Identification of internal control genes via RNA-seq analysis for data normalization in fecal RNA isolated from dairy calves

Document Type


Publication Date



American Dairy Science Association


Journal of Dairy Science




Suppl. 1






internal control genes, fecal RNA


The use of fecal material to isolate RNA and perform gene expression analysis is a novel method that has been recently used in humans, rodents, and neonatal dairy calves. In the past, our group has used several internal control genes (ICGs) to normalize gene expression with the aim to reach a suitable stability across samples. However, the identification of the most suitable ICGs, when using fecal RNA for relative mRNA expression via RT-qPCR is remaining. This study aimed to identify robust ICGs for normalization of RT-qPCR through fecal RNA-seq data. Fecal samples were collected from 6 healthy Jersey calves (5 wk old) and frozen in liquid nitrogen until RNA isolation. Total RNA was extracted from 100mg of feces through bead-beater homogenization with trizol, followed by purification through silica membrane columns. The TruSeq stranded Ribo-zero rRNA reduced library was applied to the fecal RNA samples before sequencing. The libraries were sequenced at 2 × 50 base pair read length (NGS; Illumina, NovaSeq S4) at the University of Minnesota Genomics Center. Processed reads were aligned to the bovine genome using HISAT2. Mapped genes were used as counts per million (CPM) to evaluate their stability across samples. Genes with stable expression across samples and medium RNA abundance (3.5 to 6.5 logCPM) were used to identify potential ICGs. The pairwise comparison method using geNorm software with a combination of expression ratio stability (M; the lower the M value, the higher the stability of expression ratio) and optimal number of ICG calculated by the pairwise variation (V) between the normalization factor obtained using the selected genes were used to evaluate the ICGs. The genes with the most stable expression ratio (M <0.20) among those assessed were SMS, VPS37A, ACTB, and GAPDH. The V-value (ideally <0.15) for these selected genes were V = 0.04. The level of variation and stability of these genes were clearly below the thresholds previously reported for reliability. Thus, the geometric average of the ICGs SMS, VPS37A, ACTB, and GAPDH is an appropriate method for normalization of fecal RT-qPCR data.