Identification of stable genes in the corpus luteum of lactating Holstein cows in pregnancy and luteolysis: implications for selection of RT-qPCR reference genes

Authors: MEZERA, MEGAN - University Of Wisconsin; Li, Wenli; EDWARDS, ANDREA - University Of Wisconsin; KOCH, DANIEL - University Of Wisconsin; BEARD, ADAM - University Of Wisconsin; WILTBANK, MILO - University Of Wisconsin

Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/10/2020
Publication Date: 3/27/2020
Citation: Mezera, M., Li, W., Edwards, A., Koch, D., Beard, A., Wiltbank, M. 2020. Identification of stable genes in the corpus luteum of lactating Holstein cows in pregnancy and luteolysis: implications for selection of RT-qPCR reference genes. Journal of Dairy Science. 103(5):4846-4857. https://doi.org/10.3168/jds.2019-17526.
DOI: https://doi.org/10.3168/jds.2019-17526

Interpretive Summary: Reverse transcription quantitative PCR (RT-qPCR) is a valuable and widely used technique to study changes in gene expression. The commonly used reference-gene based RT-qPCR quantification methods require the use of stably expressed “reference” genes as an internal control. The presumably stable expression of these reference genes is used to normalize the expression levels of genes of interest that are being studied. However, widely used reference genes typically come from a small pool of house-keeping genes. Unfortunately, many of these commonly used house-keeping genes can have significant expression variation among different tissues or physiological states, making them less useful as reference genes. This is particularly concerning in dynamic tissues such as the corpus luteum (CL), which is known for its rapid physiological changes during the various stages of pregnancy. Thus there is a clear need to identify stably expressed reference genes well suited for the CL over a wide range of physiological states. Whole transcriptome RNA sequencing (RNAseq) stands as an effective method to identify new reference genes by enabling the assessment of the stability of the entire pool of mRNA transcripts. We report the identification of thirteen possible new reference genes using whole transcriptome RNA sequencing in the bovine CL across a wide range of physiological states. Identification of these new reference genes will allow accurate normalization of RT- qPCR results and be of use to scientists studying physiological changes during pregnancy. Furthermore, our analysis sheds light on the effects of luteolysis and pregnancy on the stability of gene expression in the bovine CL.

Technical Abstract: In lactating dairy cattle, the corpus luteum (CL) is a dynamic endocrine tissue vital for pregnancy maintenance, fertility, and cyclicity. Understanding processes underlying luteal physiology is therefore necessary to increase reproductive efficiency in cattle. A common technique for investigating luteal physiology is reverse-transcription quantitative PCR (RT-qPCR), a value-able tool for quantifying gene expression. However, reference-gene-based RT-qPCR quantification methods require utilization of stably expressed genes to accurately assess mRNA expression. Historically, selection of reference genes in cattle has relied on subjective selection of a small pool of reference genes, many of which may have significant expression variation among different tissues or physiologic states. This is particularly concerning in dynamic tissues such as the CL, with its capacity for rapid physiologic changes during luteolysis, and likely in the less characterized period of CL maintenance during pregnancy. Thus, there is a clear need to identify reference genes well suited for the bovine CL over a wide range of physiological states. Whole-transcriptome RNA sequencing stands as an effective method to identify new reference genes by enabling the assessment of the expression profile of the entire pool of mRNA transcripts. We report the identification of 13 novel putative reference genes using RNA sequencing in the bovine CL throughout early pregnancy and luteolysis: RPL4, UQCRFS1, COX4I1, RPS4X, SSR3, CST3, ZNF266, CDC42, CD63, HIF1A, YWHAE, EIF3E, and PPIB. Independent RT-qPCR analyses were conducted confirming expression stability in another set of CL tissues from pregnancy and regression, with analyses performed for 3 groups of samples: (1) all samples, (2) samples from pregnancy alone, and (3) samples throughout the process of CL regression. Seven genes were found to be more stable in all states than 2 traditional reference genes (ACTB and GAPDH): RPS4X, COX4I1, PPIB, SSR3, RPL4, YWHAE, and CDC42. When CL tissues from pregnant animals alone were analyzed, CST3, HIF1A, and CD63 were also identified as more stable than ACTB and GAPDH. Identification of these new reference genes will aid in accurate normalization of RT-qPCR results, contributing to proper interpretation of gene expression relevant to luteal physiology. Furthermore, our analysis sheds light on the effects of luteolysis and pregnancy on the stability of gene expression in the bovine CL.