. et al., 1982, Annu. Rev. Biochem., 51, 531-554.
Berkner, K.L. and Sharp, P.A., 1983, Nucl. Acids Res.
11, 6003-6020.
Berkner, K.L., 1988, BioTechniques 6, 616-629 Berns, ~.I., l99o, Virology, 2nd Edition, Ed. by Fields, B.N., Knipe, D.M. et al., Raven Press Ltd., New York, 1743-1759.
Carpenter, G., 1984, Cell, 37, 357-358.
Cheng, S-Y. et al., 1980, Proc. Natl. Acad. Sci. USA, 77, 3425-3429.
Cotten, M., Laengle-Rouault, F., Kirlappos., H., Wagner,E., Mechtler, K., Zenke, M., Beug, H., and Birnstiel, M.L., 1990, Proc.Natl.Acad.Sci. USA 87, 4033-4037.
Ciliberto, G. et al., Cell, 1985, 41, 531-540.
Darnell, J., Lodish, H., Baltimore, D., 1975, Mol. Cell Biol., Ed. Darnell, J., Freeman, New York, 567.
Davis, B.D. and Dulbecco R., 1980, Microbiology, 3.
Edition, Ed. Davis, B.D. et al., Harper & Row, Sterilization and Disinfection, 1264-1274 Defer, C., Belin, M., Caillet-Boudin, M. and Boulanger, P., 1990, J. Virol. 64, 3661-3673.
De Wet, J., Wood, K., DeLuca, M., Helinski, D. and Subramani, S., 1987, Mol. Cell. Biol. 7, 725-737.
Dulbecco, R., 1980, Microbiology, 3. Edition, Ed. Davis, B.D. et al., Harper & Row, The Nature of Viruses, 853-884.

~ .",', X ~'~"'~;/"~

2ll~soa ~ ::

Eglitis and Anderson, 1988, Biotechniques, 6, 608-614.
Felgner, P.L., Gadek, T.R., Holm, ~., Roman, R., Chan, H., Wenz, M., Northrop, J.P., Ringold, G.M. and Danielsen, M., 1987, Proc.Natl.Acad.Sci. USA 84, 7413-7417.
Felgner, P.L., Holm, M. and Chan, H., 1989, Proc.West.Pharmacol. Soc. 32, 115.
Fields, B.N. and Knipe, D.M., 1990, Virology 2nd edition, Raven Press Ltd~, New York.
Ginsberg, H.S., 1980, Microbiology, 3. Edition, Ed.
Davis, B.D. et al., Harper & Row, Picornaviruses, 1095-1117.
Geysen et al., 1984, Proc. Natl. Acad. Sci. USA, 81, 3998. -~
Geysen et al., 1985, Proc. Natl. Acad. Sci. USA, 82, 178.
Geysen et al., 1986, Mol. Immunol., 23, 709.
Geysen et al., 1987, J. Immunol. Meth., 102, 259.
Goldstein, J.L. et al., 1982, Clin. Res., 30, 417-426.
Goldstein, J.L. et al., 1979, Proc. Natl Acad. Sci. USA, 76, 333-337.
Green, M. et al., 1989, Cell 58, 215-223.
Hearst, 3. E. and Thiry, L., 1977, Nucl. Acids Res. 4, 1 1339-1347.
¦ Heldin, C-H. et al., 1982, J. Biol. Chem., 257, 4216-~ 4221.
j Herskowitz, I., 1987, Nature 329, 219.
Hizuka, N. et al., 1981, J. Biol. Chem., 256, 4591-4597.
~ Holland, J.J., 1990, Virology, 2. Edition, Ed. Fields, 3 B.N., Knipe, D.M. et al., Raven Press Ltd., New York, Defective Viral Genomes, 151-165.
~ Horwitz, M.S., 1990, Virology, 2. Edition, Ed. Fields, 1 B.N., Knipe, D.M. et al., Raven Press Ltd., New York, Adenoviridae and their replication, 1679-1721.
Hosang, M. et al., 1987, EMBO J., 6, 1197-1202.
Huang, A.S., 1987, The Molecular Basis of Viral Replication, Hsg. Bercoff, R.P., Plenum Press New York ~ ::

211~80~

and London, The Role of Defective Interfering (DI) Particles in Viral Infection, 191-194.
Huebers. H and Finch, C., 1987, Physiol. Rev. 67, 520-582.
Hu, P.C., Collier, A.M., and Baseman, J.B., 1977, The Journal of Exp. Medicine 145, 1328-1343.
Imamura, K. et al., 1987, J. Immunol., 139, 2989-2992.
Inamine, J.M., Loechel, S. and Hu, P.C., 1988, Gene 73,175-183.
Kaplan, J. et al., 1979, J. Biol. Chem., 254, 7323-7328.
Klausner, R.D. et al., 1983, J. Biol. Chem., 258, 4715-4724.
Klausner, R.D. et al., 1983, Proc. Natl. Acad. Sci. USA
80, 2263-2266.
Kuhn, L.C. et al., 1982, Trends Biochem. Sci., 7, 299-302.
Kunkel, T.A., 1985, Proc.Natl.Acad.Sci. USA 82, 488-492.
Kurachi, K. et al., 1982, Proc. Natl. Acad. Sci. USA
1982, 79, 6461-6464.
Ledley, F.D. et al., 1989, In Biotechnology, Vol. 7b H-J
(Rehm and G. Reed, eds. (VCH Verlagsgesellschaft, Weinheim) pp. 401-457.
Loyter, A., Scangos, G.A. and Ruddle, F.H., 1982, Proc.Natl. Acad.Sci. USA 79, 422.
MacGregor, G.R. and Caskey, C.T., 1989, Nucleic Acids Res. 17, 2365.
Malim, M. et al., 1989, Cell 58, 205-214.
Malone, R.W., Felgner, P.L. and Verma, I.M., 1989, Proc.Natl. Acad.Sci. USA 86, 6077.
Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning A Laboratory Manual. Cold Spring Harbor Laboratory, 474.
Marshall, S., 1985, J Biol. Chem., 250, 4133-4144. ~
Massague, J. et al., 1986, J. Cell. Physiol., 128, 216- -222.
Maurer, R.A., 1989, Focus 11, 25.
McClure, MØ, Sommerfelt, M.A., Marsh, M. and Weiss, . ~

,, --~ 211480~

R.A., 1990, J. General Virol. 71, 767-773.
Mellman, I.S. et al., 1984, J. Cell Biol., 98, 1170-1177.
Mizel, S.B. et al., 1987, 1987, J. Immunol., 138, 2906-2912.
Pacini, D.L. et al., 1984, J. Infect. Dis. 150, 92-97.
Pastan, I., Seth, P., FitzGerald, D. and Willingham, M., 1986, in Virus attachment and entry into cells, Eds.
Crowell, R.L. and Lonberg-Holm, K., (Am.Soc.for Microbiol., Washington DC) 141-146. -Philipson, L., Lonberg-Holm, K. and Pettersson, U., 1968, J. Virol. 2, 1064-1075~
Philipson, L., 1983, Curr. Top. Microbiol. Immunol.
109, 2.
Ponder, K.P. et al., 1991, Proc. Natl. Acad. Sci USA, 88, 1217-1221.
Posner, B.I. et al., 1982, J. Cell Biol., 93, 560-567.
Precious, B. and Russell, W~C., 1985, Virology: A
practical approach, B.W.J. Mahy, ed. (IRL Press, Oxford, Washington, DC) pp. 193-205.
Roberts, M.M. et al., 1986, Science 232, 1148-1151.
Russell, W.C. et al., 1981, J. Gen. Virol., 56, 393-408.
Riordan, J.R. et al., 1989, Science, 245, 1066-1073. ~
Schalch, D.S. et al., 1986, Endocrinology, 118, 1590- -~;
1597.
Sennett, C. et al., 1981, Annu. Rev. Biochem., 50, 1053-1086.
Seth et al., Mol. Cell. Biol. 4:1528-1533 (1984).
Sly, W. et al., 1982, J. Cell Biochem., 18, 67-85.
Smith, K.A. et al., 1985, Proc. Natl. Acad. Sci. USA, 82, 864-867.
Stahl, P.D. et al., 1978, Proc. Natl. Acad. Sci. USA, 75, 1399-1403.
Sullenger, B.A. et al., 1990, Cell 63, 601-608.
Svensson, U., 1985, J.Virol., 55, 442-449.
Trono, D. et al., 1989, Cell 59, 113-120.
Uchida, Y., Tsukada, U. and Sugimori, T., 1977, J.

- 21~4800 Biochem. 82, 1425-1433.
Valerio, D. et al., 1984, Gene, 31, 147-153.
Van Scott, M.R. et al., 1986, Exp. Lung Res. 11, 75-94.
Wagner, E., Zenke, M., Cotten, M., Beug, H. and ~irnstiel, M.L., 1990, Proc.Natl.Acad.Sci. USA 87, 3410-3414.
Wagner, E., Cotten, M., Foisner, R. and Birnstiel, M.L., l991a, Proc.Natl.Acad.Sci. USA 88, 4255-4259.
Wagner, E., Cotten, M., Mechtler, K., Kirlappos, H. and Birnstiel, M.L., 1991b, Bioconjugate Chemistry 2, 226-231.
Walker, F. et al., 1987, J. Cell Physiol., 130, 255-261.
Watson, J.D. et al., 1987, Mol. Biol. of the Gene, Benjamin/Cummings Publ. Company Inc., 676-677.
Willumsen J.J., Davis, C.W. and Boucher, R.C., 1989, Am.
J. Physiol. 256, C1033-C1044.
Wood, W.I. et al., 1984, Nature, 312, 330-337.
Wu, G.Y. and Wu, C.H., 1987, J. Biol. Chem. 262, 4429-4432.
Zamecnik, P.C., Goodchild, J., Taguchi, Y. and Sarin, P.S., 1986, Proc.Natl.Acad.Sci. USA 83, 4143-4146.
Zenke, M., Steinlein, P., Wagner, E., Cotten, M., Beug, H. and Birnstiel, M.L., 1990, Proc.Natl.Acad.Sci. USA ~-;
87, 3655-3659.
Remington's Pharmaceutical Sciences, 1980, Mack Publ.
Co., Easton, PA, Osol (ed.) 3~

Claims (24)

Claims

1. New conjugate which is capable of forming complexes with a nucleic acid and consists of an internalizing factor and a substance having an affinity for nucleic acid, for introducing nucleic acid into higher eukaryotic cells, characterized in that the internalizing factor is a virus which is bound to the substance having an affinity for nucleic acid via an antibody, in such a way that said virus is capable of penetrating into the cell as part of the conjugate/nucleic acid complex and releasing the contents of the endosomes, in which the complex is located after entering the cell, into the cytoplasm.

2. The conjugate according to claim 1, characterized in that the antibody binds to a protein of said virus which does not participate in the function of the virus of penetrating into the cell and releasing the contents of the endosomes.

3. The conjugate according to claim 1 or 2, characterized in that the virus is an adenovirus.

4. The conjugate according to claim 3, characterized in that the antibody binds to an epitope in the hexon region.

5. The conjugate according to claims 3 and 4, characterized in that the adenovirus is a chimeric virus which contains the sequence coding for amino acids 914 to 928 of the Mycoplasma pneumoniae protein P1 in place of the codons 188 to 194 in the sequence coding for the hexon protein, and in that the antibody binds to the P1 region.

6. The conjugate according to one of claims 2 to 5, characterized in that the antibody is a monoclonal antibody.

7. The conjugate according to one of claims 1 to 6, characterized in that the substance with an affinity for nucleic acid is a polycation.

8. The conjugate according to claim 7, characterized in that the polycation is polylysine.

9. A complex of a nucleic acid and a conjugate of internalizing factor and substance having an affinity for nucleic acid, characterised in that it contains, as conjugate, one of the conjugates defined in claims 1 to 8.

10. The complex according to claim 9, characterized in that the nucleic acid is a gene construct which comprises a therapeutically active gene.

11. The complex according to claim 10, characterized in that the gene is a gene or gene section which is active in gene therapy.

12. The complex according to claim 10, characterized in that the gene construct comprises a section from which RNA molecules which specifically inhibit cell functions can be transcribed.

13. The complex according to claim 9, characterized in that the virus is an internalizing factor, per se, for the cell.

14. The complex according to claim 9, characterized in that the conjugate comprises a virus which, of itself, is not an internalizing factor for the cell, and the complex further comprises an internalizing factor.

15. The complex according to claim 9 or 14, characterized in that it further comprises a second conjugate of a substance having an affinity for nucleic acid and an internalizing factor which is specific for a surface receptor of a higher eucaryotic cell, wherein the virus conjugate and the internalizing factor conjugate are complexed with the nucleic acid.

16. The complex according to claim 15, characterized in that the substance having an affinity for nucleic acid of the second conjugate is an organic polycation.

17. The complex according to claim 16, wherein said polycation is polylysine.

18. The complex according to claim 15, characterized in that the second internalizing factor is transferrin.

19. The complex according to claim 15, characterised in that the second internalizing factor is a lectin.

20. A process for introducing nucleic acid into higher eucaryotic cells, characterised in that the cells are treated with one of the complexes defined in claims 9 to 19.

21. A pharmaceutical composition, comprising as active component one of the complexes defined in claims 9 to 19.

22. Transfection kit, containing a carrier unit in which there are two or more containers, a first container containing a substance having an affinity for nucleic acid, bound to an antibody, and a second container containing a virus with which the antibody is immunoreactive, the virus being capable of penetrating into a higher eukaryotic cell if it is part of a complex between the substance having an affinity for nucleic acid and a nucleic acid and wherein the virus is capable of releasing the contents of the endosomes, in which the complex is located after entering the cell, into the cytoplasm.

23. Transfection kit, containing a carrier unit in which there are one or more containers, a first container containing a substance having an affinity for nucleic acid, which is immunologically bound via an antibody to a virus which is capable of penetrating into a higher eukaryotic cell if it is part of a complex between the substance having an affinity for nucleic acid and a nucleic acid and wherein the virus is capable of releasing the contents of the endosomes, in which the complex is located after entering the cell, into the cytoplasm.

24. Transfection kit according to claim 22 or 23, characterised in that, additionally, one of the containers contains a second conjugate consisting of an internalising factor for a higher eukaryotic cell and a substance having an affinity for nucleic acid.