Subfertility in grazing dairy cows: insights from a multi-omic exploration of the uterine microenvironment : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

Author: Aranciaga, Nicolas

Date: 2021

Publisher: Lincoln University

Type: Thesis

Link to this item using this URL:

Lincoln University


Dairy cow subfertility is a worldwide issue arising from multiple factors. One of its main manifestations is around 30% early pregnancy loss by day 7 of gestation in seasonal, pasture-grazed dairy herds. Pregnancy loss has a substantial impact on the seasonal grazing dairy cow as the mating period is short (six to nine weeks in length), and cows need to conceive while under strong metabolic stress to support their peak milk production. Factors in the uterine luminal fluid (ULF), on which the early embryo depends for sustenance and growth, appear to determine a portion of early pregnancy losses, and it is hypothesised that those factors improve with increasing days postpartum. The present study examined the molecular composition of uterine luminal fluid (ULF) in day-7 pregnant dairy cows. Eighty cows were inseminated, and their uteri flushed 7 days later at two different oestrus cycles within three months of calving. Embryos recovered in those flushing samples were graded to estimate their potential viability and relate it to uterine suitability for pregnancy. The molecular milieu of ULF was investigated using three techniques: label-free quantitative proteomic analysis by LC-MS/MS, targeted metabolomic analysis by GC-MS/MS, and metabolic fingerprinting by REIMS (direct infusion mass spectrometry). For a more comprehensive perspective of the animals’ conditions, potentially relevant variables at the cow-level were also measured and analysed. Moreover, links between embryo quality and differentially abundant molecules were investigated at the biochemical pathway level and by uni- and multivariate analyses, to screen for potential biomarkers of uterine suitability and to develop a predictive modelling pipeline. Clear indications of differences across time postpartum were observed in the cow-level variables, signifying contrasting metabolic conditions between early- and mid-postpartum, with interaction between the actual time (days postpartum) and the number of oestrus events (oestrus after calving). Concomitantly, 33% more good- and excellent-quality embryos were found with increased days and oestrus cycles postpartum, reflecting a general positive effect of postpartum recovery on reproductive function. A total of 1504 proteins were detected and measured in ULF, of which 472 had not been previously reported in this fluid. The abundance of 20 proteins varied relating to embryo quality, with various suggested roles in uterine function and embryogenesis-related pathways. Some of those proteins were macrophage migration inhibitory factor, phospholipase A2, myostatin, alpha-1-antiproteinase and prostaglandin reductase 1, involved in immune and development processes. Two proteins, cystatin C and pyruvate kinase M2, were more abundant in ULF with degenerate embryos (4-16 cells) and thus were considered promising protein biomarker candidates. For biological validation, the effect of those proteins was tested on in vitro embryo culture system, together with cathepsin B, a protease potentially relevant to embryo quality. Different concentrations of each test protein were added to culture media; their effect was assessed based on development to tight morula and blastocyst stages, as well as embryo grade. Some evidence of a positive (of cystatin C) or negative (of pyruvate kinase M) effect on embryo development was observed. Targeted metabolomic analysis of ULF showed 31 compounds’ abundance varying along days and/or oestrus cycles after calving, with most (n = 25) decreasing with increasing days postpartum. These were chiefly carbohydrates (e.g. xylulose, ribose, fructose), and organic acids (e.g. malic, ethylmalonic and glyceric acids). Based on metabolomics data, pathways dysregulated at early postpartum included glycine, serine, and threonine metabolism, glycerolipid metabolism, beta alanine metabolism, pentose and glucuronate interconversions, cysteine, and methionine metabolism. Joint pathway analysis of proteomics and metabolomics data uncovered differentially regulated metabolic, signalling, and immune processes across oestrus cycles. Furthermore, dysregulation of protein metabolism and EGFR1 signalling in ULF appear to influence embryo development past the 16-cell stage. A method for rapid metabolic fingerprinting based on a novel mass spectrometry technology (rapid evaporative ionisation mass spectrometry, REIMS) was tested for potential diagnostic applications. The method was successful at obtaining a distinct spectral profile of ULF, however its implementation for assessing uterine receptivity necessitates further instrumental optimisation. Multivariate analyses (PCA and PLS-DA) of proteomic and metabolomic data suggested that the molecular microenvironment of the uterus is determined by the interaction of multiple factors at the animal level, and modelling of these intricate mechanisms is more challenging than previously thought. In conclusion, this project’s results advanced the characterisation of the molecular environment of bovine ULF. This study also provided evidence of links between molecular abundance of proteins and metabolites in the uterine environment to cow postpartum recovery and putatively to embryo quality, pinpointing metabolic and signalling pathways as potential mechanisms of action. The relevance of these molecular changes for diagnosing pregnancy suitability requires further research, of which concurrent analysis of follicular and uterine fluids at early postpartum is particularly promising.

Subjects: embryo, proteomics, metabolomics, uterine fluid, systems biology, agriculture, pastoral farm system, dairy cows, dairy cow breeding, fertility, 060803 Animal Developmental and Reproductive Biology

Copyright: (With the exceptions noted in, this metadata is available under a Creative Commons Zero license.)