Differential gene expression analysis of the whole blood transcriptome between young and old companion border collie dogs

Abstract BackgroundAging is the most significant risk factor in many diseases and for mortality alike, and it is known to be influenced by both genetic and environmental factors. Due to dogs’ importance in human societies, the study of aging in companion dogs is worthwhile in its own right. Still, dogs could also be ideal translational model animals for human aging research. In this study, our primary goal was to investigate the gene expression changes with age in whole blood samples of dogs. The advantage of using blood is its accessibility and known biomarkers for diseases.MethodsWe sequenced the poly(A)-tailed RNA fraction in the blood samples of five young and five old companion border collie dogs and implemented a differential gene expression analysis. We removed haemoglobin related reads in silico, following standard raw data quality control and alignment to the reference genome. Sample collection and preparation, sequencing and data analysis were implemented using standard laboratory methods and RNA-seq data analysis pipelines.ResultsThe estimated statistical power of the study was 98% according to an initial power analysis with a 10% false discovery rate and minimum fold change of 3. Raw sequence data quality was high with an 86% alignment rate to the CanFam 3.1 dog reference genome. Following in silico haemoglobin read removal, 23-74 million reads remained for the differential gene expression analysis. In contrast to our expectations, we could not differentiate age clusters in either a multidimensional scaling analysis or in a principal component analysis. We found only a limited number of differentially expressed genes (n=61) between young and old dogs’ blood transcriptome. The only significant gene ontology term was response to bacterium with a fold change of 8.62.ConclusionsIn contrast with the large number of differentially expressed genes in the prefrontal cortical brain region of dogs, we could identify only one blood biomarker for aging in blood. Confounding factors (e.g., diet) might have influenced our results. Future research could also use biological age instead of the animals’ chronological age.