Infection dynamics of Mycoplasma bovis in dairy herds
Mycoplasma bovis infections affect dairy cattle worldwide, yet few countries have implemented surveillance, control, or monitoring programs. Critical gaps remain in our understanding of the disease dynamics, transmission, and epidemiology. Additionally, diagnosing M. bovis is challenging due to suboptimal diagnostic tests, subclinical infections and intermittent shedding, making it difficult to identify infected animals. To address these challenges, a series of studies have aimed to improve both diagnostic protocols and understanding of transmission dynamics.
An evidence-based evaluation of carbon dioxide (CO2) requirements in culture protocols revealed that a wider range of CO2 conditions than previously described supports M. bovis growth, and CO2 may not be necessary at all. This finding offers potential for more adaptable and cost-effective culture protocols, which could improve the isolation and detection of M. bovis in both research and diagnostic settings.
Building on efforts to enhance the epidemiological understand, an age-stratified analysis of M. bovis transmission across 20 dairy herds using a Susceptible – Infected – Removed (SIR) compartmental model discovered significant heterogeneity in basic reproduction numbers (R0) between herds and highlighted the role of youngstock as a potential transmission reservoir. To further refine transmission parameters, a novel mathematical model was developed that integrated Bayesian latent class analysis with individual-level transmission modeling. This approach accounted for ‘prior knowledge’ regarding the varying diagnostic test performances, such as the poor sensitivity (Se) but high specificity (Sp). By incorporating these elements, the model provided a better fit to the available data, and yielded more accurate R0 estimates, though variability between herds persisted.
It is likely that the observed heterogeneity in transmission dynamics is influenced by strain differences. To address this, a new method was developed in-silico to identify strain differences using mock milk-samples containing pure and mixed infections. Various enrichment proportions and mixtures were tested, and the results demonstrated that Themisto and mSWEEP outperformed the widely used Kraken2 tool in accurately identifying M. bovis strains. Moreover, enriching M. bovis DNA to at least 30% of the total sequenced reads was found to be sufficient to obtain accurate PSV level data.
Finally, understanding the human element of disease control is essential for implementing effective strategies. A literature review was conducted on theories and methods used to study farmer behaviour concerning cattle disease control measures. While many studies focused on personal and interpersonal factors influencing the adoption of measures, only few addressed the broader contextual factors, and none studied actual farmer behaviour change. A deeper understanding of farmers’ motivators and barriers, and the broader sociocultural context is necessary to effectively design and implement animal disease control programs that lead to sustainable behavior change.
Together, these studies highlight the multifaceted nature of M. bovis transmission and control. By improving diagnostic methods, refining transmission models, understanding strain diversity, and considering farmer behavior, we can begin to bridge the gaps in our knowledge and develop more effective strategies to control this persistent disease.
Marit Biesheuvel recently obtained her PhD in Veterinary Epidemiology from the Faculty of Veterinary Medicine at University of Calgary in Canada. She previously completed an MSc in Animal Sciences at Wageningen University in The Netherlands. Between her studies, she worked as a Veterinary Epidemiologist at Royal GD in Deventer, the Netherlands. Currently, she works as a Technical Services Manager – Ruminants at Phibro Animal Health, Canada.
