Genome-wide Detection Of Copy Number Variations And Identification Of Chromosomal Regions For Umbilical Hernia In Pigs

Copy number variations （CNVs） refer to the deletion, insertion, and duplication of chromosomal region from 1 kb to several Mb in length. Many CNVs result in different levels of gene expression, which may account for a significant proportion of normal phenotypic variations and human diseases. Hernia is a congenital birth defect characterized by the presence of internal organs located outside of the ventral abdominal wall. Umbilical hernia is one common disease in swine and leads to huge losses in pig industry. Although a number of genes and genomic regions associated with hernia have been detected, the major gene is still unclear. Therefore a custom-made tiling oligo-nucleotide array was used for screening 12 pigs to detect porcine genome-wide CNVs. Additionally, we used SNP60 BeadChip to identify the potential causative mutations and copy number variations regions （CNVRs） associated with hernia in 48 pigs which consisted of 14 umbilical hernia pigs and 34 normal pigs. Furthermore, effect of candidate SNP on klotho （AZ） gene expression was investigated. The main results were as follows:1. Array-based comparative genomic hybridization （aCGH） was used to detect CNVs in 12 pig from 8 breeds. Two hundred and fifty-nine CNVRs across chromosomes 1-18 and X were identified, with an average size of 65.07kb and a median size of 98.74kb, covering 16.85 Mb or 0.74% of the whole genome. Concerning copy number status,93 （35.91%） CNVRs were called as gains,140 （54.05%） were called as losses and the remaining 26 （10.04%） were called as both gains and losses. Clustering results showed the dendrogram of 12 pigs was extended to two branches. The first branch was consisted of Chinese native pigs （Erhualian, Yangxin, Tongcheng）, while the second branch was extended to another two minor branches consisted of foreign pigs （Pietrain, Large White, Landrace） and crossbred pigs （DIV, Landrace x DIV）. Of all detected CNVRs,171 （66.02%） and 34 （13.13%） CNVRs directly overlapped with Sus scrofa duplicated sequences and pig quantitative trait loci （QTL）, respectively. The 259 CNVRs encompassed 372 unique transcripts. GO analysis revealed that CNVR genes belonged to these classes of genes that participated in sensory perception of smell, sensory perception of smell or chemical stimulus, sensory perception, cognition, G-protein coupled receptor protein signaling pathway, olfactory receptor activity and other basic metabolic processes. KEGG pathway analyses indicated that 50 genes involved in olfactory transduction （P< 0.05） were over-represented in the porcine CNVRs. Compared with mouse, all pig CNVR genes had dN/dS ratios significantly higher than monomorphic genes, indicating a relaxation of purifying selection. While compared with human, the pig CNVR genes with the status of gains had dN/dS ratios lower than monomorphic genes, indicating these genes subjected to stringent purifying selection compared with non-polymorphic genes. By real-time quantitative PCR （qPCR）, we successfully validated CNVR₁00 （R2=0.64, P=0.02） and CNVR₂15 （R2=0.57, P=0.05）.2. Based on the Porcine SNP60 BeadChip, we genotyped 48 pigs including 14 umbilical hernia pigs and 34 normal pigs, and identified several single nucleotide polymorphisms （SNPs） and CNVRs associated with umbilical hernia. Furthermore, we performed annotation, GO and clustering analysis of CNVRs. Seven SNPs were identified to be significantly associated with umbilical hernia by genome-wide association analysis （P< 7.27E-07）, with the most significant marker ASGA0049694 （P=1.01E-08） located at position 9520284 bp on SSC11. Five SNPs were located in star-related lipid transfer （START） domain containing 13 （STARD13） and KL gene, while the other 2 SNPs were located in inter-genic regions. One hundred and seventeen CNVs were identified and merged into 34 CNVRs, with an average size of 1.7Mb and a median size of 2.2Mb. The CNVRs encompassed 813 unique porcine genes. GO analysis revealed that CNVR genes belonged to these classes of genes which participated in sensory perception of smell and chemical stimulus, olfactory receptor activity, G-protein coupled receptor protein signaling pathway, cell surface receptor linked signal transduction and other basic metabolic processes. KEGG pathway analyses indicated that 45 genes involved in olfactory transduction （P< 0.05） were over-represented in the porcine CNVRs. The 34 CNVRs were split into 40 CNV sub_regions, with length from 30kb to 8.8Mb. Clustering results of CNV sub_regions revealed that two pigs both affected umbilical hernia clustered together and the other 46 pigs belonged to another big cluster. Twenty-eight out of 40 （70.0%） CNV sub_regions were shown to be suggestively significantly related to umbilical hernia after the P value threshold was adjusted to 0.1. The relative copy number of 2 CNVRs （CNVR IDs 2 and 39） detected by qPCR was significantly different between case group and control group（P≤0.05）. Two CNV sub_regions （CNVR IDs 2 and 39） detected from the SNP array, were selected to be validated by qPCR. The confirmed rates were 0.73 and 0.70 respectively.3. Through Promoter 2.0 online software, we attained the potential promoter regions of KL. Via double luciferase assays of KL deletions （KL-D1 （-178bp/-3bp）, KL-D2 （-418bp/-3bp）, KL-D3 （-599bp/-3bp） and KL-D4 （-835bp/-3bp）） in porcine kidney （PK） cells and swine testis （ST） cells, we confirmed that the core promoter region was KL-D2 （-418bp/-3bp）. Through double luciferase assays of reporter constructs containing the wild-type A or mutant G sequence of porcine KL intron 1 and the pig KL core promoter in PK cells and ST cells, we found luciferase activity of pGL3-D2-G was significantly higher than pGL3-D2-A（P< 0.05）. After cotransfection of siRNA of organic cation transporter 1 （OCT-1） and reconstructed plasmids in PK cells and ST cells, we found luciferase activity of pGL3-D2-A was significantly higher than pGL3-D2-G（P< 0.05）. Furthermore, compared with negative control, luciferase activity of pGL3-D2-A was significantly increased （P< 0.05）. After RNAi on OCT-1 in PK cells and ST cells, the mRNA and protein level of KL were significantly decreased （P< 0.05）, while there was no change in cellular apoptosis percentage. The binding of OCT-1 with the first intron of KL （1464bp/1476bp） was confirmed by ChIP assay.