WO2015104551A1

WO2015104551A1 - Treating susceptibility

Info

Publication number: WO2015104551A1
Application number: PCT/GB2015/050031
Authority: WO
Inventors: Thomas Moen; Jacob Seilø TORGERSEN; Nina Santi
Original assignee: AQUA GEN AS
Current assignee: AQUA GEN AS
Priority date: 2014-01-08
Filing date: 2015-01-08
Publication date: 2015-07-16
Anticipated expiration: 2016-07-08
Also published as: GB201400311D0

Abstract

The present invention relates to a composition for use in the prevention or treatment of viral infection, wherein the composition is a. an agent that blocks viral binding to cadherin; b. an agent that increases cadherin concentration at the cell-cell junction; and/or c. an agent that increases efficacy of cadherin mediated cell-cell binding.

Description

Treating Susceptibility

The present invention relates to methods for preventing (ie prophylactic treatment) and treating viral infection, and to compositions for use in such treatment.

Viruses (and virions, being the virus particles) are non-cellular infective agents that are capable of reproducing in animal host cells. A virus includes a nucleic acid (either DNA or RNA) at its core, surrounded by a proteinaceous capsid (and in some cases an outer envelope). Epithelial barriers are the main point of entry for viruses into the body. The host cells impose multiple barriers to virus entry, and virus entry pathways are largely defined by the interactions between virus particles and their receptors at the cell surface. After entry, the virus replicate inside the host cell by interacting with the host cells' replicative and protein-producing apparatus and in so doing direct the host cell to produce virions, which are released from the host cell upon its lysis or by budding of the cell membrane. Viruses are the causative agent in a number of human and animal infections, such as influenza, measles, hepatitis, rabies, foot-and-mouth disease, infectious bursal disease (IBDV), infectious pancreatic necrosis (IPN), pancreas disease (PD), heart and skeletal muscle inflammation (HSMI), infectious salmon anaemia (ISA) and cardiomyopathy syndrome (CMS).

Treatments that have been found for some viral infections are based on the use of vaccines or anti -viral therapeutics (prophylactically or otherwise). Vaccines recruit the subject's own immune responses to fight viral infection, but may not be effective if the subject's immune response is not receptive to the vaccine or can induce inappropriate immune responses that may be damaging to the subject. Anti-viral therapeutics achieve their effect in a number of potential ways (predominantly by preventing viral synthesis such as is the case for acyclovir), but resistance in virus populations to these therapeutics is known. Apart from the obvious damage caused by viruses to the human and non-human animal population, viral infections pose a severe economic threat to many agri- and aquacultural industries. For example, infectious pancreatic necrosis (IPN) is one of the major threats to the fish farming industry worldwide. The disease is caused by an aquatic birnavirus, causing necrosis of pancreatic cells and liver cells, resulting in lethargy and sudden mortality. The virus is wide-spread in nature, but does not seem to affect free-living salmon to any large extent. In aquaculture environments, the disease causes mortalities both at the fry stage, when the fish are still living in fresh water, and at the post-smolt stage, shortly after transfer to sea water. The industry-wide losses in salmon farming due to IPN have been estimated to be 8 % during the fresh water phase and 5 % during the sea phase. The IPN virus genome consists of two genomic segment (A and B), encoding four (VP2-VP5) and one (VP1) viral proteins, respectively. Cellular uptake of IPN virus depends upon recognition or a receptor protein, most likely through binding of the capsid protein VP2 to its receptor, followed by endocytosis (Granzow et al. 1997).

Infectious Bursal Disease (also known as IBD or Gumboro Disease) is a highly contagious disease infecting chickens and caused by infectious Bursal Disease virus (IBDV). IBDV can infect chickens, turkeys, ducks, geese and guinea fowl but only chickens show clinical signs of the disease. Infections in chickens with IBDV usually results in mortality at 3 to 6 weeks of age.

Pancreas disease (PD) is another viral disease of importance to the salmon industry. The responsible viral agent, salmonid alphavirus (SAV), infects farmed Atlantic salmon and rainbow trout, and clinical signs are lethargy and sudden inappetence, reduced growth and increased mortality during on-growth in seawater. Similar to IPN, the disease is characterized by necrosis of the exocrine pancreas, but also accompanied by severe cardiac and skeletal muscle lesions. Mortality rates vary from 0 to 50%, and up to 15% of survivors may fail to grow and become runts.

Heart and skeletal muscle inflammation (HSMI) is characterized by severe cardiac and skeletal muscle inflammation of some similarity to PD. The disease is associated with an infection by a piscine reovirus. (Palacios et al. 2010), and typically appears 5 to 9 months after fish are transferred from fresh water to seawater net pens. In severe cases the mortality may reach 20%. The genetic make-up of the host has a major influence on its response to pathogens and a proportion of subjects may possess natural resistance to viral infection.

There is, since treatments for viral infections are often lacking and often suboptimal, a need for treatments for currently untreatable viral infections and a need for alternative treatments for viral infections for which the aforementioned treatments are currently available.

Cadherins constitute a large family of cell surface adhesion molecules that can be grouped into several subfamilies. They participate in the maintenance of proper cell-cell contact by mediating cell adhesion at adherens junctions, and typically consist of five tandem repeated extracellular domains, a single transmembrane segment and a cytoplasmic region. Cadherins generally mediate monotypic cell-cell adhesion although heterotypic binding between different cadherin molecules. Adhesive dimerization between cadherins involves the interaction between their ECl and EC2 domains (ECl-ECl and EC1-EC2 interaction being possible), and small differences in the affinities at these inter-protein interfaces can result in highly specific cell-cell interactions. Ca2+ ions bound in the inter-domain linker region are essential for cadherin cell mediated adhesion, primarily by supporting the mechanical stability of cadherin molecules. Cell-cell adhesive strength is also determined by cadherin expression level.

Loss of cell-cell contact is a hallmark of cancer development, and epithelial cadherin (E-cadherin) has been characterized as a potent suppressor of invasion and metastasis. In general, dysregulation of cadherins and in particular epigenetic silencing contributes to different aspects of cancer progression, and they have been investigated as potential therapeutic targets. For instance, different transcriptional regulators of E-cadherin have been identified, such as Twist, Snail and ZEB family members and their respective miRNA regulators.

The inventors of the present invention have, following extensive experimentation, surprisingly found that polymorphisms in a cadherin gene (ie a gene that encodes for a cadherin protein) and variants in the amino acid sequence of cadherin (ie the protein product of the cadherin gene) are associated with resistance or susceptibility to viral infection in subjects. The inventors have gone on to demonstrate for the first time that virions can bind to cadherins, and immunofluorescence studies demonstrated interactions between cadherin, virion and clatherin; as they are co-localised in infected tissue. As cadherins are cell- surface proteins, and it is known that viruses may be internalised in a cell via clatherin-coated vesicle mechanisms, cadherins are clearly implicated in a mechanism for enabling viral cell-invasion. Furthermore, it has been demonstrated that viruses are inhibited from invading cells that expresses the resistant variant of cadherin that has been identified by the inventors. It has also be found that the polymorphism of cadherin that has been shown to confer viral resistance is located on a second extracellular domain of cadherin that interacts with a first extracellular domain in order to form cis-dimers between cadherins located on the same cell. The stability of adherence junctions formed by cadherins requires such dimerization. These findings lead directly to the understanding of how resistance to those viral infections were the virus interacts directly with cadherin is achieved. Additionally, the inventors observed another mechanism of viral resistance where viral resistance is achieved through increased activity and affinity of cadherin expressed by the subject, preventing viral access to other viral receptors present at adherens junctions and strengthening the epithelial barrier function.

These findings lead the inventors to the understanding of cadherin's role in viral infection and the fact that it may mediate cell-cell adhesion mechanisms that enhance resistance to viral infection; this has directed the inventors to new therapeutic applications for treating viral infection. In particular, the inventors are guided by this new understanding to create new vaccine therapies. Any agent, including a new vaccine, capable of blocking binding of virus to cadherin would be of therapeutic value for those viral infections were the virus interact directly with cadherin. Additionally, the inventors are guided to the understanding that any therapeutic agent that induces increased local concentrations of cadherin at cell-cell junctions, or increased efficacy of cadherin mediated cell-cell adhesion would be useful in preventing or treating viral infection by hindering viral access to other receptors present at adherens junctions and by strengthening the epithelial barrier function. To-date, there has been no suggestion of a link between cadherin or the gene encoding cadherin, and viral resistance, or between these genes/proteins and the treatment of viral infection. Consequently, it is surprising that such a link exists, and without this knowledge, therapeutic applications that work on the basis of the understanding of cadherin's role in viral infection cannot be obvious.

Accordingly, in a first aspect of the present invention, there is provided a

composition for use in the treatment of viral infection, wherein the composition comprises: -

(a) an agent that blocks viral binding to cadherin;

(b) an agent that increases cadherin concentration at the cell-cell junction; and/or

(c) an agent that increases efficacy of cadherin mediated cell-cell binding,

Viruses are strictly dependent on target host cells for their amplification, but individual viruses have unique strategies in their entry into the cell and in the further steps in replication and virion release. Specific virus surface components must be maintained to allow attachment to and penetration of host cells for infection. Reacting to these common viral components with neutralizing antibodies would allow the immune system to respond rapidly to infection and serve as a basis for a vaccine. In order to produce the most efficient vaccines, the individual viral components and host factors involved in virus entry and replication must be thoroughly investigated. In virus research, solving structural models of individual viruses using crystallography have provided new possibilities for research into structural analysis. Similarly, modeling molecular structure of host proteins interacting with viral components is possible. Specific knowledge of the molecular interactions between the virus proteins and host proteins, characteristics which normally contribute to pathogenesis, may be manipulated to treat disease, providing templates for virus vaccine and antiviral drug design. So far, a number of anti-viral drugs have been generated based on such information. The polymorphisms of the present application relate to putative binding sites for viruses on the cadherin protein molecule. Regions encompassing susceptible alleles of the polymorphisms are thought to encode good binding site for the virus. Consequently, structural characterization of interacting interfaces between virus and cadherins reveal several potent targets for the design of specific anti -viral drugs and/or vaccines. Such information has been used extensively in the effort to design vaccines that can prevent the problem of escape-mutations in HIV research. The specific polymorphisms identified in E-cadherin is an important step in the understanding of birnavirus host-pathogen interaction, and how viruses may interact with cadherins to penetrate into a cell, facilitation the spread of the viral infection within the host. Vaccines designed to initiate an antibody response specifically against the viral epitopes interacting with the extracellular domain of the cadherin are anticipated to be much more effective than current vaccines.

Consequently, an agent that blocks viral binding to a cadherin according to the present invention could be a vaccine that induces an immune response (eg antibody response) against the epitope on the virus responsible for binding to the cadherin protein. The agent may therefore be an antibody that binds to cadherin and that blocks the binding of a virus to that cadherin. The antibody may be a monoclonal or polyclonal antibody. The skilled person would be aware of methods to identify the corresponding viral epitopes by for example performing structural analysis using computer modeling, or by performing binding studies between cadherins and viral proteins. The vaccine can be produced by different means, like recombinant protein expression, specifically designed peptides or by DNA vaccine technology. Such a vaccine could block viral binding by the virus by clearing of the virus by the induced immune reaction, or by physically blocking the site on the virus required for binding to cadherin protein. The epitope, from which the vaccine may be derived, may be extrapolated from the peptide portion of the cadherin protein that is encoded by the site of the polymorphism. The antibody of the present invention may therefore be one that binds to the aforementioned epitope (optionally specifically binds). Alternatively, the agents capable of blocking viral binding to cadherin may be any agent capable of binding to monomers of cadherins available for heterodimerization with viruses like IPN. Suitable agents may be small molecules or antibodies capable of antagonizing the binding of viruses to cadherins. In a slightly different manner, cadherins can confer viral resistance through increased activity and affinity of cadherin, whereby tighter contact between cells at adherens junction prevent viral access to other viral receptors (non-cadherins) present at adherens junctions, and thereby strengthens the epithelial barrier function. The skilled person would be aware of many agents suitable for increasing the concentration of cadherin. Indeed, cancer therapies have already been designed for this purpose. For example, agents of the present invention may be capable of inhibiting the function of transcription factors that repress cadherin transcription, examples of such transcription factors follow. Several transcription factors like SNAIl/Snail 1, SNAI2/Snail 2 (also known as Slug), ZEB1, ZEB2, E47 and KLF8 (Kruppel-like factor 8) can bind to E-cadherin promoter and repress its transcription. Whereas factors such as Twist, Goosecoid, E2.2 (also known as TCF4), homeobox protein SIX1 and FOXC2 (fork-head box protein C2) repress E-cadherin indirectly. Transforming growth factor i (TGF i) induces Snaill and Twistl expressions and is another example of an indirect modulation of E-cadherin expression and so could be an agent of the present invention. Histone acetylases regulate the interaction of transcription factors with targeted sequence in the E- cadherin promoter. Numerous histone deacetylase inhibitors (HDACi) have been identified and several are in clinical use or development, for example Vorinostat (also known as, N-hydroxy-N'-phenyl-octanediamide , suberoylanilide hydroxamic acid, SAHA and Zolinza®) or Entinostat (also known as Pyridin-3-ylmethyl N-[[4- [(2-aminophenyl)carbamoyl]phenyl]methyl]carbamate, SNDX-275 and MS-275); both of which have been demonstrated to increase expression of cadherin in cancer therapies. Histone deacetylase inhibitors, including Vorinostat and/or Entinostat may therefore be an agent of the present invention. Additionally or alternatively such an agent may be a TGF-β antagonists, for example Galunisertib (also known as 4-[2-(6-Methyl-2-pyridinyl)-5,6-dihydro-4H-pyrrolo[l,2-b]pyrazol-3-yl]-6- quinolinecarboxamide and LY2157299 monohydrate), which has been demonstrated for use in cancer therapies. Alternatively, or additionally, such an agent may act by any available gene silencing process, for example by using RNA interference technology or small molecules. The agent may also be exogenous cadherin.

The skilled person would be aware of many agents capable of increasing the efficacy of cadherin mediated cell-cell binding. It is know that such binding is dependent in a dose-dependent manner on the presence of calcium ions. Consequently, the agent could be an agent capable of increasing extracellular calcium ions (particularly locally at the cell-cell junction), for example any calcium salt. The agent may therefore be a diatery Ca2+ supplement, indeed dietary calcium treatment has been shown to increase cadherin levels in intestinal cells and have shown intestinal cancer inhibitory effects in mice(see Nutr Res Pract. Oct 2012; 6(5): 396-404.

"Inhibitory effects of calcium against intestinal cancer in human colon cancer cells and ApcMin/+ mice"). In vivo, Cdhl - β-catenin complexes in intestinal adherence junctions have been investigated for their role in cancer metastasis, but not in relation to pathogen entry.

In a further aspect of the present invention, there is provided a pharmaceutical composition comprising any one or combination of the aforementioned agents and on or more pharmaceutically acceptable excipients.

In a further aspect of the present invention there is provided a composition comprising any one or combination of the aforementioned agents for use in therapy (including prophylactic therapy). In a further aspect of the present invention, there is provided a method for treating viral infection (including prophylactically), wherein the method includes the step of administering an effective amount of any one or combination of the aforementioned agents to the subject to be treated.

The family of cadherins include N-cadherin (neuronal cadherin or Cdh2), E-caderhin (epithelial cadherin on Cdh-1), P-cadherin (placental cadherin or Cdh3), R-Cadherin (retinal cadherin or CDH4), Cdh5, K-cadherin (kidney cadherin or Cdh6), Cdh7, Cdh8, Cdh9, CdhlO, Cdhll, Cdhl2, Cdhl2P2, Cdhl3, Cdhl4, Cdhl5, Cdhl6, Cdhl7, Cdhl8, Cdhl9, Cdh20, Cdh22, Cdh24, Cdh26, CELSR1, CELSR2, CELSR3, DSC1, DSC2, DSC3, DSG1, DSG2, DSG3, DSG4. As the skilled person would be aware, CELSR relates to EGF LAG seven-pass G-type receptor, DSC relates to desmocolin, and DSG relates to desmoglein. The members of the cadherin family are highly related and have been shown to provide successful cell-cell adhesion through the formation of hetero-dimers (ie dimers formed from more than one member of the cadherin family). The polymorphisms and variants identified by the inventors are thought to be associated with domains in the gene/protein that are responsible for dimerization, and would therefore be expected to have relevance to all member so the cadherin family, and so relevant to all therapeutic methods of the present invention.

Consequently, for example, the cadherin of the present invention may be any one or more of:- N-cadherin (neuronal cadherin or Cdh2), E-cadherin (epithelial cadherin on Cdh-1), P-cadherin (placental cadherin or Cdh3), R-Cadherin (retinal cadherin or CDH4), Cdh5, K-cadherin (kidney cadherin or Cdh6), Cdh7, Cdh8, Cdh9, CdhlO, Cdhll, Cdhl2, Cdhl2P2, Cdhl3, Cdhl4, Cdhl5, Cdhl6, Cdhl7, Cdhl8, Cdhl9, Cdh20, Cdh22, Cdh24, Cdh26, CELSR1, CELSR2, CELSR3, DSC1, DSC2, DSC3, DSG1, DSG2, DSG3, DSG4, or any combination thereof. Each of the aforementioned cadherins may be contemplated individually as part of the present invention. For example, the cadherin could be Chd-1. As mentioned above, the DNA polymorphisms identified by the inventors are important as the basis for arriving at the structure of any an agent that blocks viral binding to cadherin, and particularly the aforementioned vaccines. The DNA polymorphism of the present invention may be any polymorphism in the cadherin gene that can confer susceptibility to viral infection in a subject depending on the allele of the polymorphism present in the subject. A polymorphism of the present invention has been found in a domain of cadherin that is associated with

dimerization of the cadherin, that is in the extracellular domain, and that are is a cadherin tandem repeat domain (ie tandem repeat domains are known to be present in the extracellular domain of cadherins and are thought to mediate cell-cell contact when bound to calcium). Consequently, the agent of the present invention that blocks viral binding may bind to an extracellular domain of the cadherin. That agent may bind to a cadherin tandem repeat domain (e.g. any of those provided in Table 1).

As an example, Table 1 provides an overview of the positions of the cadherin tandem domains in a number of cdhl orthologs.

Table 1. Location of Cadherin tandem repeat domains (CTRD] within some Cdhl orthologs. ND = Not determined (the fifth cadherin tandem repeat domain is absent or its sequence is too different from the consensus CTRD sequence to be detected].

Species Sequence identifier of CTRD CTRD CTRD CTRD CTRD gene* 1 2 3 4 5

Atlantic BT058864.1 163- 271- 383- 501- ND salmon 263 375 487 595

Human J304360.3 158- 266- 379- 492- 598- 258 356 458 591 686

Chicken _00 039258.2 164- 272- 386- 499- 604- 264 377 489 597 690

Pig NM 0011 630 60 . 1 158- 266- 379- 492- 604- 258 371 458 591 686 Cattle AY5G8164.1 158- 266- 379- 492- 604-

258 371 482 591 686

*A11 sequences are referred to by their Genbank VERSION identifiers. The identifiers given in the table pertain to NCBI-GenBank Flat File Release 205.0. The nature of cadherins means that they are found in many animals, and in each animal cadherins perform the same role of forming cell-cell adhesions. The utility of the present invention therefore naturally extends to the use in preventing or treating resistance to viral infection in any subject that includes endogenous cadherin. Consequently, when referring to the cadherin the reference relates to the protein product of the cadherin ortholog found in the subject that is subjected to the method of the present invention. When referring to the cadherin gene the reference relates to the cadherin ortholog found in the subject that is subjected to the method of the present invention. A non-exhaustive list of potential animal subjects is as follows: - human, cow, pig, crustacean, fish or poultry. If the subject is poultry, the poultry may, for example, be a chicken or turkey. If the subject is a fish, the fish may, for example, be cod, tilapia, carp, zebra fish, an Atlantic salmon (i.e. Salmo sala ), Pacific Salmon, trout (for example, a rainbow trout [Oncorhynchus mykiss]), or another species belonging to the family Salmonidae. If the subject is a crustacean, the subject may be shrimp.

Establishing the corresponding domains in which the polymorphisms of the present invention are found in any given animal is well within the ordinary abilities of the person skilled in the art. From a known domain in any given animal, one can identify corresponding domains in other animals. Taking the Cdh-1 as an example, one can carry out multiple alignment analysis of nucleic acid encoding, or amino acid sequences for, that protein (using, for example, Clustal analysis). See Figure 1 as an example.

Not wishing to be restricted further, but in the interests of clarity, taking Cdh-1 as an example of a cadherin of the present invention, further details of this cadherin are as follows Cdhl proteins are also known by the terms E-cadherin, epithelial cadherin, uvomorulin, CD324, CAM 120/180 or Cadherin-1; but are referred to hereinafter as Cdhl. Nomenclature for abbreviated gene names differ between species; here we have adopted the nomenclature used in zebrafish, where gene names are written in lower case italics (e.g. cdhl) and protein names are written with one capital letter (Cdhl). This nomenclature will be used throughout, irrespective of the species (ie irrespective of the fact that, for example, the human orthologs of chdl and Cdhl are CDH1 and CDH1).

The extracellular region of Cdhl is known to be involved in intercellular and intracellular homeotypic interactions and clustering, whereas its cytoplasmic tail is known to associate with an array of intracellular proteins, including beta-catenin, which mediates indirect binding of Cdhl to the cytoskeleton. The binding of Cdhl to beta-catenin is controlled by the Wnt signalling pathway and hepatocyte growth factor). The gene for Cdhl is considered to have tumour suppressor qualities, with mutations in this gene being linked to the progression of a number of cancers. Cdhl is also known to act as a receptor for the bacteria Listeria monocytogenes and the fungus Candida albicans. Some viruses interact with proteins involved in junctional complexes, but none is known to interact directly with CDH1 proteins (Yoon and Spear, 2002)

Multiple copies of the cdhl gene are normally found in the genome of animals. For example, the inventors have established that in the Atlantic salmon genome, three copies of cdhl are present. One of these copies is located on Atlantic salmon chromosome 11 (ie ssall), whereas the two other copies are located within a 50,000 base pair region on Atlantic salmon chromosome 26 (ssa26). The Atlantic salmon genome has not yet been annotated, so that the three different copies of the cdhl gene do not yet have standard names. Here, the two cdhl genes on ssa26 will be referred to as cdhl-1 and cdhl-2, while the cdhl gene on ssall will be referred to as cdhl-3. The corresponding protein products will be referred to as Cdhl-1, Cdhl- 2, and Cdhl-3. Cdhl, like other members of the Cadherin family, are found in many animals. In each animal Cdhl performs the same role of forming cell-cell adhesions. The utility of the present invention therefore naturally extends to the treatment of viral infection in any subject that includes endogenous Cdhl. Consequently, when referring to Cdhl, as part of the present invention, the reference relates to the protein product of the cdhl ortholog found in the subject that is subjected to the method of the present invention. When referring to cdhl the reference relates to the cdhl ortholog found in the subject that is subjected to the method of the present invention. Establishing the ortholog for cdhl in any given animal is well within the ordinary abilities of the person skilled in the art. For example, irrespective of species, a Cdhl protein may be defined as a protein, present in epithelial cells, that contains four or five copies of the cadherin tandem repeat domain (e.g. having identifier cdll304, cd00031, or pfam00028 in the GenBank Conserved Domains database), in addition to a Cadherin cytoplasmic region (GenBank Conserved Domains identifier pfam01049), and preferably that mediates the formation of calcium dependent cell-cell junctions. Alternatively, a protein-protein BLAST (BLASTP) against the GenBank RefSeq protein database may be used in order to identify a Cdhl ortholog; if more than 50 % of the best hits (i.e. those hits having an E-value of 0.0) are annotated as a cadherin-1 protein or a cadherin-l-like protein, then the query protein is a Cdhl ortholog ), and preferably that mediates the formation of calcium dependent cell- cell junctions.

The GenBank (release 205.0) identifiers of cDNA sequences of some cdhl orthologs can be found in Table 2. Deduction of the protein sequence corresponding to a cDNA sequence is well within the ordinary abilities of the person skilled in the art.

Therefore, only cDNA sequences (not protein sequences) are listed in Table 1.

Alternatively, or additionally the ortholog may be determined to share a 99, 98, 97, 96, 95, 90, 85, 80, 75 % sequence homology with any of those sequences identified in Table 2 while mediatating the formation of calcium dependent cell-cell junctions . However, without wishing to be restricted further, but in the interest of clarity, the Cdhl or cdhl may be any of those identified in Table 1 and that correspond to the subject as indicated in Table 1. The cadherin to which the agent of the present invention blocks viral binding may therefore be any of the protein products of the genes defined in Table 2. When the agent is exogenous cadherin, it may be any of the protein products of the genes defined in Table 2.

It should be noted that in some species, CDH1 is used as an acronym for another protein which is not an epithelial cadherin, the APC/C activator protein CDHl.

Within the present application, Cdhl, and cdhl,CO\{l and CDHl always refers to epithelial cadherin proteins/genes.

Table 2 - cdhl orthologs

*A11 sequences are referred to by their Genbank VERSION identifier. The identifiers given in the table pertain to NCBI-GenBank Flat File Release 205.0 (and, since the VERSION identifier is stable, most likely to any later GenBank release).

For example, the DNA polymorphism identified by the inventors can be a DNA polymorphism in cdhl, found by the inventor to have this predictive ability. The DNA polymorphism can be located on chromosome 26 and within a cdhl gene of Atlantic Salmon. Such a DNA polymorphisms has been shown to be capable of predicting resistance to IPN, more particularly by being causal mutations in the determination of resistance or non-resistance. The Atlantic salmon contains more than one cdhl gene. In humans, only one cdhl ortholog is present, and it is located on chromosome 16 (location 16q22.1). In chicken, only one cdhl ortholog is present, and it is located on chromosome 11. The DNA polymorphism may lie within the coding region of the cdhl. For example, in the case of Atlantic salmon, the DNA polymorphism resides within exon (ie as provided in GenBank (release 205.0) sequence BT058864.1).

The DNA polymorphism may also be any DNA polymorphism that changes the amino acid sequence of the corresponding Cdhl protein. In particular, the DNA polymorphisms may be any DNA polymorphism that causes an amino acid shift in a cadherin tandem repeat domain number 2 within the Cdhl protein,. Alternatively, the DNA polymorphism may be any DNA polymorphism that causes an amino acid shift in one of the other cadherin repeat domain present in the protein in question. Table 1 contains an overview of the positions of the cadherin tandem domains in some relevant cdhl orthologs.

Consequently, the DNA polymorphism of the present invention may be a DNA polymorphism at position 1065 in GenBank sequence BT058864.1 . When cytosine is present at position 1065 in GenBank sequence BT058864.1 the allele is a resistant allele. When thymine is present at position 1065 in GenBank sequence BT058864.1 the allele is a non-resistant allele. The agent of the present invention that blocks viral binding may therefore be an agent that binds to the variant form of cadherin that is the protein product of the polymorphism that includes the non-resistant allele. That agent may specifically bind to that variant (eg, not also bind to the variant from the cadherin that is the protein product of the polymorphism that includes the resistant allele).

When the agent is exogenous cadherin, the agent may be the variant form of cadherin that is the protein product of the polymorphism that includes the resistant allele. For avoidance of doubt, the sequences of the resistance and non-resistance sequence variants (alleles) of the DNA polymorphism corresponding to position 1065 in GenBank (release 205.0] sequence are: Resistance: CCAACATGTTTGTCATCAACCCTGTGACTGGAGGGATTCGG

(SEQ ID No. 1)

Non-resistance: CCAACATGTTTGTCATCAACTCTGTGACTGGAGGGATTCGG

(SEQ ID No. 2) Where the SEQ IDs refer to the associated Patentin generated sequence listing.

For other animals, the DNA polymorphism may be selected from any of the DNA polymorphisms corresponding to position 1065 in GenBank sequence BT058864.1. Alternatively, the DNA polymorphism may be any DNA polymorphism causing an amino acid shift at an amino acid position orthologous to the amino acid position being shifted bythe DNA polymorphism corresponding to position 1065 in GenBank sequence BT058864.1 Figure 1 displays a multiple alignment of Cdhl orthologues from several species, illustrating that the positions corresponding to 1065 in GenBank sequence BT058864.1 can easily and unequivocally be determined in species such as human and chicken. Table 3 displays the positions of nucleotides, within cdhl orthologs, corresponding to position 1065 in GenBank (release 205.0) sequence BT058864.1 and the positions of other nucleotides causing amino acid changes in the amino acid positions in question. Establishing the corresponding position for the polymorphisms of the present invention in any given animal is well within the ordinary abilities of the person skilled in the art. However, without wishing to be restricted further, but in the interest of clarity, the polymorphisms may be any of those identified in Table 3 and that correspond to the subject as indicated in Table 3. The agent of the present invention that blocks viral binding may therefore be an agent that binds to the variant form of cadherin that is the protein product of the polymorphism that includes the non-resistant allele as discussed above, for example those variants derived from the sequences of table 3. That agent may specifically bind to that variant (eg, not also bind to the variant from the cadherin that is the protein product of the polymorphism that includes the resistant allele).

When the agent is cadherin, the agent may be the variant form of cadherin that is the protein product of the polymorphism that includes the resistant allele as discussed above, for example those variants derived from the sequences of table 3.

Table 3: Polymorphisms within cdhl genes, causing amino acid changes in positions orthologous to the amino acid positions that are affected (changed] by DNA poymorphisms in position 1065 of GenBank (release 205.0) sequence BT058864.1 . The positions of the polymorphisms are relative to the beginning of the corresponding mRNA GenBank sequences.

*A11 sequences are referred to by the Genbank VERSION identifier. The identifiers given in the table pertain to NCBI-GenBank Flat File Release 205.0 (and, since the VERSION identifier is stable, most likely to any later GenBank release).

^!These positions are direct orthologs of position 1065 in GenBank sequence

BT058864.1 Each of the DNA polymorphisms is contemplated individually as part of the present invention.

If an Atlantic salmon has two resistant alleles at the locus corresponding to position 1065 in GenBank sequence BT058864.1, then that animal has resistance to viral infection. If an Atlantic salmon has two non-resistant alleles at the locus

corresponding to position 1065 in GenBank sequence BT058864.1, then that animal would not be determined to have resistance to viral infection. If an Atlantic salmon carries one resistance allele and on non-resistance allele, then the animal would be determined to be intermediate in terms of resistance.

The cdhl polymorphisms of the present invention have been found to correspond to variants in the amino acid sequence of Cdhl, whichcorrelate with resistance or non- resistance to viral infection (depending on the amino acid provided in the protein at that position of variance within the amino acid chain). Consequently, agents that block viral binding to the cadherins including the non-resistant variant may be useful in the present invention.

When the agent is cadherin, the agent may be the resistant variant.

Optionally, the variant is present in an extracellular portion of CDH1 .

For example, in Atlantic salmon the amino acid provided at position 325 in the amino acid sequence for Cdhl-1 has been found to direct resistance or non- resistance to viral infection in the subject. Therefore, determining which amino acid is provided at this position (ie determining the variant at ths position) will enable the investigator to determine if the subject is likely to have viral resistance or not. For other animals, the variant may be found in the corresponding position to position 325 in the amino acid sequence for Cdhl-1 (z^'e the sequence having

GenBank (version 205.0) identifier ACN10577.1 )). Establishing the corresponding position in the ortholog for any given animal is well within the ordinary abilities of the person skilled in the art. For example, the corresponding position in the ortholog may be determined by using multiple sequence alignment programs such as Clustal Omega. Figure 1 displays a Clustal Omega multiple sequence alignment of the Atlantic salmon Cdhl-lprotein sequences, together with Cdhl orthologs from human, chicken, pig and cattle, where the amino acids corresponding to position 325 in Cdhl-1 have been highlighted. It is evident from the figure that the positions corresponding to these two positions, in the different species, can be easily and unambiguously determined by visual inspection. Human,chicken, pig and cattle are evolutionary very distant from the Atlantic salmon, yet Figure 1 demonstrates a close inter-species homology. Therefore the corresponding positions should be identified with the same precision in other all animal species. The determination of the sequence of a protein is well within the ordinary skills of the art, given that the corresponding cDNA sequence is available. If the corresponding cDNA sequence is not available, it is well within the ordinary skills of the art to obtain that sequence, for example by screening a relevant cDNA library by PCR or hybridisation, using primers or hybridisation probes derived from a cdhl ortholog from a related species, followed by DNA sequencing of the PCR- or hybridisation-positive clones and validation by nucleotide BLAST against an annotated reference database (such as the NCBI RefSeq database). Alternatively, a relevant cDNA library could be sequenced using next-generation sequencing (RNA-seq), followed by the de novo of the sequence reads into a transcriptome sequence, followed by identification validation of cdhl genes using nucleotide BLAST against an annotated reference database.

However, without wishing to be restricted further, but in the interest of clarity, the variant may be any of those identified in Table 4 and that correspond to the subject as indicated in Table 4.

Not wishing to be restricted further, but in the interests of clarity, in salmon, when the variant is determined to be proline at position 325 in the amino acid sequence for Cdhll-1, the subject is predicted to have viral resistance. When the variant is determined to be serine at position 325 in the amino acid sequence for Cdhl-1, the subject is predicted not to have viral resistance. The agent of the present invention may therefore be an agent that binds to the variant form of cadherin that has serine at position 325 in the amino acid sequence for Cdhl. The agent may specifically bind to that variant (eg, not also bind to the variant from the cadherin that is the resistant form). When the agent is cadherin, the agent may be the variant form of cadherin that has proline at position 325 in the amino acid sequence for Cdhl.

Variant identify in other animals are described in Table 4. This information can be used to assist in the design of vaccine therapies according to the present invention. The amino acid sequence of Cdhl-1 can be found in line 1 ("ssa-1") of the multiple sequence alignment of Figure 1, position 325 being highlighted in this figure. The corresponding positions within proteins from other species, being orthologues of Cdhl.l, have been similarly highlighted in Figure 1.

Table 4 Variant Positions

*Only one amino acid variant is known at each position.

**A11 sequences are referred to by their Genbank VERSION identifiers. The identifiers given in the table pertain to NCBI-GenBank Flat File Release 205.0 (and, since the VERSION identifier is stable, most likely to any later GenBank release). In any species, the DNA polymorphism, or DNA polymorphisms, predicting the design of more effective vaccines may be any DNA polymorphism(s) that cause an amino acid shift in a Cdhl ortholog. In particular, the DNA polymorphism(s) may, in any species, be DNA polymorphisms causing amino acid shifts within the first or second cadherin repeats of a Cdhl protein. The cadherin repeats are extracelluar, Ca²⁺-binding domains within the Cdhl proteins, known to be involved in protein dimerisation. Position 325 within Cdhl-1 is located within the second cadherin domain of Cdhl-1,

As mentioned above, cadherin is known to have a central role in the formation of tight cell-cell adhesions. Consequently, after realising the link with viral resistance, it is reasonable to derive from the collected understanding of the physiological role of cadherin that viral resistance may be achieved by a variant form of cadherin that is able to maintain a tighter cell-cell adhesion than other forms of the protein, and thereby restrict the ability for virions to enter into inter-cellular spaces in order to locate their docking proteins. Consequently, the present invention would be expected to have broad applicability to any virus. Not wishing to be restricted further, but in the interest of clarity, the viral infection may be bursal disease (IBDV), infectious pancreatic necrosis (IPN), pancreas disease (PD), heart and skeletal muscle inflammation (HSMI), infectious salmon anaemia (ISA) and cardiomyopathy syndrome (CMS).

All agents are to be administered in a therapeutically effective amount. If an agent that increases the cadherin concentration is administered along with an agent for blocking viral binding to cadherin, the therapeutic amount of the agent for blocking viral binding is greater than if that agent were administered alone.

The present invention may include the identification of those non-resistant individuals by predicating resistance to viral infection by determining in a subject that variant present at one or more amino acid position in the cadherin protein, and/or alleles present at one or more DNA polymorphism in the gene, and predicting whether the subject is resistant to viral infection based on the determination of the variant and/or alleles, followed by the step of administering any of the agents of the present invention to those subjects identified as being non- resistant. Alternatively, the agents are not-administered to those subject found to be resistant.

The term "ortholog" takes its normal meaning as would be well understood by the person skilled in the art. However, for the avoidance of doubt, the term may mean any gene found in one species that corresponds to a gene found in another species and wherein both genes share the same function (eg ability to form cell-cell adhesions) and share a common ancestor. The present invention will now be described by way of example with reference to the accompanying figures, in which: -

Figure 1 provides a multiple alignment (made using Clustal Omega version 1.2.0) of cadherin proteins from different species. 'Ssa-1' is the protein sequence of Cdhl-1 from Atlantic salmon, while 'Chicken', 'Human', 'Pig', and 'Cattle' are, respectively, the protein sequences of Cdhl (epihelial cadherin) orthologs from Gallus gallus, Homo sapiens, Sus scrofa and Bos Taurus. The GenBank (release 205.0) identifiers of the salmon ('Ssa-1'), Chicken, Human, Pig, and Cattle sequences of Figure 1 are, respectively, ACN1G577.1, P08641.2, NP_004351.1, NP_001156532.1, and

NP_001002763.1.

Figure 2 provides a graph showing the relationship between true genotype and assigned-to genotype (the genotype the animal was assigned to using the DNA polymorphism corresponding to position 1065 of BT058864.1 (BT058864.1_1065- TC)), for 340 animals from the AquaGen breeding nucleus.

Figure 3 shows the IPN virus levels (Ct-values after Taq-Man® real-time assay for IPNV) in fish of different genotypes (at the DNA polymorphism corresponding to position 1065 of BT058864.1 (BT058864.1J.065-TC)) sampled at day 2, 4, 6, 7, 8, 10 and 12 after experimental challenge. Given the threshold of the diagnostic test, samples are negative after 37 cycles, and for display purposes negative fish are given a value of 1. Values are normalized against the expression of the reference gene elongation factor 1 -alpha, qq = no copies of the resistance allele at BT058864.1_1065-TC, Qq = one copy of the resistance allele at BT058864.1J.065- TC, QQ = two copies of the resistance allele at BT058864.1_1065-TC.

Figure 4 shows an immunofluorescence analysis of cdhl of liver cells. Liver slices from qq and QQ Atlantic salmon (susceptible and resistant, respectively, based on analsysis using the DNA polymorphism corresponding to position 1065 within GenBank sequence BT058864.1 (BT058864.1J.065-TC)) were cultured and infected with IPNV in vitro. Visualization of IPN virions was carried out using

immunofluorescence with a polyclonal rabbit antibody specific to IPNV. Positive signals appear as white spots or aggregates in heavily infected cells. A) Susceptible fish (qq) show a widespread infection with virions on the surface cells as well as in hepatocytes located inside the tissue slice. B) In resistant animals with SNP4 (QQ), only a few IPNV particles trapped on the surface of the slice are visible. Figure 5 shows mortality of AquaGen Atlantic salmon families the 2005 year class related to an outbreak of pancreas disease at a sea site in western Norway. The families are ranked according to their breeding values for IPN resistance. 1: low, 2: medium to low, 3: medium to high or 4: high breeding values for IPN resistance. Figure 6 shows mean rate of mortalities from pancreas disease within offspring of individual parents, categorized according to parental genotypes for IPN resistance qq = no copies of the resistance allele at BT058864.1_1065-TC, Qq = one copy of the resistance allele at BT058864.1_1065-TC, QQ = two copies of the resistance allele at BT058864.1_1065-TC.

Figure 7 shows a histopathological score of heart muscle and exocrine pancreas of fish of differing genotype to the IPN QTL 28 days after challenge with PD virus (SAV- 3 variant). Figure 8 shows results of a co-immunoprecipitation study (CO-IP) that verify direct binding of Cdh-1 to IPNV. 1. Constructing a reference sequence of the QTL region

A QTL for resistance to IPN was earlier detected on Atlantic salmon chromosome 26 (Houston et al. 2008, Moen et al. 2009). At the onset of the study described in the present application, we assumed that the QTL was caused by a single, underlying (causative) mutation. Furthermore, we assumed that this causative mutation had two alleles, one allele (Q) that is associated with resistance to IPN and one allele (q) that is associated with susceptibility (non-resistance). Thus, individual animals could be assigned (through the analysis of extensive genetic data sets) a QTL genotype reflecting which combination of QTL alleles (QQ, Qq, or qq) the animals harboured.

The QTL region was defined as the region of the Atlantic salmon genome wherein it was expected, based on earlier results (Moen et al., 2009), that the causative mutation(s) underlying the QTL would be located with a close to 100 % probability. More specifically, the QTL region was defined as the region in between the SNPs ESTNV_31602_808 and GCR_cBin30387_Ctgl_91 on the Atlantic salmon SNP linkage map (Lien et al., 2011). The Atlantic salmon genome had at the time not been sequenced (although a draft of the genome sequence has recently become available). In order to identify the causative mutation(s), we therefore needed to construct a reference sequence of the QTL region. Bacterial Artificial Chromosome (BAC) clones matching the two above-mentioned SNPs were isolated from an existing BAC library (Thorsen et al., 2005). On the basis of a physical map made from this library (www.asalbase.org), a minimum tiling path of 31 BACs was made, connecting the two flanking BACs¹. Atlantic salmon genomic DNA was extracted from each BAC. An individually tagged paired-end library (with average insert size

¹ The minimum tiling path consisted of the following BACs: S0042J22, S0004K18, S0161O04, S0243D12, S0076E15, S0021H01, S0162F10, S0258L08, S0119L01, S0026N22, S0162J03, S0259M06, S0120O19, S0048P16, S0170B06, S0262M03, S0126K07, S0063G22, S0201A04, S0282P22, S0457C13, S0066E05, S0215J07, S0344A15, S0001F22, S0115B04, S0227H08, S0449E20, S0001N03, S0160J02, S0236E20. Information regarding the BACs can be found on www.asalbase.org. 350 bp) was made for each BAC DNA sample, whereupon the samples were sequenced on a HiSeq 2000 machine (Illumina Inc., San Diego, USA) to an average depth of approximately 800 times haploid genome coverage. Following removal of residual adapter sequences, discarding of too-short reads, trimming of the ends of poor quality reads, and matching of paired-end reads, a de novo assembly was made within each BAC using the 'clc_novo_assemble' program from the CLC Assemble Cell suite (CLC Bio, Aarhus, Denmark). Phrap version 1.090518 (http://phrap.org.) was then used to assemble individual BAC contig sequences into a set of contigs spanning all BACs. Finally, the contigs from this reference were combined into one contiguous genomic sequence by aligning it with scaffolds from a preliminary version of the Atlantic salmon genome sequence (which had been made in-house, using the Celera Assembler software, based on the data from the first 27 batches of sequences submitted by the sequencing project into the NCBI Trace Archive). This genomic sequence will hereafter be referred to as the 'Reference Sequence'.

2. Selection of animals for next-generation sequencing

Forty-five Atlantic salmon from the Norwegian breeding nucleus of Aqua Gen AS were chosen for massive parallel sequencing on Illumina HiSeq 2000. These 45 salmon were all selected from among the parents of the 2005 and 2008 year classes of the Aqua Gen breeding nucleus. All salmon in the Aqua Gen breeding nucleus are descendants of salmon taken from Norwegian rivers. The selection of animals was done as described in the following paragraph. Four hundred and fifty-four full-sib groups of Atlantic salmon fry were IPN- challenged in individual tanks shortly after start feeding (the protocol for the challenge test can be found in Moen et al., 2009). The parents of these full-sib groups (hereafter referred to as 'the mapping parents') would become very central to the study described in this application. Full-sib groups consisted of 103 fish (on average), and tissue samples were collected from the 10 first-to-die within the group as well as 10 survivors (or 10 last-to-die), whereupon DNA was extracted using the DNAeasy kit from QIAGEN (QIAGEN, Venlo, the Netherlands). From 206 selected full-sib groups, affected and surviving offspring were genotyped with three microsatellite markers located within the QTL region, Alu333, Ssa0384BSFU/ii and Ssa0285BSFU, whereupon the linkage phase between alleles of the three microsatellites were identified in each mapping parent using the observed co- segregation of alleles from parents to offspring (genotyping of microsatellite markers are discussed in more detail in abovementioned Moen 2009 article). This genotyping was done in an iterative fashion so that, ultimately, almost all full-sib groups that were likely to have at least one QTL-heterozygous parent (see below) were genotyped. A chi-square test was applied in order to test for co-inheritance of the three-microsatellite haplotype and the affected/resistant phenotype, leading to the identification of 110 QTL-heterozygous mapping parents. Using data from these QTL-heterozygous mapping parents, a table was created linking alleles at the three- microsatellite haplotype to QTL alleles. (If a three-microsatellite allele was found to be linked to both Q and q, only the most prevalent linkage phase was entered into the table). This table was next used to extrapolate QTL genotypes in the mapping parents found to be QTL homozygous, as well as for other animals from the Aqua Gen breeding nucleus. Twenty-two Aqua Gen animals deduced in this way to have the QTL genotype QQ (i.e. expected to provide good IPN resistance), as well as 23 Aqua Gen animals likewise found to have the qq genotype (i.e. expected to provide poor IPN resistance), were chosen for subsequent whole-genome sequencing. These sets of 22 and 23 animals were put together in such a way as to minimise the relatedness of animals within group, by maximising the diversity of three- microsatellite alleles within each group. 3. Narrowing down QTL region by next-generation sequencing

The above-mentioned 23 QQ animals and 22 qq-animals were sequenced using HiSeq2000 technology from Illumina. Individually tagged paired-end libraries were made from each sample, before samples were pooled for sequencing. A total of 264 x 10⁹ reads was produced, corresponding to a per-animal coverage of two times the haploid genome. The reads were aligned to the reference sequence using bowtie2 with default settings (both unpaired- and paired-end alignments were performed), thereafter removing alignments having Phred-scaled p-value below 30, using the 'view' function of samtools. The reads were aligned pool-wise, i.e. all reads from QQ animals were part of one assembly, and all reads from qq animals were part of another assembly. SNPs and short insertion/deletions (indels) were identified with freebayes (http://arx¾v^,.org/abs/1207.3907) using the following string to set the value of parameters relevant to the alignment: "-use-best-n-alleles 2 --read-max- mismatch-fraction 0.02 -min-alternate-total 3 -no-marginals -left-align-indels - pooled -ploidy 90" (see h ¾s: /^ for explanations of the parameters). A Fisher's exact test was used in order to test for independence between QTL allele and SNP/indel allele within reads. The SNPs with the most significant statistics from this test were genotyped in the 110 QTL- heterozygous mapping parents mentioned above, as well as in the challenge-tested offspring of those animals, and a Fisher's exact test was performed on parental haplotypes derived from this data set in order to test for independence between SNP alleles and QTL alleles. The correlation coefficient (r²) between alleles at the SNP and at the QTL, a measure of the degree of linkage disequilibirum (LD) between loci, was also calculated for each SNP, using the 'LD' function of the 'genetics' module of the R statistical program suite. These analyses identified one SNP and two indels whose genotypes were most strongly (among all the tested polymorphisms) associated with deduced genotypes at the QTL. These polymorphisms were in perfect linkage disequilibrium (LD) with each other, and their co-segregation with the QTL was highly significant (P-value = 1.62 x 10 ²⁸, Fisher's exact test for association between QTL-allele and SNP/indel-allele in haplotype copies derived from QTL-heterozygous mapping parents). The correlation coefficient (r²) between alleles at the QTL and alleles at the SNP/indel was 0.57. The SNP and the two indels resided within a 26 kbp region that contained two full-length genes, one epithelial cadherin gene [cdhl-26-1] and the gene fam96b. In addition, a truncated (lacking several exons) epithelial cadherin gene [cdhl-26-2] was noted upstream of cdhl-1.

4. Identification of BT058864.1 1065-TC Cdhl-1 is a salmon version, located on chromosome 26, of a protein known to be responsible for Ca²⁺-dependent cell-cell adhesion in epithelial tissue in a variety of species. As noted above, we identified one SNPs and two indels in close proximity to this gene. These three polymorphisms were the subject of our earlier patent application.

Cdhl-1 was amplified and sequenced, using cDNA from 29 different QQ or qq Atlantic salmon from the Aqua Gen population as template. The QTL genotype of the animals had been deduced using a three-microsatellite haplotype as described above (see section 2). The cDNA was derived from samples taken from adult salmon or from salmon fry sampled during a challenge test for IPN-resistance. Eleven different fragments were PCR-amplified, using 11 different primer pairs. The PCR primers were constructed so that there were, for every primer, at least two nucleotide mismatches between the primer sequence and the nucleotide sequence of its 'false' alignment target on chromosome 11 (i.e. the gene cdhl-11). Table N contains the sequences of all primers used as well as other details on the PCR reactions. The PCR products were direct sequenced using the the BigDye Terminator v3.1 Sequencing Kit (Applied Biosystems, Carlsbad, USA) on an Applied Biosystems 3730 DNA Analyzer. The phredPhrap.pl script, employing Phred and Phrap (http://wwvv.phr¾p.org/phredpbrapconsed,htmj). was used for base calling, quality value assignment, and alignment of reads. PolyPhred (j i ://droog\gs.wasbi gtorLedu/poiypbred/) was used for variant detection, and consed ( tt .^ w ^ was used for visualization of contigs and PolyPhred-detected variation.

Based on the DNA sequences of the amplified fragments, a SNP (BT058864.1_1065- TC) emerged as displaying a particularly large contrast between the two genotype groups. BT058864.1_1065-TC had genotype TT in all 17 qq animals, and genotype CC in all but two of the 12 QQ animals (the deviant QQ animals had genotypes TT and CT). BT058864.1_1065-TC was thereafter genotyped in 340 parents of the 2005 year class of the Aqua Gen breeding population, including most of the QTL- heterozygous mapping parents mentioned above (section 2); these 340 animals had all been given a 'true' QTL genotype as described above (section 2). Assuming that BT058864.1_1065-TC genotypes TT, CT, and CC corresponded to QTL genotypes qq, Qq, and QQ, respectively, 306 out of 340 animals were assigned to the correct genotype (figure 2). These results showed that genotypes at BT058864.1_1065-TC are strongly correlated to genotypes at the QTL.

Based on the same data, BT058864.1_1065-TC was also found to be in perfect linkage disequilibrium with the three above-mentioned DNA-polymorphisms already found to be strongly correlated to IPN (see section 3). Unlike these 3 DNA polymorphisms, however, BT058864.1_1065-TC leads to a change in the structure of CDHl-26-1. The SNP gives rise to a proline-to-serine shift on a surface-exposed part of the second extracellular cadherin (EC) domain of the protein. Proline is an amino acid that puts conformational restrictions on the proteins it is a part of, due to the unusual structure of the amino acids side chain. It is therefore likely that the proline-to-serine amino acid shift will have major consequences for the properties of the protein.

5. Demonstrating activity of Cadherin on virus in experimental challenge A challenge test was performed on Atlantic salmon fry descending from the Aqua Gen breeding population. Fry at an average weight of 0.2 grams at start feeding were transferred to the research station (Havbruksstasjonen, Troms0, Norway) and allowed one week of acclimation before challenge. The trial was conducted by Nofima AS (Troms0, Norway).

Sixteen hundred fry belonging to three different groups of fish were included in the study, which was performed in eight different tanks, 200 fry in each tank. The three groups differed in the frequency of the IPN resistance marker (ie BT058864.1_1065- TC); Resistant group, frequency of beneficial allele 0.84, Intermediate group, frequency of beneficial allele 0.45, and Susceptible group, frequency of beneficial allele 0.22. The fry were bath challenged by adding infectious pancreatic necrosis virus at a concentration of 10^s TCIDso/ml water. Normal flow was paused while the water was aerated, resuming flow after three hours. Fish were sampled at day 2, 4, 6, 7, 8, 10 and 12 post challenge. The livers of the sampled fish were carefully dissected and placed in RNAlater (Qiagen) for later determination of viral load (see below). The trial was terminated 34 days after challenge. At termination, 10 surviving fry from each of the eight tanks were sampled for determination of carrier status. The tails of all sampled fish were frozen before DNA extraction and subsequent deduction of QTN-alleles. In the Resistant group, the mortality associated with IPN was only 1.3 %. In contrast, the IPN-related mortalities in the Intermediate and Sensitive groups were 22.3 % and 52.5 %, respectively demonstrating how marker-assisted selection significantly reduce the number of IPN-induced mortalities. In order to assess whether the fish were able to resist infection or merely to survive an infection, IPN virus was quantified in survivors from the challenge test, using a Taq-Man real-time RT-qPCR assay provided by an accredited commercial lab (PatoGen Analyse AS, Alesund, Norway) and compared to the expression of a reference gene (elongation factor 2 -a). Among survivors from the Sensitive, Intermediate, and Resistant groups, 90 %, 60 %, and 10 % were carriers of the virus. A similar trend was found in samples taken from days 2, 4, 6, 7, 8, 10 and 12 post infection (dpi), with only qq fish displaying the high virus levels characteristic of diseased/moribund fish ( see Figure 3). QQ fish exclusively found in the Resistant group were predominantly negative, except for a few positive fish with very low virus levels, whereas heterozygous fish (representing both alleles) had intermediate levels. The results thus indicated that the Q alleles act by preventing the virus from infecting cells, rather than helping the fish to survive in the presence of infection.

6. Testing IPNV infection; demonstration that cadherin prevents viral infection.

The inventors have shown that Atlantic salmon that have been predicted, as described above, to be resistant to IPN, are much more resistant to IPN than Atlantic salmon that have been predicted not to be resistant to IPN. Furthermore, they have shown that the IPN virus does not enter hepatocytes cultured in vitro from resistant salmon, whereas it does enter hepatocytes similarly cultured from susceptible cells.

To test IPNV infection in vitro, resistant and susceptible salmon (identified as such by the methods of the present invention) were sacrificed and liver biopsies were embedded in 2.5% ultra-low melt agarose dissolved in Hanks buffer. Liver sections of 250 mm were prepared in ice cold Hanks buffer using a vibratome (Compressotome VF300, Precisionary Instruments, USA) and transferred to L15 Glutamax with 1% PenStrep (Life Technologies, USA). After 24 hours in culture at 15oC, 90% of the surface cells and all cells within the tissue were viable, respectively. The organotypic liver cultures were challenged with IPNV and incubated overnight, washed in ice cold Hanks buffer and then fixated using freeze substitution at -lOOoC. Briefly, the section were frozen in isobutanol at -100°C for 3 minutes and then transferred to pure ethanol, also at -100°C. The sections were slowly brought to room temperature after at least 3 days at -80°C. Immunofluorescence was carried out using antibodies against IPNV (1), Cdhl (Dako, Denmark) and Clathrin (Abeam, United Kingdom). Briefly, the sections were rehydrated and blocked with 4% dry milk in PBS with 0.4% saponin (PBSS). After overnight incubation with primary antibodies diluted in PBSS, Alexa conjugated secondary antibodies (Life Technologies) were applied for 2 hours after extensive washing. The sections were mounted with 2,2'-thiodiethanol after washing with PBSS and examined at a fluorescence microscope (Zeiss Axio Observer, Carl Zeiss Microimaging GmbH, Germany). Images of sections from susceptible fish showed a widespread IPNV infection (see Figure 4). The virus were located throughout the tissue slice and showed intracellular presence. On the surface of the hepatocytes, IPNV co-localized with cdhl and clathrin in early clathrin coated pits forming early endsomes. These findings imply that IPNV enters cells through binding to cdhl and subsequent entry depend upon clathrin mediated endocytosis. Analysis of liver section from resistant fish showed presence of only a few virus and only on the surface of the section. These virus are presumably trapped by cellular debris. IPNV were never detected within the tissue and no co-localization was observed between IPNV and cdhl or clathrin. 7. Pancreas Disease Study

9.1 Introduction pancreas disease

Pancreas disease (PD) is a viral fish disease with significant impact on salmon aquaculture in Norway, Scotland and Ireland. The disease is caused by a member of the Togaviridae family, an alphavirus, named salmonid alphavirus (SAV). Some strains of the virus are causing a related disease (sleeping disease) in farmed rainbow trout in several European countries. Outbreaks of PD can occur at all stages of the marine production cycle of Atlantic salmon (McLoughlin & Graham, 2007), and mortality rates can range from 0% up to 50% mortality. Clinical signs are characterized by sudden loss of appetite, lethargy, the appearance of fecal casts in the net pen and an increase in the mortality. A significant share of survivors will fail to grow and become runts. Histopatological examination of fish suffering from PD shows pancreatic necrosis, myocardial degeneration as well as inflammation of the heart and skeletal muscle. Vaccination has proven to give some, but not full, protection from the disease.

7.2 field study

21725 fish from 394 families were transferred to a sea site on the west coast of Norway in May 2006, in a region where the SAV-3 variant of the PD virus is known to have an endemic distribution. The families originated from approximately 400 parents, each male having been mated to two dams and vice versa. The families were ranked according to their breeding value for IPN resistance based on the results in a previous IPN challenge study, as having: 1: low, 2: medium to low, 3: medium to high or 4: high breeding values for IPN resistance. An outbreak of pancreas disease started from April 2007, resulting in the total loss of 13,6% of the fish at the site. The total mortality of the population as well as the PD specific mortalities was lowest in the families of high or medium to high IPN resistance (Figure 7). Tissue samples were harvested from approximately 50% of the mortalities. Individual samples were assigned to parental pairs using a panel of 9 microsatellite markers, 936 animals being successfully assigned to a parental pair. The within-family mortality rate ranged from 0% to 36%. Animals with two copies of the high-resistance allele for BT058864.1_1065-TC appeared to be more resistant to PD than animals carrying one or two copies of the low-resistance allele (Qq or qq) (p-value = 0.12) (Figure 6).

8. Challenge study to test the protection of the IPN QTL markers against pancreas disease

During October 2011, broodfish previously genotyped for the IPN QTL genotype markers were selected as parents for a batch of test fish(ie using the IPN resistance marker BT058864.1_1065-TC). 6 females of the qq genotype and 4 females of the QQ genotype were used along with 4 males of the qq genotype and 6 males of the QQ genotype. 100 crosses among this fish were made, resulting in batches of offspring carrying the qq, Qq or QQ genotypes. Only offspring of the qq and QQ genotypes were used for a pancreas disease challenge study. The fish were reared in the freshwater hatchery of AquaGen Kyrksaeter0ra Norway, pit-tagged and vaccinated using a commercial vaccine containing only bacterial antigens before transport to the VESO Vikan reseach station at approximately 75 grams. The challenge was performed in freshwater at 15°C. 60 fish of each genotype was kept in the same tank, along with 20% shedder fish previously injected with a Norwegian SAV-3 isolate of the PD virus. 12 fish from each genotype were sampled 18 days post challenge, and a PD-infection was confirmed by real-time RT-PCR detection of the SAV-3 virus in the hearts of fish of all of the three genotype categories. At 28 days post challenge, the heart, pancreas and muscle (red and white muscle) was sampled from the remaining fish in the tank (48 fish in each group of fish) for examination by histopathology. There was no mortality from PD during the challenge period. Samples were blinded and prepared, examined and scored for histopathology by Aquatic Veterinary Services, Belfast, Ireland, using a semi-quantitative scoring system ranging from 0 to 4 depending on the severity of the lesions. A statistical examination was performed on the results, demonstrating a significantly lower frequency of heart lesions in fish of the Qq and QQ genotypes. There was also a tendency toward lower frequency of pancreas and muscle lesions in the Qq and QQ fish(Figure 7). Thus, it was confirmed that fish carrying the IPN-QTL demonstrate a higher resistance also to the PD -virus. 9. Viral/Cadherin binding study

CO-immunoprecipitation was carried out using an IPNV specific polyclonal antibody bound to Protein A coated magnetic beads. Briefly, a protein lysate was prepared from a salmon liver and mixed with IPNV for 30 min, before adding the coated magnetic beads and incubation for 30 min. After selective extraction of protein- IPNV bound to the magnetic beads, the eluate was subjected to a SDS-PAGE gel electrophoresis together with a liver protein lysate (positive Cdhl-1 control) and a negative control co-immunoprecipitation reaction without IPNV. After gel separation the proteins were transferred to a membrane and Cdhl-1

immunoprecipitated by IPNV was detected using a specific antibody for Cdhl- 1. .The presence of Cdhl-1 in the gel clearly demonstrates that IPNV binds to the Cdhl-1 protein. See the results in figure 8. References

Granzow H, Weiland F, Fichtner D and Enzmann PJ (1997) Studies of the ultrastructure and morphogenesis of fish pathogenic viruses grown in cell culture. Journal of Fish Diseases 20: 1-10

Falk K, Namork E, Dannevig BH (1998) Characterization and applications of a monoclonal antibody against infectious salmon anaemia virus. Dis Aquat Organ 8: 77-85. Houston RD, Haley CS, Hamilton A, Guy DR, Tinch AE, Taggart JB, McAndrew BJ, Bishop SC (2008) Major quantitative trait loci affect resistance to infectious pancreatic necrosis in Atlantic salmon (Salmo salar). Genetics 178: 1109-15. Houston RD, Davey JW, Bishop SC, Lowe, NR, Mota-Velasco JC et al. (2012) Characterisation of QTL-linked and genome-wide restriction site-associated DNA (RAD) markers in farmed Atlantic salmon. BMC Genomics 13: 244. Kuznar J, Soller M, Fabias G, Espinoza JC (1995) Attachment and entry of infectious pancreatic necrosis virus (IPNV) into CHSE-214 cells. Arch Virol 140: 1833-40.

Lien S, Gidskehaug L, Moen T, Hayes BJ, Berg PR, Davidson WS, Omholt SW, Kent MP (2011) A dense SNP-based linkage map for Atlantic salmon (Salmo salar) reveals extended chromosome homeologies and striking differences in sex-specific recombination patterns. BMC Genomics 12: 615.

Madsen and Jensen (2008) DMU: a user's guide. A package for analysing multivariate mixed models, version 6, release 5.0. University of Aarhus, Tjele, Denmark.

Moen T, Hayes B, Baranski M, Berg PR, Kj0glum S, Koop BF, Davidson WS, Omholt SW, Lien S (2008) A linkage map of the Atlantic salmon (Salmo salar) based on EST- derived SNP markers. BMC Genomics 9: 223.

Moen T, Baranski M, Sonesson AK, Kj0glum S (2009) Confirmation and fine-mapping of a major QTL for resistance to infectious pancreatic necrosis in Atlantic salmon (Salmo salar): population-level associations between markers and trait. BMC Genomics 10: 368.

Shifman S, Kuypers J, Kokoris M, Yakir B, Darvasi A (2003) Linkage diseuilibrium patterns of the human genome across populations. Human Molecular Genetics 12: 771-776. Thorsen J, Zhu B, Frengen E, Osoegawa K, de Jong, PJ, Koop BF, Davidson WS, H0yheim B (2005) A highly redundant BAC library of Atlantic salmon (Salmo salar): an important tool for salmon projects. BMC Genomics 6: 50.

Yang YH, Wang JL, Miranda CL, Buhler DR (1998) CYP2M1: cloning, sequencing, and expression of new cytochrome P450 from rainbow trout liver with fatty acid (omega-6)-hydroxylation activity.

Claims

1. A composition for use in the prevention or treatment of viral infection, wherein the composition is:- a. an agent that blocks viral binding to cadherin;

b. an agent that increases cadherin concentration at the cell-cell junction; and/or c. an agent that increases efficacy of cadherin mediated cell-cell binding.

2. A composition as claimed in claim 1, wherein the agent that blocks viral binding to cadherin is a vaccine that induces an immune response against the epitope, on the virus, that is responsible for binding to the cadherin protein.

3. A composition as claimed in any of the preceding claims, wherein the agent is an

antibody or small molecule that antagonises the binding of a virus to its binding site on cadherin.

4. A composition as claimed in claim 1, wherein the agent that increases cadherin

concentration at the cell-cell junction is an inhibitor of transcription factors that supress cadherin expression.

5. A composition as claimed in claim 1, wherein the agent that increases cadherin

concentration at the cell-cell junction is a histone deacetylase inhibitor.

6. A composition as claimed in claim 1, wherein the agent that increases cadherin

concentration at the cell-cell junction is a TGF-β antagonist.

7. A composition as claimed in claim 1, wherein the agent that increases cadherin

concentration at the cell-cell junction is cadherin.

8. A composition as claimed in claim 1, wherein the agent is cdh-1 and includes proline at position 325.

9. A composition as claimed in claim 1, wherein the agent is an agent capable of increasing extracellular calcium ions.

10. A composition as claimed in claim 9, wherein the agent is a dietary source of calcium ions.

11. A composition as claimed any of the preceding claims, wherein the virus is a

birnavirus.

12. A composition as claimed in claim 11, wherein the virus is IPN.

13. A pharmaceutical composition comprising any one or combination of the compositions recited in any of the preceding claims and one or more pharmaceutically acceptable excipients.

14. A composition comprising any one or combination of the compositions recited in any of the preceding claims for use in therapy.

15. A method for preventing or treating viral infection, wherein the method includes the step of administering an effective amount of any one or combination of the compositions recited in any of the preceding claims to the subject to be treated.