In vitro method for predicting whether cattle are susceptible or resistant to paratuberculosis according to the presence of specifical deletereous genetic variants

Abstract

The present invention refers to an in vitro method for predicting whether cattle are susceptible or resistant to paratuberculosis (PTB), or for classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB; and / or for preventing susceptibility to PTB or to increase resistance to PTB.

Description

[0001] IN VITRO METHOD FOR PREDICTING WHETHER CATTLE ARE

[0002] SUSCEPTIBLE OR RESISTANT TO PARATUBERCULOSIS ACCORDING TO THE PRESENCE OF SPECIFICAL DELETEREOUS GENETIC VARIANTS

[0003] FIELD OF THE INVENTION

[0004] The present invention refers to the medical field. Particularly, the present invention refers to an in vitro method for predicting whether cattle are susceptible or resistant to paratuberculosis (PTB), or for classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB; and / or for preventing susceptibility to PTB or to increase resistance to PTB.

[0005] STATE OF THE ART

[0006] Mycobacterium avium subsp. paratuberculosis (MAP) is the causal agent of PTB, a chronic intestinal inflammatory disease that affects cattle worldwide and has important economic implications. PTB is associated with weight loss, decrease in milk production, and premature culling and causes estimated losses of US$198M in the USA and USS364M in Europe. Furthermore, there is evidence suggesting that MAP might act as an environmental trigger of chronic inflammatory diseases such as Crohn s disease and autoimmune diseases in humans, including multiple sclerosis, Type-1 diabetes mellitus, and rheumatoid arthritis. This potential threat to human health has stimulated interest in the development of more sensitive PTB diagnostic and control methods.

[0007] PTB pathogenesis can be classified into four stages: silent, subclinical, clinical, and advanced clinical. MAP infects domestic and wild ruminants by ingestion of contaminated water or feces and is phagocytosed by host macrophages which present MAP antigens to CD4 naive T cells via the major histocompatibility complex II (MHCII). The activation of CD4+ cells results in the induction of a Th 1 immune response, interferon-gamma (IFNy) secretion, and recruitment of more macrophages and lymphocytes to the site of infection. However, MAP can persist within infected macrophages due to its capacity to interfere with the binding of phagosomes and lysosomes, and to inhibit apoptosis and antigen presentation through down-regulation of MHCII. After the establishment of MAP within host macrophages, the subclinical stage begins and is characterized by the development of small focalized lesions formed by 5 to 30 macrophages and a few lymphocytes. After 2-5 years, the granulomatous lesions become more severe and an inflammatory infiltrate causes a marked thickening of the wall and the appearance of clinical signs such as diarrhea, weight loss, and decreased milk production. Host immune response against MAP infection depends on several factors including mycobacterial factors (load, strain), host factors (genetic susceptibility or resistance, breed, immunity status), and environmental factors (stress factors, management). A better understanding of the molecular mechanisms responsible for the host response against MAP is needed to design new strategies to reduce MAP prevalence. One generally accepted control strategy is the breeding of cattle that are genetically resistant to MAP infection. Genome-wide association studies (GWAS) have been used to identify single nucleotide polymorphisms (SNPs), loci, candidate genes, molecular functions, and metabolic pathways associated with susceptibility, tolerance, and resistance to MAP infection. Although SNPs associated with several PTB phenotypes have been identified, they are normally located in non-coding regions, so pinpointing the direct causal effect of these SNPs is still a challenge. In this context, the identification of functional mutations in gene coding regions specifically present in cows without lesions or with a specific type of PTB-associated lesion could shed light on which candidate genes, are associated with a specific type of response against MAP infection.

[0008] RNA sequencing (RNA-Seq) is a next-generation sequencing technique that allows the global quantification of mRNA levels in a cell or tissue and is used to compare gene expression levels across phenotypically divergent groups. Other uses of RNA-Seq data include non-coding RNA analysis, novel mRNA isoforms discovery, and alternative splicing analysis. RNA-Seq generates sequences on a very large scale at a reduced cost than the Sanger sequencing. Taking this into account, RNA-Seq has been used to identify SNPs in expressed coding regions of different animal species, including cattle. Several methods for SNP calling using RNA-Seq data have been developed and compared such GATK, Samtools, SNPiR, and CLC BioGenomics Workbench. Additionally, it has been demonstrated the improved sensitivity of joint genotyping calling using GATK compared to individual calling. This approach offers the possibility of calling DNA variants on cohorts of samples.

[0009] SNPs in the coding region of genes (cSNPs) are divided into two types: synonymous and nonsynonymous. Nonsynonymous SNPs are divided into two types: missense and nonsense. Missense cSNPs produce a variation in the amino acid sequence and have the ability to alter the structure and function of a protein, thereby affecting disease outcomes. However, for the vast majority of cSNPs no experimental evidence is currently available to substantiate their deleterious effects. Missense cSNP prediction tools try to ascertain if a cSNP will affect a protein's function using public databases. Prediction tools used a large range of structure attributes to separate deleterious from neutral cSNPs. However, structural attributes are not available for all cSNPs. Sequence attributes usually identify important residues using information from homologous proteins. With enough homologs (~10), sequence attributes can often compete effectively with structural approaches and allow the identification of cSNPs affecting protein structure and function.

[0010] In a nutshell, there is an unmet need of finding reliable strategies aimed at predicting whether cattle are susceptible or resistant to PTB, classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB, preventing susceptibility to PTB or increasing resistance to PTB. The present invention is focused on solving this problem by identifying functional and deleterious genetic variants associated with susceptibility and resistance to MAP infection by means of RNA-Seq data from peripheral blood (PB) and ileocecal valve (ICV) from Holstein cows with distinct PTB -associated lesions in gut tissues or without lesions.

[0011] DESCRIPTION OF THE INVENTION

[0012] Brief description of the invention

[0013] The present invention refers to an in vitro method for predicting whether cattle are susceptible or resistant to PTB, or for classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB; and / or for preventing susceptibility to PTB or to increase resistance to PTB.

[0014] The inventors of the present invention have identified SNPs occurring in protein-coding regions (cSNPs) that can alter the amino acid sequence of the encoded proteins and have deleterious effects in coding proteins underlying disease or resistance to PTB. The present invention uses RNA-Seq data obtained from blood and ileocecal valve samples collected from control Holstein cattle without lesions, and with focal and diffuse PTB-associated lesions in gut tissues to identify deleterious SNPs that were unique to each group of animals. For this purpose, the RNA-Seq reads were aligned against the bovine ARS-UCD1.2.109 reference genome using the STAR aligner, and SNPs were called using BCFtools applying a very stringent SNP quality threshold. Variant effect prediction and functional analyses using the candidate genes affected by the identified deleterious variants were performed with VeP and ClusterProfiler, respectively. From the 856, 625, and 603 cSNPs specifically identified in the whole transcriptome of control cows and cows with focal and diffuse lesions; 31, 15, and 31 variants had deleterious effects, respectively. Pathways enrichment analysis using the deleterious SNPs-associated candidate genes revealed an enrichment of genes involved in the negative regulation of apoptosis and cellular metabolic process in cows with focal lesions and antigen recognition and presentation in cows with without lesions or with diffuse lesions. Thus, the present invention provides insight into the underlying host response to MAP infection and might be useful in breeding strategies to improve PTB resistance in dairy cattle. By focusing on a specific number of cSNPs (see Tables 1, 2 and 3) a substantial amount of genotyping and analysis workload could be reduced, and quick SNP -profiling assays could be developed.

[0015] Table 1. Deleterious cSNPs specifically identified in the groups of cows with focal lesions

[0016] Table 2. Deleterious cSNPs specifically identified in the groups of cows with diffuse lesions

[0017] Table 3. Deleterious cSNPs specifically identified in the groups of cows without lesions

[0018] Note: For the avoidance of doubt regarding genomic positions, it is noted that the SNP calling was performed using the ARS-UCD1.2 bovine reference genome, while the Variant Effect Prediction (VEP) analysis was carried out using Ensembl database version 111 (corresponding to ARS-UCD1.3). In this latter assembly, the BoLA gene (ENSBTAG00000002069) spans the genomic region 23:28,523,535-28,724,601, which fully encompasses the five SNPs disclosed herein. The apparent discrepancy in genomic coordinates results exclusively from differences between genome assemblies, and all the claimed SNPs are consistently located within the BoLA gene in ARS-UCD 1.3. Although any of the cSNPs included in Tables 1, 2 and 3 affect the structure and function of the corresponding gene encoded protein which in turn, affects resistance or susceptibility to PTB, the Bos taurus MHCII gene (BoLA) gene was the only candidate commonly affected by different cSNPs in the three groups of cows (Figure 3).

[0019] So, the first embodiment of the present invention refers to an in vitro method for predicting whether cattle are susceptible or resistant to paratuberculosis (PTB), or for classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB, the method comprising the identification of the alternative allele of at least a single nucleotide polymorphisms (SNPs) in the BoLa gene, in a biological sample obtained from cattle.

[0020] The second embodiment of the present invention refers to the in vitro use of a SNP in the BoLa gene for predicting whether cattle are susceptible or resistant to PTB, or for classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB.

[0021] In a preferred embodiment, the identification of the alternative allele of the SNP G / A at position 23:28667355-28667355, the SNP C / A at position 23:28723238-28723238, the SNP G / T / C at position 23:28571907-28571907 and / or the SNP G / C at position 23:28666367-28666367 in the BoLa gene predicts that the cattle are susceptible to PTB.

[0022] In a preferred embodiment, the identification of the alternative allele of the SNP G / A at position 23:28667355-28667355 and / or the SNP C / A at position 23:28723238-28723238 in the BoLa gene predicts that the cattle may develop focal lesions is response to an infection caused by Mycobacterium avium subsp. paratuberculosis (MAP); or wherein the identification of the alternative allele of SNP G / T / C at position 23:28571907-28571907 and / or the SNP G / C at position 23:28666367-28666367 in the BoLa gene predicts that the cattle may develop diffuse lesions is response to an infection caused by MAP.

[0023] In a preferred embodiment, the identification of the alternative allele of the SNP A / G at position 23:28721926-28721926 in the BoLa predicts that the cattle are resistant to PTB.

[0024] In a preferred embodiment, the method comprises the identification of the alternative allele of all the SNPs of Tables 1, 2 or 3 in a biological sample obtained from cattle.

[0025] In a preferred embodiment, the method comprises: a) Identification of the alternative allele of all the single nucleotide polymorphisms (SNPs) of Tables 1 and / or Table 2 in a biological sample obtained from cattle, b) wherein the identification of the alternative allele of all the SNPs of Table 1 and / or Table 2 predicts that the cattle are susceptible to PTB. In a preferred embodiment, a) the identification of the alternative allele of all the SNPs of Table 1 predicts that the cattle may develop focal lesions is response to an infection by MAP, or b) wherein the identification of the alternative allele of all the SNPs of Table 2 predicts that the cattle may develop diffuse lesions in response to an infection caused by an infection with MAP.

[0026] In a preferred embodiment, a) the identification of the alternative allele of all the single nucleotide polymorphisms (SNPs) of Table 3 in a biological sample obtained from cattle, b) wherein the identification of the alternative allele of all the SNPs of Table 3 predicts that the cattle are resistant to PTB.

[0027] In a preferred embodiment, the cattle are selected from the group comprising: a cow, a bull, heifers, bullocks, steers, beef cattle and dairy cattle.

[0028] In a preferred embodiment, the method is carried out by using DNA extracted from a biological sample obtained from the cattle.

[0029] The third embodiment of the present invention refers to a genetic editing tool designed to edit a codon sequence within the BoLA gene for preventing susceptibility to PTB or to increase resistance to PTB.

[0030] In a preferred embodiment, the codon sequence within the BoLA gene comprises the SNP G / A at position 23:28667355-28667355, the SNP C / A at position 23:28723238-28723238, the SNP G / T / C at position 23:28571907-28571907, the SNP G / C at position 23:28666367-28666367 and / or SNP A / G at position 23:28721926-28721926.

[0031] In a preferred embodiment, the genetic editing tool is selected from: CRISPR-Cas9, TALENs (Transcription Activator-Like Effector Nucleases), ZFNs (Zinc Finger Nucleases), Base Editors or Prime Editing.

[0032] Alternatively, the present invention also refers to a method for predicting whether cattle are susceptible or resistant to paratuberculosis (PTB).

[0033] For the purpose of the present invention the following terms are defined:

[0034] • The term "comprising" means including, but not limited to, whatever follows the word "comprising". Thus, use of the term "comprising" indicates that the listed elements are required or mandatory, but that other elements are optional and may or may not be present. • The term "consisting of’ means including, and limited to, whatever follows the phrase “consisting of’. Thus, the phrase "consisting of’ indicates that the listed elements are required or mandatory, and that no other elements may be present.

[0035] Description of the figures

[0036] Figure 1. Workflow diagram to identify functional cSNPs and predict corresponding variant effects. Identification of cSNPs uniquely present in cows with a specific type of PTB- associated lesion (focal or diffuse) and without lesions using RNA-Seq data from peripheral blood (PB) and ileocecal valve (ICV) samples of 14 Holstein cattle. PB and ICV samples from animals with and without lesions were merged.

[0037] Figure 2. Density plots showing the identified cSNPs. Graphs show the number of cSNPs per Mbp in the y-axis versus chromosome and position in the x-axis for the comparisons of c cows without lesions vs cows with focal lesions (A) and cows without lesions vs cows with diffuse lesions (B).

[0038] Figure 3. Unique and common candidate genes in cows without lesions and with focal and diffuse PTB-associated lesions. Venn diagram showing the genes affected by missense cSNPs in cows with focal lesions, in blue; cows with diffuse lesions, in red; and cows without lesions, in green. Only candidate genes with a recognized gene code in Ensembl are included in the figure.

[0039] Figure 4. Enriched QTLs in Holstein cows with focal or diffuse PTB-associated lesions and without lesions. Bubble plots representing all significantly enriched QTLs associated with health traits that overlap cSNPs with deleterious effects identified in the transcriptomes of cows with focal (A) or diffuse lesions (B) and without lesions (C). The areas of the bubbles represent the number of observed QTLs and the color represents the P-value scale (darker color = smaller P -value). The richness factor for each QTL represents the ratio of the number of QTL and the expected number of QTLs.

[0040] Detailed description of the invention

[0041] The present invention is illustrated by means of the Examples set below without the intention of limiting its scope of protection.

[0042] Example 1. Material and methods

[0043] Example 1.1. RNA-Seq data RNA-Seq data used in this study (Gene expression omnibus public repository, GEO Accesion: GSE137395) was obtained from fourteen Holstein cows from a single commercial dairy farm in Asturias (Spain). Cows were tested by histopathological analysis and Ziehl Neelsen (ZN) staining of gut tissues, ELISA for antibodies against MAP, and fecal and gut tissues PCR and bacteriological culture. RNA was extracted from ileocecal valve (ICV) tissue and peripheral blood (PB) from the control cows without lesions (N = 4), and with focal (N = 5) and diffuse (N = 5) lesions in their intestinal tissues. Animals were 18 months old or older (mean: 4.6 years old). Briefly, PB samples were collected from the coccygeal vein in PAXgene Blood tubes at the time of slaughter and RNA was extracted using the PAXgene Blood RNA kit, according to manufacturer’s instructions (Qiagen, Hilden, Germany). ICV samples were collected in RNA later (Sigma, St Louis, MP), and RNA extracted using the RNeasy Mini Kit, according to the manufacturer’s instructions (Qiagen, Hilden, Germany). RNA-Seq libraries were generated with the Illumina NEBNext® Ultra Directional RNA Library preparation kit (Illumina Inc, CA, US), and were single-end sequenced (1 x 75) with an Illumina NextSeq500 sequencer at the Genomic Unit of the Scientific Park of Madrid, Spain.

[0044] Example 1.2. RNA-Seq data analysis

[0045] The overall study workflow is presented in Figure 1. Identification of cSNP variants unique for each group (with focal or diffuse lesions, or without lesions) was performed using a pipeline that was adopted from a previous study that determined an optimized pipeline to detect missense cSNPs from RNA-Seq data from multiple samples per animal. Quality control of the RNA-Seq raw data was performed using FastQC 0.12 to identify sequencing read artifacts including sites with low-quality reads (Phred score < 30), duplicated reads, uncalled bases, and potentially contaminated reads. Next, reads were trimmed to remove Illumina adapters and low-quality bases at the start and end of reads using Trimmomatic 0.38. Additionally, reads with an average quality score below 20, within a sliding window of 5 nucleotides, and with a length less than 75 nucleotides were removed. Next, the trimmed reads for each sample were aligned to the most recent Bos taurus annotation file (Ensembl ARS-UCD1.2.109, downloaded on 15 / 02 / 2023) using STAR 2.7.0a. The alignment options consisted of allowing any number of mismatches as long as the ratio of mismatches to read length was less than 0.04 and allowing a maximum of one alignment per read. Reads mapping in more than one place of the genome were considered unmapped. After alignment, read duplicates were removed with PICARD tools 2.12. Finally, the aligned reads (.bam files) for each group of animals were merged using Samtools 1.4 leading to a total of three files (.bam).

[0046] Example 1.3. Variant calling analysis

[0047] Variant calling was performed using bcftools 1.11. This involves the conversion of .bam files into a binary call format (.bcf) file containing variant information including genomic position, alternative allele detected, and quality of the SNP call. Very stringent criteria were applied to maximize the accuracy and prevent false-positive SNP detection as previously described. Variants with a minimum read depth below 10 and a minimum of 2 supporting reads for the alternative allele were filtered. SNPs within 5 bp of an insertion / deletion were also removed. Variants with quality values Phred score below 30 of the alternative alleles, a minor allele frequency of the alternative allele lower than 20%, and a maximum proportion of missing alleles of 80%, were filtered. Finally, bcf files from cows with focal lesions vs without lesions, diffuse lesions vs without lesions, focal vs diffuse lesions, diffuse vs focal lesions, and without lesions vs any type of lesions were compared. Unique SNPs to each of the three groups of cows (without lesions or with focal or diffuse lesions) were pooled and considered for further analysis.

[0048] Example 1.4. Variant effect prediction and candidate genes identification

[0049] Variants unique to cows with PTB-associated lesions (focal or diffuse) or without lesions were analyzed for functional consequences using the Variant Effect Predictor (VEP) (Ensembl database version 111). The VEP is a powerful toolset for the analysis, annotation, and prioritization of genomic variants in coding and non-coding regions. The VEP annotates two broad categories of genomic variants: (1) sequence variants with specific and well-defined changes (including SNPs, insertions, deletions, multiple base pair substitutions, microsatellites, and tandem repeats); and (2) larger structural variants (greater than 50 nucleotides in length), including those with changes in copy number or insertions and deletions of DNA. For all input variants, the VEP returns an indication of the effect (low, moderate, or high) of the amino acid change using protein biophysical properties. Variant impact “high” is defined as the disruptive impact in the protein that would lead to protein truncation, loss of function, or tissue-nonsense- mediated delay. In addition, Sorting Intolerant From Tolerant (SIFT) available in Ensembl was used. The SIFT algorithm uses sequence homology to compute the likelihood that an amino acid substitution will have an adverse effect on protein function. The underlying assumption is that evolutionarily conserved regions tend to be less tolerant of mutations, and therefore amino acid substitutions in these regions are more likely to affect function. The SIFT workflow begins with a query protein that is searched against a protein database to obtain homologous protein sequences. The chosen sequences are aligned, and for a particular position, SIFT looks at the composition of amino acids and computes the score. A SIFT score is the normalized probability of observing the new amino acid at that position, and ranges from 0 to 1. A score between 0 and 0.05 is predicted to affect protein function. cSNPs were selected for further analysis if either their functional consequence was predicted as High by the VEP algorithm or if the SIFT algorithm predicted them as deleterious (score range: 0-0.5). Genes affected by a missense cSNPs were considered as candidate genes. The function of the identified candidate genes was searched in GeneCards by searching their gene symbol. To visualize unique and common candidate genes in each specific group of animals, the R library VennDiagram 1.7.3 was used.

[0050] Example 1.5. Quantitative trait loci (QTL) identification and enrichment analysis

[0051] The R package Genomic functional annotation in livestock for positional candidate loci (GALLO 1.3) was used to annotate QTLs within an interval of 500 Kbp of the identified missense cSNPs. Overlapping QTLs were merged to create a single QTL. To determine which of the annotated QTLs overlapped with QTLs previously described in the QTL database release 52, a QTL enrichment analysis was performed using GALLO. QTLs with an FDR-adjusted p-value equal to or less than 0.05 were considered enriched.

[0052] Example 1.6. Functional analysis for enriched gene ontologies (GO) and metabolic pathways.

[0053] Using the candidate genes where missense cSNPs were located, an enrichment analysis of GO terms (biological process, molecular function, and cellular compartment) and metabolic pathways was performed for each group of animals with the ClusterProfiler 4.10.0 package in R (R core team, 2023) for each group of animals. To avoid false positives, only GO terms and pathways with an FDR-adjusted p-value equal to or less than 0.05 were considered enriched.

[0054] Example 2. Results

[0055] Example 2.1. SNP calling

[0056] Identification of cSNPs was performed using the workflow described in Figure 1. The quality of raw reads was verified using FastQC and trimmed with Trimmommatic. Overall, an average of 90.22% of the reads mapped to one location of the reference genome, while 7.30% mapped in multiple regions and were considered unmapped. The mean number of uniquely mapped reads identified in PB (21,331,835) and ICV (19,506,829) were similar. Previous studies highlighted that merging RNA-Seq data from several tissues and individuals of the same group increases the number of aligned reads, improves SNP detection, and read depth coverage, and, therefore, reduces potential false positives. According to this, mapped reads from PB and ICV samples of animals with a specific type of lesion or without lesions were merged.

[0057] SNP calling allowed the identification of cSNPs associated with the first group of animals in each of the comparisons: control vs focal lesions (N= 2,128), control vs diffuse lesions (N= 2,194), focal lesions vs control (N= 2,388), focal vs diffuse lesions (N= 1,939), diffuse lesions vs control (N= 1,956), and diffuse vs focal lesions (N= 1,656) (Table 4).

[0058] Table 4. Results of the SNP calling for all tested comparisons. cSNPs; SNPs in coding regions. Control cows correspond to the group of animals without PTB-associated lesions. (+) cSNPs common in the two comparisons cSNPs identified in the comparisons of control cows vs cows with focal lesions (N= 2,128) and control cows vs cows with diffuse lesions (N= 2,194) are presented in Figure 2. We observed that cSNPs in expressed regions were distributed along the entire genome but there were genomic regions located in the chromosomes 18 and 23 that accumulated larger number of cSNPs.

[0059] Subsequently, cSNPs unique for the controls cows (N= 856) or cows with focal (N= 625) or diffuse lesions (N= 603) were selected. After performing a variant effect prediction analysis with VEP and SIFT, missense cSNPs with a deleterious effect in the corresponding coding proteins were identified. A total of 31, 15, and 31 missense cSNPs affecting 29, 14, and 29 genes were identified in the cows without lesions, and with focal and diffuse lesions, respectively. When looking at the candidate genes affected by the identified cSNPs (Figure 3), we found that the Bos taurus MHCII gene (BOLA) was affected by two different deleterious cSNPs located on BTA23 in the control cows and in cows with focal or diffuse lesions. This was the only gene commonly affected by different missense cSNPs in the three groups of cows.

[0060] Example 2.2. QTLs enrichment analysis

[0061] QTLs within an interval of 500 Kbp of the identified missense cSNPs for each group of cows were annotated. The identified QTLs overlapped with QTLs previously described in the cattle database related to health, production, reproduction, meat and carcass, and milk. In the control cows without lesions, 19 QTLs overlapped with QTLs previously associated with somatic cell score, 14 with bovine leukemia virus (BLV) susceptibility, 8 with bovine tuberculosis susceptibility, 7 with the bovine respiratory disease susceptibility, 7 with MAP susceptibility, 4 with white-blood cell number, 2 with cell-mediated immune response, and 2 with antibody- mediated immune response, among others. Generally, a lower somatic cell count (leukocytes cells) indicates better animal health. In the cows with focal lesions, 27 QTLs overlapped with QTLs previously associated with bovine leukemia virus (BLV) susceptibility, 24 QTLs overlapped with MAP susceptibility, 17 QTLs overlapped with bovine respiratory disease susceptibility, 8 with bovine tuberculosis, 7 with somatic cell score, 7 with white-blood cell number, 4 with cell-mediated immune response, and 7 with antibody -mediated immune response. These results suggest that the cows in the subclinical stage of MAP infection (with focal lesions) share a high number of genetic regions with cows susceptible to viral infections caused by BLV and responsible for bovine respiratory disease. In cattle, most BLV-infected animals are asymptomatic (70%); leukemia is rare (about 5% of infected animals), but lymphoproliferation is more frequent (30%). Lymphoprofileration is also characteristic of the subclinical stage of MAP infection. When the asymptomatic stage was compared with the clinical stage of PTB, there was a decrease in the number of CD4" T cells, whereas the CD8+T cells remained at comparable frequencies.

[0062] Finally, in the cows with diffuse lesions, 34 QTLs overlapped with QTLs previously associated with bovine tuberculosis susceptibility, 22 with MAP susceptibility, 17 with BLV susceptibility, 14 with somatic cell score, 12 with antibody-mediated immune response, 7 with bovine respiratory disease susceptibility, 8 with white-blood cell number, and 2 with cell- mediated immune response. These results suggested that the cows in more advanced stages of infection (with diffuse lesions) share a high number of gene coding regions with deleterious cSNPs with cattle infected with other pathogenic Mycobacteria such as Mycobacterium bovis. For the identified QTLs that overlapped with health-associated QTLs previous described in the databases, a QTL enrichment analysis was performed. In the three groups of animals, a significant enrichment of QTLs previously associated with BLV susceptibility and adaptative- mediated immune responses was observed highlighting the important role of the genetic regions associated with these traits in MAP infection susceptibility and resistance (Figure 4). In addition, in the cows with focal lesions and without lesions, enrichment of QTLs associated with cell-mediated immune responses was observed. Enrichment of QTLs associated with whole blood cell number was observed in the cows without lesions and with diffuse lesions.

[0063] Example 2.3. Functional analysis for enriched gene ontologies (GO) and metabolic pathways using candidate genes Using the candidate genes where missense cSNPs were located, an enrichment analysis of GO terms and metabolic pathways was performed. As seen in Table 5, the enrichment analysis using the candidate genes affected by missense cSNPs unique to cows without lesions showed an enrichment of genes involved in 11 GOs and 6 metabolic routes. Table 5. Enrichment analysis using the candidate genes affected by cSNPs with deleterious effects that were unique to cows without lesions.

[0064] Enrichment of genes involved in antigen presentation and processing like the BOLA, clathrin- coated vesicles (Clathrin Assembly Protein Complex 3 Beta-1 Large Chain, AP3B1 and chromaffin granules (Chromogranin A, CHGA) was observed. AP3B1 is essential for trafficking cargo proteins, including Clathrin (CLTA) and lysosomal-associated membrane protein 1 (LAMP-1) to the phagosome and lysosome-related organelles. CHGA is the precursor of three biologically active peptides; vasostatin, pancreastatin, parastatin, catestatin and chromofungin which have direct or indirect effects on intestinal inflammation. Therefore, missense cSNPs in the CHGA coding region might control tissue homeostasis.

[0065] In the group of cows with focal lesions, (Table 6) we observed an enrichment of genes including the Sphingolipid Biosynthesis Regulator 3 (0RMDL3) and Ankyrin Repeat Domain- Containing Protein (KANK2) involved in 11 GOs including the negative regulation of programmed cell death (G0:0043069) and cellular metabolic process (G0:0031324) among others. Table 6. Enrichment analysis for the candidate genes affected by cSNPs with deleterious effects that were unique to cows with focal lesions.

[0066] Our results suggested that missense cSNPs causing modifications in the structure and function of these apoptosis regulators might cause a negative regulation of apoptosis and favor MAP persistence within infected macrophages during the long subclinical stage of MAP infection.

[0067] Finally, we observed a significant enrichment of candidate genes involved in 27 metabolic pathways including antigen processing and presentation and Thl and Th2 cell differentiation (BOLA-DQB / BOLA-DQA5 / BOLA) and bile secretion and antifolate resistance (ABCG2 / LOC 100299180) in animals with diffuse lesions of PTB (Table 7).

[0068] Table 7. Enrichment analysis using the candidate genes affected by cSNPs with deleterious effects that were unique to cows with diffuse lesions.

[0069] Exacerbated pro-inflammatory immune responses mediated by the presence of SNPs in human homologs to the BOLA-DQB / BOLA-DQA5 / BOLA genes have been associated with susceptibility to numerous human autoimmune and inflammatory human diseases like allograph rejection (btaO533O), type I diabetes mellitus (bta04940), graft-versus-host disease (btaO5332), autoimmune thyroid disease (bta05320), viral myocarditis (bta05416), tuberculosis (bta05152), toxoplasmosis (bta05145), asthma (btaO531O), inflammatory bowel disease (bta05321), Epstein-Barr virus infection (bta05169), Human T-cell leukemia virus 1 (bta05166), Leishmaniasis (bta05140), Rheumatoid arthritis (btaO5323 , Staphylococcus aureus infection (bta05150), Herpes simplex virus 1 infection (bta05168), systemic lupus erythematosus (bta05322), and Influenza A (bta05164).

Claims

22CLAIMS1. In vitro method for predicting whether cattle are susceptible or resistant to paratuberculosis (PTB), or for classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB, the method comprising the identification of the alternative allele of at least a deleterious single nucleotide polymorphism (SNPs) in the Bos taurus MHCII (BoLa) gene, in a biological sample obtained from cattle; wherein the identification of the alternative allele of the deleterious SNP G / A at position 23:28667355-28667355, the deleterious SNP C / A at position 23:28723238-28723238, the deleterious SNP G / T / C at position 23:28571907-28571907 and / or the deleterious SNP G / C at position 23:28666367-28666367 in the BoLa gene predicts that the cattle are susceptible to PTB; or wherein the identification of the alternative allele of the deleterious SNP A / G at position 23 :28721926-28721926 in the BoLa predicts that the cattle are resistant to PTB.

2. In vitro use of a deleterious SNP in the BoLa gene for predicting whether cattle are susceptible or resistant to PTB, or for classifying and / or selecting cattle according to whether they are susceptible or resistant to PTB; wherein the identification of the alternative allele of the deleterious SNP G / A at position 23:28667355-28667355, the deleterious SNP C / A at position 23:28723238-28723238, the deleterious SNP G / T / C at position 23:28571907-28571907 and / or the deleterious SNP G / C at position 23 :28666367-28666367 in the BoLa gene predicts that the cattle are susceptible to PTB; or wherein the identification of the alternative allele of the deleterious SNP A / G at position 23:28721926-28721926 in the BoLa predicts that the cattle are resistant to PTB.

3. In vitro method, or in vitro use, according to any of the previous claims, wherein the identification of the alternative allele of the deleterious SNP G / A at position 23:28667355-28667355 and / or the deleterious SNP C / A at position 23:28723238- 28723238 in the BoLa gene predicts that the cattle may develop focal lesions is response to an infection caused by Mycobacterium avium subsp. paratuberculosis (MAP); or wherein the identification of the alternative allele of deleterious SNP G / T / C at position 23:28571907-28571907 and / or the deleterious SNP G / C at position 23:28666367-28666367 in the BoLa gene predicts that the cattle may develop diffuse lesions is response to an infection caused by MAP.

4. In vitro method, or in vitro use, according to any of the previous claims, the method comprising the identification of the alternative allele of all the deleterious SNPs of Tables 1, 2 or 3 in a biological sample obtained from cattle.

5. In vitro method, or in vitro use, according to any of the previous claims, which comprises: a) Identification of the alternative allele of all the deleterious SNPs of Tables1 and / or Table 2 in a biological sample obtained from cattle, b) wherein the identification of the alternative allele of all the deleterious SNPs of Table 1 and / or Table2 predicts that the cattle are susceptible to PTB.

6. In vitro method, or in vitro use, according to any of the previous claims, wherein: a) The identification of the alternative allele of all the deleterious SNPs of Table 1 predicts that the cattle may develop focal lesions is response to an infection by MAP, or b) wherein the identification of the alternative allele of all the deleterious SNPs of Table2 predicts that the cattle may develop diffuse lesions in response to an infection caused by an infection with MAP.

7. In vitro method, or in vitro use, according to any of the previous claims, which comprises: a) Identification of the alternative allele of all the deleterious SNPs of Table3 in a biological sample obtained from cattle, b) wherein the identification of the alternative allele of all the deleterious SNPs of Table 3 predicts that the cattle are resistant to PTB.

8. In vitro method, or in vitro use, according to any of the previous claims, wherein the cattle are selected from the group comprising: a cow, a bull, heifers, bullocks, steers, beef cattle and dairy cattle.

9. In vitro method, or in vitro use, according to any of the previous claims, wherein the method is carried out by using DNA extracted from a biological sample obtained from the cattle.

Images