EP1075524A2 - Attenuated influenza viruses - Google Patents

Abstract

An attenuated influenza virus carrying a genomic nucleic acid segment which comprises 5' and 3' non-coding regions providing a mutated duplex region of an influenza virus RNA genomic segment operably-linked to a protein coding sequence for an influenza viral protein or a functional modification thereof, wherein said duplex region has at least one base-pair substitution such that expression of said protein-coding sequence in cells infected by said virus is reduced to give an attenuated phenotype. The attenuated influenza virus can be used in a vaccine.

Description

ATTENUATED INFLUENZA VIRUSES

The present invention relates to modified viruses, in particular attenuated influenza viruses which may be employed as an influenza virus vaccine. Modified viruses of the invention also include recombinant attenuated influenza viruses suitable for use as viral vectors for expression of heterologous sequences in target cells.

Influenza remains a constant worldwide threat to human health. While inactivated influenza virus vaccines have been available for many years, such vaccines provide only limited protection. Previous efforts to provide a safe, live attenuated influenza vaccine have focussed primarily on cold-adapted influenza viruses. Thus, attenuated influenza viruses have previously been obtained by extensively passaging influenza virus at low temperatures. As a result of adaptation to growth at low temperature, influenza viruses which have lost their ability to replicate at higher temperatures (about 39 °C) are obtained. The replication of such cold-adapted (CA) viruses is only slightly restricted in the cooler upper respiratory tract, but highly restricted in the warmer lower respiratory tract, the major site of disease-associated pathology. Sequence comparisons between wild-type and CA influenza viruses have revealed both silent mutations and non-silent mutations leading to amino acid changes in the coding regions of several gene segments. Most amino acid changes were found to be the result of point mutations. The genetic instability of point mutations, and the level of immunogenicity of C A influenza viruses, remain as perceived potential problems in use of CA influenza viruses as vaccines for worldwide general use. Another approach to obtaining attenuated influenza viruses which has been investigated is the construction of chimeric influenza viruses in which a non-coding region of an influenza virus genomic segment is substituted by a non-coding region from a genomic segment of an influenza virus of a different type. Such attenuated chimeric A/B influenza viruses are discussed, for example , in Muster et al., Proc. Natl. Acad. Sci. USA (1991) 88, 5177-5181, Luo et ai, J. Virology (1992) 6_.6, 4679-

4685 and Bergmann and Muster, J. General Virology (1995) 76 3211-3215. Three types of influenza virus are known designated as types A, B and C. Each of these types has many strains. The genome of an influenza virus is a segmented genome consisting of a number of negative sense RNAs (8 in the case of types A and B and 7 in the case of type C), which encode (in the case of type A) 10 polypeptides: the RNA-directed RNA polymerase proteins (PB 1, PB2 and PA) and nucleoprotein (NP) which form the nucleocapsid, the matrix proteins (Ml, M2), two surface glycoproteins which project from the lipoprotein envelope (hemagglutinin (HA) and neuraminidase (NA)) and the non-structural proteins NS 1 and NS2. The majority of the genomic RNA segments are monocistronic. Thus, in the case of influenza virus of type A, 6 of the 8 genomic RNA segments are monocistronic and encode HA, NA, NP and the viral polymerase proteins, PB1, PB2 and PA.

During the replication cycle of an influenza virus, the viral genome (vRNA) is transcribed into mRNA and replicated into complementary RNA (cRNA) molecules, which in turn are used as templates for vRNA synthesis. These processes are known to be catalyzed by the viral polymerase complex consisting of three subumts formed by the PB 1, PB2 and PA polypeptides. mRNA synthesis is initiated by capped RNA primers, which are cleaved from host cell mRNA by an endonuclease associated with the viral polymerase complex. The synthesis of mRNA is prematurely terminated at a run of uridines, in the case of an influenza A virus 16 or 17 nucleotides away from the 5' end of the vRNA template, and subsequently a poly(A) tail is added. On the other hand, cRNA synthesis is believed to be initiated in the absence of primer resulting in full-length precise copies of the vRNA segments. The nucleoprotein has been implicated as a switching factor, which acts as an antiterminator during cRNA synthesis. Influenza vRNA segments may be prepared in vitro by transcription from plasmid DNA and mixed with viral polymerase proteins and nucleoprotein to form ribonucleoprotein complexes (RNPs) having all the components necessary for transcription and replication. Such RNPs can be incorporated into viable influenza virus particles in cell packaging systems, e.g. employing a helper virus. The development of RNP reconstitution and transfection systems has permitted detailed characterization of the RNA signals in influenza A vRNAs involved in the regulation of transcription initiation, termination, and polyadenylation (4, 20-22, 25, 32, 34). All these signals are known to reside in the terminal sequences of vRNA segments (19). The 5¹ and 3' ends contain 13 and 12 conserved nucleotides respectively, which have the ability to form a partially double-stranded panhandle/RNA-fork or corkscrew structure (6, 7, 13). Initial in vitro transcription studies with model RNA templates implied that vRNA and cRNA promoters were located exclusively in the 3' terminal sequences (25, 32) and that the panhandle had no apparent role in the initiation of transcription in vitro. However, detailed mutagenesis studies of the terminal sequences subsequently showed that the 5' end forms an integral part of the promoter. These findings were based on binding experiments of the RNA polymerase to the putative promoter RNA (7, 33) and, more importantly, on in vitro transcription studies with mutant model template RNAs (7, 8, 28). In addition, activation of the viral polymerase-associated endonuclease requires interaction of the polymerase complex with the 5' as well as the 3' terminal sequences of vRNA segments (11).

The postulated double-stranded region of the promoter of an influenza A vRNA segment is now recognised to consist of 5 to 8 base-pairs. The first 3 base-pairs, those formed by nucleotides 11' to 13' at the 5' end and nucleotides 10 to 12 at the 3' end, are strictly conserved among different vRNA segments of all influenza A viruses. Sequencing studies have shown that the 3' and 5' non-coding terminal sequences of influenza B and C vRNA segments are also highly conserved and show partial inverted complementarity (36, 37). Consequently, it is believed that the capability of base-pairing of nucleotides of the non-coding regions to form a panhandle structure is important for proper functioning of all influenza vRNAs. The term duplex region of an influenza vRNA segment as used hereinafter will be understood to refer to the region which is formed by such base-pairing.

Kim et al. (14) have previously used a choloramphenicol acetyhτansferase (CAT) reporter gene construct in which negative sense CAT RNA is flanked by the non-coding sequences of an influenza A virus NS gene to determine the effect of mutations in the postulated duplex promoter region on CAT expression in

Madin-Darby bovine kidney (MDBK) cells. Negative-sense CAT RNA constructs were incorporated into RNP complexes, which were then used to transfect monolayers of MDBK cells infected with a helper influenza virus and CAT activity assayed. Using this model system, single mutations of the conserved residues at positions 11 and 12 of the 3' terminus and at positions 12' and 13' of the 5' terminus of the CAT gene construct were found to abolish or virtually abolish CAT activity.

The introduction of second complementary mutations into such constructs so as to restore the capability for Watson-Crick base-pairing was found, however, to partially restore CAT activity. Thus, the constructs with the base-pair substitutions of U12-A13* for C12-G13^* and Al 1-U12' for Cl 1-G12' were found to express CAT at 31% and 22% respectively compared to the control construct with wild-type influenza A gene non-coding regions.

The same CAT reporter gene system was also used to investigate the effect of mutations of the U10-A11' base-pair. Single mutations, U10 to G10 and Al 1' to Cl 1', significantly decreased CAT activity, but both mutants exhibited detectable activity. A combination of the two mutations to introduce a Gl 0-C 11 ' base-pair did not give improved CAT activity. It was therefore suggested that the properties of the base-pair at positions 10-11' might be different from those at positions 11-12' and 12-13'.

Such experiments merely test the effect of influenza vRNA duplex region mutations on the expression of a heterologous CAT reporter gene in cultured human cells. It is not possible to predict from such studies whether mutations which allow some CAT activity will, when incorporated into an influenza vRNA genomic fragment, permit rescue of that fragment into a viable virus. Equally, it is not possible to predict, even if such mutations give rise to viable virus, whether such viruses will be attenuated. Indeed, this is supported by the finding of the inventors that the base-pair substitution of C12-G13' by U12-A13' in the NA gene vRNA segment of an influenza A virus can be rescued into a viable influenza A virus which does not show significant attenuation on MDBK cells (see the Examples).

In contrast, it has now been established that substitution of A for C and U for G at position 11-12' in the duplex region of the NA-specific vRNA of an influenza A virus does lead to attenuation on MDBK cells and also other cell types in culture. It has also been shown that influenza A virus with the same base pair substitution is attenuated in vivo and can give rise to protective immunity against wild-type influenza A virus. Evidence suggests that such attenuation arises from reduced polyadenylation of the NA-specific mRNA. Base-pair substitution in the duplex region of a vRNA segment is thus proposed as a new general strategy for achieving attenuation of influenza viruses. Such base-pair substitution can be selected by application of known rescue systems for incorporating genetically-engineered influenza vRNA segments into viable influenza viruses as further discussed below. In one aspect, the present invention thus provides an attenuated influenza virus carrying a genomic nucleic acid segment which comprises 5' and 3' non-coding regions providing a mutated duplex region of an influenza virus RNA genomic segment operably-linked to a protein coding sequence for an influenza viral protein or functional modification thereof, wherein said duplex region has at least one base-pair substitution such that expression of the said protein-coding sequence in cells infected by the said virus is reduced to give an attenuated phenotype.

Mutated duplex region of an influenza virus RNA genomic segment will be understood to exclude any native influenza virus vRNA duplex region derived from a vRNA of a wild-type influenza virus of a different type.

The term "cells" in this context may encompass human and/or animals cells in vivo normally infected by influenza viruses. For the purpose of selection of attenuated viruses of the invention, the same term will be understood to refer to cells of a single cell type or more than one type, e.g. cultured human or non-human animal cells of one or more than one type. They may be in vivo cells, e.g. cells of an animal model. Cultured cells which may prove useful in the selection of attenuated viruses of the invention in vitro include one or more of MDBK cells, Madin-Darby canine kidney (MDCK) cells and Vero (African green monkey kidney) cells.

While an attenuated virus of the invention may have a single base-pair substitution in the duplex non-coding region of a genomic segment, it will be appreciated that such a virus may have more than one such substitution, either on the same genomic segment or different genomic segments, e.g. 2 base pair substitutions in the same genomic segment duplex region. The duplex base-pair substitution(s) will desirably result in some, e.g. at least about one log, reduction in plaque titre compared to the parent wild-type virus on MDBK cells. The duplex base-pair substitution(s) will desirably provide an attenuated virus exhibiting some, e.g. at least about one log, more preferably at least about 3 to 4 log, reduction of plaque titre on MDCK cells and Vero cells compared to the parent wild-type virus. An attenuated virus of the invention may, for example, exhibit as much as about 5 log reduction of plaque titre compared to the parent wild-type virus on Vero cells arising from the vRNA non-coding region base substitutions. Such an attenuated virus is exemplified by influenza A/WSN/33 having an NA-specific vRNA segment incorporating the base-pair substitution A 11 -U 12' for C-G at position 11 - 12' of the duplex region and additionally having the base-pair substitution G10-C11' for U10-A11' (mutant Dl/2 referred to in the examples). Other influenza A viruses incorporating the same base-pair substitutions, either in the NA-specific vRNA segment or a vRNA segment encoding another influenza virus protein, also exemplify the invention. As indicated above, attenuated viruses of the invention also include influenza

A/WSN/33 having the single base-pair substitution Al 1-U12' in the NA-specific vRNA segment (mutant D2 referred to in the examples) and other influenza A viruses having the same base-pair substitution in the NA-specific vRNA segment or another viral protein-encoding vRNA segment. Thus, in one embodiment the present invention provides an attenuated influenza virus of type A carrying a mutated influenza A virus genomic RNA segment having the mutation C to A at position 11 from the 3' terminus of the native parent segment and the mutation G to U at position 12' from the 5' terminus of the native parent segment, or functionally equivalent substitutions such as modified base substitutions at the same positions, so as to provide an attenuating base-pair substitution in the non-coding duplex region.

Additionally, in a further embodiment, the present invention provides such an attenuated virus of type A which in the same vRNA segment has the mutation U to G at position 10 from the 3' terminus of the native parent segment and the mutation A to C at position 11' from the 5' terminus of the native parent segment, or functionally equivalent substitutions at the same positions, so as to provide an additional base-pair substitution in the non-coding duplex region. Such a virus may be a wild-type virus which has been attenuated by introduction of one or more base-pair substitutions as above into the non-coding duplex region, or a recombinant attenuated virus carrying a heterologous coding sequence as further discussed below. Desirably, for example, the attenuating base-pair substitution(s) will be introduced into the genomic nucleic acid segment encoding NA or a functional modification of that surface glycoprotein.

Although the invention is further illustrated hereinafter with particular reference to influenza A/WSN/33, the invention is not confined to influenza viruses of the A-type. Functionally equivalent mutations to the D2 or D 1/2 mutations, i.e. attenuating base-pair substitutions, in viruses of the B and C types may be analogously identified by reference to available sequence information and application of known rescue systems applicable to any genetically-engineered influenza vRNA segment suitable for providing the characteristic of attenuation to a complete influenza virus.

Thus, a further embodiment of the invention, is an influenza virus of type B carrying a mutated influenza B virus genomic RNA segment, e.g. NA-encoding segment, having an attenuating base-pair substitution in the non-coding duplex region at a functionally homologous position to the base-pair substitution in influenza A/WSN/33 designated above as D2. The invention also extends to influenza viruses of type C carrying such a base-pair substitution in a mutated influenza C virus genomic RNA segment, e.g. a mutated NA-encoding segment.

Brief Description of the Figures

Figure 1 is a representation of the conserved sequences of an influenza A virus vRNA in the panhandle/RNA-fork conformation (7, 13). Conserved base-pairs in the double-stranded region of the RNA-fork, involving both the 5' and 3' ends of the RNA segment, are boxed. Numbering of residues starts from the 3' end and from the 5' end. The 5' end numbers are distinguished by prime ('). Base-pairs in the conserved double-stranded region of the modified NA-encoding vRNA of the transfectant viruses designated DI, D2, D3 and Dl/2 in the examples are shown. Changed base-pairs are highlighted.

Figure 2 shows growth curves of transfectant viruses on MDBK cells. Confluent cells in 35 mm dishes were infected with wild-type influenza A/WSN/33 (wild-type; WT) virus, and with the transfectant DI, D2, D3 or D 1/2 viruses at a multiplicity of infection (m.o.i.) of 0.01. At the indicated time points, infectious particles present in the media were titrated by plaque assay in MDBK cells. The presented values are averages from duplicate experiments.

Figure 3 shows the nucleotide sequence of the plasmid pT3NAml containing the full-length cDNA of the NA gene of influenza A/WSN/33 (positions 2412-3820) flanked by a unique Bbsl restriction site at one end (position 2404) and a bacteriophage T3 RNA polymerase promoter at the other end (positions 3821-3836) in the background of the pUC19 cloning vector between the EcoRl (position 2398) and Hind III (position 3837) restriction sites (9). This plasmid was employed to obtain the mutant versions of the NA-encoding vRNA of influenza A/WSN/33 present in the DI, D2, D3 and Dl/2 viruses (see Example 1).

Figure 4 shows the time course of pathogenicity of wild-type, DI, D2, D3 and Dl/2 viruses in mice when intranasally infected with 10³ plaque-forming units (pfu)

(see Example 13).

Figure 5 shows body weight following intranasal infection of mice with wild-type, DI, D2, D3 and Dl/2 viruses at 10³ pfu.

Figure 6 shows the time course of pathogenicity of wild-type, DI, D2, D3 and Dl/2 viruses in mice when intranasally infected with 3xl0⁴ pfu.

Figure 7 shows body weight following intranasal infection of mice with wild-type, DI, D2, D3 and Dl/2 viruses at 3xl0⁴pfu.

Figure 8 shows the time course of pathogenicity of wild-type, DI, D2, D3 and Dl/2 viruses in mice when intranasally infected with 10⁶ pfu. Figure 9 shows body weight following intranasal infection of mice with wild-type, DI, D2, D3 and Dl/2 viruses at 10⁶ pfu.

Figure 10 shows viral titres (log pfu per ml) on lungs of mice at 3 days (left) and 6 days (right) post-infection , following intranasal infection with wild-type (WT) and DI, D2, D3 and Dl/2 viruses at 10³ pfu (see Example 14). Figure 11 shows body weight of D2-immunised mice (3 dose levels: 10⁶,

3x10⁴ and 10³ pfu) following challenge with 10⁶ pfu wild-type virus (see Example 15).

Figure 12 shows body weight of Dl/2-immunised mice (3 dose levels: 10⁶, 3x10⁴ and 10³ pfu) following challenge with 10⁶ pfu of wild-type virus.

A nucleic acid segment of a virus of the invention incorporating an attenuating base-pair substitution as discussed above, and DNAs capable of transcription to provide such a nucleic acid, also constitute additional aspects of the invention. A nucleic acid of the invention may preferably correspond to a mutated native influenza virus RNA genomic segment having an appropriate attenuating base-pair substitution in the non-coding duplex region. Such an RNA may have additional modifications, for example, one or more additional nucleotides added at the 3' and/or 5' terminus or internally which do not destroy function. It may be a chimeric RNA.

A DNA capable of transcription in vitro to provide an RNA nucleic acid segment of the invention may be initially constructed in a plasmid by application of conventional techniques and isolated from that plasmid by restriction endonuclease digestion. As illustrated by plasmid pT3NAml referred to above, for this purpose a cDNA of a native influenza virus vRNA segment may be inserted into a plasmid flanked by an appropriate promoter and a restriction endonuclease site. The cDNA may then be subjected to site-directed mutagenesis by, for example, PCR-directed mutagenesis employing appropriate mutagenic primers to provide a sequence encoding the desired mutated vRNA segment for transcription. Alternatively, a genomic nucleic acid segment of the invention may be synthesized.

For preparation of an attenuated virus of the invention, a genomic nucleic acid segment having at least one attenuating base-pair substitution as defined above may be complexed in vitro with influenza viral polymerase proteins and nucleoprotein to form a RNP complex. Such RNP complexes, which constitute a still further aspect of the present invention, may be prepared in conventional manner as previously employed for incorporation of genetically-engineered influenza vims RNA genomic segments into RNA complexes for viral rescue in cells (4, 5, 38).

RNP complexes of the invention may be transfected into cultured cells, e.g. MDBK cells, MDCK cells or Vero cells, again using conventional techniques. Methods commonly employed for this purpose include DEAE-dextran transfection and electroporation (19, 39).

In yet another aspect, the present invention provides a method of preparing an attenuated influenza vims of the invention which comprises providing in a host cell the genomic nucleic acid segments for said vims under conditions whereby said segments are packaged into a viral particle. For this purpose, the genomic nucleic acid segments may be provided in the host cell by plasmids. Alternatively, RNP complexes of the invention as hereinbefore described may be transfected into host cells that have previously been infected with an influenza helper vims to complement the RNP complexes and enable selection of the desired attenuated viral particles. A number of helper vims-based cellular rescue systems for particular influenza vims genes have previously been described and have been reviewed by Muster and Garcia-Sastre (56). Such gene specific rescue systems are briefly summarized below.

Helper vims based influenza gene rescue systems

Helper based rescue systems have been reported allowing the genetic manipulation of influenza A vRNAs for NA and HA surface antigens, the non- structural proteins, NP, PB2 polymerase protein and the M proteins.

NA gene specific rescue system

The most commonly employed helper vims based influenza gene rescue system is limited to the NA of influenza A/WSN/33 vims (4, 5). This method is based on the observation that only influenza vimses with an NA gene from influenza A/WSN/33 are able to grow on MDBK cells in the absence of trypsin. In this rescue system, the helper vims is a reassortant containing seven gene segments from influenza A/WSN/33 and a NA gene from a vims other than influenza A/WSN/33. Generally A/WSN-HK, which has an NA gene from influenza A/HK/8/68, is used as the helper vims. In this system, the NA gene of influenza A/WSN/33 is transfected into cells infected with the helper vims. The vims is then selected by growing on

MDBK cells in the absence of exogenous proteases. NA genes can also be rescued by using a NA-deficient mutant vims as a helper vims. Such a helper vims requires exogenous neuraminidase to grow in tissue culture. The NA-gene is transfected into cells infected with the helper vims. The vims is then selected by growing on cells in the absence of neuraminidase (43).

NS gene specific rescue system

A temperature-sensitive influenza vims with a defect in the NS 1 protein is used as the helper vims of a NS gene specific rescue system. The NS gene segment carries two overlapping genes coding for the NS1 and NS2 proteins. This rescue system allows the rescue of a NS gene segment encoding an NS 1 protein which has activity at the non-permissive temperature. In this system, the NS gene segment which is to be rescued is transfected into cells infected with the temperature-sensitive vims. The vims with the transfected NS gene segment is selected by growing the vims at the non-permissive temperature as described by Enami et al. (40).

PB2 gene specific rescue system

A vims with an avian influenza A vims PB2 gene can be used as the helper vims in a PB2 gene specific rescue system. The avian influenza A vims PB2 gene restricts the replication of the helper vims in mammalian cells. Therefore, this rescue system can rescue a PB2 gene which allows replication of influenza vims in mammalian cells. The PB2 gene which is to be rescued is transfected into cells infected with the helper vims. The vims with the transfected PB2 gene is selected by growing the vims in mammalian cells. Subbarao et al. (41) have used such an avian influenza A vims PB2 gene based system to rescue the PB2 gene of wild-type influenza A/ Ann Arbor/6/60 vims.

M gene specific rescue system

An amantidine-sensitive influenza vims carrying an M gene of influenza A/equine/Miami/1/63 vims can be used as a helper vims of an M gene specific rescue system. The rescue system allows the rescue of an M gene which confers amantidine resistance to a vims. In this system, the M gene which is to be rescued is transfected into cells infected with the helper vims. The vims with the transfected M gene is selected by growing the vims in the presence of amantidine. Castrucci and Kawaoka (42) have used such an amantidine-sensitive M gene based system to rescue the M gene of influenza A/PR/8/34 vims.

Antibody-based rescue systems

These systems depend on the binding or non-binding of the transfectant vims to a particular antibody (5, 52). Such antibody is a neutralising antibody which binds to influenza vims and impairs its growth in tissue culture. The helper vims may, for example, carry a gene which encodes an influenza surface protein which displays the antibody epitope. This system can therefore be used to select for transfectant vims which does not carry such a gene, but which of course is viable. This type of rescue system thus allows the rescue of a gene encoding an influenza surface protein. The gene to be rescued is transfected into cells infected with the helper vims. The vims with the transfected gene is selected by growing the vims in the presence of the antibody. Such a system was used by Enami and Palese (5) to rescue a transfected synthetic HA segment.

NP gene specific rescue system Li and coworkers (39) reported a reverse genetics system for the rescue of the influenza A vims nucleoprotein gene. In this system, a temperature-sensistive (ts) mutant ts56 is used as a helper vims. RNA complexes are reconstituted in vivo as described before (5) and are then introduced by electroporation into ts56 helper virus infected cells. Transfectant vimses with a rescued NP-encoding vRNA segment are selected at the non-permissive temperature by plaquing on MDBK cells.

Influenza B vims rescue system

Barclay and Palese (44) have additionally described the rescue of HA genes in an influenza B vims.

The preparation of an attenuated vims of the invention may alternatively be achieved using the expression vector-based influenza gene rescue strategy developed by Pleschka et al. (45). In contrast to the RNP transfection system referred to above, this eliminates the need for purification of the viral NP and polymerase proteins which is required for in vitro reconstitution of RNP complexes. Expression vectors are co-transfected into host cells which will provide the NP and P proteins and also a genomic segment of the invention incorporating an attenuating base-pair mutation. In this case, RNP complexes of the invention are formed intracellularly. The cells may then be infected with an influenza helper vims as previously described to select for the required attenuated influenza vims . An RNA complex of the invention may also be rescued in host cells into a viable attenuated vims by transfecting into the host cells additional complementing RNA complexes thereby eliminating the need for a helper vims. This may be achieved in accordance with the general rescue strategy for influenza vims genes more recently described by Enami (46 ). This strategy involves purifying RNPs from an appropriate influenza vims and treating the RNPs in vitro with RNase H in the presence of a cDNA which hybridizes to the influenza vims gene to be rescued. In this way specific digestion of that gene by the RNase H is achieved. The gene depleted RNPs are then co-transfected into cells with the RNP-complex containing the nucleic segment to provide the attenuating base-pair substitution. The cells are then overlaid with agar and transfectant attenuated vimses obtained by direct plaque formation. This strategy, unlike the above described helper vims-based gene rescue strategies, can be applied to any influenza gene from any influenza vims. It can thus be applied to obtain an attenuated vims or gene of the invention of any influenza type. Since reversion of a base-pair mutation requires two specific mutations, attenuated influenza vimses of the invention are expected to be highly stable (see Example 12). Hence, such vimses may be particularly favoured for use as influenza vims vaccines.

As indicated above, a vims of the invention may additionally contain a heterologous coding sequence capable of being expressed in target cells. Such a heterologous coding sequence may encode an antigenic peptide or polypeptide capable of stimulating an immune response (either an antibody response or a cell-mediated immune response) to a pathogenic agent. Representative examples of such pathogenic agents are vimses, e.g. other influenza vimses or non-influenza vimses such as HIV, bacteria, fungi, parasites, eg. malarial parasites, and disease-causing cells such as cancer cells.

Thus, in yet another aspect, the present invention provides a vaccine comprising a vims of the invention. Particularly preferred are such vaccines wherein the attenuated influenza vims acts as a combined vaccinating agent against more than one pathogenic agent, e.g. an influenza vims and a second pathogenic agent other than an influenza vims. Such vaccines may be formulated and administered in accordance with known methods for this purpose.

Thus in a still further aspect, the present invention provides a method of stimulating an immune response against an influenza vims, e.g. an influenza vims of Type A, either alone or together with stimulation of an immune response against one or more further pathogenic agents, which comprises administering in an immunising mode an attenuated influenza vims of the invention capable of inducing said immune response(s). Intranasal immunisation with an attenuated influenza vims of the invention may, for example, be preferred. Such immunisation may be carried out as illustrated by the immunisation studies with recombinant influenza vimses expressing an HTV-epitope reported by Muster et al. (49) and Ferko et al. (53) (see also Example 15). A suitable immunisation dose may be, for example, in the range of 10³-10⁹ pfu. Booster immunisations may be given following an initial immunisation with a vims having the same functional characteristics, but of a different subtype or type. Methods for incorporating heterologous coding sequences into an influenza vims have previously been described, for example, in Published International Application WO91/03552 (Palese et al) and are also reviewed by Muster and Garcia-Sastre in Textbook of Influenza 1998 (56). The heterologous coding sequence may be on a genomic segment incorporating an attenuating base-pair substitution or on a different genomic segment. It may be carried by an additional nucleic acid segment also incorporating a gene for an influenza viral protein to provide selection pressure. It has previously been reported, for example, that an influenza vims can be constmcted carrying at least 9 different vRNA segments (40). Use of attenuated recombinant influenza vimses of the invention as vectors to express foreign antigens for vaccinating purposes is an attractive therapeutic strategy since:

(i) Antibodies to the different subtypes show little cross-reactivity. One drawback with the use of a vims as a vaccine is that an immune response will be produced to the vims. It is often desired that one or more booster immunisations comprising the same antigen are given after the initial immunisation. However, the immune response to the vims reduces the effectiveness of subsequent immunisations with the same vims. Since antibodies to different influenza subtypes show little cross-reactivity, subsequent immunisations with an influenza vims of a different subtype but which expresses the same antigen should overcome this effect. (ii) Influenza vimses have been shown to induce strong cellular and humoral responses.

(iii) Influenza vimses have been shown to induce strong mucosal responses.

Intranasal immunisation with influenza vims has been shown to induce long lasting responses in genital and intestinal mucosa. (iv) Influenza vimses are non-integrating and non-oncogenic.

(v) As previously noted above, attenuated influenza vimses of the invention can be anticipated to be attenuation stable. For vaccinating purpose, a heterologous coding sequence may be provided in an attenuated vims of the invention encoding an antigen of a pathogenic agent or a modification thereof capable of stimulating an immune response. The heterologous coding sequence may be inserted into a viral gene to provide a fusion protein which retains the function of the parent viral protein. One site which has previously been found to tolerate insertions of foreign antigens (epitope grafting) is the antigenic B site of HA. Antigenic site B of that surface protein consists of an exposed loop structure located on top of the protein and is known to be highly immunogenic.

Manipulation of the HA gene of an influenza vims to insert a viral epitope in the HA protein B site has previously been reported (see again the studies of Muster et al. reported in 49 and the studies of Li et al. reported in 48). The same strategy has also previously been employed by Rodrigues et al. to express B-cell epitopes derived from a malaria parasite (50). Heterologous coding sequences for an antigenic polypeptide may also, for example, be preferably inserted into an influenza vims NA gene. Strategies for epitope grafting into influenza viral proteins have also previously been described, for example, in WO91/03552.

Epitope grafting of a foreign sequence into an influenza vims protein may result in a non-functional chimeric viral protein and make the rescue of a viable transfectant vims impossible. A different strategy for expressing foreign sequences by recombinant influenza vimses, which may be applied to attenuated vimses of the present invention, involves the engineering of gene segments containing an additional open reading frame. A recombinant genomic segment may be constmcted which provides an internal ribosome entry site for a heterologous coding sequence. This approach has previously been used, for example by Garcia-Sastre et al. to obtain an influenza vims vRNA segment which encodes both a tmncated form of gp41 of HIV and NA (9). Alternatively, a heterologous coding sequence may be fused in frame to a viral protein coding sequence to encode a chimeric polyprotein capable of autoproteolytic protease cleavage to give the viral protein and a desired second polypeptide, e.g. a viral antigen. This strategy has been shown by Percy et al. to be suitable for expressing non-influenza proteins up to 200 amino acids in length (51).

It will be appreciated that a recombinant attenuated vims of the invention may be employed as a vehicle for expression of heterologous coding sequences in target cells for a variety of therapeutic purposes in addition to vaccination. Such a recombinant vims may, for example, have a genomic segment encoding any of the following:

- a cytokine such as an interferon or an interleukin,

- a toxin,

- a palliative capable of inhibiting a function of a pathogenic agent either directly or indirectly, e.g. a viral protease inhibitor

- an enzyme capable of converting a compound with little or no cytotoxicity to a cytotoxic compound, e.g. a viral enzyme such as Herpes simplex thymidine kinase capable of phosphorylating purine and pyrimidine analogues to active toxic forms, - an antisense sequence, - a ribozyme.

Sequences encoding such agents may be incorporated into an attenuated influenza vims of the invention by any of the techniques previously referred to above in connection with providing attenuated vimses of the invention expressing foreign epitopes. A heterologous coding sequence in an attenuated recombinant vims of the invention may be under the control of a tissue-specific and/or event-specific promoter. A recombinant vims of the present invention may be employed for gene therapy.

A recombinant vims of the invention may be administered directly or used to infect cells ex vivo which are then administered to a patient.

Thus, in still further aspects, the present invention provides a pharmaceutical composition comprising a recombinant vims of the invention in combination with a pharmaceutically acceptable carrier or diluent for delivery of a heterologous coding sequence to target cells. It also provides ex vivo cells infected by a vims of the invention and such cells hosting a recombinant influenza vims of the invention formulated for administration with a pharmaceutically acceptable carrier or diluent. In yet another aspect, the present invention provides a method of delivering a heterologous coding sequence to cells which comprises infecting said cells with an attenuated recombinant influenza vims of the invention carrying said sequence. Vimses of the invention may also find use as a helper vims to rescue genes which can substitute for the gene(s) affected by the attenuating mutation(s) to provide vimses showing increased growth on a selected cell type. For this purpose, an attenuated vims will preferably be chosen which exhibits at least about a 3-4 log, preferably at least about a 5 log, reduction in growth compared to the corresponding wild-type vims on one or more cell types. Thus, in yet another embodiment, the present invention provides use of a vims of the invention as a helper vims to rescue an influenza vims genomic nucleic acid segment in cells, wherein vimses produced containing said segment are selected on the basis of increased growth compared with the helper vims on cells of a selected type. For example, an influenza A vims of the invention having an attenuating base-pair substitution in the non-coding duplex region of its NA-encoding vRNA may be usefully employed to rescue an

NA-encoding vRNA or functional modification thereof derived from a second influenza A vims. A typical protocol for this purpose will comprise the steps of:

1. infecting cells with the helper vims,

2. transfection of an RNP complex containing the gene(s) to be rescued into the helper vims infected cells, and

3. selection of rescued vimses, either on the same cell type or a different cell type on which the helper vims shows increased attenuation.

The cell type in step 3 will be chosen such that only vimses which have acquired the transfected gene(s) are expected to grow to high titre. For example, the D 1 /2 mutant version of influenza A/WSN/33 referred to above is particularly favoured as a helper vims for use to rescue NA genes originating from other influenza vimses of the A-type. In this case, MDBK cells may, for example, be initially infected with the Dl/2 helper vims and Vero cells preferably used for selection of vimses carrying an NA gene containing vRNA without an attenuating mutation. The D2 mutant derived from influenza A/WSN/33 may similarly be employed.

Influenza A/WSN/33 is known to exhibit in mice neurovimlence associated with the surface antigen NA (54). For this reason, the attenuated modified versions of that vims referred to above are not regarded as suitable for direct vaccine use. However, by using, for example, the Dl/2 mutant as a helper vims as above, NA vRNAs may be obtained for site-directed mutagenesis to constmct alternative attenuated influenza A vimses according to the invention more suitable for therapeutic, e.g. vaccine, use.

The following examples illustrate the invention. Example 1

Introduction of mutations into the duplex region of the NA-encoding vRNA of an influenza vims of type A

In order to produce NA-encoding viral genomic RNA with mutations in the 5' and 3' non-coding regions, plasmids were constmcted which contained the corresponding cDNA with the desired mutations.

The starting plasmid for site-directed mutagenesis was pT3NAml (see Figure 3) which, as previously noted above, contains the full length cDNA of the NA gene of influenza A/WSN/33 vims (positions 2412-3820) flanked by a unique Bbsl restriction site at one end (position 2404) and a bacteriophage T3 RNA polymerase promoter at the other end (positions 3821-3836) in the background of the pUC19 cloning vector between the EcoRl (position 2398) and Hind III (position 3837) restriction sites (9). Samples of influenza A/WSN/33 for preparation of the NA-encoding cDNA insert in plasmid pT3NAml are obtainable, for example, from the W.H.O. Collaborating Centre, Division of Virology, National Institute for

Medical Research, London, U.K.

An alternative plasmid which may be employed to construct DNA templates for transcription of mutant NA-encoding vRNA segments of influenza A/WSN/33 is the pUC19-derived plasmid pT3NAv, whose constmction is described in WO91/03552 (Palese, P. et al.). Plasmid pT3NAv also contains the full length cDNA of the NA gene of influenza A/WSN/33 flanked by a promoter specifically recognised by bacteriophage T3 RNA polymerase and a restriction endonuclease cleavage site.

PCR products were made using pT3NAml as a template and the following primers modified to provide mutations as specified in Fig. 1 :

5'-CGGAATTCGAAGACGCAGCAAAAGCAGGAGTTTAAATGAATCC-3' (primer 1) and 5'-

CCAAGCTTATTAACCCTCACTAAAAGTAGAAACAAGGAGTTTTTTGAA C-3' (primer 2) (the residues at which mutations were introduced are underlined, e.g. for constmction of the DI mutant cDNA, in both primers 1 and 2 the first underlined

A nucleotide was substituted by a C nucleotide). The PCR products were digested with EcoRI and Hindlll restriction enzymes and they were cloned into pT3NAml cut with the same enzymes. NA genes and the flanking sequences in the modified plasmids were sequenced with an automated sequencer (Applied Biosystems). The following double-mutations were introduced into the NA gene of influenza A WSN/33 vims: U-A-G-C (10-11') (mutant DI), C-G-A-U (11-12')

(mutant D2), and C-G-U-A (12-13') (mutant D3) (Fig. 1). In addition, six NA genes with the corresponding single-mutations were constmcted (U-G10, A-Cl 1', C→Al 1, G->U12', C-U12, and G-A13').

Example 2

Production of and transfection of ribonucleoprotein (RNP complexes.

Transfectant vimses were prepared as described by Enami and Palese (5). NA-specific RNP complexes were reconstituted in vitro and transfected into MDBK cells infected with A/WSN-HK helper vims (5). Synthetic RNAs were obtained by T3 RNA polymerase transcription of modified pT3NAml plasmids linearized with Bbsl restriction enzyme. RNAs were reconstituted into RNP complexes using RNA polymerase and NP protein isolated from influenza X-31 vims. Influenza X-31 vims is a reassortant of influenza A/HK/8/68 and A/PR 8/34 vimses and was supplied by Evans Biological, Ltd., Liverpool, England. The RNP complexes were transfected by the DEAE-dextran transfection method into MDBK cells infected with WSN-HK helper influenza vims grown in 10-day embryonated chicken eggs. The MDBK cells were grown in reinforced minimal essential medium. For subsequent experiments, influenza A/WSN/33 wild-type vims was also grown in MDBK cells in reinforced minimal essential medium. Rescued transfectant vimses were plaque purified three times in

MDBK cells. A single plaque was used for preparing a stock vims for further analysis.

Example 3 Sequencing of the NA genes of transfectant vimses.

The presence of the mutations in the transfectants was confirmed by sequence analysis of the 3' and 5' terminal sequences of the NA gene. Viral RNA for sequencing was isolated by phenol-chloroform extraction from transfectant vimses purified by centrifugation through a 30 % sucrose cushion. In some cases, total RNA isolated with RNAzol B (Tel-Test, Inc., Friendswood, TX) from infected cells was used. Sequences of the 5' end were obtained either by direct RNA sequencing or by 5'

RACE. Direct sequencing of the 5' ends was performed using a primer complementary to nucleotide positions 1280 to 1299 (5'-

TGGACTAGTGGGAGCATCAT-3') of the influenza A/WSN/33 NA gene and an RNA sequencing kit (United States Biochemical Corporation, Cleveland, OH) following the manufacturer's instmctions. For 5' RACE, viral RNA was reverse transcribed using a primer complementary to nucleotide positions 879 to 898 (5'-GGGTGTCCTTCGACCAAAAC-3') of the influenza A/WSN/33 NA gene. The reverse transcription product was extended with terminal deoxynucleotidyl transferase (TdT) (Gibco BRL, Gaithersburg, MD) and amplified by PCR with the primer used for direct RNA sequencing (see above) and the 5' RACE abridged anchor primer (Gibco BRL). PCR products, cut with Spel restriction enzyme, were cloned into the Xbal site of pUC18 and sequenced with a DNA sequencing kit (United States Biochemical). In order to sequence the 3' end of the NA gene of transfectant vimses, viral RNA was 3'-polyadenylated using poly(A) polymerase (Gibco BRL). The polyadenylated RNA was reverse transcribed using the primer

5*-GCGCAAGCTTCTAGATTTTTTTTTTTTTT-3' and the cDNA was amplified by PCR with a primer containing nucleotides corresponding to positions 115 to 98 (5'-GCGCAAGCTTTATTGAGATTATATTTCC-3') of the influenza A/WSN/33 NA gene and the primer used for reverse transcription. PCR products digested with Hindlll were cloned into pUC 18 and sequenced with the DNA sequencing kit.

Transfection of all three NA genes with double-mutations resulted in rescue of transfectant vimses (DI, D2, and D3). On the other hand, only three out of the six single-mutant constmcts were rescued, carrying mutations at positions 10, 11', and 13' (Fig. 1). In three attempts, none of the other three constmcts (with mutations at positions 11, 12, and 12') was rescued.

Confirmation of mutations in the two single mutant transfectants at positions 10 and 11^* was more difficult since they were unstable. Specifically, cloning of the 3' end of the NA vRNA of the U→GIO mutant resulted in one clone with mutant and two clones with wild-type sequences. Direct RNA sequencing of the 5' end of the NA-specific vRNA from purified A→CH' transfectant, following three plaque to plaque passages, revealed a wild-type sequence. However, when NA-specific vRNA from MDBK cells infected with the original plaque of this transfectant was sequenced, the presence of the mutation was confirmed. Thus it seems likely that the transfectant reverted to wild-type during the plaque purification steps. This interpretation is supported by the observation that the transfectant initially produced small plaques, but showed larger plaques upon passaging. Taken together, sequencing data of the single mutants showed that transfectant vimses with single mutations, at least those with mutations at positions 10 and 11', are unstable.

Example 4 Growth properties of the DI. D2, Dl/2 and D3 mutants

DI, D2, and D3 were grown on MDBK cells. Confluent monolayers of MDBK cells were infected at low m.o.i. (0.01) and the amount of infectious vims released into the medium was assayed at different time points by plaque assay on MDBK cells (Fig. 2). The D2 transfectant vims showed approximately one log reduction in plaque titre compared to the wild-type vims. However, DI and D3 transfectant vimses were not significantly affected by the mutations. Consistently, the plaque size of D2 was reduced, but both DI and D3 vimses showed plaque sizes similar to that of the wild-type.

The growth properties were also investigated of mutant influenza A/WSN/33 having multiple double-mutations in the NA-specific vRNA. A constmct incorporating double-mutations from both DI and D2 transfectants was successfully rescued (Dl/2) (Fig. 1) into infectious vims. The Dl/2 transfectant was plaque purified three times and the presence of mutations was confirmed by sequencing. This vims showed similar reduction in plaque titres (Fig. 2) and plaque size on MDBK cells as the D2 transfectant. The effect of the Dl/2 mutations on viral growth was more dramatic on MDCK and Vero cells where reductions of at least three to four logs in plaque titres were observed (see Examples 10 and 11 below).

Example 5

Measurement of NA levels in transfectant vimses The level of NA expressed by the vimses was determined to see if it corresponded to growth levels. Influenza A/WSN/33 and transfectant vimses were grown in MDBK cells and purified by 30 to 60% sucrose gradient ultracentrifugation. About 10 μg of viral proteins were denatured with 0.5%o SDS and 1% β-mercaptoethanol at 100 °C for 10 minutes and digested with 400 u of PNGase F (New England Biolabs, Inc., Beverly, MA) for 20 h at 37 °C in a reaction buffer containing 50 mM sodium phosphate, pH 7.5, 1% NP-40, and 5 mM Pefabloc (Boehringer Mannheim Corporation, Indianapolis, IN). The PNGase F treatment removes N-linked carbohydrate chains from NA and HA. This gives a better resolution of the NA band which migrates closely to NP and HA on gels. Proteins were analyzed by 12% SDS-PAGE and staining with Coomassie Brilliant Blue.

Both D2 and Dl/2 virions showed a dramatic reduction in NA content compared to that of the wild-type vims or the DI and D3 transfectants.

In order to quantitate NA levels of the D2 and Dl/2 vimses, neuraminidase activity was measured. About 2 μg, 0.5 μg, 0.125 μg, and 0.031 μg (4 fold dilutions) of proteins from purified vims were incubated for 10 minutes at 37 °C in 150 mM phosphate buffer, pH 6.0, 1 mM CaCl₂, containing 50 nmols of 2'-(4- methylumbelliferyl)-α-D-N-acetylneuraminic acid (MU-ΝAΝA) as substrate in a total volume of 100 μl (27). Then 2 ml of stop buffer (0.5 M glycine/ΝaOH, pH 10.4) were added and the released 4-methylumbelliferone was determined by spectrofluorometry. 0.1 mM solution of 4-methylumbelliferone was used as a standard control. ΝA activity was expressed as nmoles of 4-methylumbelliferone released in 1 minute per μg of viral proteins.

ΝA activity associated with the wild-type vims was 2.18 nmol min^"1 μg^'1. However, the transfectant vimses D2 and Dl/2 exhibited only 0.24 and 0.25 nmol min^"1 μg^"1 activity, respectively. Thus, the transfectant vimses showed approximately a 10 fold reduction in ΝA activity compared to the wild-type vims which is in agreement with the reduced NA levels observed in SDS-PAGE.

Example 6

NA-specific vRNA levels in purified transfectant vimses Viral RNA from wild-type and transfectant vimses purified through a 30% sucrose cushion was extracted with phenol/chloroform. The viral RNAs purified from wild-type and transfectant vimses were analyzed by PAGE and the RNA segments were visualized by silver-staining. The NA segment was present in all transfectant vimses at levels comparable to that of the wild-type vims. In order to quantify NA-specific vRNA levels, a primer extension analysis was performed using vRNA extracted from purified vimses.

Primer extension analysis of NA and NS vRNA levels was performed as previously described (2). Briefly, 100 ng of viral RNA was transcribed with 200 u of Superscript (Gibco BRL) for 1 h at 42 °C in the presence of 3 x 10⁵ cpm of ³²P-labelled NA- and NS-specific primers. The NA-specific primer,

5'-GTGGCAATAACTAATCGGTCA-3', is complementary to nucleotides 1151 to 1171 of the NA vRNA. The NS-specific primer,

5'-GGGAACAATTAGGTCAGAAGT-3', is complementary to positions 695 to 715 of the NS vRNA. Primer extension reactions were stopped by adding an equal volume of 90% formamide and 10 mM EDTA followed by heating to 95 °C for 3 minutes. Extension products were analyzed on 5% polyacrylamide gels in the presence of 7 M urea and quantitated by phosphorimager analysis of dried gels (Molecular Dynamics).

The NS gene was used as an internal control. The amounts of NA-specific vRNA segments in the transfectant vimses were similar (±20%) to that of the wild-type vims in two experiments.

Example 7

NA-specific vRNA levels in cells infected with the D2 or Dl/2 transfectant vimses. MDBK cells were infected with wild-type or transfectant vimses at an m.o.i. of 2 and total RNA was isolated from cells at 3.0, 5.5, 8.0, and 10.5 h postinfection with RNAzol B (Tel-Test). NA-specific vRNA levels in total RNA were measured by primer extension assay as described above in Example 6 using 5μg of total RNA. Cells infected with the D2 transfectant vims contained NA-specific vRNA levels similar (±10%) to those infected with the wild-type vims. Although cells infected with the Dl/2 transfectant vims showed a 28 to 53% reduction in NA-specific vRNA levels (results obtained by phosphorimager analysis in two experiments at 5.5, 8.0, and 10.5 h postinfection), this decrease cannot account for the ten-fold reduction of NA protein levels.

Example 8

NA-specific mRNA and cRNA levels in cells infected with the D2 or Dl/2 transfectant vimses.

Since NA-specific vRNA levels were not dramatically affected by the mutations in the D2 and Dl/2 transfectant vimses, the 10 fold reduction in NA levels (see above) could result from a reduction in mRNA levels and/or from a defect in translation. In order to distinguish between these possibilities, the amounts of NA-specific mRNA in cells infected with D2 or Dl/2 transfectant vimses were measured by using a primer extension assay. MDBK cells were infected at an m.o.i of 2 with wild-type or transfectant vimses and total RNA was isolated at 3.0, 4.5, 6.0, and 7.5 h postinfection.

Primer extension analysis of NA and HA mRNA and cRNA levels in total RNA from infected cells was performed under the same conditions as described in Example 6. The primer for NA-specific mRNA and cRNA, 5'-GCGCAAGCTTTATTGAGATTATATTTCC-3 '. contains 18 nucleotides (underlined) corresponding to positions 115 to 98 of the NA gene. The primer for the extension of HA-specific mRNA and cRNA,

5'-CATATTGTGTCTGCATCTGTAGCT-3', corresponds to positions 94 to 71 of the HA gene.

Since total RNA from infected cells contains both mRNA and cRNA, which differ only at their termini, signals for both species of RNAs were expected in the same primer extension assay. Due to the presence of a heterologous 10 to 15 nucleotides long capped primer at the 5' end of mRNA molecules, the signal for mRNA on gels appears as a multiple band containing DNA species of different sizes. On the other hand, the signal for cRNA appears as a single band, which is approximately 10 to 15 nucleotides shorter than the signal for mRNA. NA-specific mRNA levels in cells infected with either D2 or Dl/2 transfectant vims were below detection levels. NA-specific cRNA levels were apparently unaffected in these transfectant vimses. An additional band running slightly faster than the NA-specific cRNA band, detected in all samples, represents a nonspecific signal, since it was also detected in RNAs extracted from uninfected cells.

The observed attenuation of NA-specific mRNA levels in cells infected with the D2 transfectant is consistent with the previous findings of Kim et al. (14) that an A-U(l 1-12') base-pair mutation in the context of a vRNA-like CAT reporter gene resulted only in 22% reporter activity compared to a wild-type control. However, the G-C(10-l 1') and U-A(12-13') base-pair mutations, which had no effect on the expression levels of the neuraminidase of the DI and D3 transfectants, resulted in only 20 and 31% activities, respectively, in a CAT reporter gene system (14). It is thus clear that base-pair mutations in the context of a CAT reporter gene system and a rescued native NA gene containing vRNA segment have different effects.

Example 9

In vitro transcription of NA-specific ribonucleoprotein complexes.

In theory, the reduction of mRNA levels observed as above could have been caused by a decrease in mRNA stability or by a decrease in mRNA synthesis. The interference with mRNA synthesis may occur at the point of initiation, e.g. capped

RNA primer binding or endonuclease activity could be inhibited. Alternatively, termination or polyadenylation of viral mRNA could be affected. In order to distinguish between all these possibilities, in vitro transcription assays were performed. Wild-type influenza A/WSN/33 vims, D2, and Dl/2 transfectants were grown in MDBK cells and purified on a 30% sucrose cushion. Twelve 15 cm dishes were used for each vims. The purified vimses were resuspended in 200 μl of PBS and dismpted by adding 50 μl of 5x dismption buffer (500 mM Tris-HCl [pH 7.4], 500 mM NaCl, 25 mM MgCl₂, 5 mM DTT, 25% glycerol, 2.5% NP-40, 2.5% Triton X-100, 50 mg ml^"1 lysolecithin) and incubation at 37 °C for 30 min. The dismpted vimses were fractionated by centrifugation on a discontinuous glycerol gradient (70%, 50%, and 30%, 150 μl of each) in 100 mM Tris-HCl (pH 7.4), 100 mM NaCl, 5 mM MgCl₂, and 1 mM DTT. The gradients were centrifuged for 4 h at 15 °C in 0.8 ml tubes at 45,000 rpm in a Beckman SW55 rotor with adaptors. Fractions collected from the bottom of the tubes were analyzed by 12% SDS-PAGE and those enriched in RNPs were used in transcription assays.

In vitro transcriptional activity was measured using globin mRNA as primer. Transcription reactions were performed by using 6 μl of RNPs in a total reaction volume of 20 μl containing 50 mM Tris-HCl (pH 7.8), 50 μM KC1, 10 mM NaCl, 5 mM MgCl₂, 5 mM DTT, 1 mM ATP, 0.5 mM each GTP and CTP, 50 μM UTP, 0.1 μM [α-³²P] UTP (3,000 Ci mmol^"1), 20 u of RNase inhibitor (Boehringer

Mannheim Corporation, Indianapolis, LN), 0.6 μg of rabbit globin mRNA (Gibco BRL). After incubation at 31 °C for 1.5 h, transcription products were extracted with phenol/chloroform and precipitated in the presence of 5 μg of carrier yeast RNA.

NA-specific transcription products were synthesized from both the wild-type and the transfectant RNPs. However, there was a significant difference in the pattern of the bands. The wild-type NA-specific transcription product appeared as a wide band corresponding to RNA species with poly(A) tails of different sizes. On the other hand, the NA-specific transcription products of both the D2 and Dl/2 transfectants produced less diffuse bands, which implied that these products might not be polyadenylated. In order to characterize the transcription products, they were analyzed by oligo(dT)-cellulose chromatography.

The fractions depleted of poly(A)-containing molecules showed higher levels of NA-specific transcription products for the D2 and Dl/2 transfectants, but lower levels for the wild-type control. On the other hand, fractions enriched in poly(A)-containing molecules showed lower levels of the NA-specific transcription products for the D2 and Dl/2 transfectants, but higher levels for the wild-type vims. This seems to confirm that there is a large proportion of NA-specific transcription products of the D2 and Dl/2 transfectants which lack poly(A) tails.

It is thus proposed that the mutations in the NA-specific vRNA of D2 and Dl/2 interfere with polyadenylation of mRNA transcripts. The observed low levels of mRNA in cells infected with these vimses is fully consistent with this conclusion, since non-polyadenylated capped transcripts are most likely rapidly degraded in the cell (30).

Example 10 Growth of transfectant vimses on MDCK cells.

MDCK cells in 96-well plates were infected with 5xl0⁴ pfu and 10 times dilutions of wild-type influenza A/WSN/33 vims, or transfectant DI, D2, D3, and Dl/2 vimses. Four wells were used for each vims. Infected cells were maintained in 100 μl of Dulbecco's minimal essential medium (DMEM) supplemented with 10% bovine semm albumin and 1 μg/ml of trypsin. After 72 h, 50 μl of the medium was tested for hemagglutination with 50 μl of 1.5% red blood cells and LD₅₀ was calculated for each vims. ΓD₅₀ is defined as the dose at which 50% of the medium of the infected cells gives a positive haemagglutination signal. It was found that the LD₅₀ for the wild-type vims and the DI transfectant was 5 pfu. On the other hand, the LD₅₀ of the D3 transfectant was 20 times higher. The LD₅₀ of the D2 and D 1/2 transfectant was approximately 3000 times higher than that of the wild-type or the DI transfectant.

Example 11 Growth of the Dl/2 transfectant on Vero cells.

Confluent Vero cells in 35 mm dishes were infected at an m.o.i. of 0.01 with wild-type influenza A/WSN/33 vims or Dl/2 transfectant in duplicates. Cells were maintained in DMEM supplemented with 2% FBS for 72 h and vims present in the medium was titrated by plaque assay on MDBK cells. The wild-type vims reached 5xl0⁷ pfu/ml, but there was less than 5xl0² pfu/ml of infectious vims in the medium from the cells infected with the Dl/2 transfectant. Taken together, the data in Examples 4, 10 and 11 show that base-pair mutations in the double-stranded region of the promoter of an influenza A vims vRNA can lead to reduced growth of influenza vims in tissue culture. As noted above, the D2 and Dl/2 transfectant vimses showed approximately one log reduction in growth in MDBK cells, while both the DI and D3 vimses grew like the wild-type.

A more dramatic reduction in growth was observed for the D2 and Dl/2 vimses on MDCK and Vero cells. Interestingly, the D3 transfectant showed reduced growth on MDCK cells compared to the wild-type. Both D2 and Dl/2 transfectants exhibited approximately four log reduction on MDCK cells, and the Dl/2 transfectant 5 log reduction on Vero cells. Such results are indicative that influenza A vimses having the D2 and Dl/2 mutations will exhibit effective attenuation in vivo.

Example 12

Passage of transfectant vimses and sequencing to determine the stability of the DI. D2 and D3 mutations

Stocks of DI, D2, and D3 transfectant vimses with confirmed double-mutations were plaqued on MDBK cells and individual plaques were passaged ten times on MDBK cells at a low m.o.i. After ten passages, the vimses were plaqued and single plaques were used to prepare vims stocks for sequencing. Stocks of passaged vimses were purified through a 30 % sucrose cushion and viral

RNA was isolated by phenol-chloroform extraction. In order to sequence the 3' end of the NA gene, viral RNA was 3'-polyadenylated using poly(A) polymerase (Gibco BRL, Gaithersburg, MD). The polyadenylated RNA was reverse transcribed using the primer 5'-GCGCAAGCTTCTAGATTTTTTTTTTTTTT-3' and the cDNA was amplified by PCR with a primer containing nucleotides corresponding to positions

115 to 98 (5'-GCGCAAGCTTTATTGAGATTATATTTCC-3') of the influenza A/WSN/33 NA gene and the primer used for reverse transcription. PCR products digested with Hindlll were cloned into pUC18 and sequenced with a DNA sequencing kit (United States Biochemical, Corporation, Cleveland, OH). Three clones originating from three individually passaged plaques of the DI transfectant showed the presence of the U-G10 mutation. All clones obtained from 5 individually passaged plaques of the D2 transfectant had the expected C-Al 1 mutation. In addition, two of the clones showed a U-^C change at position 4 which is a natural variation observed among different influenza A vims isolates. In two of the clones, we have also found a U-C mutation at position 23 adjacent to the initiation codon for the neuraminidase which changes the second amino acid of NA from an asparagine to an aspartate. Only two of the clones obtained from the D3 transfectant showed the C→U12 mutation. The third clone had a wild-type sequence indicating that this base-pair mutation might not be stable. A reversion of A→G13' could result in a viable vims with a U-G(12-13') base-pair, which could then revert to the wild-type C-G(12-13') base-pair by a U→C12 change. Due to the presence of different residues such a reversion cannot occur at the other two studied base-pairs.

In summary, the mutations in the 3' end of DI and D2 transfectants were preserved during ten passages. Preliminary data confirms the presence of the mutations also in the 5' end of the NA segment of the passaged transfectant vimses. It can be assumed that transfectant vimses with double-mutations should be stable since two specific mutations would have to occur simultaneously in order to revert to the wild-type sequence. It did not prove possible to rescue any transfectant vimses with C→Al 1 or G-U12' single mutations which suggests that such vimses might be severely impaired or not viable at all.

Example 13

Attenuation of D2 and Dl/2 vimses in mice

Influenza A/WSN/33 wild-type and transfectant vimses DI, D2, D3 and Dl/2 were grown at 37°C in Madin-Darby bovine kidney (MDBK) cells in reinforced minimal essential medium. Plaque assays were performed on MDBK cells.

Groups of five female BALB/c mice were used for influenza vims infection at 6 to 12 weeks of age. Intranasal (i.n.) inoculations were performed in mice under ether anesthesia using 50μl of PBS containing 10⁶, 3xl0⁴ or 10³ plaque forming units (pfu) of DI, D2, D3 or Dl/2 vims. As controls, mice were infected with wild-type influenza A/WSN/33 vims using the same pfu of vims. This vims was rescued by ribonucleoprotein transfection of a wild-type NA gene as previously described by Enami and Palese (4). Animals were monitored daily and sacrificed when observed in extremis. All procedures were in accord with NIH guidelines on care and use of laboratory animals. The results are shown in Figures 4 to 9.

All mice infected with wild-type vims developed signs of disease and died by day 15 post-infection. However, all mice infected with the D2 or Dl/2 vimses survived. Only those D2 or Dl/2 vims-infected animals lost weight which were infected with the high dose of vims (10⁶ pfu); they lost 10 to 20% of body weight by day 3 post-infection, but they quickly recovered in the following days. The vimlence of the DI vims was indistinguishable from the vimlence of wild-type vims in these experiments. The D3 vims showed a slightly attenuated phenotype in mice.

Example 14

Impaired replication of the D2 and Dl/2 vimses in mouse lungs

Groups of 6 BALB/c mice were infected intranasally as above with 10³ pfu of wild-type, DI, D2, D3 or Dl/2 vimses. Three days post-infection, three mice per group were sacrificed, their lungs were extracted and homogenized in 2 ml of PBS, and vims titres were measured by plaque assay in MDBK cells. Six days post-infection, the rest of the mice were also sacrificed and viral titres were determined in their lungs by the same protocol. The results are shown in Fig. 10. The wild-type and the DI vimses grew to high titres in the lungs of the infected mice (approximately 10⁶ and 10⁷ pfu/ml at days 3 and 6 post-infection, respectively). Titres in the lungs of mice infected with the D3 vims were approximately one and a half logs lower. By contrast, viral titres were not detectable or very low (less than 10³ pfu/ml) in the lungs of the D2 or Dl/2 infected mice. The results demonstrate that replication of the D2 and Dl/2 vimses is highly impaired in mouse lungs.

Example 15

Induction of protective immunity by D2 and Dl/2 vimses Sera from the groups of surviving mice which were intranasally infected with

D2 or Dl/2 vims as above was collected and pooled 3 weeks after infection. The sera were treated with receptor destroying enzyme (Sigma) to eliminate unspecific inhibitors of influenza vims-mediated haemagglutination as previously described by Burnet and Stone (55). The haemagglutination inhibition (HI) titres were determined as the highest semm dilution that was able to neutralize the haemagglutination activity of a preparation of influenza A/WSN/33 vims with an HA titre of 8. In these assays, 0.5% chicken red blood cells were used.

All pools of sera which were tested were found to contain antibodies against influenza A/WSN/33 vims with HI activity. HI titres were higher in the animals immunized with the higher vims doses (see Table 1 below).

In addition, all mice which were intranasally infected with D2 or Dl/2 vims were observed to be protected against death and disease (as measured by body weight loss) when challenged with a lethal infection dose (more than 1000 LD₅₀s) of wild-type A/WSN/33 vims (see Table 1 and Figures 11 and 12).

Table 1

Protection against wild-type influenza vims infection in mice immunized with D2 and Dl/2 vimses

Example 16

Use of the Dl/2 transfectant vims as a helper vims to rescue NA genes

As noted above, the Dl/2 transfectant vims showed approximately 5 log reduction in growth on Vero cells compared to wild-type influenza A/WSN/33. It can therefore be employed to provide an alternative rescue system for rescue of NA-encoding vRNA segments of influenza A vimses. An appropriate protocol for this consists of the following steps:

1. infection of MDBK cells with Dl/2 helper vims ;

2. treatment of the infected MDBK cells with DEAE-dextran/DMSO transfection reagent;

3. transfection of a synthetic NA ribonucleoprotein complex into Dl/2 helper vims infected and DEAE-dextran/DMSO-treated MDBK cells; and

4. selection of rescued vimses on Vero cells. Only vimses which acquire the transfected NA gene grow to high titre on Vero cells.

REFERENCES

1 Muster, T. Subbarao, E.K., Enami, M., Murphy, B.R. and Palese, P. 1991

An influenza A vims containing influenza B vims 5' and 3' non-coding regions on the neuraminidase gene is attenuated in mice. Proc. Natl. Acad. Sci. USA 88, 5177-

5181.

2. Luo, G., Bergmann, M., Garcia-Sastre, A., and Palese, P. 1992. Mechanism of attenuation of a chimeric influenza A B transfectant vims. J. Virol. 66, 4679- 4685.

3. Bergmann, M. and Muster, T. 1995. The relative amount of an influenza A vims segment present in the viral particle is not affected by a reduction in replication of that segment. J. Gen. Virol. 76, 3211-3215

4. Enami, M., W. Luytjes, M. Krystal, and P. Palese. 1990. Introduction of site specific mutations into the genome of influenza vims. Proc. Natl. Acad. Sci. USA 87: 3802-3805. 5. Enami, M., and P. Palese. 1991. High-efficiency formation of influenza vims transfectants. J. Virol. 65: 2711-2713.

6. Flick, R., G. Neumann, E. Hoffmann, E. Neumeier, and G. Hobom. 1996. Promoter elements in the influenza vRNA terminal structure. RNA 2: 1046-1057.

7. Fodor, E., D. C. Pritlove, and G. G. Brownlee. 1994. The influenza vims panhandle is involved in the initiation of transcription. J. Virol. 68: 4092-4096.

8. Fodor, E., D. C. Pritlove, and G. G. Brownlee. 1995. Characterization of the RNA-fork model of virion RNA in the initiation of transcription in influenza A vims. J. Virol. 69: 4012-4019.

9 Garcia-Sastre, A., T. Muster, W. S. Barclay, N. Percy, and P. Palese. 1994. Use of a mammalian internal ribosomal entry site element for expression of a foreign protein by a transfectant influenza vims. J. Virol. 68: 6254-6261.

10 Gubareva, L. V., R. Bethell, G. J. Hart, K. G. Murti, C. R. Penn, and R. G. Webster. 1996. Characterization of mutants of influenza A vims selected with the neuraminidase inhibitor 4-guanidino-Neu5Ac2en. J. Virol. 70: 1818-1827.

11 Hagen, M., T. D. Y. Chung, J. A. Butcher, and M. Krystal. 1994 Recombinant influenza vims polymerase: requirement of both 5' and 3' viral ends for endonuclease activity. J. Virol. 68: 1509-1515.

12. Honda, A., and A. Ishihama. 1997. The molecular anatomy of influenza vims RNA polymerase. Biol. Chem. 378: 483-488.

13. Hsu, M., J. D. Parvin, S. Gupta, M. Krystal, and P. Palese. 1987 Genomic RNAs of influenza vimses are held in a circular conformation in virions and in infected cells by a terminal panhandle. Proc. Natl. Acad. Sci. USA 84: 8140-8144. 14. Kim, H-J., E. Fodor, G. G. Brownlee, and B. L. Seong. 1997. Mutational analysis of the RNA-fork model of the influenza A vims vRNA promoter in vivo. J. Gen. Virol. 78: 353-357.

15 Krug, R. M., F. V. Alonso-Caplen, I. Julkunen, and M. G. Katze. 1989 Expression and replication of the influenza vims genome, p.98-152. In R. M. Kmg (ed.), The Influenza Vimses. Plenum, New York.

16. Li, X., and P. Palese. 1992. Mutational analysis of the promoter required for influenza vims virion RNA synthesis. J. Virol. 66: 4331-4338.

17. Li, X., and P. Palese. 1994. Characterization of the polyadenylation signal of influenza vims RNA. J. Virol. 68: 1245-1249.

18. Luo, G., W. Luytjes, M. Enami, and P. Palese. 1991. The polyadenylation signal of influenza vims RNA involves a stretch of uridines followed by the RNA duplex of the panhandle structure. J. Virol. 65: 2861-2867.

19 Luytjes, W., M. Krystal, M. Enami, J. D. Parvin, and P. Palese. 1989 Amplification, expression, and packaging of a foreign gene by influenza vims. Cell. 59: 1107-1113.

20 Martin, J., C. Albo, J. Ortin, J. A. Melero, and A. Portela. 1992 In vitro reconstitution of active influenza vims nucloeprotein complexes using viral proteins purified from infected cells. J. Gen. Virol. 73: 1855-1859.

21. Mena, I., S. de la Luna, C. Albo, J. Martin, A. Nieto, J. Ortin, and A. Portela. 1994. Synthesis of biologically active influenza core proteins using a vaccinia-T7 RNA polymerase expression system. J. Gen. Virol. 75: 2109-2114. 22. Neumann, G., and G. Hobom. 1995. Mutational analysis of influenza vims promoter elements in vivo. J. Gen. Virol. 76: 1709-1717.

23. O'Neill, R. E., J. Talon, and P. Palese. 1998. The influenza vims NEP (NS2 protein) mediates the nuclear export of viral ribonucleoproteins. EMBO J. 17: 288- 296.

24. Palese. P. 1977. The genes of influenza vims. Cell 10: 1-10.

25. Parvin, J. D., P. Palese, A. Honda, A. Ishihama, and M. Krystal. 1989. Promoter analysis of the influenza vims RNA polymerase. J. Virol. 63: 5142-5152.

26. Piccone, M. E., A. Fernandez-Sesma, and P. Palese. 1993. Mutational analysis of the influenza vims vRNA promoter. Vims Res. 28: 99-112.

27 Potier, M., L. Mameli, M. Belisle, L. Dallaire, and S. B. Melancon. 1979. Fluorometric assay of neuraminidase with a sodium (4-methylumbelliferyl- -D-N- acetylneuraminate) substrate. Anal. Biochem. 94: 287-296.

28 Pritlove, D. C . , E. Fodor, B. L . Seong, and G. G. Brownlee. 1995 In vitro transcription and polymerase binding studies of the termini of influenza A vims complementary RNA: evidence for a cRNA panhandle. J. Gen. Virol. 76: 2205-2213.

29. Pritlove, D. C, L. L. M. Poon, E. Fodor, J. Sharps, and G. G. Brownlee.

1998. Polyadenylation of influenza vims mRNA transcribed in vitro from model virion RNA templates: requirement for 5' conserved sequences. J. Virol.72: 1280-1287. 30. Proudfoot, N. J., and E. Whitelow. 1988. Termination and 3' end processing of eukaryotic RNA, p. 97-129. In D. M. Glover and B. D. Hames (ed ), Frontiers in molecular biology - transcription and splicing. LRL Press, Oxford.

31. Robertson, J. S., M. Schubert, and R. A. Lazzarini. 1981. Polyadenylation sites for influenza vims mRNA. J. Virol. 38: 157-163.

32. Seong, B. L., and G. G. Brownlee. 1992. A new method for reconstituting influenza polymerase and RNA in vitro: A study of the promoter elements for cRNA and vRNA synthesis in vitro and viral rescue in vivo. Virology 186: 247-260.

33 Tiley, L. S., M. Hagen, J. T. Matthews, and M. Krystal. 1994 Sequence- specific binding of the influenza vims RNA polymerase to sequences located at the 5' ends of the viral RNAs. J. Virol. 68: 5108-5116.

34. Yamanaka, K., N. Ogasawara, H. Yoshikawa, A. Ishihama, and K. Nagata.

1991. In vivo analysis of the promoter structure of the influenza genome using a transfection system with an engineered RNA. Proc. Natl. Acad. Sci. USA 88: 5369-5373.

35. Zheng, H., P. Palese, and A. Garcia-Sastre. 1996. Nonconserved nucleotides at the 3' and 5' ends of an influenza A vims RNA play an important role in viral RNA replication. Virology 217: 242-251.

36. Desselberger, U., Racariello, V.R., Zazra, J.J. and Palese, P. 1980. The 3' and 5'-terminal sequences of influenza A, B and C vims RNA segments are highly conserved and show partial inverted complementarity. Gene 8, 315-328.

37. Lee, Y-S and Seong, B.L. 1996. Mutational Analysis of Influenza B vims RNA transcription in vitro. J. Virol. 70, 1232-1236. 38. Garcia-Sastre, A. and Palese, P. 1993. Genetic manipulation of negative-strand RNA vims genomes. Ann. Rev. Microbiol. 47, 765-90.

39. Li, S., Xu, M. and Coelingh, K. 1995. Electroporation of ribonucleoprotein complexes for rescue of the nucleoprotein and matrix genes. Vims Res. 37, 153-161.

40. Enami, M., Sharma, G., Benham, G. and Palese, P. 1991. An influenza vims containing nine different RNA segments. Virology 185, 291-8.

41. Subbarao, E. K., Park, E.J., Lawson, CM., Chen, A.Y. and Murphy, B.R.

1995. Sequential addition of temperature-sensitive missense mutations into the PB2 gene of influenza A transfectant vims can effect an increase in temperature sensitivity and attenuation and permits the rational design of a genetically engineered live influenza A vims vaccine. J. Virol. 69, 5969-77.

42. Castrucci M. R. and Kawaoka, Y. 1995. Reverse genetics system for generation of an influenza A vims mutant containing a deletion of the carboxyl-terminal residue of M2 protein. J. Virol. 69, 2725-8.

43. Liu, C. and Air, G. M. 1993. Selection and characterisation of a neuraminidase-minus mutant of influenza vims and its rescue by cloned neuraminidase genes. Virology 194, 403-7.

44. Barclay, W.S. and Palese, P. 1995. Influenza B vimses with site-specific mutations introduced into the NA gene. J. Virol. 76, 3211-5.

45. Pleschka, S., Jaskunas, R., Engelhardt, O.G., Zurcher, T., Palese P. and Garcia-Sastre, A. 1996. A plasmid-based reverse genetics system for influenza A vims. J. Virol. 70, 4188-92. 46. Enami, M. 1997. Improved technique to genetically manipulate influenza vims. In Frontiers of RNA Vims Research p.19, The Oji International Seminar in Natural Science, Kyoto, Japan 1997.

47. Published International Application WO 91/03552 (Palese, P. et al.)

48. Li, S., Polords, V., Isobe, H. et al. 1993. Chimeric influenza vims induces neutralising antibodies and cytotoxic T cells against human immunodeficiency vims type l. J. Virol. 67, 6659-66.

49. Muster T., Ferko B., Klima, A. et al. 1995. Mucosal model of immunisation against human immunodeficiency vims type 1 with a chimeric influenza vims.

J. Virol. 69, 6678-86.

50. Rodrigues, M., Li, S., Murata, K., Rodrigues D. 1994. Influenza and vaccinia vimses expressing malaria CD8+ T and B cell epitopes. J. Immunol. 153, 4636-48.

51. Percy, N., Barclay, W. S., Garcia-Sastre, A. and Palese, P. 1994. Expression of a foreign protein by influenza A vims. J. Virol 68₃ 4486-92.

52. Horimoto, T. and Kawaoka, Y. 1994. Reverse genetics provides direct evidence for a correlation of haemagglutinin cleavability and vimlence of an avian influenza A vims. J. Virol. 68, 3120-3128.

53. Ferko, B., Egorav, A. et al. 1997. Influenza vims as a vector for mucosal immunisation. In Frontiers of RNA Vims Research, p.18. The Oji International Seminar in Natural Science, Kyoto, Japan. 54. Li et ai. 1993. Glycolysatio of neuraminidase determines the neurovimlence of influenza A/WSN/33. J. Virol. 67, 6667-73.

55. Burnet, F.M. and J.D. Stone. 1947. The receptor-destroying enzyme of V. cholerae. J. Exper. Med. Sci. 25: 227-233.

56. Muster, T. and Garcia-Sastre, M. June 1998. Textbook of Influenza, Blackwell Science Ch. 9, p.93-106, Genetic Manipulation of Influenza Vimses.

Claims

CLALMS

1. An attenuated influenza vims carrying a genomic nucleic acid segment which comprises 5' and 3' non-coding regions providing a mutated duplex region of an influenza vims RNA genomic segment operably-linked to a protein coding sequence for an influenza viral protein or a functional modification thereof, wherein said duplex region has at least one base-pair substitution such that expression of said protein-coding sequence in cells infected by said vims is reduced to give an attenuated phenotype.

2. A vims as claimed in claim lwhich exhibits a reduction in plaque titre compared to the parent wild-type vims on cells of one or more type selected from Madin-Darby bovine kidney (MDBK) cells, Madin-Darby canine kidney (MDCK) cells and Vero cells.

3. A vims as claimed in claim 2 which exhibits at least about one log reduction in plaque titre compared to the parent wild type vims on MDBK cells.

4. A vims as claimed in claim 2 or claim 3 which exhibits at least about 3 to 4 log reduction in plaque titre compared to the parent wild type vims on MDCK cells and Vero cells.

5. A vims as claimed in any one of claims 1 to 4 wherein said genomic nucleic acid segment is a mutated native influenza vims genomic RNA segment.

6. A vims as claimed in any one of claims 1 to 5 which is an attenuated influenza vims of type A, wherein said nucleic acid segment is a mutated influenza A vims genomic RNA segment having the mutation C to A at position 11 from the 3'-terminus of the native parent segment and the mutation G to U at position 12' from the 5'-terminus of the native parent segment, or functionally equivalent substitutions at the same positions, so as to provide an attenuating base-pair substitution in the non-coding duplex region.

7. A vims as claimed in claim 6 wherein said nucleic acid segment also has the mutation U to G at position 10 from the 3' terminus of the native parent segment and the mutation A to C at position 11' from the 5' terminus of the native parent segment, or functionally equivalent substitutions at the same positions, so as to provide an additional base-pair substitution in the non-coding duplex region.

8. A vims as claimed in claim 6 or claim 7 wherein said nucleic acid segment encodes neuraminidase (NA) or a functional modification thereof.

9. A vims as claimed in any one of claims 1 to 8 which is a wild-type vims which has been attenuated by said base-pair substitution(s).

10. A vims as claimed in any one of claims 1 to 8 which additionally comprises a heterologous coding sequence capable of being expressed in target cells.

11. A vims as claimed in claim 10 wherein said heterologous coding sequence encodes an antigenic peptide or polypeptide capable of stimulating an immune response to a pathogenic agent.

12. A vims as claimed in claim 9 which is attenuated influenza A/WSN/33 having a NA-encoding nucleic acid segment as defined in claim 8.

13. A nucleic acid as defined in claim 1 or any one of claims 5 to 8.

14. A DNA capable of transcription to provide a nucleic acid according to claim 13.

15. A plasmid containing a DNA as claimed in claim 14.

16. A ribonucleoprotein (RNP) complex wherein a nucleic acid as claimed in claim 13 is complexed with polymerase proteins and nucleoprotein of an influenza vi s for use in preparing an attenuated vims as claimed in any one of claims 1 to 12.

17. An ex vivo cell infected by a vims as claimed in any one of claims 1 to 12.

18. A vaccine comprising a vims as claimed in any one of claims 1 to 11.

19. A vaccine as claimed in claim 18 which comprises a vims as claimed in claim 11 and which is capable of stimulating an immune response to an influenza vims and a second pathogenic agent other than an influenza vims.

20. A pharmaceutical composition comprising a vims as claimed in claim 10 in combination with a pharmaceutically acceptable carrier or diluent for delivery of said heterologous coding sequence to target cells.

21. A pharmaceutical composition comprising cells infected with a vims according to claim 10 or claim 11 in combination with a pharmaceutically acceptable carrier or diluent.

22. A method of preparing a vims according to any one of claims 1 to 12 which comprises providing in a host cell the genomic nucleic acid segments for said vims under conditions whereby said segments are packaged into a viral particle.

23. Use of a vims as claimed in any one of claims 1 to 12 as a helper vims to rescue an influenza vims genomic nucleic acid segment in cells, wherein vimses produced containing said nucleic acid segment are selected on the basis of increased growth compared with the helper vims on cells of a selected type.

24. Use of an influenza A vims as claimed in claim 8 as a helper vims in accordance with claim 23 to rescue an NA-encoding influenza A vims genomic nucleic acid segment or a functional modification thereof.

25. Use as claimed in claim 24 of attenuated influenza A/WSN/33 having mutations as defined in claim 7 in the NA-encoding genomic RNA segment, wherein selection of vimses carrying the nucleic acid segment to be rescued is carried out on Vero cells.

26. A method of stimulating an immune response against an influenza vims, optionally together with stimulation of an immune response against one or more further pathogenic agents, which comprises administering in an immunising mode an attenuated influenza vims as claimed in any one of claims 1 to 11.

27. A method of delivering a heterologous coding sequence to cells which comprises infecting said cells with a vims according to claim 10 carrying said sequence.