Genetic selection of cattle for reduced bovine tuberculosis transmission

Abstract

Bovine tuberculosis (bTB) is a major cattle disease with significant economic impact on production in multiple countries. In the United Kingdom, bTB remains a critical challenge, especially in high incidence areas despite control programmes being in place. These programmes consist of regular skin testing and culling of test positive cattle, movement restrictions, wildlife control, and other biosecurity measures. Nevertheless, successful eradication of the disease has yet to be achieved. Previous studies have suggested that breeding cattle for enhanced bTB resistance can complement existing eradication efforts. However, breeding for increased resistance alone may not be sufficient to help achieve the national target to eradicate bTB in the next decade. Therefore, genetic selection for low bTB infectivity, in addition to high bTB resistance, has been proposed as a possible solution to accelerate this process. However, the genetics of bTB infectivity need to be investigated in order to demonstrate the feasibility of including infectivity into national breeding programmes.

The overall aim of this thesis is to investigate statistical evidence of genetic variation underlying bTB infectivity of cattle in Great Britain (GB) and assess the feasibility and implications of considering infectivity as an additional breeding goal.

The primary objective of this thesis is to define a bTB infectivity phenotype, which is crucial for understanding how infection spreads among cattle. Additionally, the thesis aims to examine various factors affecting bTB infectivity, including environmental, management, and host genetic influences that might contribute to variation in disease transmission. Another key objective is to estimate genetic parameters related to bTB infectivity. Using these estimates, the thesis derives and assesses estimated breeding values (EBVs) for bTB infectivity. Finally, the thesis investigates the effect of selecting for reduced bTB infectivity.

Chapter 1 presents a literature review on bTB, its transmission and pathogenesis, diagnosis and current control strategies. Furthermore, the thesis aim, objectives and outline are presented.

Chapter 2 explores bTB data from Great Britain on both phenotypic and genetic levels and introduces the concept of the "index case approach" to define novel bTB infectivity phenotypes. Index case here refers to the first single positively tested animal in a herd that signals the start of a bTB breakdown. bTB infectivity is then defined as the number of secondary cases (NSC) attributed to the index case. Linear mixed models and generalized linear mixed models (GLMMs) are used to explore the effect of multiple factors and derive estimates of the genetic variance of bTB infectivity. The results produce the first estimates of genetic variation and heritability in the bTB infectivity. However, more advanced statistical models need to be explored to improve model fit and provide deeper insights into bTB infectivity and transmission dynamics.

In Chapter 3, Markov Chain Monte Carlo (MCMC) techniques are applied to fit GLMMs that can account for potential overdispersion and zero inflation issues in the data. Four different GLMMs (specifically Poisson, Zero-Inflated Poisson (ZIP), Hurdle Poisson, and Geometric models) are employed to detect and estimate genetic variation in infectivity. Factors affecting bTB infectivity from Chapter 2 are included in these models. The results show that genetic variation in bTB infectivity exists and is estimable. Furthermore, sire estimated breeding values (EBVs) are derived for bTB infectivity. Based on the estimated posterior mean genetic variances obtained, sire selection leading to a reduction in infectivity by one genetic standard deviation would result in a 32 - 44% decrease in the expected NSC per index case.

Chapter 4 focuses on the implications of incorporating bTB infectivity into breeding programmes. The potential reduction in the number of secondary cases is calculated from hypothetical genetic gains achieved from selection based on sire EBVs from Chapter 3. Results show that, in order to achieve 10% reduction in number of secondary cases using the infectivity phenotype introduced in this thesis, strong genetic progress would be required. This may involve high selection intensity, accurate EBV evaluations, and consistent use of genetically superior animals for breeding. Furthermore, correlations of sire EBVs for bTB infectivity with bTB resistance and other economically important traits are examined, using the infectivity EBVs estimated in Chapter 3 and existing EBV estimates for the other traits. Results show no antagonistic correlations with the other traits, suggesting no potential adverse effects from selecting for reduced infectivity.

In conclusion, this thesis proposes and examines a proxy infectivity phenotype for bTB. Exploration of the GB data reveals that there is underlying genetic variation in bTB infectivity, suggesting that the trait can be improved with genetic selection. Our results also suggest that implementing selection to decrease bTB infectivity is feasible and effective, and does not have any expected negative effect on other traits that are considered in the current breeding programme. Further studies are recommended to refine the infectivity phenotype for more accurate genetic evaluation and effective genetic selection for low bTB transmission.