Growth curves for cattle, and the effect of ammonia-treatment of cereal grains on bovine production performance and faecal proteome

Huang, Yixin (2023) Growth curves for cattle, and the effect of ammonia-treatment of cereal grains on bovine production performance and faecal proteome. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2023HuangYPhD.pdf] PDF
Download (3MB)


The rate of growth of livestock is an important determinant of profitable and environmentally sustainable production systems. Slowly growing animals usually have higher fixed costs of production than rapidly growing animals, and their greenhouse gas emissions intensity (GHG EI) is higher. This thesis investigates the growth of cattle – firstly by investigation of mathematical models of growth, and then by the investigation of the effects of a recently developed method for cereal grain preservation using enzymecatalysed urea. Finally, the thesis considers faecal proteomic examination as a potential tool to detect health and growth performance differences in cattle. The thesis had two main aims: (1) to find the most suitable growth models for cattle at different life stages using high-density bodyweight data and determine whether the best fitting model(s) improved parameter estimation in comparison with the traditional linear model; (2) to describe the effect of the enzyme-catalysed ammonia treatment of cereal grains on the growth performance of finishing beef cattle and their faecal proteome.

Chapter 2 describes investigations into animal growth models for cattle. Animal growth models can be used to quantify animal growth rates, inform about animal health status, and can serve as the basis for strategies to improve animal productivity and genetic selection. However, model fitting for retrospective performance of animal growth has mostly used relatively infrequent bodyweight (BW) observations. Recent technical developments have enabled multiple records of BW for every single animal every day, automatically weighing animals when they drink milk replacer or water, during milking, or when moving between yards or pens. The logistic, Brody, Gompertz, von Bertalanffy and Richards models were applied to high-density but intermittent bodyweight data from the whole of life of Holstein-Friesian cows; the linear, quadratic, cubic, power and exponential models were fitted to near-continuous BWs of Holstein and Holstein-cross calves for the first 100 days or so of life; the linear, logarithmic and negative exponential models were fitted to nearcontinuous bodyweights of mixed-breed beef cattle for the last 100 days or so before slaughter. Model comparisons were based on goodness-of-fit statistics and estimations of biological parameters. Generalized linear models were fitted to compare the strength of association between known correlates of growth rate and observed growth rate using two alternative models of growth. Full data sets from the calves and reduced (sparse) data sets were used for comparison of parameter estimate precision derived from the best fitting model and the traditional linear model. The von Bertalanffy model was the best growth model to describe the whole of life of the lactating Holstein-Friesian cows, the exponential model was the most suitable model for calves, and both the linear and the negative exponential models performed well for fitting to the growth of the finishing beef cattle. Application of the exponential model in calves improved the precision of parameter estimation and the sensitivity of analysis of known correlates of growth compared with the traditional linear model, with both high-density and sparse data.

Chapter 3 describes two trials: one was conducted on a Scottish beef finishing unit where the cattle were fed on a diet based on barley that was treated with ammonia (ATB) or propionate preservatives (PTB), and the other was conducted in a typical Italian beef fattening system where the cattle were fed on a maize-based diet with ammonia treatment (ATM) or without the ammonia treatment (UTM). Consistent with the findings in Chapter 2 that there was little consistent benefit in using non-linear models rather than traditional linear models for analysing growth of cattle in the fattening period, the growth rate estimates in chapter 3 used the traditional method - average daily gain (ADG). Growth performance of animals from groups in each trial were compared, and effects of the treatment on ruminal volatile fatty acids and faeces were studied. The enzyme-catalysed ammonia treatment of grain decreased FCR and faecal starch concentrations, having similar effects of improving the growth performance of cattle in beef fattening systems to those previously reported in studies u sing direct insufflation with anhydrous ammonia. This method of processing cereal grains has the potential to increase nutrient utilization on commercial cattle farms.

Chapter 4 addresses a potential problem with faecal proteomic studies for cattle: whether highly abundant high molecular weight glycoproteins have detrimental effects on protein identification from faeces that were prepared by filter-aided sample preparation (FASP) method. Therefore, an in-gel sample preparation method (IGSP) was developed, which increased the number of bovine faecal protein identifications.

In Chapter 5, to follow up on the apparent differences in faecal starch and occurrence of diarrhoea in the cattle that were fed ATB or PTB in Chapter 3, the faecal proteomes of a subset of animals were investigated. Faecal samples were analysed by nanoflow ultrahighperformance liquid chromatography-electrospray ionisation-tandem mass spectrometry (nUHPLC-ESI-MS/MS, Orbitrap Elite) after IGSP as developed in Chapter 4, trypsin digestion and TMT labelling. Data were assigned using the Sequest HT search engine to interrogate sequences in bovine, barley, bacterial (Clostridium, Bacteroides, Ruminococcus, Prevotella and Eubacterium) and archaeal (30 methanogenic genera) databases in Swissprot and TrEMBL using Proteome Discoverer (PD). Pairwise protein abundance ratios for animals on the two diets were calculated, with the hypothesis test as the background-based t-test in PD. Antibodies to bovine serum albumin and barley serpin Z4 were used in western blots (WB) to validate the results from proteomics. In total, 281 bovine proteins, 199 barley proteins, 176 bacterial proteins and 190 archaeal proteins were identified in the bovine faeces. The presence of bovine serum albumin and barley serpin Z4 were confirmed by WB. Mucin 2 was the most abundant host protein identified in the faeces, and many host digestive enzymes and protease inhibitors were also found. Barley serpin Z4 was the most abundant barley protein identified in the faeces. Many microbial proteins were identified in the faecal samples, with a large proportion of bacteria from Clostridium, and Methanobrevibacter was the dominant archaeal genus. The host proteins were significantly over-represented in biological processes such as microtubule-based movement, defence response to Gram-positive bacterium, negative regulation of endopeptidase activity, cell migration and proteolysis. The microbial proteins were over-represented in biological processes including carbohydrate metabolism, gluconeogenesis, glucose metabolism and glycolysis. Thirty-nine proteins were differentially abundant in the two treatment groups, the majority being more abundant in the PTB group compared to the ATB group (28 vs 11).

The investigations described in this thesis identified the best models for growth of cattle at different life stages, and demonstrated the potential to improve the precision of parameter estimations by the most suitable nonlinear models as an alternative to the traditional linear model for calves up to about four months old. The enzyme-catalysed ammonia treatment improved the production performance of finishing beef cattle in two distinct systems. New protocols for faecal proteomic investigation were developed and the effects of enzymecatalysed ammonia treatment on the faecal proteome were examined. The host, dietary, and microbial proteins of bovine faecal samples in this system were identified, providing a foundation for the future study of cattle GI-related diseases and optimizing diets for cattle to improve performance.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: Q Science > QR Microbiology
S Agriculture > SF Animal culture
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Biodiversity, One Health & Veterinary Medicine
Supervisor's Name: Jonsson, Professor Nicholas, Eckersall, Professor David and Johnson, Dr. Paul
Date of Award: 2023
Depositing User: Theses Team
Unique ID: glathesis:2023-83447
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 22 Feb 2023 11:42
Last Modified: 22 Feb 2023 14:39
Thesis DOI: 10.5525/gla.thesis.83447
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year