Genomic and transcriptomic characterization of pig obesity models

Obesity is a major risk factor for globally prevalent serious diseases like cardiovascular diseases, type 2 diabetes, and cancer, which are leading causes of worldwide mortality of over 4 million people dying each year according to the World Health Organization (WHO). Therefore, there is a great need to understand the complexities of obesity and its associations with diseases to form the basis of developing diagnostic tools and interventions to treat and/or prevent obesity and maintain healthy body weight. To achieve this, animal models faithfully mirroring the complex human state of obesity are pivotal, i.e., human translational animal models in which a complete metabolic syndrome and development of co morbidities are obtained upon diet-induced obesity. Pigs are good models as they are comparable to humans with respect to e.g., the cardiovascular system, inflammation phenotypes and mechanisms and organ sizes. Regarding obesity related metabolic syndrome, the Ossabaw pig is an outstanding large animal model because of its “thrifty genotype” with a solid ability to convert food into body fat and gain weight with high energy high-fat dieting, presenting as robust metabolic syndromes. Given the remarkable importance of this breed, we performed whole genome sequencing and genomewide RNA sequencing of this pig. We also carried out RNA sequencing of Göttingen minipig challenged with swine adapted influenza virus (swH1N2) to study the response of viral infectious disease in lean and obese individuals and find potential biomarkers for acute kidney injury (AKI).

Paper 1 presents the whole genome sequencing and assembly of Ossabaw pig. Using single tube long fragment reads (stLFR) and Nanopore sequencing technologies, we obtained the first high quality genome of the Ossabaw pig with Contig N50 of ~6.03 Mb, which is significantly higher than most published pig genomes. We found that the Ossabaw genome comprised 4,130 heterozygous genes enriched in important pathways. Genomic comparison to Duroc (the pig reference genome) reveals that variations identified in Ossabaw pig may be related to its “thrifty” phenotype. Gene LEPR (leptin receptor gene) contained two positively selected sites which may lead to pseudogenization of this gene or transcripts with possible significant effects on obesity and obesity induced chronic inflammation. Paper 2 presents the high depth whole genome resequencing of 49 Ossabaw pigs to find new population mutations and natural selection genes, as validation and complementarity to paper 1. 3,688,878 SNPs were shared among the 49 Ossabaw individuals. Genomic selective-sweep regions of the Ossabaw population harbor genes crucial to lipid metabolism. Inbreeding event estimation for all pigs reveals that Ossabaw individuals have much higher numbers of homozygosity fragments than other five pig breeds. Numerous high-frequency ROH (runs of homozygosity) regions were detected, which contain important genes closely related to obesity, insulin resistance and diabetes, such as CD36, AQP3, AQP7, and two well-known obesity related genes - POMC and PPARGC1A. And 849 genes were detected with novel and Ossabaw-specific non-synonymous mutations, for example, the well-known obesity related gene - leptin (LEP, three mutations). These candidate genes have the potential to be biomarkers and therapeutic targets for obesity and obesity related comorbidities. Paper 3 presents the high depth genome-wide RNA sequencing of Ossabaw, Production (a crossbreed between Landrace and Yorkshire) and Göttingen minipigs of four tissue types to identify differentially expressed genes (DEGs) of the Ossabaw compared to other pig breeds and check the expression of genes emphasized in Paper 2. Thousands of DEGs were detected for Ossabaw. DEGs related to lipid metabolism and strongly associated with obesity and diseases such as DLK1, ABCC8, ADCYAP1, FABP3 and ACOT7 were found, especially the candidate genes such as LEP, LEPR, PPARGC1A and PPARGC1B identified in Paper 1 and Paper 2. Gene co-expression analysis gave us the hub genes including such as lncRNA genes, ABCC8 and IGF1 of the expression matrix, providing another candidate genes that may be related to obesity tendency of Ossabaw pigs. Paper 4 presents genome-wide RNA sequencing of kidney tissue samples from lean and obese Göttingen minipig challenged with influenza A virus (swH1N2) to see the renal response to viral infection in lean and obese pigs and find potential biomarkers for acute kidney injury (AKI). We identified 522 differentially expressed genes between obese and lean individuals infected with swH1N2. Almost all identified immune-related genes, for example, complement components and their receptors genes, were upregulated in obese animals. Long non-coding RNA (lncRNA) and gene-coexpression analysis showed that lncRNAs, central for neuromodulation and cell cycle related genes are also differentially expressed.