WO1998004710A9 - Live recombinant bhv/brsv vaccine - Google Patents

Abstract

The present invention refers to synthetic Bovine Respiratory Syncytium virus genes. Also the invention relates to live attenuated Bovine Herpesvirus recombinants carrying such synthetic genes. Furthermore, the invention relates to vaccines based on these live attenuated recombinants, for the protection of cattle against both Bovine herpesvirus infection and against Bovine Respiratory Syncytium virus infection. Also the invention relates to methods for the preparation of such live attenuated recombinants and to methods for the preparation of such vaccines. Furthermore, the invention relates to live Bovine herpesvirus recombinant particles carrying a class II membrane glycoprotein and to live Bovine herpesvirus recombinants carrying a heterologous gene fused to a class II membrane glycoprotein membrane anchor, to vaccines based thereon, and to methods for the preparation of such recombinants, such particles and such vaccines.

Description

LIVE RECOMBINANT BHV/BRSV VACCINE

T e present invention refers to synthetic Bovine Respiratory Syncytium virus genes, live attenuated Bovine Heφesvirus recombinants, live attenuated Bovine Heφesvirus recombinants carrying such genes, vaccines based on these live attenuated recombinants, methods for the preparation of such live attenuated recombinants and to methods for the preparation of such vaccines

Bovine Respiratory Syncytium virus (BRSV), a member of the paramyxoviruses, is a cause of respiratory tract infections in cattle

BRSV infection occurs world-wide and can cause severe disease especially in the lower respiratory tract, similar to the disease caused by human respiratory syncytial virus (HRSV) in children (Kimman and Westenbπnk, Archives of Virology, 1990 112, 1-25)

Although mortality vanes between 1 % and 30%, (Stott et al , J Hygiene 85. 251-261

(1980), Verhoeff, et al. Vet Rec 115 488-492 (1984)), the morbidity is very high (Baker et al., Am. J. Vet Res 46: 891-892 (1985), Baker et al., Vet. Clin N Am.: Food Animal

Practice 1. 259-275 (1985))

In one study it was found that more than 95% of cattle over 2 years of age were infected with BRSV (Van der Poel et al., Archives of Virology 1993, 133, 309-321) Reinfections occur frequently in both species, but in cattle usually without causing clinical signs (Kimman and Westenbπnk, Archives of Virology 1990, 112, 1-25) which suggests that a natural infection protects against clinical signs after reinfection

In cattle, expeπmentally infected through the natural route, it was found that the virus replicated in the cells of the nasal mucosae, the pharynx and the lungs (McNulty et al , Am

J. Vet Res 44" 1656-1659 (1983),Thomas et al , Bπtish J EΞxp Pathol. 65" 19-28 (1984)

Castleman et al., Am. J. Vet. Res. 46: 547-553 (1985))

Since BRSV-infection leads to high economic losses, there is a clear need for vaccines against BRSV.

However, vaccine development has been hampered because it is not known how a protective immune response can be induced First attempts to vaccinate children against

HRSV lead to enhanced disease after natural infection, which suggests that vaccination may even be harmful (Anderson et al., Journal of Infectious Diseases 1995, 171 , 1-7) It is known, however, that antibodies against two major surface glycoproteins , gF (shortly F; a fusion protein) and gG (shortly G; an attachment protein), play a key role in protection (Kimman and Westenbrink, Archives of Virology 1990, 112, 1-25).

Since, as mentioned above, it is not known how a protective immune response can be induced, it is important for potential vaccines to at least mimic the natural infection as closely as possible.

Therefore, it is tempting to include the immunologically relevant genes into live recombinant vector viruses.

These vector viruses mimic the natural infection in the sense that they do infect host cells and express, next to their own genetic material, the additional genetic information cloned into their genome.

An obvious vector for essaying the possible expression of both F and G protein under laboratory conditions is vaccinia virus. This virus has successfully been used as an expression vector for a multitude of different genes for many years.

See e.g. Paoletti, USP. 4,603,112 ( the general principle for integration of foreign DNA into vaccinia virus to produce a modified virus capable of expressing foreign genes.)

For a general review about vaccinia virus vectors, see e.g. Jolly, D.J.; Semin-lmmunol.

1990 Sep; 2(5): 329-339 (1990)

Expression of both the F and the G protein in vaccinia has been shown:

HRSV gG-expression in vaccinia virus as live recombinant vector virus was shown by Stott et al., J. Virol. 60: 607-613 (1986). Moreover, it was shown by Stott, that mice vaccinated with the HRSV gG vaccinia live recombinant vector virus were protected against challenge with wild-type HRSV.

Expression of the HRSV F-protein in vaccinia was shown by Wertz et al., J. Virol. 61 : 293-

301 (1987). Mice vaccinated with the live HRSV gF vaccinia recombinant vector were also protected against infection with wild-type HRSV.

For experimental purposes, the use of vaccinia virus as a live recombinant vector virus is attractive, because much is known about the virus and many tools for making vaccinia recombinants are available. Vaccinia virus, however, is known to have an extremely broad host range. It is known to be infectious for all mammalian species tested so far, ranging from rabbits, mice, raccoons, sheep, goats and camels to humans (Jolly, D.J.; Semin-lmmunol. 1990 Sep; 2(5): 329-339

(1990)).

From an environmental point of view, this broad host range of live recombinant viruses in general is an unwanted situation, especially where only bovine animals are target animals for BRSV-vaccination.

Vaccinia virus, although a bovine pathogen, would therefore certainly not be the vector of preference for use in animal vaccines for the protection of bovine animals against BRSV for the following reason:

Promising live recombinant vector virus candidates however would be the live attenuated bovine herpesviruses (BHV), since such bovine herpesviruses, if expressing the BRSV F- and/or G-gene, would have the following advantages:

• they are host-specific: they only infect bovine species.

• they protect against two different diseases: Bovine herpesvirus infection and BRSV- infection.

Live attenuated recombinant BHV-recombinants are known to be good potential expression vectors, and vaccine viruses at the same time.

BHV-1 has e.g. been used for the expression of LacZ (Schmitt et al., J. Virol. 70: 1091-1099

(1996)) and bovine interleukins 2 and 4 (Kϋhnle et al.; J. Virol. 77, in press (1996))

BHV-4 has been used e.g. for the expression of the LacZ-gene (Vanderplasschen et al.,

Virology 213: 328-340 ('95)), and the herpes simples virus I thymidine kinase gene (Keil,

G.M., VIII^th International Congress of Virology, Berlin, 26-31 August, 1990)

It has been suggested in e.g. EP0663403, in WO 95/21261 and WO 93/02104, to use live attenuated BHV-viruses as carriers for e.g. the BRSV G- and F-gene, because of the advantageous characteristics of BHV.

Until now, however, no cases have come to our knowledge of the actual expression of

BRSV-genes cloned in BHV as a recombinant carrier virus, let alone showing in vivo protection by such a BHV-recombinant. Two main problems are encountered when expression of BRSV-genes is tested in BHV-1 :

• BHV/BRSV-recombinant viruses in which the BRSV-F- and/or BRSV-G-gene is inserted in BHV, make no stable BRSV-RNA transcripts (as shown in Example 1).

• The presence of genes encoding wild-type BRSV-G- and F-protein makes it impossible to obtain BHV/BRSV-recombinant viruses that propagate in a stable manner. (Rijsewijk et al., (9^th BHV-workshop, July '95, Groningen, The Netherlands)).

The objective of the present invention is, to provide a BHV/BRSV-recombinant virus that overcomes the problems mentioned above.

Surprisingly it was found now, that if the original BRSV- gene is replaced by a synthetic BRSV-gene having a not naturally occurring nucleic acid sequence, but still encoding the naturally occurring amino acid sequence, both problems mentioned above are solved. The synthetic gene of the present invention can be efficiently expressed in BHV- recombinant viruses under the control of eukaryotic promoters, and the so obtained BHV/BRSV-recombinant viruses can be propagated in a stable manner.

A naturally occurring nucleic acid sequence is understood to be a nucleotide sequence as it is found in naturally occurring viruses, such as the viruses isolated from the field.

The synthetic gene according to the present invention has a nucleotide sequence that, albeit different from the original nucleotide sequence (i.e. the nucleotide sequence found in virus isolated from the field), still encodes exactly the same amino acid sequence, and thus encodes the native protein.

This is possible due to the degeneracy of the genetic code: all amino acids (with the exception of Met and Trp) are coded for by at least two different triplets. In the case of Leu, there are even 6 different triplets encoding this amino acid.

Thus, there is a large number of possibilities to modify the nucleic acid sequence encoding a certain gene, without at the same time modifying the amino acid sequence.

This principle is applicable to any gene, e.g. the BRSV-F- or G-gene.

In a preferred form of the invention, the BRSV-gene encodes the BRSV-G protein. This protein, as motivated above, is known to play an important role in the induction of an immunological response against BRSV-infection. In an equally preferred form of the invention, the BRSV-gene encodes the BRSV-F protein. This protein, as also motivated above, also plays an important role in the induction of an immunological response against BRSV-infection.

The degeneracy of the genetic code also allowed e.g. the introduction of a number of restriction enzyme cleavage sites in the nucleic acid sequence encoding the G-protein gene, without modifying the amino acid sequence of the G-protein.

As an example may serve the modification of the BRSV-G-gene made from position 55 onwards (see table 4), where the sequence (GC)CTCA was replaced by (GC)TAGC, thus providing the Nhel restriction site at position 55, without changing the two amino acids

Alanine and Serine encoded by these two triplets.

A relatively large number of new restriction sites, nine of which are unique, was thus introduced at regular distances in the gene encoding the BRSV gG-protein. The unique sites are underlined in table 4.

They are also shown in figure 5, where their location is compared with the restriction sites found in the original BRSV-G gene.

Synthetic DNA is defined as a DNA that is made by synthesis, instead of being isolated from a natural source.

This does not necessarily mean that each and every nucleotide of the DNA is synthesized. It is also possible to modify an existing DNA by replacing part thereof by a part with another nucleic acid sequence using recombinant DNA technology. The result of this modification is also considered to be a synthetic DNA

The synthetic DNA can be made in a number of different ways, all known in the art. One useful method for modifying a nucleotide sequence, is e.g. site-directed mutagenesis. With this generally known method, modifications are made deliberately at predetermined sites. It is also possible to replace small or longer fragments by fragments with an alternative sequence. Many different techniques for DNA-manipulation are currently available. One possible method of replacing (parts of) a naturally found sequence with a synthetically made sequence, is e.g. to cut the nucleic acid sequence with restriction enzymes to remove the sequence to be replaced, followed by ligation of a synthetically made fragment with the same restriction sites. Alternatively, the whole gene can be replaced by synthetic DNA-fragrhents. Techniques for site-directed mutagenesis, and for DNA-synthesis are known in the art. Also, DNA can be made fully synthetically in a DNA-synthesizer.

Given the fact that native BRSV-genes, when expressed in BHV do not give stable RNA transcripts, it is surprising that RNA from synthetic BRSV-genes is so stable.

An explanation for the surprisingly stable RNA formed by synthetic genes may be the following: RNA, synthesized in the cytoplasm, as is the case with BRSV-genes after natural infection, is not subjected to the process of RNA-splicing. RNA-splicing is restricted to RNAs synthesized in the nucleus, such as the nuclear RNAs.

The use of BHV (replicating in the nucleus) for the expression of genes originating from

BRSV, normally replicating in the cytoplasm, may lead to the formation of unstable (BRSV-

)RNA due to the presence of obvious and/or cryptic splice-donor and/or -acceptor sites, that are recognized by the nuclear RNA-splicing mechanism.

Splice-sites have small consensus sequences for both splice-donor sites (AGGU as a consensus sequence) and splice acceptor sites (a UC-rich region). In e.g. the nucleotide sequence of the wild-type G-gene at least one potential splice-donor site at position 531-

534 (see table 4) matching the AGGU consensus sequence was found, that is not present in the synthetic G-gene.

At least one potential splice-acceptor site is located from position 268 on, where a UC-rich region is found.

Therefore, in a more preferred form, the BRSV-gene is modified in such a way, that at least one possible splice-donor or acceptor site, found in the naturally occurring nucleic acid sequence, is removed in the synthetic BRSV-gene.

Consensus sequences of both the splice-donor- and -acceptor as known in the art, have been mentioned above. Splice-sites can be removed by replacing one or more nucleotides of the potential splice-site within the framework of what is allowed by the degeneracy of the genetic code. This can be done by e.g. site-directed mutagenesis, or replacement of small or longer fragments by fragments with an alternative sequence. Alternatively, the whole gene can be replaced by synthetic DNA-fragments.

Techniques for site-directed mutagenesis, and for DNA-synthesis are known in the art. It may not be necessary to remove all potential splice sites, since in many cases, the removal of only a splice-donor, or -acceptor site will suffice to suppress splicing.

Comparison of the nucleotide sequence of the synthetic G-gene and the original G-gene shows, that the GC-content of the synthetic gene is much higher than the GC-content of the native G-gene.

Comparison of the GC-content of BRSV-genes and BHV-genes showed the following: the average GC-content of both the BRSV-F- and -G-gene is about 42%, whereas the codon usage of e.g. the BHV-gD-gene is about 72%.

There appears to be an evolutionary advantage for BHV, being a virus replicating in the nucleus, to select for a high genomic GC-content.

This may point in the direction of a different codon usage in the nucleus, in which case the

BRSV-RNA, even if this is correctly transcribed and not destroyed due to splicing-events, can not be correctly translated.

It was therefore decided to make a comparison of the codon usage of the BHV-gD-gene and the BRSV-G-gene. This comparison showed e.g. that of e.g. the codons encoding the amino acid Arginine, BRSV never uses CGC and 6 times uses AGA, whereas BHV uses

CGC 15 times and AGA only once. For glycine, the BRSV-gene uses both GGT and GGC twice, whereas the BHV-gD-gene uses GGT only three times and GGC even 23 times.

This shows, that there indeed is a remarkably high preference for a high GC-percentage in the codons used, in BHV, whereas BRSV, normally replicating in the cytoplasm has a low

GC-content.

Therefore it is desirable to raise the GC-content of the synthetic BRSV-gene by replacing the nucleotides A and T by G and C, as far as possible within the framework of what is allowed by the degeneracy of the genetic code. This can be done by e.g. site-directed mutagenesis, or replacement of small or longer fragments, or all of the gene by synthetic

DNA-fragments. Thus, in an even more preferred form, the synthetic BRSV-gene has been modified in such a way, that the GC-content is at least 50%.

In an still even more preferred form, the GC-content is at least 60 %

In a most preferred form, the synthetic BRSV-gene has the sequence presented in table 4, second line.

It is clear, that the BRSV-gene, in order to be expressed, must be placed under the control of a promoter.

A large number of suitable promoters is known in the art, that are recognized for their very efficient level of expression. Such promoters are e.g. the Pseudorabies gX-promoter, the Pseudorabies TK-promoter, the Adenovirus Major Late promoter, the Retroviral Long Terminal Repeat, the SV40 Early and Late promoters, the MCMViel promoter, the MCMVel promoter, the HCMViel promoter, and the BHV-gE promoter.

From these, the MCMViel promoter, the MCMVel promoter, the HCMViel promoter, and the BHV-gE promoter, are preferred promoters.

Therefore, in a preferred embodiment, the synthetic BRSV-gene is placed under the control of one of the promotors of the group of promotors consisting of the MCMViel promoter, the MCMVel promoter, the HCMViel promoter, and the BHV-gE promoter.

Another embodiment of the invention relates to live attenuated BHV-recombinant viruses carrying a synthetic BRSV-gene according to the present invention.

As motivated above, BHV-viruses are the carriers of choice for BRSV-genes.

In a preferred form, as a BHV-recombinant, the BHV-1 virus is used.

This virus is a very commonly found pathogen in cattle, also (as is the case with BRSV) causing high economical losses.

The most common manifestation of Bovine Herpes Virus-I infection is bovine rhinotracheitis which varies from a mild respiratory disease to a severe infection of the entire respiratory tract. From an economical point of view, IBR is also the most dramatic manifestation of BHV-1 infection.

Morbidity rate in IBR is usually close to 100%.

If morbidity is < 100% then the animals are protected by antibodies. Fatalities are rare however, in the case of IBR. In young calves however, death as a result of BHV-I induced encephalitis is sometimes found. Next to this, several strains of BHV-I have shown to induce abortion.

Due to secondary bacterial infections, pneumoniae and enteritis may occur (Bielefeld-

Ohmann et al; J. Infect. Diseases 151 : 937-947 (1985))

Therefore, using BHV-1 as the live attenuated recombinant virus for carrying and expressing a BRSV-gene is very efficient: vaccines based on such a live attenuated recombinant virus protect against both BRSV and BHV-1. Animals so vaccinated are protected against the two most frequently found causes of respiratory disease in cattle.

A large number of potential insertion sites can be used for the insertion of the synthetic

BRSV-gene in the BHV-recombinant. The most suitable technique for such an insertion is homologous recombination, known in the art and frequently used.

In principle it suffices to integrate in a vector-DNA that comprises the gene to be inserted, a sufficiently large DNA-fragment from BHV, at positions left and right of the gene to be integrated. Homologous recombination will then allow the gene to be inserted into the BHV- genome at a previously determined site.

Preferably, the BRSV-gene to be inserted should be cloned between left and right fragments of a non-essential BHV-gene, since in that case, a viable recombinant is obtained that is not disturbed in essential functions.

Several non-essential genes are known for BHV. The genes coding for the (glyco)proteins gE, gl, gG, and US2 (=PrV 28k) are e.g. very suitable as integration sites.

In a more preferred form, the BRSV-gene to be expressed is integrated in the gE-gene of

BHV.

In an equally more preferred form, the BRSV-gene to be expressed is integrated in the gl- gene of BHV-1. A BHV-I recombinant according to the invention may, next to a BRSV-gene, comprise other genes encoding antigens from microorganisms or viruses that are pathogenic for cattle. A very attractive live attenuated BHV-recombinant is e.g. a BHV-recombinant that comprises both the BRSV-G-gene and the BRSV-F-gene, or one BRSV-gene and a non- BRSV-gene.

A BHV-recombinant comprising and expressing both the BRSV-G-gene and the BRSV-F- gene is also within the scope of the invention.

It is also possible that the BHV-recombinant comprises, next to one or more BRSV-genes, genes encoding antigens from other microorganisms or viruses that are pathogenic for cattle.

Therefore, in a preferred form, the present invention provides BHV-recombinants comprising, next to a BRSV-gene, a gene encoding an antigen from microorganisms or viruses that are pathogenic for cattle.

In an even more preferred form of this embodiment, the gene is chosen from the group of cattle pathogens, consisting of Bovine Rotavirus, Bovine Viral Diarrhoea virus, Parainfluenza type 3 virus, Bovine Paramyxovirus, Foot and Mouth Disease virus, Bovine Respiratory Syncytium virus and Pasteurella haemolytica.

Also, a gene may be introduced into the BHV-recombinant according to the invention, that encodes a cytokine. Several cytokines, e.g. interferons are known to play an important role as immune modulators. Thus it may be advantageous to include genetic information for this kind of molecules into said section.

Still another embodiment of the present invention refers to vaccines for the protection of cattle against virus infection, based upon live attenuated BHV-recombinants expressing a

BRSV-gene, according to the invention.

Vaccines based thereon have the advantage that they mimic the natural infection of not only BHV, but to a large extend also of BRSV, as motivated above.

Therefore, they provide protection against more than one pathogen, in this case against at least BHV-infection and BRSV-infection. Vaccination is in many cases at least a two-step process: a first immunisation with an antigen triggers the immune response, and a second immunisation; the booster, actually enhances both the speed and the strength of the immune response. After a first vaccination, immunity against both the carrier itself and the heterologous gene carried by the carrier is triggered. This is the result of the fact that the recombinant carrier infects a cell and during viral replication both the viral proteins and the heterologous protein of the carried gene are expressed and become presented at the cell membrane. There they are detected by the immune system. A known problem however, when using viruses as live recombinant carriers is the following: antibodies against the recombinant carrier raised during first immunisation prevent a successful second round of infection with the recombinant carrier. Thus no expression of proteins takes place, and therefore the immune system will not see the encoded proteins a second time. For the carrier virus this problem can be circumvented by just giving a higher dose of virus particles, since the viral proteins are perse present on the virus particle. In this approach the virus particles act as an inactivated vaccine and as such stimulates the immune system. Infection is however necessary for the heterologous gene to be expressed. Therefore, since no second round of infection occurs, the heterologous gene carried by the carrier is not expressed a second time. Thus no booster immunisation against the heterologous gene product will be generated.

It was now surprisingly found, that the BRSV gG protein is incorporated into the envelope of the BHV-1 virus particles during the maturation of the virus. This is contrary to what could be expected, since BRSV gG protein belongs to the class II membrane glycoproteins, whereas the alphaherpesviruses do, as far as known, not make any class II membrane glycoproteins. Class II membrane glycoproteins are characteristic in that they have an N- terminal membrane anchor. Membrane glycoproteins having an N-terminal membrane anchor are e.g. the hemagglutinin-neuraminidase of the paramyxovirus Simian Virus 5, the Influenza virus neuraminidase and the G-protein of the Human Respiratory Syncytial virus. (Hiebert et al., J. Virol. 54: 1-6 (1985), Bos et al., Proc. Natl. Acad. Sci. 81 : 2327-2331 (1984), Wertz et al., Proc. Natl. Acad. Sci. 82: 4075-4079 (1985)). Since BRSV gG lacks any herpesspecific targeting signals, BRSV gG could not be expected to be incorporated in the envelope of the BHV-1 particle. The unexpected incorporation of BRSV gG in the envelope of BHV means that the problem mentioned above can be circumvented: a booster immunisation with a high dose of BHV virus particles carrying the BRSV gG protein on their envelope causes a second immune response to be triggered against both the BHV envelope proteins and the BRSV gG protein on the envelope, without infection being necessary.

We found the unexpected incorporation of BRSV gG in the envelope to be dependent on the presence of the BRSV gG membrane anchor. Such a membrane anchor has the same characteristics in all class II membrane glycoproteins (C-ll MGs). This means that in general class II membrane glycoproteins for which the gene is carried by BHV, will be incorporated into the envelope of the BHV-particles. The use of BHV as a carrier for C-ll MGs thus circumvents the problem addressed above concerning the inefficiency of booster-reactions against heterologous genes in carriers.

One embodiment of the invention therefore relates to live attenuated BHV-recombinant virus particles, carrying a heterologous class II membrane glycoprotein.

It could moreover be shown that the incorporation of heterologous non-Class-ll-proteins into the envelope of BHV virus particles can also be obtained, provided that these proteins are coupled to a C-ll MG membrane anchor.

This means that the advantages mentioned above are not restricted to Class II membrane proteins, but can be obtained for any antigenically important protein. Coupling of such a protein to a C-ll MG membrane anchor now allows for the first time to target important proteins to the BHV membrane by using this class II anchor. A convenient way to make this coupling is to clone the nucleotide sequence encoding a C-ll MG membrane anchor at the 5' terminus of the heterologous gene to be expressed. Standard recombinant DNA technology suffices to obtain such clones. Merely as an example: above, the live attenuated BHV-recombinant carrying the synthetic gG of BRSV according to the invention is described. This BHV-recombinant can easily be used for targeting a non-C-ll protein into the envelope of BHV virus particles, by simply replacing all of the cloned gG-gene except for the first 213 base pairs encoding the membrane anchor, with the coding sequence of a non-C-ll protein-encoding gene. Therefore, still another embodiment of the invention relates to live attenuated BHV- recombinants carrying a heterologous gene fused to a class II membrane glycoprotein membrane anchor. In a preferred form the invention relates to live attenuated BHV-recombinants carrying a heterologous gene fused to the BRSV-gG membrane anchor.

The membrane anchor of the gG protein comprises the region from amino acid number 1 to 71. In principle, the whole membrane anchor sequence should be present in order to assure a stable integration into the virus envelope. It is however possible to use only part of the membrane anchor, i.e. less than the above-mentioned 71 amino acids, provided that integration of the final construct into the envelope can still be obtained. Furthermore, the present invention relates to method for the preparation of live attenuated recombinant BHV particles carrying a heterologous protein incoφorated in the particle envelope. Such methods comprise growing a live attenuated BHV-recombinant carrying a heterologous gene that encodes a class II membrane glycoprotein membrane anchor or is fused to a class II membrane glycoprotein membrane anchor.

It is clear, that the above-mentioned live attenuated BHV-recombinant particles are very suitable as a basis for vaccines for combating at the same time both BHV-infection and infections caused by microorganisms having a class II membrane glycoprotein that plays an important role in raising protective immunity.

Moreover, the above-mentioned live attenuated BHV-recombinants are very suitable as a basis for vaccines for combating at the same time both BHV-infection and infections caused by any microorganisms having a protein that plays an important role in raising protective immunity when fused to a class II membrane glycoprotein membrane anchor.

The vaccine according to the present invention may comprise a pharmaceutically acceptable carrier. One possible carrier is a physiological salt-solution. Another pharmaceutically acceptable carrier is for instance the solution in which an adjuvant is provided.

If desired, an adjuvant and possibly one or more emulsifiers such as Tween(^) and

Span(R) are also incorporated in the live or inactivated vaccine according to the invention. Suitable adjuvants are for example vitamin-E acetate solubilisate, aluminum hydroxide, - phosphate or -oxide, (mineral) oil emulsions such as Bayo ^R) and Marcol52(^R), and saponins. Incoφoration of the antigens in Iscoms is also a possible way of adjuvation.

It is advantageous to add a stabilizer to live or inactivated viruses, particularly if a dry composition of live viruses is prepared by lyophilisation. Suitable stabilizers are, for example, SPGA (Bovarnik et al., J. Bacteriology 59, 509, 1950), carbohydrates (such as sorbitol, mannitol, trehalose, starch, sucrose, dextran or glucose), proteins (such as albumin or casein), or degradation products thereof, and buffers (such as alkali metal phosphates). If desired, one or more compounds with adjuvant activity as described above can also be added.

A vaccine according to the invention may be administered by intramuscular or subcutaneous injection or via intranasal, intratracheal, oral, cutane, percutane or intracutane administration.

The natural route of infection is through the respiratory tract. Therefore, in a preferred form, the vaccine is administered intranasally.

Alternatively, the DNA isolated from the live attenuated BHV-recombinant according to the present invention can be administered. Vaccination methods using naked DNA instead of viruses have been described ia. by Cohen (Science 259: 1691-1692 (1993)), by Pardoll (Immunity 3: 165-169 (1995)) and by Montgomery (Current Biology 5: 505-510 (1994))

The useful effective amount to be administered will vary depending on the age, weight, and mode of administration. A suitable dosage can be for example about 10³ - 10^{π 0} pfu/animal.

Another embodiment of the present invention relates to methods for the preparation of vaccines according to the invention. Such methods comprise admixing a live attenuated BHV-recombinant according to the present invention and a pharmaceutically acceptable carrier. For the preparation of a live vaccine the BHV-I mutant according to the present invention can be grown on susceptible cells. Such susceptible cells are e.g. Madin Darby Bovine Kidney cells (MDBK-cells) or bovine embryonic cells.

The viruses thus grown can be harvested by collecting the tissue cell culture fluids and/or cells. The live vaccine may be prepared in the form of a suspension or may be lyophilized.

Still another embodiment of the present invention refers to methods for the preparation of live attenuated BHV-recombinants according to the invention.

Such methods comprise bringing together in a suitable host cell BHV-DNA and a vector comprising the BRSV-gene to be expressed, placed under the control of a suitable promoter and flanked by 3' and 5' flanking regions that share homology with BHV- sequences, in order to facilitate homologous recombination.

This can be done by first transfecting the suitable host cell with the vector, followed by infection with the BHV-virus.

Alternatively, both the vector and the isolated viral DNA may transfected together into a suitable host cell.

EXAMPLES

Example 1.

Cloning of the native G-gene (Gori) and the native F-gene (Fori) of BRSV under the control of the Pseudorabies virus (PrV) gp50 promoter.

For the expression of both the F-gene and the G-gene, the Pseudorabies virus gp50 promoter was chosen. This promoter is known to be a versatile and efficient promoter. PrV belongs just like BHV-1 to the alpha-herpesviruses, a group of herpesviruses of which the members show extensive structural and functional homology.

Gori was cloned under the control of the gp50-promoter as follows: Plasmid pRSV02 (see figure 1 , and kindly provided by Dr. P. Sondermeijer, Intervet Int., Boxmeer, The Netherlands) was cleaved with EcoRI, the ends filled with Klenow- polymerase, and the fragment containing Gori was isolated using gel-electrophoresis. Plasmid Tn77 (Klupp et al.; Virology 182: 732-741 (1991)) as depicted in figure 2, containing the PrV gp50 promoter was cleaved with Xbal (in the multicloning site downstream the promoter) and the ends filled with Klenow-polymerase. Subsequently the Gori fragment was ligated into the cleaved Tn77. See figure 3 for cloning scheme.

Fori was cloned under the control of the gp50-promoter as follows: Plasmid pRSV02 (see figure 1) was cleaved with Bglll, the ends filled with Klenow- polymerase, and the fragment containing Fori was isolated using gel-electrophoresis. Plasmid Tn77 was digested with Xbal and the ends filled with Klenow-polymerase. Subsequently the Gori fragment was ligated into the cleaved Tn77. See figure 3 for cloning scheme.

Cloning of Tn77-Gori-ORF and Tn77-Fori-ORF in the transfer-vector pAT-glV.

Vector pAT-glV has been described extensively in EP 0.663.403. It should be mentioned here that the vector pAT-glV is also referred to as pAT-gD, due to the new nomenclature of the herpesvirus (glycoprotein-)genes.

In order to clone the Gori-ORF and Fori-ORF brought under the control of the gp50- promoter, into the transfer-vector pAT-gD, the respective Tn-plasmids were cleaved with

Smal, and the respective inserts were isolated from agarose gels. pAT-gD was cleaved with

Ncol, the ends were filled with Klenow-polymerase and the respective inserts cloned in the now blunt-ended cleavage-sites.

The resulting plasmids pAT-gD Gori and pAT-gD Fori are shown in figure 4.

Integration of the respective plasmids pAT-gD Gori and pAT-gD Fori into BHV.

Madin Darby Bovine Kidney (MDBK) cells were transfected with 5 μg DNA of each of the above-mentioned plasmids 24 hours after seeding using the calcium phosphate precipitation technique. Transfected cells were shocked with glycerol 4 h after addition of the DNA and cultures were infected with phenotypically complemented glV^" (=gD) virus BHV-1/80-221 (Fehler et al, J. of Virol., 66: 831-839 (1992)) with a multiplicity of 10 PFU per cell.

At 48 h p.i. the cells were harvested and the cell culture supernatant was used for the infection of MDBK cells. Under these conditions only genotypically glV positive virions can grow productively and form plaques on the monolayers. At 3 days p.i. virions from single plaques were isolated and DNA from MDBK cells was purified, cleaved with Hindlll, size separated by agarose gel electrophoresis and transferred to nitrocellulose filters. Filters were hybridized with ³²P-labeled DNA either from the MCMV-ie2 polyadenylation signal, the β-galactosidase-ORF, Gori and Fori. The results demonstrate that the isolated virions lack the β-galactosidase sequences, and that the plasmids mentioned above are integrated.

It was concluded that insertion of the above mentioned plasmids indeed occurred via homologous recombination. Isolated virions were plaque purified twice to ascertain homogenous preparations. The activity of the gp50-promoter was checked, by making constructs in which the BRSV-G gene or BRSV-F gene were replaced by the luciferase gene or the native gp50 gene. In both cases, expression was found as expected, indicating the correct functioning of the gp50-promoter.

Southern-blotting of restriction-enzyme digests of the recombinants followed by probing with both BHV-specific and BRSV-G- or BHSV-F-specific DNA probes was done in order to check whether the recombinants still comprised the inserted BRSV-G- or BRSV-F-gene. These inserts appeared to be present fully as expected.

The transcription of BRSV-G- or BHSV-F-specific RNA was tested. Northern-blots of RNA obtained from cells infected with the various recombinants were probed with BHV-specific and BRSV-G- or BRSV-F-specific DNA probes.

From these experiments it was concluded that although expression of BHV-related RNAs was normal, no full-length BRSV-G- or BRSV-F-specific RNA could be found.

Example 2.

Construction of BRSV Gsyn ORF from synthetic oligonucleotides.

The cDNA sequence of the wild-type BRSV-G gene was determined by Lerch (J. Virol. 64: 5559-5569 (1990)). This sequence was used as the guiding sequence to compose a synthetic gene called BRSV Gsyn ORF or shortly Gsyn, that has many more convenient restriction sites compared to the wild-type BRSV-G gene, but still codes for a gG-protein that has an amino acid sequence that is fully identical to the wild-type gG-protein (further called Gori).

As can be seen from figure 5, the synthetic gene has 10 different restriction-sites, regularly divided over the whole gene, whereas the original gene has only three restriction-sites, two of which are moreover located almost at the same site . The synthetic gene therefore, contrary to the original gene, facilitates the construction of a whole range of smaller and larger deletion mutants. This allows the determination of sites that negatively interfere with the^" expression of BRSV- genes in vector viruses that replicate in the nucleus. To reach this goal, the following steps were taken:

A total of 24 synthetic oligonucleotides as indicated in table 1 was used for the construction of Gsyn. Table 2 shows fragments and cloning strategy.

Table 1 : synthetic oligonucleotides used for synthesis of Gsyn

Complementary strands were mixed in equimolar amounts in 10 mM Tris-HCL, pH 7.5, boiled for 5 minutes and then slowly cooled down to room temperature. This provided the 7 double-stranded fragments depicted in table 3.

Table 2, the fragments used for the various cloning steps are indicated

Fragment 1:

AAGCπACAAGTATGAGCAACCACACGCACACGCACCACCTG GTTCAAGACGCTGAAG TGCATTCGAATGTTCATACTCGTTGGTGTGCGTGTGCGTGGTGGACTTCAAGTTCTGCGACTTC Hindlll

CGCGCGTGGAAGGCTAGCAAGTACTTCATCGTCGGCCTGAGCTGCCTGTACAAGTTCAACCTGA GCGCGCACCTTCCGTACGTTCATGAAGTAGCAGCCGGACTCGACGGACATGTTCAAGTTGGACT

AGAGCCTGGTCCAGACGGCGCTGAGCACGCTCGCGAG TCTCGGACCAGGTCTGCCGCGACTCGTGCGAGCGCTCCTAG

Nrul

Fragment 2:

CGATGATCACGCTGACGAGCCTGGTCATCACGGCGATCATCTACATCTCCGTGGGCAACGCG GCTACTAGTGCGACTGCTCGGACCAGTAGTGCCGCTAGTAGATGTAGAGGCACCCGTTGCGC

AAGGCGAAGGCGAAGCCGACGTCG TTCCGCTTCCGCTTCGGCTGCAGCCTAG Aatll

Fragment 3:

CGAAGCCGACGATCCAGCAGACGCAGCAGCCGCAGAACCACACGAGCCCGTTCTTCACGG TGCAGCTTCGGCTGCTAGGTCGTCTGCGTCGTCGGCGTCTTGGTGTGCTCGGGCAAGAAGTGCC

AGCACAACTACAAGAGCACGCACACGAGCATCCAGAGCACGACCTTAAG TCGTGTTGATGTTCTCGTGCGTGTGCTCGTAGGTCTCGTGCTGGAA TCCTAG

Aflll

Fragment 4:

GCITAAGCCAGCTGCTGAACATCGACACGACGCGCGGCATCACGTAGGCCACAGCACGA ACGTCGAATTCGGTCGACGACTTGTAGCTGTGCTGCGCGCCGTAGTGCATACCGGTGTCGTGCT Aflll

ACGAGACGCAGAACCGCAAGATCAAAGGCCAGAGCACGCTGCCGGCGACGCGCAAGCCGCCGAT TGCTCTGCGTCTTGGCGTTCTAGTTTCCGGTCTCGTGCGACGGCCGCTGCGCGTTCGGCGGCTA

CAACCCGAGCGGAT GTTGGGCTCGCCTAGC

Fragment 5:

GAAGCTTATCGATACCGCCGGAGAACCACCAGGACCACAACAACTTCCAGACGCTGCC ACGTCTTCGAATAGCTATGGCGGCCTCTTGGTGGTCCTGGTGTTGTTGAAGGTCTGCGACGG Clal

GTACGTCCCGTGCAGCACGTGCGAGG CATGCAGGGCACGTCGTGCACGCTCCCATTG

Fragment 6:

GGTAACCTGGCGTGCCTGAGCCTGTGCCACATCGAGACGGAGCGCGCGCCGAGCCGGG ACGTCCATTGGACCGCACGGACTCGGACACGGTGTAGCTCTGCCTCGCGCGCGGCTCGGCCC BstEII

CCCCGACGATCACGCTGAAGAAGACGCCGAAGCCGAAGACGACGAAGAAGCCGACGAAGACG GGGGCTGCTAGTGCGACTTCTTCTGCGGCTTCGGCTTCTGCTGCTTCTTCGGCTGCTTCTGC

ACGATCCACCACCGCA TGCTAGGTGGTGGCGTGATC

Fragment 7:

GACTAGTCCGGAGACGAAGCTCCAGCCGAAGAACAACACGGCGACGCCGCAGCAAGGC ACGTCTGATCAGGCCTCTGCTTCGAGGTCGGCTTCTTGTTGTGCCGCTGCGGCGTCGTTCCG Spel

ATCCTGAGCAGCACGGAGCACCACACGAACCAGAGCAGGACGCAGATCTGAATTCA TAGGACTCGTCGTGCCTCGTGGTGTGCTTGGTCTCGTGCTGCGTCTAGACTTAAGTTCGA

EcoRI

Table 3 Fragments 1, 2 and 3 given in table 3 were cloned adjacently in pUC18, yielding the plasmid pUC18-F1-3.

This was done by first cleaving pUC18 with Aatll and BamHl. Fragment 1 was ligated into this DNA via its cohesive ends, resulting in plasmid pF1 in which the Aatll cleavage site of pUC18 was destroyed. Plasmid pF1 was cleaved with Nrul and BamHl and received fragment 2 to obtain pF2, in which fragment 3 was ligated after digestion with Aatll and BamHl to yield pF1-3.

Fragments 7, 6, 5 and 4 given in table 3 were cloned in that order in pUC19, yielding pUC19-F4-7.

This was done by first cleaving pUC19 with Hindlll and Pstl, followed by insertion of fragment 7. The resulting plasmid pF7 was cleaved with Pstl and Spel to receive fragment 6 to generate pF6. Plasmid pF6 was digested with Pstl and BstEII and after insertion of fragment 5 the obtained pF5 was treated with Pstl and Clal, after which fragment 4 was integrated, resulting in plasmid pF4-7.

Finally, pUC19-F4-7 was digested with Xbal, filled with Klenow-Polymerase, and cleaved with Aflll.

Plasmid pUC18-F1-3 was also digested with Xbal, filled with Klenow-Polymerase, and cleaved with Aflll, whereafter the insert was isolated trough gel-electrophoresis.

Finally, the insert was cloned in the cleaved pUC19-F4-7, to yield pBRSV Gsyn-ORF, containing the reconstructed BRSV Gsyn-ORF, shortly called Gsyn, flanked by EcoRI sites.

This is depicted in figure 6.

The sequence of the Gsyn-ORF is depicted in table 4.

ATG TCC AAC CAT ACC CAT CAT CTT AAA TTC AAG ACA TTA AAG AGG GCT TGG AAA 54

AG. ... ..C ..G ..C ..C ..G ..G ... ... ..G C.G ... C.C ..G ... . -G

M S N H T H H K F K T L K R A W K 18

GCC TCA AAA TAC TTT ATA GTA GGA TTA TCA TGT TTA TAT AAG TTC AAT TTA AAA 108

■ •X-ΔGC ..G ..c . -C ..C ..C C.G AGC ..c C.G ..C ... ... ..C C.G ..G

A S K Y r I V G L S c L Y K F N L K 36

TCC CTT GTC CAA ACG GCT TTG TCC ACC CTA GCA ATG ATA ACC TTG ACA TCA CTC 162

AG. ..G ... ..G ... ..G c.. AG. ..G ..c ..G ... ..C ..G c.. ..G AGC ..G

S L V Q T A L S T A M I T L T S L 54

GTC ATC ACA GCC ATT ATT TAC ATT AGT GTG GGA AAT GCT AAA GCC AAG CCC ACA 216

TCC ..£_ ..G

V I T A I I Y I S V G N A K A K P T 72 CC AAA CCA ACC ATC CAA CAA ACA CAA CAG CCC CAA AAC CAT ACC TCA CCA TTT 270

--G AGC ..G ..C

S K P T I Q Q T Q Q P Q N H T S P F 90

TTC ACA GAG CAC AAC TAC AAA TCA ACT CAC ACA TCA ATT CAA AGC ACC ACA CTG 324

AGC ..G ..G AGC

F T ε H N Y K S T H T S I Q S T T L 108

TCC CAA CTA CTA AAC ATA GAC ACT ACT AGA GGA ATT ACA TAT GGT CAC TCA ACC 378

AG. ..G ..G ..G ..C ..G ..G C.C ..C ..C ..G ..C AGC ..G s Q L N I D T T R G I T Y G H S T 126

AAC GAA ACC CAA AAC AGA AAA ATC AAA GGC CAA TCC ACT CTA CCC GCC ACC AGA 432

AG. C.C

N E T Q N R K I K G Q S T L P A T R 144

AAA CCA CCA ATC AAT CCA TCG GGA AGC ATC CCC CCT GAA AAC CAT CAA GAC CAC 486

AGC ... ΪCG ..A ..G ..G ..G ..c ..G

K P P I N P S G s I P P E N H Q D H 162

AAC AAC TTC CAA ACA CTC CCC TAT GTG CCT TGC AGT ACA TGT GAA GGT AAT CTT 540

N N F Q T P Y V P C S T C E G N 180

GCT TGC TTA TCA CTC TGC CAT ATT GAG ACG GAG AGA GCA CCA AGC AGA GCC^" CCT 594

..G C.G AGC C.C C.G . _{n η} -^■ G

A C S L C H I E T E R A P S R A P 198

ACA ATC ACC CTC AAA AAG ACT CCA AAA CCC AAA ACC ACT AAA AAG CCA ACC AAG 648

T I T L K K T P K P K T T K K P T K 216

ACA ACA ATC CAC CAC AGA ACC AGC CCT GAA ACC AAA CTG CAA CCT AAA AAC AAC 702

..G

T T I H H R T S P E T K Q P K N N 234

ACA GCA ACT CCA CAA CAA GGC ATC CTC TCT TCA ACA GAA CAT CAC ACA AAT CAA 756

..G ..G ..G ..G ..G ..G AGC AGC

T A T P Q Q G I L S S T E H H T N Q 252

TCA ACT ACA CAG ATC TAG 774

AGC . .G . .G . . .GA

S T T Q 257

Table 4.

Example 3.

CONSTRUCTION AND ANALYSIS OF BHV1-BRSV-G RECOMBINANTS

Cloning of the MCMV promoters ie1 and e1 in pAT-gD PLUS.

In order to test the expression of the Gsyn gene under various conditions, it was decided to clone the gene under the control of various alternative promoters. First of all, a multlcloning site was inserted in pAT-gD between the MCMVie2-polyA fragment and the gD-polyA fragment, and this new plasmid was named pAT-gD PLUS. Starting with pAT-gD PLUS, two constructs were made:

Construction of pROME: plasmid pAMB33 (Dorsch-Hasler et al., Proc. Natl. Acad. Sci 82: 8325-8329 (1985) was digested with Hpal and Xbal. Fragments were filled with Klenow- polymerase and isolated from agarose gels. A 1.4 kB fragment comprising the MCMV ie1 promoter was obtained.

The plasmid pAT-gD PLUS was cleaved in its polylinker site with EcoRV. In this site, the isolated MCMV ie1 fragment was cloned, to yield pROMI (See figure 7).

Construction of pROMI: plasmid MM354, comprising the MCMV e1 promoter (sequence of the Murine Cytomegalovirus early 1 gene is present in the EMBL Gene Bank at Heidelberg, Germany) was digested with Smal and Hindlll. Fragments were filled with Klenow- polymerase and isolated from agarose gels. A 1.1 kB fragment comprising the MCMV e1 promoter was obtained.

The plasmid pAT-gD PLUS was cleaved in its polylinker site with EcoRV. In this site, the isolated MCMV e1 fragment was cloned, to yield pROME (See figure 7).

Cloning of BRSV Gori ORF in pROME and pROMI.

Plasmid pRSV02 was cleaved with Bglll, and a 0.8 kB fragment was isolated from agarose gel.

Both the plasmids pROME and pROMI were digested with Bglll and the isolated 0.8 kB fragment was cloned to yield pROME-GoriΔ and pROMI-GoriΔ. Since a very short (19 bp) fragment at the 3' end of the G-gene is missing due to the use of Bglll as the restriction enzyme of choice, this missing fragment was made synthetically and cloned in the Ncol-site of pROME-GoriΔ and pROMI-GoriΔ to finally yield pROME-Gori and pROMI-Gori. (See figure 8).

Cloning of BRSV Gsyn ORF in pROME and pROMI.

Plasmid BRSV Gsyn ORF was cleaved with EcoRI and the resulting fragments were filled with Klenow-polymerase. A 0.8 kB fragment comprising Gsyn was isolated from agarose gel.

Both the plasmids pROME and pROMI were digested with Bglll and the isolated 0.8 kB fragment was cloned to yield pROME-Gsyn and pROMI-Gsyn.

(See figure 9).

Cloning of BRSV Fori ORF in pROME and pROMI.

Plasmid pRSV02 was cleaved with Bglll, and a 2.0 kB fragment comprising Fori was isolated from agarose gel.

Both the plasmids pROME and pROMI were digested with Bglll and the isolated 2.0 kB fragment was cloned to yield pROME-Fori and pROMI-Fori.

(See figure 10).

Integration of the respective plasmids pROME-Gori, pROMI-Gori, pROME-Gsyn, pROMI-Gsyn, pROME-Fori and pROMI-Fori in BHV.

MDBK cells were transfected with 5 μg DNA of each of the above-mentioned plasmids 24 hours after seeding using the calcium phosphate precipitation technique. Transfected cells were shocked with glycerol 4 h after addition of the DNA and cultures were infected with phenotypically complemented glV- virus BHV-1/80-221 (Fehler et al, J. of Virol., 66: 831- 839 (1992)) with a multiplicity of 10 PFU per cell.

At 48 h p.i. the cells were harvested and the cell culture supernatant was used for the infection of MDBK cells. Under these conditions only genotypically glV positive virions can grow productively and form plaques on the monolayers. At 3 days p.i. virions from single plaques were isolated and DNA from MDBK cells was purified, cleaved with Hindlll, size separated by agarose gel electrophoresis and transferred to nitrocellulose filters. Filters were hybridized with ³²P-labeled DNA either from the MCMV-ie2 polyadenylation signal, the β-galactosidase-ORF, ORF-1 or the sequences downstream ORF-1 presented for homologous recombination. The results demonstrate that the isolated virions lack the β- galactosidase sequences, that ORF-1 is deleted and that the plasmids mentioned above are integrated.

It was concluded that insertion of the above mentioned plasmids indeed occurred via homologous recombination. Isolated virions were plaque purified twice to ascertain homogenous preparations.

Transcription of the BRSV Gori and BRSV Gsyn ORFs in recombinant BHV-1 infected cells.

To test for the expression of the BRSV G ORFs by the BHV-1 recombinants, MDBK cells were infected with BHV-1/eGsyn (the BHV-1 recombinant in which the Gsyn gene is integrated under the control of the MCMVel promoter) (Figure 11 , lanes 1, 3, 5, and 7) and BHV-1/eGori (the BHV-1 recombinant in which the Gsyn gene is integrated under the control of the MCMVel promoter) (Figure 11, lanes 2, 4, 6 and 8). Cytoplasmic RNA was isolated at 6 h p.i., size separated by agarose gel electrophoresis, and transferred to nitrocellulose. A ³³P-labeled DNA probe representing the BRSV Gsyn ORF detected an RNA of 1.3 kb after infection with BHV-1/eGsyn (Figure 11 , lane 1) and, after extended exposure, a transcript of 1.9 kb (Figure 11, lane 5) whose synthesis is initiated approximately 600 bp upstream the MCMV e1 promoter (Grzimek and Keil unpublished). Even after longer exposure no transcripts were unequivocally detected by ³²P-labeled DNA from the BRSV Gori ORF (Figure 11 , lanes 2 and 6). Hybridization of the filters with ³²P- labeled DNA encompassing the BHV-1 gD ORF showed that comparable amounts of gD RNA were detectable in BHV-1/eGsyn (Figure 11, lanes 3 and 7) and BHV-1/eGori (Figure 11 , lanes 4 and 8) infected cells, demonstrating that absence of stable transcripts from the BRSV Gori ORF was not due to degradation of the RNA from BHV-1/eGori infected cells. Substantially, the same results were obtained with RNAs isolated at 16 h after infection with BHV-1/eGsyn and BHV-1/eGori in absence or presence of PAA (not shown). To elucidate whether the lack of stable RNAs from the BRSV Gori ORF in BHV-1/eGori infected cells was due to inactivation of the MCMV e1 promoter, promoter activities in BHV- 1/eGori and BHV-1/eGsyn infected MDBK cells were analyzed by nuclear run-on transcription assays. Labeled RNA synthesized in nuclei isolated at 6 h p.i. was hybridized to BRSV Gori, BRSV Gsyn, BHV-1 gD and plasmid vector pSP73 sequences, dotted on nitrocellulose membranes (Figure 12). Transcription from all ORFs was detectable, demonstrating that the MCMVel promoter was active in BHV-1/eGori infected cells. From these results we conclude, that the transcription activities of the BRSV G ORFs are comparable and that RNAs transcribed from BRSV Gori ORF are unstable in infected cells.

Identification and characterization of BHV-1 expressed BRSV-G

To test for the protein expression of BRSV-G in BHV-1/eGsyn infected cells, BRSV-G specific monoclonal antibody 20 (MAb 20) and a polyclonal antiserum (anti-VacGsyn) were used to stain plaques in infected cell cultures by indirect immunofluorescence. The anti- VacGsyn serum was raised in rabbits after infection with VacGsyn, a recombinant vaccinia virus which expresses the BRSV Gsyn ORF. MDBK cells were infected with wild type BHV- 1 strain Schόnbόken (BHV-1/Schό), BHV-1/eGsyn or BHV-1/eGori. After development of plaques cultures were fixed and incubated with the BHV-1 gD specific MAb 21/3/3 (Fehler et al, J. of Virol., 66: 831-839 (1992)), MAb 20 or the anti-VacGsyn serum. Antibodies bound on the cell surface were visualized with DTAF-conjugated anti-species immunoglobulin G by fluorescence microscopy. As shown in Figure 13, MAb 20 and the anti-VacGsyn serum bound only to cells in plaques generated by BHV-1/eGsyn whereas MAb 21/3/3 reacted with gD on the surfaces of cells infected with BHV-1/Schό, BHV- 1/eGsyn and BHV-1/eGori. Similar fluorescence was obtained when cells were permeabilized with Triton X 100 prior incubation with the respective antibodies. This result demonstrates that the BRSV Gsyn ORF encoded protein is expressed on the surface of the BHV-1/eGsyn-infected cells.

To further characterize the BRSV Gsyn ORF encoded gene product, proteins from infected cells (Figure 14, lanes 1-4), purified virions (Figure 14, lanes 5-7) and the cell culture medium from BHV-1/eGsyn infected cells (Figure 14, lane 8) were analyzed by immunoblotting using MAb20. This antibody did not specifically bind to proteins from cells infected with BHV-I/Schό (Figure 14, lane 1, ) and BHV-1/eGori (Figure 14, lane 2), proteins from purified BHV-1/Schό and BHV-1/eGori virions (Figure 14, lanes 5 and 6) and proteins released into the culture medium of BHV-1 /eGsyn infected cells (Figure 14, lane 8) but strongly reacted with proteins with apparent molecular masses of 38 and 43 kDa among BHV-1/eGsyn infected cell proteins (Figure 14, lanes 3). In addition, several weaker bands ranging in size from 30 to 100 kDa were detected. To elucidate possible precursor/product relationships among the MAb 20 reactive polypeptides, BHV-1 /eGsyn infected cells were incubated fro 2 h with cycloheximide prior to lysis to inhibit cfe novo protein synthesis. This treatment resulted in a considerable decrease of the signal at 38 kDa and slight reduction of the band intensity at 43 kDa, whereas the higher molecular weight bands increased in size and abundance (Figure 14, lane 4). Polypeptides around 38 kDa were no longer detectable and the protein migrating at 30 kDa appeared unaffected. Prolongation of the cycloheximide incubation time did not significantly alter this pattern but resulted in a general loss of the intensities of all bands (not shown). After electrophoresis of proteins from purified BHV-1/eGsyn virions MAb 20 bound to polypeptides which migrated as a diffuse band with an apparent molecular mass between 70 and 120 kDa (Figure 14, lane 7). From these results we conclude that the 38 kDa and 43 kDa proteins represent precursor molecules which are modified to the high molecular weight form. Association of the high molecular weight proteins, but no the putative precursor molecules, with purified virions indicated that they represent mature BHV-1 expressed BRSV-G.

Cloning of the synthetic BRSV-G gene in the BHV-1 recombination cassette p175 behind the promoter of glycoprotein E (gE)

The synthetic BRSV-G gene Gsyn has also been cloned behind the BHV1 gE-promoter. As an insertion site for the construct, the glycoprotein E (gE) gene has been chosen because it is not essential for virus replication and because gE deletion mutants have good vaccine properties (Van Engelenburg et al., Journal of Virology 75: 2311-2318 (1994); Patent Application WO92/21751). In order to replace BHV1 gE by Gsyn, the Gsyn gene has first been cloned into BHV1 recombination cassette p175. Recombination cassette p175 is constructed by cloning the flanking regions of the BHV1 gE gene into pUC18 using standard methods (Maniatis, T. et al, in "Molecular cloning; a laboratory manual, (1982) ISBN 0-87969-136-0). The left hand flanking region starts at the Pstl site in the coding region of glycoprotein I (gl) and ends at the BstBI site (=Asull-site) 17 nucleotides upstream of the start of the gE open reading frame. This 1.1 kb Pstl-BstBI fragment contains the gE promoter that directs the transcription of the gE gene.

The right hand gE flanking region starts at the EcoNI site at the stop codon of the gE open reading frame and ends at the first Smal site downstream of the gE open reading frame that is located in the terminal repeat region. (The left and right flanking region have been described in WO92/21751.)

The Gsyn gene has been liberated from plasmid BRSV Gsyn ORF (see figure 6) using restriction enzymes Accl and Ncol and has been ligated in the proper orientation between the BstBI and EcoNI site, using standard methods (Maniatis, T. et al, in "Molecular cloning; a laboratory manual,(1982) ISBN 0-87969-136-0).

The resulting recombination construct has been named p531 (see Figure 15).

Cloning of the synthetic Gsyn gene in the BHV1 recombination cassette phCMV175 behind the promoter of the immediate early 1 gene of human cytomegalovirus (hCMVie)

To enhance the expression of the BRSV-G gene in the gE locus of BHV1, the promoter of the immediate early 1 gene of human cytomegalovirus (hCMVie-promoter: Peeters et al.,; J. Virol. 66: 894-905 (1992)) has been cloned upstream of the BRSV-G gene. To recombine this hCMVie-BRSV-G combination into the BHV1 genome, the BRSV-G gene has first been cloned into the phCMV175 recombination cassette. The phCMV175 recombination cassette has been derived from the cassette p175 by inserting a 720 bp Asel fragment, containing most of the hCMVie promoter cis-regulatory signals, between the BstBI site and the EcoNI site of p175. The 720 bp Asel fragment has downstream of the hCMVie promoter sequences a polylinker region and a polyadenylation signal. The synthetic BRSV-G gene has been liberated with restriction enzymes Accl and Ncol and cloned into the Smal site in this polylinker region using standard methods (Maniatis, T. et al, in "Molecular cloning; a laboratory manual, (1982) ISBN 0-87969-136-0). The resulting construct has been named pr608 (see Figure 16). Recombination of p531 and pr608 into the genome of BHV1.

To recombine the Gsyn gene under the control of the gE promoter and under the control of the hCMVie promoter into the genome of BHV1 the cotransfection method has been used as described by (Van Engelenburg et al. (1994), Journal of Virology 75, 2311-2318). Microgram quantities of genomic DNA of BHV1 strain Lam and linearized plasmid DNA of either construct p531 or construct pr608 have been cotransfected into cultured monolayers of Embryonic bovine trachea cells (EBTr) using the transfection method of F.L. Graham and A.J. van der Eb (1973, Virology 75, 456-467). 48 hours after the cotransfection the cultured EBTr cells have been freeze-thawed to liberate the emerged wild type and recombinant viruses. The freeze-thawed cell debris has been pelleted and 100 microliters of the supernatant has been used to infect monolayers of Ebtr cells on a 96 well plate to enrich for recombinant virus. After three days of virus replication the culture medium of each well has been transferred to a second 96 well plate and stored at -20°C. The remaining infected monolayers have been immuno-histologically stained applying the immuno-peroxidase monolayer assay (IPMA) method as described by Wensvoort et al., (1986, Veterinary Microbiology 12, 1011-08) and anti-BRSV-G monoclonal antibody 20 (MAb20) as described by Langedijk et al., (1996, Journal of General Virology 77, 1249-1257). The culture medium of Mab20 positive wells (see Figures 20 and 21) has been used to infect another 96 well plate with EBTr monolayers and the procedure has been repeated until virtually all BHV1 infected cells express BRSV-G. The virus stocks enriched for recombinant virus have been used for three rounds of plaque purification and the resulting BHV-1-BRSV-G recombinants have been named respectively 531 and 608 (see Figures 17 and 18).

Restriction enzyme analysis of BHV1 -BRSV-G recombinants

To analyze whether besides the intended recombination no gross rearrangements have taken place in the genome of the BHV1 constructs The DNA of recombinants 531 and 608 have been isolated and analyzed with restriction enzyme Hindlll. No gross rearrangements could be found (see Figure 19). Expression of BRSV-g in BHV1 -BRSV-G recombinants in vitro

To compare the BRSV-G expression by BHV1 -BRSV-G construct 531 and BHV1 -BRSV-G construct 608, 96 well plates with monolayers of EBTr cells have been infected with a 0,5 x

10⁵ or 0,5 x 10⁴ TCID₅₀ per well. At specific time points after infection the cultures have been frozen and stained using the IPMA procedure with a the soluble substrate 5- aminosalicyclic acid (5AS) of which the extinction was measured using a Titertek multiscan at 405 nm as described by Westenbrink et al., J. Gen. Virol. 7: 591-601 (1989).

See figures 16 and 17 for immuno-histochemical detection of BRSV-G protein expressed by recombinant virus 531 and 608 in EBTr cells.

See figure 22 for the expression kinetics of recombinant virus 531 and 608 in EBTr cells.

As can be seen from these figures, expression of the BRSV-Gsyn gene is found in vitro.

Example 4.

Vaccination challenge experiments with the BHV-1 /BRSV-Gsyn recombinant pr608 in animals.

Animals

Twelve specific-pathogen-free calves were obtained by caesarean section, deprived of colostrum and reared in isolation. Calves were vaccinated twice with 3 weeks interval, starting at an age of 7 weeks. The calves were free of antibodies against bovine heφesvirus 1 (BHV-1) and antibodies against BRSV before the start of the experiment, determined in the gB-blocking ELISA (Kramps et al, 1994, Journal of Clinical Microbiology, 32, 2175-2181) and in the BRSV F-ELISA (described by Schrijver et al., 1996, Clinical and Diagnostic Laboratory Immunology, in press, adapted from Westenbrink et al, Research in Veterinary Science 38: 334-340 (1985)). The calves were allotted randomly in two groups of 6 animals, and housed separately in conditioned isolation stables. Vaccinations

At an age of 7 weeks, vaccinations started. Six calves were vaccinated twice with 5x10⁷ 50% tissue culture infective dose (TCID_S0) of the BHV-1/BRSV-Gsyn recombinant pr608 per animal, in a volume of 20 ml, of the recombinant gE-BHV-1 virus expressing the G-protein of BRSV (recBHV-G). The animals were sedated with 0,3 ml xylazine (20 mg/ml) intravenously immediately before the vaccination. Then 5 ml of recBHV-G diluted in HMEM containing 2% foetal bovine calf serum (FCS) was sprayed into each nostril using a vaccination nozzle attached to a syringe. Subsequently, 10 ml was injected intratracheally with a 25 gauge needle. In the laboratory the remainder of the inoculum was titrated on susceptible embryonic bovine tracheal (EBTr) cells, to determine the exact titer of the inoculum. The second vaccination was performed after 21 days following the same procedure. The 6 control calves were mock-vaccinated with medium.

Challenge.

All animals were challenged 6 weeks after the first vaccination by intranasal instillation of 1 ml of 10³⁶ TCIDso/ml of a none-cell-culture passage BRSV strain Odijk (1996, Van der Poel et al., Veterinary Quarterly, in press) in each nostril. The calves were sedated with 0,2 ml of xylazine (20 mg/ml) intravenously immediately before the challenge. The challenge inoculum consisted of clarified lung lavage fluid of experimentally BRSV-infected SPF- calves with Hank's MEM (1996, Van der Poel et al., Veterinary Quarterly, in press). The inoculum was found free of bovine virus diarrhea virus and mycoplasmata.

Clinical observations, virus isolations and antibody titrations

Recording of clinical signs, excretion and infectious BRSV in nasopharyngeal swabs and lung lavage fluid, and antibody titrations were determined as described previously (1996, Van der Poel et al., Veterinary Quarterly, in press). Clinical observations were performed daily during 12 days after challenge, starting 2 days before the challenge. Nasopharyngeal swabs were taken, and lung lavages were performed, at day -5, 2, 5, 7, 9 and 12 after challenge. Infectious BRSV titrated on susceptible bovine foetal diploid lung (BFDL) cells. Antibodies against the G protein of BRSV were determined by means of a G peptide ELISA as described previously (Schrijver et al., 1996, Clinical and Diagnostic Laboratory Immunology, in press). Results

Titration of infectious BRSV in nasopharyngeal swab fluid and in lavage fluid has been completed. The results are indicated in Figures 23 and 24. It is clear that the amount of infectious virus is greatly reduced in the calves vaccinated with the recBHV-G vaccine, compared to the greater amount of infectious BRSV in mock-vaccinated calves.

From the reduction of infectious BRSV, it can be concluded that: a the synthetic G protein is adequately expressed by the BVH-1 vector in vivo. b. the recombinant BVH/BRSV-Gsyn vaccine protects against BRSV infection in calves.

Example 5

Virus neutralisation assays

In order to demonstrate the presence of BRSV gG in the viral envelope, the following experiment was done: MDBK cells were infected with BHV-1/Schό, BHV-1/G_ori and BHV- 1/G_Sy_n. (See fig. 25). Proteins from infected cells, harvested at 10 h.p.i. (lanes 1-4), from purified virions (lanes 5-7) and cell culture medium (lane 8) were analysed by immunoblotting with BRSV G-specific MAb 20. Proteins shown in lane 4 were from cells incubated with cycloheximide (100 μg/ml) for 2 h before lysis.

The immunoblot analysis shown in Fig. 25 indicated that BHV-1/Gsyn virus particles contain the G glycoprotein and therefore should be susceptible to inactivation by gG specific antibodies. In order to demonstrate that BHV-1/Gsyn virus particles can indeed be inactivated by gG-specific antibodies, BHV-1/G_orj and BHV-1/G_syn virions were tested for complement dependent and independent neutralisation by an anti-VacG_syn serum raised against vaccinia virus expressing G_syn and an anti-BRSV serum, raised in gnotobiotic calves after infection with BRSV.

Approximately 200 PFU of wild-type and recombinant BHV-1 /gG strains were incubated with serial dilutions of monoclonal antibodies or sera against gG in a final volume of 100 μl cell culture medium containing 20% FCS with or without 5% normal rabbit serum as a source of complement and plated onto MDBK-Bu100 cells after 1 h at 37°C. Cultures were overlaid with semisolid medium 2 h later and plaques were counted 2 to 3 days after infection. % neutralisation resistant infectivity was calculated from controls incubated without antibody. The anti-VacG_Syn serum specifically neutralised BHV-1/G_syn in presence of complement with 50% neutralisation at a dilution of about 1 :1 ,000 (Fig. 26). Both sera had no significant effect on the infectivity of BHV-1/G_ori (Fig. 10) and BHV-1/Schό. The results of this experiment show that indeed the gG protein is present at the viral envelope of BHV-1/Gs_yn virus particles. The results moreover show, that it is possible to neutralise these virus particles with antiserum specifically directed against BRSV.

Penetration kinetics

The penetration behaviour of BHV-1/G_syn was analysed to test whether the presence of the

G glycoprotein influences the entry of BHV-1/G_Sy_n virions into the target cells.

Penetration kinetics were determined essentially as described previously (Fehler et al, J. of

Virol., 66: 831-839 (1992)) using low pH inactivation of extracellular virions at different times after a shift of infected cells from 4 to 37°C.

Fig. 27 shows the result of a representative experiment. BHV-1/Schό, BHV-1/G_ori and BHV-

1/G_Sy_n were completely protected from low pH mediated inactivation after 60 min. A slight, but in all experiments consistent difference was observed for 50% penetration for which

BHV-1/G_Sy_n required approximately 5 min more than BHV-1/Schό and BHV-1/G_ori.

The results of this experiment show that association of BRSV G glycoprotein with BHV-1 virus particles has no considerable effect on the penetration process.

LEGENDS TO THE FIGURES

Figure 1

Physical map of plasmid pRSV02. The plasmid contains a 2.8 kB cDNA fragment that comprises the native BRVS-G-gene and the BRSV-F-gene.

Figure 2

Physical map of plasmid Tn77. This plasmid comprises a 0.24 kb Ndel-Nlalll fragment from

Pseudorabies virus, comprising the PrV gp50 promoter.

Figure 3

Cloning strategy of the BRSV-G-gene and the BRSV-F-gene into the plasmid Tn77. In this figure, the various cloning steps are stepwise indicated.

Figure 4

Cloning strategy of the BRSV-G-gene and the BRSV-F-gene, brought under the control of the PrV gp50 promoter, into the plasmid pAT-gD. In this figure, the various constructs are indicated.

Figure 5

Restriction map of both the native BRVS-G-gene (Gori) and the synthetic BRSV-G-gene

(Gsyn). The unique restriction sites are indicated by vertical bars.

Figure 6

Physical maps of the constructs pUC18-F1-3 and pUC19-F4-7, and physical map of BRSV-

Gsyn-ORF, which was constructed from pUC18-F1-3 and pUC19-F4-7.

Figure 7

Physical map of the plasmids pROME and pROMI. The location of the respective promoters

MCMViel and MCMVel are indicated.

Figure 8 Physical map of the plasmids pROME-Gori and pROMI-Gori. The construction of these plasmids from pROME and pROMI, and the stepwise cloning of fragments constituting the Gori-genes are cloned is indicated.

Figure 9

Physical map of the plasmids pROME-Gsyn and pROMI-Gsyn. The construction of these plasmids from pROME and pROMI is indicated.

Figure 10

Physical map of the plasmids pROME-Fori and pROMI-Fori. The construction of these plasmids from pROME and pROMI is indicated.

Figure 11

Identification of transcripts. Whole cell RNA from cells infected with BHV-1/eGsyn (lanes 1 , 3, 5 and 7) and BHV-1/eGori (lanes 2, 4, 6 and 8) was prepared at 6 h p.i. and 5 μg were transferred to nitrocellulose after 1% agarose gel electrophoresis. Filters were hybridized to ³²P-labeled DNA from the BRSV Gsyn ORF (lanes 1 and 5), the BRSV Gori ORF (lanes 2 and 6) and the BHV-1 gD ORF (lanes 3, 4, 7 and 8). Bound radioactivity was visualized by autoradiography. Lanes 5-8 show longer exposures of lanes 1-4. Transcript sizes are indicated in kb.

Figure 12

Run-on transcription. At 6 h p.i., nuclei from MDBK cells infected with BHV-1/eGori and BHV-1/eGsyn were isolated, and transcripts, elongated in presence of α-[³²P]CTP were hybridized to dot blots containing 1.5 pmol of plasmids pSPGori or pSPGsyn, pgD and plasmid vector pSP73 (indicated by an open circle). Bound radioactivity was visualized by autoradiography. To determine the ratios of transcription between the BRSV-G ORFs and BHV-1 gD, radioactivity in the respective dots was determined by Cerenkov counting.

Figure 13

Cell surface expression of BRSV-G. MDBK cell cultures were infected with approximately

20 PFU of BHV-1/Schό, BHV-1/eGsyn and BHV-1/eGori fixed after development of plaques and used for indirect immunofluorescence with BHV-1 gD specific MAb 21/3/3, BRSV-G specific MAb 20 and the anti-VacGsyn serum. Bound antibodies were visualized by staining with DTAF-conjugated goat anti-species immunoglobulin G.

Figure 14

Identification of the BHV-1/eGsyn expressed BRSV-G. MDBK cells were infected with BHV- 1/Schδ (lanes 1 and 5), BHV-1/eGori (lanes 2 and 6) and BHV-1/eGsyn (lanes 3, 4, 7 and 8). Proteins from infected cells, harvested at 10 h p.i. (lanes 1-4) from purified virions (lanes 5-7) and cell culture medium (lane 8) were analyzed by immunoblotting with the BRSV-G specific MAb 20. Proteins shown in lane 4 were from cells incubated with cycloheximide (100 μg/ml) for 2 h before lysis.

Figure 15

Construction of p531.

In order to clone the synthetic BRSV-G gene into the BHV1 recombination vector, the BRSV-G gene has been isolated from the BRSV-Gsyn-ORF vector using restriction enzymes Accl and Ncol. Shown at the top. This Accl-Ncol fragment has been cloned between the BstBI and the EcoNI site of recombination cassette p175. Shown in the middle. Recombination cassette p175 contains the flanking regions of the BHV1-gE gene. The right hand 1.1 kb Pstl -BstBI fragment covers the C-terminal portion of the gl gene and the gE promoter region. The left hand 1 kb EcoNI-Smal fragment covers the US9 gene and part of the terminal repeat region (TR). The resulting construct has been named p531 and is shown at the bottom.

Figure 16

Construction of pr608.

In order to clone the synthetic BRSV-G gene behind the hCMVie promoter the BRSV-G gene has been isolated from the BRSV-Gsyn-ORF vector using restriction enzymes Accl and Ncol. Shown at the top. This Accl-Ncol fragment has been cloned into vector phCMV175, shown in the middle. Recombination cassette phCMV175 has been derived from recombination cassette p175 (see Figure 11). To construct phCMV175 a 720 bp Asel fragment has been cloned between the BstBI and EcoNI site of p175. This 720 bp contains most of the hCMVie promoter and has downstream of the transcription initiation site a polylinker sequence and a polyadenylation signal (Poly-A sign). Shown in the middle. The BRSV-G fragment has been cloned into the Smal site of this polylinker sequence and the resulting construct has been named pr608 and is shown at the bottom.

Figure 17

Diagram of BHV1-BRSV-G recombinant 531. The top line represents the BHV1 genome with its two segments: a long unique segment (L) and a short segment (S). The S segment has a central unique short region that is bordered by a direct repeat indicated by the black boxes. The gE locus is located in the unique short region. In the bottom line shows a part of the unique short region and a bordering repeat has been shown in more detail. Indicated are the position of the glycoprotein I (gl ) gene, the gE promoter (gE-P), the synthetic BRSV-G gene and the US9 gene.

Figure 18

Diagram of BHV1 -BRSV-G recombinant 608. The top line represents the BHV1 genome with its two segments: a long unique segment (L) and a short segment (S). The S segment has a central unique short region that is bordered by a direct repeat indicated by the black boxes. The gE locus is located in the unique short region. The bottom line shows a part of the unique short region and a bordering repeat has been shown in more detail. Indicated are the position of the glycoprotein I (gl) gene, the promoter of the human cytomegalovirus (hCMV-P), the synthetic BRSV-G gene, a polyadenylation signal (Poly-A sign) and the US9 gene.

Figure 19

Immuno-histochemical detection of BRSV-G protein expressed by recombinant virus 531. EBTr cells have been infected with a mixture of wild type BHV1 and the BHV1-BRSV-G recombinant and after three days stained with the IPMA method using anti-BRSV-G Mab20. The chromogenic substrate used for the peroxidase reaction is 3-amino-9-ethylcarbazole (AEC). This substrate gives a non-soluble precipitate. The stained regions are infected with the BHV1-BRSV-G recombinant 531 and the non-stained regions are infected with wild type BHV1. Amplification 400 times. Figure 20

Immuno-histochemical detection of BRSV-G protein expressed by recombinant virus 608.

For technical details see Figure 5.

Figure 21

Restriction enzyme analysis of the genomes of BHV1 -BRSV-G recombinant viruses 531 and 608. Genomic DNA has been isolated according to standard methods (see Van Engelenburg et al., 1994, Journal of Virology 75, 2311-2318) and digested with restriction enzyme Hindlll. The obtained DNA fragments have been separated on a 0.7% agarose gel according to standard methods (Maniatis, T. et al, in "Molecular cloning; a laboratory manual, (1982) ISBN 0-87969-136-0), stained with ethidium bromide using an UV transilluminator and photographed using a Polaroid camera with a 667 Polaroid film. In lane 1 genomic DNA of phage lambda digested with Hindlll has been separated as a molecular size marker. In lanes 2, 3 and 4 digests of genomic DNA of respectively BHV1 parent strain Lam, BHV1-BRSV-G recombinant 531 and BHV1-BRSV-G recombinant 608 have been separated.

Figure 22

Quantification of BRSV-G expression.

50 microliters of 10⁶ or 10⁵ TCID_S0/ml BHV1-BRSV-G recombinant viruses 531 or 608 have been added to each well of a 96 well plate with monolayers of EBTr cells. At specific time points after infection the cultures have been frozen and stained using the IPMA procedure with the soluble substrate 5-amino-salicylic acid (5AS) of which the extinction was measured using a Titertek multiscan at 450 nm.

Figure 23

Excretion of BRSV virus from lung tissues from BRSV challenged calves as recovered in lavage fluid. The indicated titers (in log₁₀TCID₅o) are the mean values (+/- the standard deviation) found in each group of calves (N=6). The recBHV-G group had been vaccinated twice with the BHV1 -BRSV-G recombinant 608. The control group has been twice mock infected.

Figure 24 Excretion of BRSV virus from nasal and tracheal tissues from BRSV challenged calves as recovered in nasopharyngeal swabs. The indicated titers (in log₅₀ TCIDso' are the mean values (+/- the standard deviation) found in each group of calves (N=6). The recBHV-G group had been vaccinated twice with the BHV1 -BRSV-G recombinant 608. The control group has been twice mock infected.

Figure 25

Identification of the BHV-1 /G_syn expressed BRSV-G. MDBK cells were infected with BHV- 1/Schδ (lanes 1 and 5), BHV-1/G_ori (lanes 2 and 6) and BHV-1/G_syn (lanes 3, 4, 7 and 8). Proteins from infected cells, harvested at 10 h.p.i. (lanes 1-4), from purified virions (lanes 5- 7) and cell culture medium (lane 8) were analysed by immunoblotting with BRSV G-specific MAb 20. Proteins shown in lane 4 were from cells incubated with cycloheximide (100 μg/ml) for 2 h before lysis.

Figure 26

BHV-1/Gsy_n virions are susceptible to neutralisation by antibodies against the G glycoprotein. Approximately 200 PFU of BHV-1/eG_ori were incubated with 20% fetal calf serum and complement or serial dilutions of anti-VacGsyn with complement (open squares) or the anti-BRSV hyperimmune serum with complement (open circles) and about 160 PFU of BHV-1/G_Sy_n were incubated with 20% fetal calf serum and complement or with serial dilutions of anti-VacGsyn with complement (closed squares) the anti BRSV hyperimmune serum without complement (triangles, no complement added to the FCS control) or the anti- BRSV hyperimmune serum with complement (closed circles). After 60 min at 37°C virions were plated on MDBK cells and cultures were overlaid with semi-solid medium. Plaques were counted 3 days later. The plaque count of each FCS control was set as 100% neutralization resistant infectivity. Serum dilutions are indicated. Data from one representative experiment are shown.

Figure 27

Penetration kinetics. Stocks of BHV-1/Schδ (asterix), BHV-1/G_ori (open squares) and BHV- 1/G_Sy_n (closed squares) were diluted to yield approximately 200 PFU per culture dish. Cells were preincubated at 4°C for 15 min. and the respective viruses were allowed to adsorb for 2 h at 4°C. After a temperature shift to 37°C non penetrated virions were inactivated by low pH treatment at the times indicated. Cells were overlaid with semi-solfd medium and plaques were counted 3 days later. Percent penetration was calculated from respective values obtained without inactivation.

Claims

Claims

1. Synthetic BRSV-gene, characterised in that said BRSV-gene has a not naturally occurring nucleotide sequence encoding a naturally occurring amino acid sequence.

2. Synthetic BRSV-gene according to claim 1 , characterised in that said BRSV-gene encodes the BRSV gG protein.

3. Synthetic BRSV-gene according to claim 1 , characterised in that said BRSV-gene encodes the BRSV gF protein.

4. Synthetic BRSV-gene according to claims 1-3, characterised in that in said BRSV-gene at least one naturally occurring putative splice-donor- or -acceptor site is removed.

5. Synthetic BRSV-gene according to claims 1-4, characterised in that said BRSV gene has a GC-content of more than 50%.

6. Synthetic BRSV-gene according to claim 5, characterised in that said BRSV gene has a GC-content of more than 60%.

7. Synthetic BRSV-gene according to claim 1 , characterised in that said BRSV gene has the nucleic acid sequence presented in table 4, second line.

8. Synthetic BRSV-gene according to claims 1-7, characterised in that said BRSV gene is under the control of the MCMViel -promoter.

9. Synthetic BRSV-gene according to claims 1-7, characterised in that said BRSV gene is under the control of the MCMVel -promoter.

10. Synthetic BRSV-gene according to claims 1-7, characterised in that said BRSV gene is under the control of the HCMViel -promoter.

11. Synthetic BRSV-gene according to claims 1-7, characterised in that said BRSV gene is under the control of the control of the gE-promoter.

12. Live attenuated BHV-recombinant, characterised in that it carries a synthetic BRSV- gene according to claims 1-11.

13. Live attenuated BHV-recombinant according to claim 12, characterised in that said BHV-recombinant is a BHV-1 recombinant.

14. Live attenuated BHV-recombinant according to claim 13, characterised in that the synthetic BRSV-gene is inserted in the gE-gene of said BHV recombinant.

15. Live attenuated BHV-recombinant according to claim 13, characterised in that the synthetic BRSV-gene is inserted in the gl-gene of said BHV recombinant.

16. Live attenuated BHV-recombinant according to claims 12-15, characterised in that It comprises another gene encoding an antigen from micro-organisms or viruses that are pathogenic for cattle.

17. Live attenuated BHV-recombinant according to claim 16, characterised in that the gene is chosen from the group of cattle pathogens, consisting of Bovine Rotavirus, Bovine Viral Diarrhoea virus, Parainfluenza type 3 virus, Bovine Paramyxovirus, Foot and Mouth Disease virus, Bovine Respiratory Syncytium virus and Pasteurella haemolytica.

18. Live attenuated BHV-recombinant, characterised in that it carries a heterologous gene fused to a class II membrane glycoprotein membrane anchor.

19. Live attenuated BHV-recombinant according to claim 18, characterised in that said class II membrane glycoprotein membrane anchor is the BRSV gG membrane anchor.

20. Live attenuated BHV-recombinant virus particle, characterised in that it carries a heterologous class II membrane glycoprotein.

21. Method for the preparation of live attenuated recombinant BHV particles carrying a heterologous protein incorporated in the particle envelope, said method comprising growing a live attenuated BHV-recombinant carrying a heterologous gene that encodes a class II membrane glycoprotein membrane anchor or is fused to a class II membrane glycoprotein membrane anchor.

22. Vaccine for the protection of cattle against virus infection, characterised in that said vaccine comprises a live attenuated BHV-recombinant according to claims 12-19 and/or comprises a live attenuated BHV-recombinant virus particle according to claim 20.

23. Vaccine according to claim 22, characterised in that it is administered intranasally.

24. Method for the preparation of a vaccine according to claim 22 or 23, characterised in that said method comprises the admixing of a live attenuated BHV-recombinant according to claims 12-19 and/or a live attenuated BHV-recombinant virus particle according to claim 20 and a pharmaceutically acceptable carrier.

25. Method for the preparation of a live attenuated BHV-recombinant according to claims 12-17, said method comprising bringing together in a suitable host cell isolated BHV-DNA and a vector comprising the synthetic BRSV-gene, placed under the control of a suitable promoter and flanked by 3' and 5' flanking regions that share homology with BHV- sequences.

26. Method for the preparation of a live attenuated BHV-recombinant according to claims 12-17, said method comprising bringing into a suitable host cell a vector comprising the synthetic BRSV-gene, placed under the control of a suitable promoter and flanked by 3' and 5' flanking regions that share homology with BHV-sequences, followed by infection of said suitable host cell with BHV.