WO2015025165A1 - T cell epitopes of classical swine fever virus - Google Patents

Abstract

The invention relates to peptides for use in the vaccination of animals against classical swine fever (CSF) and to the treatment thereof. It also encompasses nucleotides encoding such peptides, pharmaceutical compositions and methods of treatment or vaccination.

Description

T CELL EPITOPES OF CLASSICAL SWINE FEVER VIRUS

Field of the Invention

The invention relates to peptides for use in the vaccination of animals against classical swine fever (CSF) and to the treatment thereof.

Background to the Invention

Classical swine fever (CSF) is a severe and often lethal viral disease of domestic pigs and wild boars. The aetiological agent is the classical swine fever virus (CSFV), a small, enveloped, positive-sense, single-stranded RNA virus belonging to the pestivirus genus of the Flaviviridae family [1,2]. The disease is endemic in South East Asia, parts of Central and South America and the Russian Federation. Despite the stringent controls adopted in the EU, the virus continues to be an epizootic threat with recent outbreaks in Lithuania (2009 and 2011) and Latvia (2012) [3]. CSF is amenable to control by vaccination and live attenuated C-strain vaccines are highly efficacious. However, the inability to differentiate vaccinated animals from those infected with CSFV limits its utility as a control tool in outbreak settings in the EU [4]. Control of CSF outbreaks via a stamping-out policy is expensive, because large numbers of animals have to be culled including those slaughtered pre-emptively. Public resistance against such drastic measures is also growing. As a consequence, there is increased pressure to develop and adopt alternative strategies, like marker vaccines, to aid the control of CSF outbreaks [4].

C-strain vaccine induced IFN-γ responses have been correlated to rapid protection against the disease [5] and CSFV-specific IFN-γ secreting CD8 T cells are detected in the blood early after vaccination [6]. Determining the viral proteins that are the targets of the CD8 T cell response in immune animals would provide an important step towards developing a next generation marker vaccine capable of providing rapid protection against CSFV. CSFV has four structural proteins (the core protein and the envelope glycoproteins Erns, El and E2) and eight non-structural proteins (Npro, p7, NS2, NS3, NS4A, NS4B, NS5A, NS5B) [1]. E2 and

NS3 have been described as targets of the T cell response and both proteins induce IFN-γ release [6-9] and cytotoxic activity by T cells from vaccinated pigs [7-10]. A T cell epitope was identified on NS4 [9] and our group recently reported NS5B as a putative target of IFN-γ secreting T cells from C-strain vaccinated pigs [6]. Epitopes may be located on other viral proteins, since peptides pooled to represent Erns, El, NS2, NS4B and NS5A were able to induce PBMC proliferation in vaccinated pigs, but their ability to elicit an IFN-γ or cytotoxic response was not tested [9]. Most of these studies utilised inbred homozygous pigs, so were focussed on a single haplotype [7,9, 10] and the phenotype of the responding T cells/MHC restriction was not or only partially characterized [6-10].

Knowledge of epitopes within viral proteins that are targeted by CD8 T cell is also necessary to ensure that genetically attenuated or sub-unit DIVA vaccines include these regions. As the major target of neutralizing antibody responses, the structural protein E2 has been used to create subunit or chimeric vaccines [11,12]. Additional evidence that this protein is also able to target the cellular immune response comes from a recent study which showed that a DNA vaccine expressing E2 induced a cellular immune response, characterized by IFN-γ releasing T cells, before the appearance of neutralizing antibodies [11]. Moreover, a chimeric vaccine CP7_E2alf, where the E2 protein of the CSFV strain Alfort 187 is inserted in the backbone of the bovine viral diarrhoea virus (BVDV) strain CP7, can fully protect pigs from challenge before the appearance of neutralizing antibodies [12]. However, it remains to be determined whether CSFV E2 specific or BVDV cross-reactive antigen specific T cells are involved in mediating this protection.

The ability of subunit vaccines to trigger T cell responses that contribute to protection has been observed in other viruses belonging to the family Flaviviridae, such as hepatitis C virus (HCV) [13-15] and dengue virus (DENV) [16, 17]. Both a HCV peptide-vaccine, including 5 MHC-class I and 3 MHC class-II-restricted epitopes, a DNA vaccine encoding NS3 and NS4 and an adenovirus-based vaccine expressing NS3 of HCV induce a strong CD8 T cell response in vaccinated individuals [13-15]. DNA vaccines based on the NS3 protein from DENV and an adenovirus expressing DENV NS 1 induce a peptide-specific IFN-γ response by CD8 T cells from vaccinated mice which correlated with protection [16, 17].

With a view to better define the protein and epitope targets of the CD8 T cell response in pigs protected against CSF by vaccination with a live-attenuated C-strain vaccine; the inventors screened a peptide library spanning the CSFV proteome. Five T cell antigens and the corresponding antigenic regions/epitopes were identified. The conservation of these peptides across CSFV strains was assessed as was the restricting MHC class I haplotypes and the CSFV-specific T cells responding were phenotypically and functionally characterized. Summary of the Invention

According to the invention, there is provided an isolated peptide comprising an epitope having an amino acid sequence comprising at least 8, 9, 10 or 11 contiguous amino acids taken from one of the following sequences:

L SRVDN ALLKF ;

LISTVTGIFLI;

RYYEPRDSYFQ;

VEYSFIFLDEY;

PESRKKLEKALLAWA,

or a variant or functional fragment thereof.

As used herein, a "variant" means a peptide in which the amino acid sequence differs from the base sequence from which it is derived in that one or more amino acids within the sequence are substituted for other amino acids. The variant is a functional variant, in that the functional characteristics of the peptide from which the variant is derived are maintained. For example, a similar immune response is elicited by exposure of an animal, or a sample from an animal, to the variant polypeptide. In particular, any amino acid substitutions, additions or deletions preferably do not alter or significantly alter the tertiary structure of one or more epitopes contained within the peptide from which the variant is derived. The skilled person is readily able to determine appropriate functional variants and to determine the tertiary structure of an epitope and any alterations thereof, without the application of inventive skill.

Amino acid substitutions may be regarded as "conservative" where an amino acid is replaced with a different amino acid with broadly similar properties. Non-conservative substitutions are where amino acids are replaced with amino acids of a different type.

By "conservative substitution" is meant the substitution of an amino acid by another amino acid of the same class, in which the classes are defined as follows:

Class Amino acid examples

Nonpolar: Ala, Val, Leu, He, Pro, Met, Phe, Trp

Uncharged polar: Gly, Ser, Thr, Cys, Tyr, Asn, Gin

Acidic: Asp, Glu Basic: Lys, Arg, His.

For example, variants of the 1 lmer VEYSFIFLDEY which contain conservative substitutions include include VEYSTIFLDEY and VEYSFITLDEY.

As is well known to those skilled in the art, altering the primary structure of a peptide by a conservative substitution may not significantly alter the activity of that peptide because the side-chain of the amino acid which is inserted into the sequence may be able to form similar bonds and contacts as the side chain of the amino acid which has been substituted out. This is so even when the substitution is in a region which is critical in determining the peptide's conformation.

As mentioned above, non-conservative substitutions are possible provided that these do not disrupt the tertiary structure of an epitope within the peptide, for example, which do not interrupt the immunogenicity (for example, the antigenicity) of the peptide.

Broadly speaking, fewer non-conservative substitutions will be possible without altering the biological activity of the polypeptide. Suitably, variants may be at least 85% identical, 87.5% identical, for example at least 88% identical, at least 88.9% identical, at least 90% identical or at least 90.9% identical to the base sequence.

As used herein, the term "epitope" refers to the amino acids (typically 8-11 amino acids) within a peptide sequence which are essential in the generation of an immune response. The epitopes referred herein immune response are those which can be detected by means of a cell- mediated immunity (CMI) assay and are recognisable by a T cell by binding of a T cell receptor to the epitope presented by an MHC class I molecule.

The epitope may comprise consecutive amino acids, or the amino acids forming the epitope may be spaced apart from one another. In the latter case, the nature of the amino acids between the amino acids forming the epitope may not be crucial to the activity and may be varied, provided that the tertiary structure of the epitope is maintained, for example so that an immune response such as a cell-mediated immune response can occur in response to the presence of the epitope. Determination of the amino acids which form an epitope or part of an epitope can be undertaken using routine methods. For example, one of a series of small mutations such as point mutations may be made to a peptide and the mutated peptide assayed to determine whether the immunogenic or diagnostic activity has been retained. Where it has, then the variant retains the epitope. If activity has been lost, then the mutation has disrupted the epitope and so must be reversed.

The peptide or fragment may comprise or consist of the epitope. The peptide may, therefore, be the same length, shorter or longer than the epitope. For example, the peptide may be at least 8, 9, 10, 11, 12, 13, 14, 15, 18, 20, 25, 30 or 35 amino acids in length.

In one embodiment, the peptide or fragment is 8 amino acids in length. In that embodiment, the peptide or fragment may have substantial homology to one of the following sequences: STVTGIFL or YEPRDSYF. Alternatively, the peptide or fragment is longer than 8 amino acids in length, but comprises an amino acid sequence having substantial homology to or consisting of STVTGIFL or YEPRDSYF.

In another embodiment, the peptide or fragment is 9 amino acids in length. In that embodiment, the peptide or fragment may have substantial homology to the following sequence: RDNALLKF or to RVDNALLKF. Alternatively, the peptide or fragment is longer than 9 amino acids in length, but comprises an amino acid sequence having substantial homology to or consisting of RDNALLKF or to RVDNALLKF.

Alternatively, the peptide or fragment may comprise or consist of an amino acid sequence having substantial homology to any of the sequences shown in Table A.

Also provided is an isolated nucleic acid molecule encoding the peptide of the invention, a vector comprising such a nucleic acid molecule and a host cell comprising such a vector. The host cell may be a cell other than a human embryonic stem cell.

Using the standard genetic code, a nucleic acid encoding an epitope or peptide may readily be conceived and manufactured by the skilled person. The nucleic acid may be DNA or RNA, and where it is a DNA molecule, it may comprise a cDNA or genomic DNA. The invention encompasses fragments and variants of the isolated nucleic acid, where each such fragment or variant encodes a peptide with antigenic properties as defined herein. Fragments may suitably comprise at least 15, for example at least 30, or at least 60 consecutive bases from the basic sequence. The term "variant" in relation to a nucleic acid sequences means any substitution of, variation of, modification of, replacement of deletion of, or addition of one or more nucleic acid(s) from or to a polynucleotide sequence providing the resultant peptide sequence encoded by the polynucleotide exhibits at least the same properties as the peptide encoded by the basic sequence. In this context, the properties to be conserved are the ability to form one or more epitopes such that an immune response is generated which is equivalent to that of the diagnostic reagent peptide or isolated peptide as defined herein. The term, therefore, includes allelic variants and also includes a polynucleotide which substantially hybridises to the polynucleotide sequence of the present invention. Such hybridisation may occur at or between low and high stringency conditions. In general terms, low stringency conditions can be defined a hybridisation in which the washing step takes place in a 0.330-0.825M NaCl buffer solution at a temperature of about 40-48°C below the calculated or actual melting temperature (T_m) of the probe sequence (for example, about ambient laboratory temperature to about 55°C), while high stringency conditions involve a wash in a 0.0165-0.0330M NaCl buffer solution at a temperature of about 5-10°C below the calculated or actual T_m of the probe (for example, about 65°C). The buffer solution may, for example, be SSC buffer (0.15M NaCl and 0.015M tri-sodium citrate), with the low stringency wash taking place in 3 x SSC buffer and the high stringency wash taking place in 0.1 x SSC buffer. Steps involved in hybridisation of nucleic acid sequences have been described for example in Sambrook et al. (1989; Molecular Cloning, Cold Spring Harbor Laboratory Press, Cold Spring Harbor). Typically, variants have 60% or more of the nucleotides in common with the nucleic acid sequence of the present invention, more typically 65%, 70%, 80%, 85%, or even 90%, 95%, 98%) or 99% or greater sequence identity.

Nucleotide variants are provided in Table B, C, D, E and F..

Peptides may be prepared synthetically using conventional peptide synthesisers. Alternatively, they may be produced using recombinant DNA technology or isolated from natural sources followed by any chemical modification, if required. In these cases, a nucleic acid encoding the peptide is incorporated into suitable expression vector, which is then used to transform a suitable host cell, such as a prokaryotic cell such as E. coli. The transformed host cells are cultured and the peptide isolated therefrom. Vectors, cells and methods of this type form further aspects of the present invention. Sequence identity between nucleotide and amino acid sequences can be determined by comparing an alignment of the sequences. When an equivalent position in the compared sequences is occupied by the same amino acid or base, then the molecules are identical at that position. Scoring an alignment as a percentage of identity is a function of the number of identical amino acids or bases at positions shared by the compared sequences. When comparing sequences, optimal alignments may require gaps to be introduced into one or more of the sequences to take into consideration possible insertions and deletions in the sequences. Sequence comparison methods may employ gap penalties so that, for the same number of identical molecules in sequences being compared, a sequence alignment with as few gaps as possible, reflecting higher relatedness between the two compared sequences, will achieve a higher score than one with many gaps. Calculation of maximum percent identity involves the production of an optimal alignment, taking into consideration gap penalties.

Suitable computer programs for carrying out sequence comparisons are widely available in the commercial and public sector. Examples include the Gap program (Needleman & Wunsch, 1970, J. Mol. Biol. 48: 443-453) and the FASTA program (Altschul et al, 1990, J. Mol. Biol. 215: 403-410). Gap and FASTA are available as part of the Accelrys GCG Package Version 11.1 (Accelrys, Cambridge, UK), formerly known as the GCG Wisconsin Package. The FASTA program can alternatively be accessed publically from the European Bioinformatics Institute (http://www.ebi.ac.uk/fasta) and the University of Virginia (http://fasta.biotech.virginia.edu/fasta_www/cgi). FASTA may be used to search a sequence database with a given sequence or to compare two given sequences (see http://fasta.bioch.virginia.edu/fasta_www/cgi/search_frm2.cgi). Typically, default parameters set by the computer programs should be used when comparing sequences. The default parameters may change depending on the type and length of sequences being compared. A sequence comparison using the FASTA program may use default parameters of Ktup = 2, Scoring matrix = Blosum50, gap = -10 and ext = -2.

The peptide of the invention comprises a region that is antigenic, specifically a region that can provoke an antigenic or immune response to the CSFV. The peptide may comprise more than one antigenic region and in that case, the second antigenic region may also provoke an immune response to CSFV, or may stimulate a different immune response. When the peptide contains more than one region that is antigenic to CSFV, the regions may direct or provoke an immune response to the same or different proteins from CSFV. For example, one region may relate to one of the envelope glycoproteins and the other to a different envelope glycoproteins or with a non- structural protein. Alternatively, they may both relate to a non- structural protein.

Also provided is a pharmaceutical composition comprising a peptide or a nucleotide according to the invention.

The pharmaceutical composition may comprise one or more pharmaceutically or otherwise biologically active agents in addition to the peptide of the invention. For example, the composition may include another immunogenic agent or a therapeutic agent.

The pharmaceutical composition may include any appropriate carriers, adjuvants etc. In particular embodiments, the pharmaceutical composition may contain the peptide in synthetic or recombinant form, along with one or more adjuvants. In other embodiments, the peptide may be delivered in a delivery system, such as a nanoparticle, or other particle in which the peptide may be encapsulated or otherwise delivered. When the pharmaceutical composition comprises a nucleotide it may be in the form of, for example, plasmid DNA, a vector, such as a viral vector or a combination of vectors

The pharmaceutical composition may, for example, be a vaccine. When it is a vaccine, it may be for administration by any appropriate route, such as orally or by injection, especially subcutaneous or intra muscular injection.

Also provided is the peptide of the invention, or a fragment thereof, for use in therapy. In particular, the therapy may be the treatment of or vaccination against classical swine fever.

The invention further provides a method of vaccinating a subject against or treating a subject having classical swine fever comprising administering a peptide or fragment thereof or pharmaceutical composition of the invention to the subject.

The subjects may be any subject that may be infected with CSFV. In particular, it may be swine. The invention will now be described in detail, by way of example only, with reference to the figures.

Brief description of the drawings Figure Legends

Figure 1. Vaccination with C-strain and challenge with virulent CSFV induces predominately a virus-specific IFN-γ CD8 T cell response. Pigs were vaccinated on day -5 and then challenged with CSFV Brescia strain on days 0 and 28 post-challenge. Seven days after the second challenge, PBMC from vaccinated/challenged (V/C) and control animals stimulated with CSFV and IFN-γ release measured by ELISpot assay. Panel A shows the mean mock-virus stimulated corrected IFN-γ spot-forming cells (SFC)/5xl0⁵ PBMC for each group and error bars represent SEM. Panel B shows the gating strategy used to interrogate responses in singlet, live CD8 T cells and memory CD4 T cells. Plots in panel C shows the mean mock-virus stimulated corrected % IFN-γ expressing CD4 T cells and CD8 T cells +/- SEM for 8 V/C animals and 3 control animals. Values of control and V/C groups were compared using a two-tailed un-paired t-test and significance is indicated by **p<0.01, *p<0.05.

Figure 2. CD8 T cells from pigs vaccinated with C-strain and challenged with virulent CSFV display distinct profiles of antigen reactivity. Fourteen days after re-challenge with CSFV Brescia strain, PBMC from four selected pigs were stimulated with synthetic peptide pools representing the 12 CSFV proteins and IFN-γ expression was assessed by flow cytometry. Graphs show the mean unstimulated corrected %∑FN-y⁺ CD8 T cells for each animal in response to peptide pools +/- SEM. Statistical analyses were performed using a one-way ANOVA followed by a Dunnett's multiple comparison test versus the unstimulated cells; ***p<0.001, **p<0.01.

Figure 3. Identification of putative antigenic peptides recognised by CSFV specific CD8 T cells. Twenty one days after re-challenge, PBMC from pigs AN5, AN7, AN11 and AN13 were stimulated with synthetic peptides pooled in a 2-way matrix for proteins E2 (pig AN11), NS2 (pig AN7), NS3 (pigs AN5 and AN11) and NS5A (pigs AN5 and AN13). Peptides representing the core protein were screened individually (pig AN13). IFN-γ expression by CD8 T cells was assessed by flow cytometry as described above. The mean unstimulated corrected % IFN-γ expressing CD8 T cells are presented and error bars represent SEM. Statistical analyses were performed using a one-way ANOVA followed by a Dunnett's multiple comparison test versus unstimulated cells; ***p<0.001, **p<0.01, *p<0.05.

Figure 4. Identification of minimal length antigenic peptides recognised by CSFV specific CD8 T cells. PBMC from pigs AN5, AN7, AN11 and AN13, collected at day 28 or cryopreserved at later time-points were stimulated with the identified 15mer or consensus l lmer antigenic peptides and the truncated derivatives of NS5A LSRVDNALLKF, NS2 LISTVTGIFLI and E2 RYYEPRDSYFQ and IFN-γ expression by CD 8 T cells assessed by flow cytometry. Panels A, B and C shows the mean unstimulated corrected % IFN-γ expressing CD8 T cells +/- SEM. Statistical analyses were performed using a one-way ANOVA followed by a Dunnett's multiple comparison test versus the previously identified l lmer antigenic peptide; ***p<0.001, **p<0.01, *p<0.05. Panels D-H show reactivity against a logio dilution series of the identified minimal length antigenic peptides (NS5A RVDNALLKF, NS2 STVTGIFL and E2 YEPRDSYF) or antigenic regions (core PESRKKLEKALLAWA and NS3 VEYSFIFLDEY).

Figure 5. Recognition of the identified epitopes by CD8 T cells from C-strain vaccinated pigs challenged with divergent CSFV strains. PBMC from pigs vaccinated with C-strain CSFV and challenged with CSFV isolate UK2000/7.1 (A) or CBR/93 (B), collected 9 or 12 days post-challenge, were stimulated with antigenic peptides NS2 STVTGIFL and NS3 VEYSFIFLDEY. IFN-γ expression by CD8 T cells assessed by flow cytometry and graphs show the mean unstimulated corrected % IFN-γ expressing CD8 T cells +/- SEM. Statistical analyses were performed using a one-way ANOVA followed by a Dunnett's multiple comparison test versus the un-stimulated control; **p<0.01, *p<0.05.

Figure 6. Phenotypic characterization of CSFV epitope-specific CD8 T cells.

Cryopreserved PBMC from pigs AN5, AN7, ANl l and AN13 were stimulated with the identified antigenic peptides and the phenotype of IFN-γ expressing CD8 T cells determined by flow cytometry. Histograms shows representative dot plots of the expression of IFN-γ versus CD 107a, CD25 and CD27 by singlet, live CD 8 T cells. The table reports the mean % expression of these markers by peptide-specific∑FN-y⁺ CD8 T cells from triplicate cultures +/- SEM.

Figure 7. Characterization of polyfunctional cytokine expression by CSFV epitope- specific CD8 T cells. Cryopreserved PBMC from pigs AN5, AN7, ANl l and AN13 were stimulated with the identified antigenic peptides and the expression of IFN-γ, TNF-a and IL- 2 by CD8 T cells was simultaneously assessed by flow cytometry. Panel A shows representative dot plots of the expression of TNF-a and IL-2 by peptide specific IFN-γ expressing CD8 T cells and Panel B the relative proportions of IFN-γ secreting CD8 T cells expressing either of the two other cytokines. In panel C, the mean fluorescence intensity (MFI) of IFN-γ staining in each of the populations is presented. Values represent the mean values for triplicate cultures +/- SEM. Statistical analyses were performed using a one-way ANOVA followed by a Dunnett's multiple comparison test against IFN-γ only secreting CD8 T cells; ***p<0.001, **p<0.01, *p<0.05.

Detailed description of the invention

The invention will now be described in detail by way of example only.

1. Materials and Methods

Viruses

A commercial live attenuated C-strain CSFV (genotype 1.1, AC Riemser^® Schweinepestvakzine, Riemser Arzneimittel AG, Riems, Germany), the highly virulent Alfort 187 (genotype 1.1, AHVLA CSF reference laboratory virus archive, Addlestone, UK) and Brescia (genotype 1.1, kindly provided by Dr Alexandra Meindl-Bohmer, University of Veterinary Medicine, Hannover, Germany) strains, the moderately virulent UK2000/7.1 (a genotype 2.1 isolate from the UK [18]) and CBR/93 (a genotype 3.3 isolate kindly provided by S. Parchariyanon [19]) strains of CSFV were propagated in porcine kidney (PK15) cells maintained in Eagle's Minimum Essential Medium (E-MEM) (Life Technologies, Paisley, UK) with 10% FBS (Autogen Bioclear, Calne, UK) and antibiotics (lOOU/ml penicillin, lOOmg/ml streptomycin, both from Life Technologies). CSFV was harvested from cultures after 4 days and TCID₅₀ titres were determined according to standard protocols [20]. Mock virus supernatants were prepared in an identical manner from uninfected PK15 cells. Vaccination with CSFV

Experiment 1 : An experimental vaccination/challenge study was performed to assess the specificity of CSFV-specific CD8 T cell IFN-γ responses. Eleven Large White/Landrace pigs, 6 months of age, were utilized; eight animals were vaccinated 5 days before challenge (day -5) by intramuscular inoculation of 10⁵ TCID₅₀ of C-strain CSFV and challenged on day 0 and 28 by intranasal inoculation of 10⁵ and 10⁶ TCID₅₀ of CSFV Brescia, respectively (1 ml divided equally between each nostril and administered using a mucosal atomization device MAD-300, Wolfe Tory Medical, Salt Lake City, USA). Three negative control pigs received similar inoculations of mock virus supernatant on each occasion. Virus back titrations of the inocula confirmed the vaccine and challenge doses were those expected.

Experiment 2: A second vaccination/challenge study assessed recognition of identified T cell epitopes by additional C-strain vaccinated pigs. Animals vaccinated and challenged in two independent, previously described, experiments [5] were utilised. Large White/Landrace cross male pigs, 9 weeks of age, were vaccinated intramuscularly with 2 ml of reconstituted C-strain vaccine, as described by the manufacturer (Riemser Arzneimittel AG) and after 3 or 5 days were challenged by intranasal inoculation of 10⁵ TCID₅₀ CSFV UK2000/7.1 or CBR/93 strains as described above.

Clinical and haematological methods

Temperatures and clinical scoring were monitored for 7 days before and after each of the CSFV Brescia inoculations for Experiment 1 and from day -5 to 15 days post-challenge for Experiment 2 [5]. Nine parameters relevant for indication of CSF (liveliness, body tension, body shape, breathing, walking, skin, eye/conjunctiva, appetite/leftover at feedings, defecation) were examined and scored as 0 (normal), 1 (slightly altered), 2 (distinct clinical sign) or 3 (CSF symptoms). A total clinical score for each animal was assigned twice daily and temperatures were monitored by daily rectal thermometer readings [21]. Peripheral blood leukocyte counts were monitored by volumetric flow cytometry during both experiments [21]. In brief, 50μ1 EDTA blood was incubated with 5μ1 of anti-porcine CD45-FITC monoclonal antibody (mAb) (O. lmg/ml, K252-1E4, AbD Serotec, Oxford, UK) for 10 minutes at room temperature (RT) in the dark. FACS Lysing solution (945μ1) (BD Biosciences, Oxford, UK) was added for 10 minutes at RT to lyse erythrocytes and fix leukocytes. Cell counts were obtained on a volumetric flow cytometer (MACSQuant Analyzer, Miltenyi Biotec, Bisley, UK) by gating FITC positive events and leukocyte counts per μΐ/blood were obtained by multiplying the leukocyte density by the dilution factor (x 20).

Purification and cryopreservation of peripheral blood mononuclear cells (PBMC)

Heparinised venous blood was collected on days -5, 0 and every 7 days until day 56 post- challenge for Experiment 1 and every three days from day -5 to 15 post-challenge for Experiment 2. PBMC were prepared by diluting 20ml of blood in 10 ml of PBS (Life Technologies), layering over 20ml of Histopaque-1077 (Sigma- Aldrich, Poole, UK) and centrifuged at 1455 x g for 30 minutes at room temperature (RT), without braking, in a rotating bucket centrifuge. PBMC were aspirated from the plasma- Histopaque-1077 interface and washed three times in PBS by centrifugation at 930 x g for 5 minutes at 4°C. PBMC were re-suspended in RPMI-1640 medium (Life Technologies) supplemented with 10% FBS, lOOU/ml penicillin and lOOmg/ml streptomycin (cRPMI) and cell densities determined using a volumetric flow cytometer (Miltenyi Biotec) and gating on events with typical forward scatter (FSC) and side scatter (SSC) for PBMC. Cells were used directly for phenotypic/functional analysis or cryopreserved for subsequent analysis.

For cryopreservation, PBMC were adjusted to a density l-2xl0⁷ cells/ml, re-suspended in cold (4°C) 10% DMSO (Sigma- Aldrich) in FBS and transferred to pre-cooled (4°C) labeled cryotubes. These tubes were immediately transferred to pre-cooled (4°C) Cryo 1°C Freezing Container (Nalgene, Fisher Scientific, Loughborough, UK) pre-filled with 250 ml of 100% isopropyl alcohol, which was placed in a -80°C freezer for a minimum of 4 hours to a maximum of 24 hours. Cryotubes were then transferred to a liquid nitrogen storage container.

Synthetic CSFV peptides

A synthetic overlapping peptide library was designed which comprised pentadecamer peptides off-set by four residues. The peptide sequences were designed using the predicted polyprotein of CSFV C-strain Riems (GenBank accession number AY259122.1). The synthesized library of 945 peptides (JPT Peptide Technologies, Berlin, Germany) were reconstituted in sterile lOmM HEPES (Life Technologies) buffered 40% acetonitrile (Sigma- Aldrich) at a concentration of 2mg/ml. For initial screening, peptides were combined into pools representing the structural and non- structural proteins of CSFV and diluted in cRPMI and used at a final total peptide concentration of 1 μg/ml, unless otherwise stated. In order to identify the antigenic peptides from positive pools, a two-way matrix system was adopted to screen peptides representing the non- structural proteins: NS5A, NS3, NS2 and the structural protein E2. Matrix pools were designed so that each peptide was uniquely present in 2 different pools. The 121 NS5A peptides were prepared in 22 peptide pools (A-V), the 110 NS2 peptides in 21 peptide pools (A-U), the 168 NS3 peptides 26 peptide pools (A-Z) and the 90 E2 peptides in 19 matrix peptide pools (A-S). After identification of the antigenic pentadecamers, truncated 8-l lmers of consensus antigenic sequences were designed, synthesised (JPT Peptide Technologies), reconstituted and prepared as described above.

Porcine IFN-γ ELISpot assay

ELISpot plates (96 well Multiscreen-IP Filter Plates; Millipore, Watford, UK) were prepared by pre-wetting each well with 15ml of 35% ethanol for 1 min then washing 3 times with sterile PBS. The capture antibody (anti-porcine IFN-γ mAb, P2G10, BD Biosciences, Oxford, UK), prepared at 0.5mg/ml in PBS, was added at 50ml/well and the plates incubated at 4°C overnight. Capture antibody was then decanted and plates washed 3 times with unsupplemented RPMI-1640 medium. Plates were blocked by addition of lOOml/well cRPMI and incubation for at least lhr at 37°C. Freshly isolated PBMC were suspended at 5xl0⁶/ml in cRPMI and ΙΟΟμΙ of cells was added to each well. CSFV Alfort 187 strain was diluted in cRPMI and added at a multiplicity of infection (MOI) of 1. Concanavalin A (Sigma- Aldrich) at lOmg/ml and mock-infected PK-15 cell cryolysate supernatant were used as positive and negative controls, respectively. All conditions were tested in triplicate and plates were incubated at 37°C in a 5% C0₂ humidified atmosphere for 18 hours. Well contents were discarded and 100ml of cold water was added to each well and incubated for 5 min. Wells were then washed 5 times with PBS containing 0.05% Tween 20 (ELISpot Wash Buffer). Biotinylated anti-porcine IFN-γ mAb (P2C11, BD Biosciences, Oxford, UK), diluted to 0.167mg/ml in PBS, 0.05% Tween 20, 1% FBS was added (50ml/well) and plates incubated at 4°C overnight. Plates were washed 3 times with ELISpot Wash Buffer and incubated with streptavidin-HRP (R&D Systems, Abingdon, UK; 0.5mg/ml, 50ml/well) for lhr at 37°C. After plates were washed 5 times, BCIP/NBT substrate (R&D Systems, Abingdon, UK, ΙΟΟμΙ/well) was added and plates incubated at RT in the dark until spots became visible, typically 15 - 60 min. The substrate was then discarded and the plates washed extensively with water and left to dry in the dark. Spots were visualized using an automated ELISpot reader (Autolmmun Diagnostika, Strassberg, Germany). Multi-parameter cytofluorometric analysis of PBMC responses

Both freshly isolated and cryopreserved PBMC were used in this study. To resuscitate cryopreserved cells, cryovials were rapidly thawed in a 37°C water bath and cells transferred to tubes containing 10ml of pre-warmed (37°C) cRPMI. Cells were washed by centrifugation, 930 x g for 5 minutes at RT, and resuspended in fresh warm cRPMI. Cell densities were calculated using flow cytometry as described above and adjusted to 1 x 10⁶ cells/ml and ΙΟΟμΙ transferred to wells of a 96-well round bottom microtitre plate (Costar, Fisher Scientific). PBMC were stimulated with ΙΟΟμΙ of CSFV Alfort -187 strain, at a MOI=l, or with ΙΟΟμΙ of peptide pools or individual peptides at 1 mg/ml, unless otherwise stated. Mock- virus PK15 supernatants/cRPMI medium and ConA (10μg/ml) were used as negative and positive controls, respectively. Cells were incubated at 37°C for 2 (peptide-stimulation) or 14 (virus-stimulation) hours and then brefeldin A (GolgiPlug, BD Biosciences) was added (0.2μ1Λνε11) and cells were further incubated for a further 16-18 or 6 hours following stimulation with peptide or CSFV, respectively. To detect cytotoxic degranulation, CD107a- Alexa Fluor 647 or IgGl isotype control-Alexa Fluor 647 mAbs (both AbDSerotec, Oxford, UK; ΙΟμΙ/well) and monensin (Golgi Stop, BD Biosciences; 0.2 μΐ/well) were added in conjunction with Brefeldin A.

Cells were washed in Dulbecco's PBS without Mg²⁺ and Ca²⁺ (DPBS; Life Technologies) and stained with Near Infra-Red Fixable Live/Dead Viability Dye (Life Technologies) for 30 minutes at 4°C. Cells were washed twice with DPBS supplemented with 2% FBS and 0.09% sodium azide (FACS buffer) and stained with mAbs specific for surface markers for lOmin at RT: CD8a-PE (76-2-11, BD Biosciences) and CD4-PerCP-Cy5.5 (74-12-4, BD Biosciences), CD25 (K231.3B2, AbD Serotec) and CD27 (b30c7, kindly provided by Dr Wilhelm Gerner, University of Veterinary Medicine, Vienna, Austria [22]). Cells were washed twice with FACS buffer and staining with CD25 and CD27 was visualized by incubation of cells with APC-conjugated rat anti-mouse IgGl (BD Biosciences) for lOmin at RT and then washed twice with FACS buffer. Surface stained cells were fixed and permeabilised using CytoFix/CytoPerm Solution (BD Bioscience) for 20min at 4°C. After two washes in BD Perm/Wash Buffer (BD Biosciences), PBMC were incubated with cytokine specific mAbs at RT for 10 minutes in the dark. mAbs used for cytokine staining were: IFN-y-FITC or -Alexa Fluor 647 (CC302, AbD Serotec), TNF-a-Pacific Blue (MAbl l, Biolegend, Cambridge Bioscience, Cambridge, UK) and IL-2 (A150D 3F1, Life Technologies), labelled using Zenon Alexa Fluor 647 mouse IgG2a labelling kit (Life Technologies). IgGl-FITC or -Alexa Fluor 647 isotype control mAbs were used to control staining with IFN-γ mAbs. Un-stained cells were used as control for IL-2 and T F-a. The cells were given two final washes in BD Perm/Wash buffer and re-suspended in FACS buffer prior to flow cytometric analysis on a MACSQuant Analyzer (Miltenyi Biotec) or CyAn ADP (Beckman Coulter, High Wycombe, UK) flow cytometers. Cells were analyzed by exclusion of doublets, followed by gating on viable cells (Live/Dead Fixable Dead Cell Stain negative) in the lymphocyte population, then defined lymphocyte subpopulations were then gated upon and their expression of cytokines assessed. Gates were set using the corresponding isotype/unstained controls and values were corrected by subtraction of the % positive events in the biological negative control (cRPMI or mock-virus supernatant stimulated). The number of singlet live lymphocytes acquired for analysis was approximately 400,000.

Sequence analysis of T cell antigenic peptides/epitopes

The sequences of the identified T cell antigenic regions/epitopes were aligned against the predicted full-length polyprotein sequences of 14 CSFV isolates (GenBank accession numbers shown in parentheses): Genotype 1.1 - Brescia (AF091661), Alfort/187 (X87939.1), KC Vaccine (AF099102), ALD (D49532), GPE^" (D49533), Alfort-A19 (U90951), cF114 (AF333000), Shimen (AF092448), Koslov (HM237795) and SWH (DQ 127910); Genotype 2.1 - Penevezys (HQ148063); Genotype 2.3 - Borken (GU233731) and Alfort Tubingen (AAA43844); the BVDV reference strain NADL (AJ133738) and the border disease virus (BDV) reference strain BD31 (U70263), using the clustal W algorithm on MegAlign (DNAStar Lasergene 9 Core Suite, Madison, WI, USA).

The sequences of the identified T cell antigenic regions/epitopes were also aligned against the corresponding sequences of the CSFV isolates UK2000/7.1 and CBR/93. These were generated using RNA from CSFV strains UK2000/7.1 and CBR/93 and creating cDNA by reverse transcription as previously described [21]. A 5μ1 aliquot of cDNA was used as template for PCR amplification with high fidelity Platinum Taq in a reaction mix containing: Ι μΐ dNTP (lOmM of each dNTP, Promega, Southampton, UK), 5μ1 5x PCR buffer, 1.5μ1 MgSC-4 (50 mM), 0.2μ1 Platinum Taq (5U/ml) (all Life Technologies) and 2μ1 (20μΜ) of each primer (Sigma- Aldrich). The following primers (5 '-3') were used to amplify the selected regions: Core-F-954 AGAGCATGAGAAGGACAGYA and Core-R-1211 GTGCCRTTGTCACTYAGGTT, E2-F-2397 GTGCAAGGTGTGRTATGGC and E2-R- 3613 GTGTGGGTRATTAAGTTCCCTA, NS2-F-3840 TAGTAGTCGYYGTGATGTTR and NS2-R-4248 GCCCACATCGTAAAMACCA, NS3-F-5974

AGGATAGGKGAGATGAAGG and NS3-R-6332 TTCATCTCCTCTACTGGTATC, NS5A-F-9278 ACTATGACTACAGGGGAG and NS5A-R-9857

GATATCTTTGTGGAGTCTGT. The following thermal profile was adopted: 94°C for 2 minutes, 40 x (94°C 30 seconds; 53°C 30 seconds for E2/ 49°C 30 seconds for core and NS3/ 46°C 30 seconds for NS2 and NS5A; 72°C 60 seconds); 72°C 1 minute: 4°C. Sequencing reactions using purified amplicons were set up in both directions using standard Sanger sequencing (CSU-AHVLA, UK) and the amino acid sequences deduced from the assembled amplicon sequences were aligned with the matching antigenic peptide.

MHC class I haplotype determination by low-resolution (Lr) PCR-based analysis

Pigs used in this study were genotyped for their swine leukocyte antigen (SLA) class I haplotypes by low-resolution PCR screening assays (PCR-SSP) on PBMC-derived genomic DNA as previously described [23].

Data analysis and statistics

Graphical and statistical analysis was performed using GraphPad Prism 5.04 (GraphPad Software Inc, La Jolla, USA). Data was represented as means with standard error of means (SEM) quoted to indicate the uncertainty around the estimate of the group mean. A two-tailed unpaired t-test or a one-way-analysis-of-variance (ANOVA) followed by a Dunnett's multiple comparison test was used and a p-value < 0.05 was considered statistically significant.

Results

Detection of CSFV-specific CD8 T cell responses following vaccination and challenge

Animals vaccinated in Experiment 1 were solidly protected against both challenge infections, at 5 and a further 28 days post vaccination, with no evidence of leukopenia or clinical signs of the disease being observed (data not shown). One week after the second challenge inoculation a significant IFN-γ response following in vitro stimulation with CSFV was observed in PBMC from vaccinated/challenged (V/C) animals, but not control pigs (Fig. 1 A). Using the flow cytometric gating strategy displayed in Fig. IB, CSFV-specific IFN-γ CD8 T cells (CD4^"CD8a^high) and 'memory' CD4 T cells (CD4⁺CD8a^low) were detected in PBMC from V/C animals, but not control pigs, with the former population dominating the IFN-D response (Fig. 1C). From the 8 V/C pigs, the greatest CD8 T cell response was identified in pigs AN5, AN7, AN11 and AN13, and these animals were selected for an in-depth analysis of the specificity of their CD8 T cell responses.

Screening of a CSFV proteome-wide peptide library to identify CD8 T cell antigens

To identify CSFV T cell antigens, PBMC from the four selected V/C pigs were stimulated in vitro with pools of overlapping 15mer peptides representing the 12 CSFV proteins and CD8 T cell reactivity was screened using IFN-γ detection by flow cytometry (Fig. 2). Significant IFN-γ responses were observed in all V/C animals and each animal reacted against a unique profile of antigens. Pig AN5 mounted a significant IFN-γ response against NS3 and NS5A peptides, pig AN7 reacted significantly against the NS2 peptide pool, pig AN11 responded to peptides representing the E2 and NS3 proteins, and pig AN13 mounted the greatest response to the core peptides with significant reactivity also observed against NS5A.

Identification of CD8 T cell epitopes on CSFV

To identify the peptide targets of these CD8 T cell IFN-γ responses, a two way matrix system was adopted to screen peptides representing the non- structural proteins NS5A, NS3, NS2 and the structural protein E2, whereas the peptides spanning the core protein were screened individually. The matrix pools were designed so that each peptide was uniquely present in 2 defined pools. PBMC from pigs AN5 and AN13 were stimulated in vitro with the NS5A matrix peptide pools and∑FN-y⁺ CD8 T cells were identified using flow cytometry. Pools C, K, N and Q induced a significant IFN-γ response in CD8 T cells from pig AN5 (Fig. 3). By analysing the peptide constituents of the reacting matrix pools, four potential antigenic 15mers were identified for further testing: peptides #25, 33, 58 and 66. Pig AN13 did not mount a significant CD8 T cell IFN-γ response to any of the NS5A matrix pools. PBMC from pig AN7 were stimulated with the NS2 matrix peptide pools and a CD8 T cell IFN-γ response was observed against pools B, I, O, P (Fig. 3) identifying 4 putative antigenic peptides: #35 42, 46 and 53. PBMC from pigs AN5 and AN11 were stimulated in vitro with the NS3 matrix peptide pools and peptides present in the pools D, F, G, X, Y and Z induced a statistically significant CD8 T cell IFN-γ response, which indicated that peptides #134, 136, 137, 147, 149, 150, 160, 162 and 163 were potentially antigenic. Pig AN5 did not mount a significant response to the NS3 matrix pools. E2 matrix peptide pools were screened with PBMC from pig AN11 leading to the identification of significant reactivity against pools E, G, O and P; suggesting reactivity against peptides 45, 47, 55 and 57. The identified putative antigenic peptides were next screened individually to assess their recognition by CD8 T cells. PBMC from pig AN5 were found to show significant reactivity to overlapping NS5A peptides 33 and 58, but not 25 and 66, suggesting that at least one epitope lay in the l lmer consensus region LSRVDNALLKF. Of the four possible antigenic NS2 peptides, the inventors observed that the overlapping peptides 42 and 46, with the consensus sequence LISTVTGIFLI, but not 35 and 53, induced a statistically significant greater number of IFN-γ expressing CD8 T cells from pig AN7 compared to un-stimulated controls. PBMC from pig ANl l showed significant reactivity against two pairs of overlapping peptides, E2 peptides 47 and 55 and NS3 peptides 134 and 163 with l lmer consensus sequences of RYYEPRDSYFQ and VEYSFIFLDEY, respectively. Due to its short length, peptides spanning the core protein were screened individually using PBMC from pig AN13. Significant CD8 T cell IFN-γ responses were observed against the 15mer peptide #20 (PE SRKKLEK ALL AW A) (Fig. 3).

The length of CD8 T cell epitopes can vary between 8 and 11 amino acids. In order to identify the minimal length antigenic peptides and define the natural epitopes, we synthesised the two possible lOmer, three 9mer and four 8mer sequences for each consensus l lmer sequence of overlapping peptide pairs (NS5A-LSRVDNALLKF NS2-LISTVTGIFLI, E2- RYYEPRDSYFQ). Due to a limitation in the numbers of PBMC cryopreserved from pigs AN13 and ANl l, we were unable to identify the minimal length antigenic peptides for NS3 i902-i9i2 VEYSFIFLDEY (pig AN11) and core₂₄i-255 PESRKKLEKALLAWA (pig AN13) regions. Following stimulation of PBMC from pig AN5 with the NS5A l lmer LSRVDNALLKF and truncated derivatives (Fig. 4 A), one lOmer, two 9mers and the four 8mers induced a statistically significant lower number of IFN-γ producing CD8 T cells than the l lmer, suggesting that the 9mer RVDNALLKF was the minimal length antigenic peptide. Similar analysis with pig AN7 (Fig. 4B) identified the minimal length antigenic peptide on NS2 to be the 8mer STVTGIFL. A 8mer, YEPRDSYF, located on E2, was identified as being the shortest peptide recognised by CD8 T cells from pig ANl l (Fig. 4C). Responses to all antigenic peptides were all shown to be dose-dependent and the threshold concentration at which responses became undetectable was proportional to the frequency of the peptide specific T cell population (Fig. 4D-H).

The sequences of CD8 T cell epitopes are well conserved among CSFV strains Using the Clustal W protein alignment tool, the inventors investigated the conservation of the identified CD8 T cell epitopes/antigenic regions among different CSFV strains (Table 1). We observed that the antigenic region PESRKKLEKALLAWA, located on core protein, showed only one amino acid substitution in the genotype 3.3 strain CBR/93. The NS3 l lmer VEYSFIFLDEY displayed one amino acid substitution in the genotype 2.1 strain Penevezys strain and in CBR/93. The E2 epitope YEPRDSYF was 100% conserved across all the CSFV strain analysed. The NS2 epitope STVTGIFL was conserved in all the CSFV strains analysed except the Penevezys strain, where there was a single amino acid substitution. The NS5A epitope RVDNALLKF was conserved among all genotype 1.1 strains tested except Shimen, where a single amino acid substitution was observed. A different single amino acid substitution was observed in the genotype 2 strains and 2-3 substitutions were observed in the two genotype 3 strains. When compared to the reference BVDV and BDV strains, the antigenic regions/epitopes on core, NS2 or E2 showed no or single amino acid substitutions 2-3 substitutions were observed in the NS2 epitope and in the NS5A epitope was poorly conserved with only three amino acids being shared.

Differences in CD8 T response specificity is associated with distinct MHC class I haplotype expression

The inventors investigated whether the different specificity of CD8 T cell responses between pigs was due to them bearing different MHC class I haplotypes. The porcine swine leukocyte antigen (SLA) class I (SLA-1, SLA-2 and SLA-3) haplotypes of the four pigs were determined using a PCR-SSP-based typing assay. Each animal was heterozygous and no two haplotypes were shared between these animals (Table 2). To assess the recognition of the identified T cell epitopes and antigenic peptides by a larger number of C-strain vaccinated pigs, PBMC collected during Experiment 2 were assayed. Pigs were vaccinated with the C- strain and challenged after 5 or 3 days with the virulent CSFV strain UK2000/7.1 (n=9) or CBR/93 (n=7). These animals were solidly protected against challenge with no evidence of leukopenia, viraemia or clinical signs of the disease being observed (as described previously, [5]). Four C-strain vaccinated/UK2000/7.1 challenged pigs (AD53, AD56, AD62 and AD65) and two C-strain vaccinated/CBR/93 challenged pigs (AE15 and AE17) reacted against the NS2 8mer STVTGIFL, with a significantly greater number of IFN-D expressing CD8 T cells compared to unstimulated cells (Fig. 5). A significant CD8 T cell IFN-γ response against the NS3 l lmer VEYSFIFLDEY was detected in three C-strain vaccinated/UK2000/7.1 challenged pigs (AD57, AD64 and AD66) and three C-strain vaccinated/CBR/93 challenged pig (AE4, AE8 and AE16). No responses were observed in these animals against the core, E2 or NS5A peptides (data not shown). To assess whether recognition of the two antigenic peptides was associated with the previously identified MHC class I haplotypes, all animals were typed for SLA class I using the PCR-SSP-based assay (Table 2). No haplotypes were found to be shared between these pigs and the animals used to define the antigenic peptides, but all pigs that reacted against NS2 8mer STVTGIFL displayed the haplotype Lr-22.0 and pigs that reacted against the NS3 l lmer VEYSFIFLDEY carried the Lr-01.0 haplotype. Interestingly three NS3 reactor animals also expressed Lr22.0 but no response was detected against NS2. Since the inventors had shown that the NS3 l lmer VEYSFIFLDEY had a single amino acid substitution in the CBR/93 sequence (a tyrosine at position 4), we assayed responses to both peptides and found they stimulated comparable CD8 T cell IFN-γ responses in SLA-I Lr-01.0 pigs challenged with CSFV CBR/93 (data not shown).

Functional and phenotypic characterization of CSFV peptide-specific CD8 T cells

The ability of the five identified antigenic peptides to elicit cytotoxic activity was investigated by assessment of surface mobilisation of CD 107a, a marker of degranulation, by IFN-Y⁺ CD 8 T cells. Over 90% of IFN-y⁺ CD 8 T cells expressed CD 107a on their surface after peptide stimulation, suggesting cytotoxic activity of these cells against the peptide- presenting cells (Fig. 6). With the aim to further characterize the epitope-specific CD8 T cell populations, the expression of the activation markers CD25 and CD27 on IFN-y⁺ CD8 T cells were investigated using flow cytometry. Interestingly, the majority of peptide-specific IFN-Y⁺ CD8 were CD25^" and CD27⁺ (Fig. 6). However, differences were observed in levels of CD27 expression between animals/antigenic peptides with the majority of CD8 T cells responding to NS5-RDNALLKF expressing high levels of CD27 whereas the others showed the largest proportion of IFN-y⁺ CD8 T cells in the CD27^low population (Fig. 6). We finally analysed the ability of CD8 T cells to produce other cytokines in response to peptide stimulation. The ability of IFN- D⁺ CD8 T cells to express IL-2 and TNF-a after peptide stimulation was assessed, using the gating strategy illustrated in Fig. 7A. While variability was observed in responses between animals/antigenic peptides, the majority of peptide specific CD8 T cells expressed either IFN-γ alone or IFN-γ and TNF-a (Fig. 7B). Only a small proportion expressed IFN-γ and IL-2 or all three cytokines, with the largest populations being observed in response to the E2 and core peptides. Interestingly, the cells co-expressing TNF-a or TNF- a/IL-2 expressed incrementally greater amounts of IFN-γ compared to cells which expressed IFN-γ alone or IFN-γ and IL2 (Fig. 7C), indicative of a greater 'quality' of cytokine response.

Discussion

Despite presenting the porcine immune system with a polyprotein of almost 4000 amino acids in length, this study has shown that the CD8 T cell response to CSFV is highly focussed, dominated by only one or two epitopes each located on five different viral antigens (core, E2, NS2, NS3 and NS5A). Moreover, the specificity of the response varied between animals dependent upon their MHC class I haplotype, which present a high degree of polymorphism in pigs [24]. The antigenic sequences identified in this study showed no overlap with those previously mapped [9, 10]. The most likely reason for this is the MHC haplotype of the animals studied, which in the case of the two earlier studies were inbred pigs homozygous for SLA class I haplotype 4a.O. Interestingly, low resolution SLA class I typing suggested that pig AN5 carried haplotype 4.0 and this was confirmed by allele specific PCR (data not shown). It therefore seems likely that the immunodominant response to NS5 A was restricted by one of the Lr24.0 alleles at the expense of peptides restricted by the SLA-1 4a.0 alleles. In support of this, the present data suggested that responses to NS2_1223-1230 and NS3 i902-i9i2 peptides were additionally restricted by the Lr-22.0 and 01.0 haplotypes, respectively, yet three animals that were heterozygous for these haplotypes mounted responses solely against NS3. Two of the peptides identified in this study, NS31902-1912 and E2₉96-ioo3, were homologous to antigenic regions we previously identified using BVDV peptides to screen CSFV-specific T cell responses [6].

To the inventors knowledge this is one of only a few reports on the definition of minimal length antigenic peptides recognised by porcine CD8 T cells. A recent study used the NetMHCpan prediction algorithm [25] to identify a 9mer peptide (MTAHITVPY) from the PI capsid antigen of the foot-and-mouth disease virus (FMDV) that bound the SLA allele SLA-1 *0401 [26, 27]. It was subsequently shown that a SLA- 1 *0401 /MTAHITVPY tetramer stained CD8 T cells from SLA matched FMDV vaccinated pigs and the size of the stained population correlated with cytotoxic responses in these animals [26]. In support of this immunoinformatics approach, the CSFV polyprotein was screened for binding to potential restricting alleles (based on the SLA class I Lr typing results) and two of the peptides NS2i223-i23o and NS31902-1912 were predicted by NetMHCpan (www.cbs.dtu.dk/services/NetMHCpan/) to bind strongly to at least one of the class I alleles potentially present in each of the restricting haplotypes: SLA-2* 12.01 for NS2_1223-1230 (potentially present in both haplotypes Lr-38.0 and Lr-22.0) and NS3i₉₀2-i9i2 was predicted to strongly bind SLA-*08.01 (Lr-07-0), SLA-3*07.01 (Lr-28.0);and SLA-2 01.01 and SLA-2 01.02 (Lr-01.0) (data not shown).

Immunodominance in CD8 T cell responses is thought to arise primarily as a consequence of the limitations of peptides to bind with high-affinity to available MHC class I molecules, with additional limitations in antigen processing and the CD8 T cell receptor repertoire also playing a role. Only approximately 1/2000 of the peptides within an antigen can achieve immunodominant status with a given MHC class I allele [28]. Immunodominance is thought to be critical for immunity since numerically prominent CD8 T cells have been shown to confer more effective protection than T cells specific for subdominant epitopes. It has also been shown that the efficacy of peptides in providing protection against a viral challenge is proportional to their binding affinity for the restricting MHC class I molecule [29]. In HIV-1 infection, it has been shown that immunodominant CD8 T cell responses may limit virus replication and they are the primary targets of escape mutations [30, 31]. Immunodominance of CD8 T cell responses to Flaviviruses has also been described, with disparity in the degree depending on the virus/host system studied [32, 33]. A recent study, utilising a similar strategy to the present one, assessed the specificity of human T cell response to Dengue virus. Using a cohort of 25 patients, 21 novel CD8 T cell epitopes were identified, with NS3 and NS5 being the most antigenic and the majority of epitopes being recognised by single patients [32].

Immunodominance may also be a key factor in determining the strain specificity of immunity if directed against polymorphic epitopes. The identified epitopes in this study were well conserved amongst CSFV isolates, although for most of these variants it remains to be determined whether these substitutions could affect T cell recognition. NetMHCpan predicted that the mutation of tyrosine in place of a phenylalanine at position 1906 of the NS3 l lmer VEYSFIFLDEY would not affect the binding to SLA class I alleles and we were able to show this mutation did not affect T cell reactivity. Moreover, the antigenic core₂₄i-255 peptide and the E2₉9₆-ioo3 and NS31902-1912 epitopes are well conserved between CSFV, BVDV and BDV. Such conserved T cell epitopes could enhance the efficacy of a BVDV/CSFV chimeric vaccine. It has recently been reported that the chimeric vaccine CP7_E2alf induces rapid protection, comparable to the C-strain vaccine [12, 34]. Since this protection precedes the appearance of neutralizing antibodies, it may be that T cell responses against epitopes conserved between BVDV and CSFV are contributing to the protective effect. Support for this hypothesis also stems from our recent studies that showed a close temporal correlation between the induction of CSFV-specific T cell IFN-γ responses and rapid protection induced by the C-strain vaccine [5] and that T cells from these animals cross-react with BVDV derived peptides [6].

Several studies have identified E2 and NS3 as targets of the T cell response against CSFV [6- 10]. However, our results provide strong evidence that at least in the context of responses induced by C-strain CSFV, E2 and NS3 are not necessarily the major T cell antigens and other CSFV proteins may be important targets of the CD8 T cell response. These results however may not limit the design of a marker vaccine based on only one or two immunodominant antigens. A study on the related yellow fever virus showed that, in case of abrogation of the dominant CD8 T cell epitope, the frequencies of T cells recognizing the subdominant CD8 T cell epitope increased dramatically [33]. A previous study showed that vaccination with a IFN-γ adjuvanted subunit CSFV E2 vaccine fully protected pigs against a challenge administered only 7 days later, suggesting that a T cell response directed solely against E2 may be protective [35]. A recent evaluation of dendrimeric peptide formulations of linear B cell epitopes from E2 and a CD8 T cell antigenic peptide from NS3, induced the production of both neutralizing antibodies and IFN-γ producing cells, which resulted in a degree of protection against CSFV [36]. Significantly, the protection achieved was greatest when the T cell antigenic peptide was included together with the B cell epitopes [37]. Incorporation of T cell epitopes identified in the present study might further enhance the development of such a marker vaccine approach.

The inventors observed that all the identified peptides were able to induce CD 107a mobilization in∑FN-y⁺ CD8 T cells. Translocation of lysosomal-associated membrane protein 1 (LAMP 1 /CD 107a) to the cell membrane has been validated as a marker of cytotoxic degranulation by CD8 T and K cells [38]. The data support previous studies, which report the ability of the viral antigens E2 and NS3 to elicit cytotoxic activity in vaccinated pigs [7- 10] and suggest that these T cells possessed cytotoxic activity in addition to cytokine release. The inventors also assessed the polyfunctionality of the peptide-specific CD8 T cell populations, by assessing co-expression of IFN-γ, T F-a and IL-2, since this may be central to their protective capacity. Studies on HCV have shown that vaccination with vectors expressing NS3 and NS4a proteins induce specific polyfunctional CD8 T cells which are associated with protection [14, 15]. However, the majority of peptide-specific CD8 T cells expressed IFN-γ alone or IFN-γ and T F-α, with only a small percentage co-expressing IFN- γ, T F-α, and IL-2. Interestingly, as observed in other systems [39, 40], the subset expressing all three cytokines showed the highest 'quality' of response producing more IFN-γ on a per cell basis.

Analysis of activation markers showed that the majority of the CD8 T cells releasing IFN-D after peptide stimulation did not express the alpha chain of the IL-2 receptor (CD25). In contrast, a previous study reported variable expression of this marker on CSFV-specific IFN- γ CD8 T cells [41]. We speculate that the absence of CD25 expression is not indicative of a naive state of CD8 T cells, but could reflect a memory phenotype of these cells since a previous study reports that after antigen stimulation some human CD8 T cells express high levels of CD25, whereas others display low levels of this marker and the latter preferentially differentiate into long-lived functional memory cells [42]. Similar to what has been described for human CD8 T cell responses to other Flaviviruses, variable levels of expression of CD27 were observed on peptide-specific CD8 T cells [15, 43]. Based on data from human CD8 T cells this may indicate that the majority of CSFV epitope-specific CD8 T cell populations may be T effector memory (T_EM) cells [44]. Our results suggest that vaccine strategies should be adopted in order to drive the generation of peptide-specific polyfunctional CD8 T_EM cells. A peptide-based vaccine targeting dendritic cells to provide a strong signal to naive CD8 T cells should be designed, with the inclusion of cytokines (IL-12 and IFN-γ) or pathogen- recognition receptor agonists as adjuvants.

By utilising a comprehensive approach to determine the antigen specificity of the CD8 T cell response, we have provided strong evidence for immunodominance in the T cell response to CSFV. CD8 T cell responses of individual animals were uniquely focussed on only one or two epitopes which were mapped on the core, E2 and non- structural proteins NS2, NS3 and NS5A. The individual responses were associated with the expression of distinct MHC class I haplotypes, and for two of the peptides there was evidence that they are presented by alleles present in other haplotypes. The five identified antigenic peptides were highly conserved across CSFV isolates, and for some were also well conserved when aligned against the other pestiviruses. The responding CD8 T cells displayed evidence of cytotoxic function, with the majority of IFN-y⁺ cells co-expressing the cytotoxicity marker CD 107a and populations also releasing T F-α and/or IL-2. The antigens and epitopes identified and characterised in this study are therefore useful in the preparation of vaccines and treatments for CSFV.

References

1. Thiel H, Stark R, Weiland E, Rumenapf T, Meyers G: Hog cholera virus: molecular composition of virions from a pestivirus. J Virol 1991, 65:4705-4712.

2. Moennig V: Introduction to classical swine fever: virus, disease and control policy. Vet Microbiol 2000, 73:93-102.

3. http://www.oie.int/wahis_2/public/wahid.php/Diseaseinfomiation/statusdetail

4. Beer M, Reimann I, Hoffmann B, Depner K. Novel marker vaccines against classical swine fever. Vaccine. 2007 25:5665-5670.

5. Graham SP, Everett HE, Haines FJ, Johns HL, Sosan OA, Salguero FJ, Clifford DJ, Steinbach F, Drew TW, Crooke HR: Challenge of pigs with Classical swine fever viruses after C-strain vaccination reveals remarkably rapid protection and insights into early immunity. PLoS One 2012, 7:e29310

6. Graham SP, Haines F, Johns H, Sosan OA, La Rocca SA, Lamp B, Rumenapf T, Everett H, Crooke H: Characterization of vaccine-induced, broadly cross-reactive IFN-g secreting T cell responses that correlate with rapid protection against classical swine fever virus. Vaccine 2012, 30:2742-2748.

7. Ceppi M, de Bruin MG, Seuberlich T, Balmelli C, Pascolo S, Ruggli N, Wienhold D, Tratschin JD, McCullough KC, Summerfield A: Identification of classical swine fever virus protein E2 as a target for cytotoxic T cells by using niRNA- transfected antigen presenting cells. J Gen Virol 2005, 86:2525-2534.

8. Rau H, Reverts H, Balmelli C, McCullough K, Summerfield A: Immunological properties of recombinant classical swine fever virus NS3 protein in vitro and in vivo. Vet Res 2006, 37: 155-168.

9. Armengol E, Wiesmuller KH, Wienhold D, Buttner M, Pfaff E, Jung G, Saalmuller A: Identification of T-cell epitopes in the structural and non-structural proteins of classical swine fever virus. J Gen Virol 2002, 83:551-560.

10. Pauly T, Elbers K, Konig M, Lengsfeld T, Saalmuller A, Thiel HJ: Classical swine fever virus-specific cytotoxic T lymphocytes and identification of a T cell epitope. J Gen Virol 1995, 76:3039-3049.

11. Tarradas J, Argilaguet JM, Rosell R, Nofrarias M, Crisci E, Cordoba L, Perez-Martin

E, Diaz I, Rodriguez F, Domingo M, Montoya M, Ganges L: Interferon-gamma induction correlated with protection by DNA vaccine expressing E2 glycoprotein against classical swine fever virus infection in domestic pigs. Vet Microbiol 2010, 142: 51-58.

Koenig P, Lange E, Reimann I, Beer M: CP7_E2alf: a safe and efficient marker vaccine strain for oral immunization of wild boar against Classical swine fever virus (CSFV). Vaccine 2007, 25:3391-3399.

Schlaphoff V, Klade CS, Jilma B, Jelovcan SB, Cornberg M, Tauber E, Manns MP, Wedemeyer H, IC41 Study Group: Functional and phenotypic characterization of peptide-vaccine-induced HCV-specific CD8 T cells in healthy individuals and chronic hepatitis C patients. Vaccine 2007, 25:6793-6806.

Lang Kush KA, Ginsberg AA, Yan J, Wiseman RW, Khan AS, Sardesai NY, O'Connor DH, Weiner DB: Hepatitis C virus NS3 NS4A DNA vaccine induces multiepitope T cell responses in rhesus macaques mimicking human immune responses. Mol Ther 2012, 20:669-678.

Mikkelsen M, Hoist PJ, Bukh J, Thomsen AR, Christensen JP: Enhanced and sustained CD8 T cell responses with an adenoviral vector-based hepatitis C virus vaccine encoding NS3 linked to the MHC class II chaperon protein invariant chain. J Immunol 2011,186: 2355-2364.

Costa SM, Yorio AP, Goncalves AJ, Vidale MM, Costa EC, Mohana-Borges R, Motta MA, Freire MS, Alves AM: Induction of a proctive response in mice by the dengue virus NS3 protein using a DNA vaccine. PLos One 2011, 6:e25685.

Gao G, Wang Q, Dai Z, Calcedo R, Sun X, Li G, Wilson JM: Adenovirus-based vaccines generate cytotoxic T lymphocytes to epitopes of NS1 from dengue virus that are present in all major serotypes. Hum Gene Ther 2008, 19:927-936.

Sandvik T, Drew T, Paton D: CSF virus in East Anglia: where from? Vet Rec 2000, 147:251.

Parchariyanon S, Inui K, Damrongwatanapokin S, Pinyochon W, Lowings P, Paton D: Sequence analysis of E2 glycoprotein genes of classical swine fever viruses: identification of a novel genogroup in Thailand. Dtsch Tierarztl Wochenschr 2000, 107:236-238.

Drew TW: Classical swine fever (hog cholera). In: OIE Manual of Diagnostic Tests and Vaccines for Terrestrial Animals (mammals, birds and bees), Chapter 2.8.3, vol.2. 6^th edition Paris, France: World Organization for Animal Health (OIE); 2008. pp 1092-1106. Everett HE, Salguero FJ, Graham SP, Haines FJ, Johns H, Clifford D, Nunez A, La Rocca SA, Parchariyanon S, Steinbach F., Drew TW, Crooke H: Characterization of experimental infection of pigs with genotype 2.1 and 3.3 isolates of classical swine fever virus. Vet Micro 2010, 142:26-33.

Reutner K, Leitner J, Essler SE, Witter K, Patzl M, Steinberger P, Saalmuller A, Gerner W: Porcine CD27: identification, expression and functional aspects in lymphocyte subset in swine. Dev Comp Immunol 2012, 38:321-331.

Essler SE, Ertl W, Deutsch J, Ruetgen BC, Groiss S, Stadler M, Wysoudil B, Gerner W, Ho CS, Saalmueller A: Molecular characterization of swine leukocyte antigen gene diversity in purebred Pietrain pigs. Anim Genet 2013, 44:202-205.

Lunney JK, Ho CS, Wysocki M, Smith DM: Molecular genetics of the swine major histocompatibility complex, the SLA complex. Dev Comp Immunol 2009, 33:362- 374.

Hoof I, Peters B, Sidney J, Pedersen LE, Sette A, Lund O, Buus S, Nielsen M: NetMHCpan, a method for MHC class I binding prediction beyond humans. Immunogenetics 2009, 61: 1-13.

Patch JR, Pedersen LE, Toka FN, Moraes M, Grubman MJ, Nielsen M, Jungersen G, Buus S, Golde WT: Induction of foot-and-mouth disease virus-specific cytotoxic T cell killing by vaccination. Clin Vaccine Immunol 2011, 18:280-288.

Pedersen LE, Harndahl M, Nielsen M, Patch JR, Jungersen G, Buus S, Golde WT: Identification of peptides from foot-and-mouth disease virus structural proteins bound by class I swine leukocyte antigen (SLA) alleles, SLA-1*0401 and SLA- 2*0401. Animal Genetic 2012, ahead of print.

Yewdell JW, Bennink JR: Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses. Annu Rev Immunol 1999, 17:51-88.

Oukka M, Manuguerra JC, Livaditis N, Tourdot S, Riche N, Vergnon I, Cordopatis P, Kosmatopoulos K: Protection against lethal viral infection by vaccination with nonimmunodominant peptides. J Immunol 1996, 157:3039-3045.

Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, Feeney M, Tang Y, Holmes EC, Allen T, Prado JG, Altfeld M, Brander C, Dixon C, Ramduth D, Jeena P, Thomas SA, St John A, Roach TA, Kupfer B, Luzzi G, Edwwards A, Taylor G, Lyall H, Tudor- Williams G, Novelli V, Martinez-Picado J, Kiepiela P, Walker BD, Goulder PJ: HIV evolution: CTL escape mutation and reversion after transmission. Nat

Med 2004, 10:282-289.

Ammaranond P, Zaunders J, Satchell C, van Bockel D, Cooper DA, Kelleher AD: A new variant cytotoxic T lymphocyte escape mutation in HLA-B27-positive individuals infected with HIV type 1 AIDS Res Hum Retroviruses 2005, 21:395- 397.

Rivino L, Kumaran EA, Jovanovic V, Nadua K, Teo EW, Pang SW, Teo GH, Gan VC, Lye DC, Leo YS, Hanson BJ, Smith KG, Bertoletti A, Kemeny DM, MacAry PA: Differential Targeting of Viral Components by CD4+ versus CD8+ T Lymphocytes in Dengue Infection J Virol 2013; 87:2693-2706

Van der Most RG, Harrington LE, Giuggio V, Mahar PL, Ahmed R: Yellow fever virus 17D envelope and NS3 proteins are major targets of the antiviral T cell response in mice. Virology 2002, 296: 117-124.

Renson P, Dimna M, Keranflec H A, Cariolet R, Koenen F, Potier MF. CP7_E2alf oral vaccination confers partial protection against early classical swine fever virus challenge and interferes with pathogeny-related cytokine responses. Vet Res 2013, 44:9.

Toledo J, Barrera M, Farnos O, Gomez S, Rodriguez M, Aguero F, Ormazabal V, Parra N, Suarez L, Sanchez O: Human alFN co-formulated with milk derived E2- CSFV protein induce early fully protection in vaccinated pigs. Vaccine 2010, 28:7907-7914.

Monso M, Tarradas J, de la Torre B, Sobrino F, Ganges L, Andreu D: Peptide vaccine candidates against classical swine fever virus: T cell and neutralizing antibody responses of dendrimers displaying E2 and NS2-3 epitopes. J Peptide Science 2011, 17:24-31.

Tarradas J, Monso M, Fraile L, de la Torre BG, Munoz M, Rosell R, Riquelme C, Perez LI, Nofrarias M, Domingo M, Sobrino F, Andreu D, Ganges L: A T-cell epitope on NS3 non-structural protein enhances the B and T cell responses elicited by dendrimeric constructions against CSFV in domestic pigs. Vet Immunol Immunopathol 2012, 150:36-46.

Betts MR, Brenchley JM, Price DA, De Rosa SC, Douke DC, Roeder M, Koup RA: Sensitive and viable identification of antigen-specific CD8T cells by a flow cytometric assay for degranulation. J Immunol Methods 2003, 281:65-78. Whelan AO, Villarreal-Ramos B, Vordermeier HM, Hogart PJ: Development of an antibody to bovine IL-2 reveals multifunctional CD4 T_EM cells in cattle naturally infected with bovine tuberculosis. PLos One 201 1, 6: 1-8.

Seder RA, Darrah PA, Roederer M: T-cell quality in memory and protection: implications for vaccine design. Nat Rev Immunol 2008, 8:247-258.

Suradhat S, Sada W, Buranapradiktun S, Damrongwatanapokin S: The kinetics of cytokine production and CD25 expression by porcine lymphocyte subpopulations following exposure to classical swine fever virus (CSFV). Vet Immunol Immunopathol 2005, 106: 197-208.

Kalia V, Sarkar S, Subramaniam S, Haining WN, Smith KA, Ahmed R: Prolonged interleukin-2Ralpha expression on virus-specific CD8 T cells favours terminal- effector differentiation in vivo. Immunity 2010, 32:91-103.

Tomiyama H, Takata H, Matsuda T, Takiquchi M: Phenot pic classification of human CDS T cells reflecting their function: inverse correlation between quantitative expression of CD27 and cytotoxic effector function. Eur J Imm 2004, 34:999-1010.

Tian J, Zeng G, Pang X, Liang M, Zhou J, Fang D, Liu Y, Li D, Jiang L: Identification and immunogenicity of two new HLA-A*0201-restricted CD8 T- cell epitopes on dengue NS1 protein Int Immunol 2012, 24: 207-218

Table A. Conservation of identified CD8 T cell epitopes/antigenic regions among different CSFV isolates and the related pestiviruses, bovine viral diarrhoea (BVDV) and border disease virus (BDV).

Virus Strain Geno ype GenBank: ΟΟΓβ241-255 NS3_1S E2 996-1003 NS2 NS5A3070-30

CSFV C-strain Riems AY259122.1 PESRKKLEKALLAWA VEYSFIFLDEY YEPRDSYF STVTGIFL RVDNALLK CSFV Brescia AF091661

CSFV Alfort/187 X87939.1

CSFV KC Vaccine AF099102

CSFV ALD D49532

CSFV GPE^" D49533

CSFV Alfort-A19 U90951

CSFV CF1 14 AF333000

CSFV Shimen AF092448 K

CSFV Koslov HM237795

CSFV SWH DQ127910

CSFV Penevezys 2. HQ148063 . M ...T CSFV UK2000/7.1 2. K..T CSFV Borken 2.3 GU233731 ...T CSFV Alfort Tiibingen 2.3 AA43844 ...T CSFV CBR/93 3.3 K.. DT...

CSFV 94.4/IL/94/TWN 3.4 AY646427 K.. K.

BVDV NADL 1a AJ133738 , Y. L. S . V. D.. PE. SE

BDV BD31 1a U70263 , Y. I . S ... D. EQD.. E

Table B. Conservation of nucleotide sequences encoding the identified CD8 T cell antigenic region on core protein among different CSFV isolates and the rel pestiviruses, bovine viral diarrhoea virus (BVDV) and border disease virus (BDV).

CO re 1094-1138

Virus Strain Genotype GenBank Accession No.

CCAGAATCTAGGAAGAAATTAGAAAAAGCCCTATTGGCATGGGCG

CSFV Alfort 187 1 .1 X87939 CCAGAATCTAGGAAGAAATTAGAAAAAGCCCTATTGGCATGGGCG

CSFV KC Vaccine 1 .1 AF099102 A

CSFV ALD 1 .1 D49532

CSFV Brescia 1 .1 AF091661

CSFV GPE- 1 .1 D49533

CSFV Alfort A19 1 .1 U90951

CSFV CF1 14 1 .1 AF333000

CSFV Riems 1 .1 U45477 C

CSFV Shimen 1 .1 AF092448

CSFV Koslov 1 .1 HM237795 G

CSFV SWH 1 .1 DQ127910

CSFV Penevezys 2.1 HQ148063 G..C C T

CSFV UK2000/7.1 2.1 C C C....T A

CSFV Borken 2.3 GU233731 G..C C G T

CSFV Alfort Tiibingen 2.3 AA43844 G..C C G T

CSFV CBR/93 TTG C A ... C .... T

CSFV 94.4/IL/94/TWN AY646427 G GC C A BVDV NADL AJ133738 .AG AC . C C . G AT . G G A BDV BD31 U70263 .CT..G A..A G..G A..G T T

Table C. Conservation of nucleotide sequences encoding the identified CD8 T cell epitope on NS2 protein among different CSFV isolates and the rel pestiviruses, bovine viral diarrhoea virus (BVDV) and border disease virus (BDV).

Virus Strain Geno ype GenBank Accession No.

AGC AC AG TGAC AGG TATC TT TTA

CSFV Alfort 187 X87939 AGCACAGTGACAGGTATCTTTTTA CSFV KC Vaccine AF099102 C..G CSFV ALD D49532

CSFV Brescia AF091661

CSFV GPE- D49533

CSFV Alfort A19 U90951

CSFV CF1 14 AF333000

CSFV Riems U45477

CSFV Shimen AF092448

CSFV Koslov HM237795

CSFV SWH DQ127910

CSFV Penevezys 2 HQ148063 . T .. G T..A CA.G

CSFV UK2000/7.1 2 .... G T..A CC.G

CSFV Borken 2.3 GU233731 .... G T..A CC.G

CSFV Alfort TUbingen 2.3 AA43844 G T .. A CC.G

CSFV CBR/93 3.3 .A. , CC.

CSFV 94.4/IL/94/TWN 3.4 AY646427 ,C.

BVDV NADL 1 AJ133738 , . CTG .. AT . T .. GG . G .. C . . G

BDV BD31 1a U70263 . T . T . .. . T . T A.. .C.T

Table D. Consen ation of nucleotide sequences encoding the identified CD8 T cell epitope on E2 protein among different CSFV isolates and the related pesti bovine viral diarrhoea virus (BVDV) and border disease virus (BDV).

Virus Strain Genotype GenBank Accession No.

TATGAGCCCAGGGACAGCTACTTC

CSFV Alfort 187 1 .1 X87939 TATGAGCCCAGGGACAGCTACTTC

CSFV KC Vaccine 1 .1 AF099102 T...

CSFV ALD 1 .1 D49532

CSFV Brescia 1 .1 AF091661

CSFV GPE- 1 .1 D49533

CSFV Alfort A19 1 .1 U90951

CSFV CF1 14 1 .1 AF333000

CSFV Riems 1 .1 U45477

CSFV Shimen 1 .1 AF092448

CSFV Koslov 1 .1 HM237795

CSFV SWH 1 .1 DQ127910

CSFV Penevezys 2.1 HQ148063 T...

CSFV UK2000/7.1 2.1 T...

CSFV Borken 2.3 GU233731

CSFV Alfort Tubingen 2.3 AA43844 A T

CSFV CBR/93 3.3 . . . . A A T . . T . . .

CSFV 94.4/IL/94/TWN 3.4 AY646427 . . . . A

BVDV NADL 1 AJ133738 T A T

BDV BD31 1 a U70263 A . . . . A . . . . .

Table E. Conservation of nucleotide sequences encoding the identified CD8 T antigenic region on NS3 protein among different CSFV isolates and the rel pestiviruses, bovine viral diarrhoea virus (BVDV) and border disease virus (BDV).

Virus Strain Genotype GenBank Accession No.

GTTGAGTACTCCTTCATATTTCTTGATGAGTAC

CSFV Alfort 187 1.1 X87939 GTTGAGTACTCCTTCATATTTCTTGATGAGTAC

CSFV KC Vaccine 1.1 AF099102

CSFV ALD 1.1 D49532

CSFV Brescia 1.1 AF091661 T

CSFV GPE- 1.1 D49533

CSFV Alfort A19 1.1 U90951

CSFV CF114 1.1 AF333000

CSFV Riems 1.1 U45477

CSFV Shimen 1.1 AF092448

CSFV Koslov 1.1 HM237795

CSFV SWH 1.1 DQ127910

CSFV Penevezys 2.1 HQ148063 . .A T. .T.A.. .C.. .T.A..C

CSFV UK2000/7.1 2.1 . .A T C

CSFV Borken 2.3 GU233731 . .A. .A..T T C. .A T

CSFV Alfort TUbingen 2.3 AA43844 .A. .A.. T T C. .A T

CSFV CBR/93 3.3 .A T . . T . A .. . C A..C

CSFV 94.4/IL/94/TWN 3.4 AY646427 . G T T C

BVDV NADL 1 AJ133738 .A. .A A. A CT.A A.. .

BDV BD31 1a U70263 .. . .A.. .A.T.A.. .T..C. .A

Table F. Conservation of nucleotide sequences encoding the identified CD8 T cell epitope on NS5A protein among different CSFV isolates and the rel pestiviruses, bovine viral diarrhoea virus (BVDV) and border disease virus (BDV).

Virus Strain Genotype GenBank Accession No.

AGGGTTGATAATGCTCTATTGAAATTT

CSFV Alfort 187 1 .1 X87939 AGGGTTGATAATGCTCTATTGAAATTT

CSFV KC Vaccine 1 .1 AF099102 C

CSFV ALD 1 .1 D49532

CSFV Brescia 1 .1 AF091661

CSFV GPE- 1 .1 D49533

CSFV Alfort A19 1 .1 U90951

CSFV CF1 14 1 .1 AF333000

CSFV Riems 1 .1 U45477

CSFV Shimen 1 .1 AF092448 .A

CSFV Koslov 1 .1 HM237795

CSFV SWH 1 .1 DQ127910

CSFV Penevezys 2.1 HQ148063 C .. C . CC C . T

CSFV UK2000/7.1 2.1 .A...C....CC C

CSFV Borken 2.3 GU233731 ..A..C....CC GC

CSFV Alfort TUbingen 2.3 AA43844

CSFV CBR/93 3.3 .A... G ... G ..A.. ....A

CSFV 94.4/IL/94/TWN 3.4 AY646427 . A ... C .. C .. A .. C . GC .... G ...

BVDV NADL 1 AJ133738 GA .. C .. CCC .. AGC . GTCTG .. A. G

BDV BD31 1 a U70263 GA .. G ... C . A . ACT .. C .. G . G . A .

Claims

Claims

1. An isolated peptide comprising an epitope having an amino acid sequence comprising at least 8, 9, 10 or 11 contiguous amino acids taken from one of the following sequences:

L SRVDNALLKF ;

LISTVTGIFLI;

RYYEPRDSYFQ;

VEYSFIFLDEY; and

PESRKKLEKALLAWA; or a variant or functional fragment thereof.

2. An amino acid sequence having substantial homology to any of the sequences shown in Table A.

3. An isolated nucleic acid molecule encoding the peptide of claim 1 or claim 2.

4. A vector comprising a nucleic acid molecule according to claim 3.

5. A host cell comprising such a vector.

6. An isolated nucleic acid molecule comprising a nucleotide sequence having significant identity with one or more of the sequences provided in Table B.

7. A pharmaceutical composition comprising a peptide according to claims 1 or 2, or a nucleotide according to claim 3 or 6, or a vector according to claim 4.

8. A pharmaceutical composition according to claim 7 comprising another biologically or therapeutically active agent.

9. A peptide according to claims 1 or 2, or a nucleotide according to claim 3 or 6, or a vector according to claim 4 for use in therapy.

10. A peptide according to claims 1 or 2, or a nucleotide according to claim 3 or 6, or a vector according to claim 4 for use in the treatment of or vaccination against classical swine fever.

11. A method of vaccinating a subject against or treating a subject having classical swine fever comprising administering a peptide according to claims 1 or 2, or a nucleotide according to claim 3 or 6 or a pharmaceutical composition according to claim 10 to the subject.