Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. The Diagnosis of an Outbreak of Mycoplasma bovis Clinical Mastitis in a Multi-farm North Otago Farming Operation A thesis presented in partial fulfilment of the requirements for the degree of Master of Veterinary Science at Massey University, Manawatū, New Zealand. Kevin Patrick Kearney 2021 i Abstract Mycoplasma bovis (M. bovis) causes a multitude of disease syndromes in dairy cattle including clinical mastitis (CM), arthritis and pneumonia. The detection in July 2017 of M. bovis, for the first time in New Zealand (NZ), on a South Island dairy farm, prompted a national animal disease response. This descriptive study aims to describe the clinical and diagnostic test findings of an outbreak of M. bovis CM, on a large multi-farm dairy enterprise where there was a single hypothesised infection source and date. Samples were collected as part of surveillance activities on-farm and at slaughter, together with farmer- selected CM cows, to provide results from real-time polymerase chain reaction (qPCR) and enzyme- linked immunosorbent assay (ELISA) tests of bulk tank milk (BTM), individual cow serum ELISA tests, quarter milk samples (QMS), and palatine tonsils qPCR tests. Post-mortem sampling of the mammary glands of M. bovis CM cases was also performed. Positive BTM PCR, supported by BTM ELISA, confirmed infection in two of the four dairy herds in the enterprise and herd-level serology (serum ELISA) confirmed infection in a third herd. There was a common clinical presentation in infected herds of an unusually high incidence of apparent treatment failure (ATF) of non-systemically ill, multiple quarter CM cases, from some of which M. bovis was detected. Individual CM cases were found in the main to be QMS M. bovis qPCR positive, serum ELISA positive and palatine tonsil qPCR positive. In approximately 70% of M. bovis CM cases, M. bovis was found to be the sole pathogen. A smoothed function model between serum ELISA and time from first diagnosis of CM, from which M. bovis was detected, predicted that the average interval between clinical diagnosis and a serum ELISA test positive result was five days. The higher observed agreement between the serum ELISA and palatine tonsil qPCR, was for M. bovis CM cows sampled on-farm compared with cows sampled at slaughter. Gross lesions of fibrosis, caseous necrosis and cystic dilation in the udders of M. bovis positive CM cows were seen together with granulomatous and suppurative inflammatory patterns histologically. High immunoreactivity in immunohistochemistry for the M. bovis antigen was also present. From the key diagnostic test findings, M. bovis was likely to have been one of several pathogens which caused individual cases of CM on the farming enterprise, and in many cases may have been the sole cause of CM cases. The results of this study can raise awareness of and provide information to aid dairy farmers and veterinarians determine if M. bovis has a role in CM outbreaks with unexpectedly increased numbers of treatment failures and can inform the regulatory response for surveillance and testing of herds and individual cattle for M. bovis. ii Acknowledgements During the learning and writing phases of this thesis, I have had tremendous help and support from many different people from several different fields. Collectively everyone has aided the completion of this study. The farm owners, of this North Otago dairy operation, allowed me to be very involved with the outbreak from the outset, and permitted me access to their herds, in a very stressful farming and family situation. My three supervisors at Massey University, Palmerston North, NZ - Christopher Compton (chief supervisor), Fernanda Castillo-Alcala and Richard Laven have offered outstanding academic support, expertise in their respective fields, patience with the constant stream of emails, encouragement and have been very generous with their time. Veterinary epidemiologists at MPI, Amy Burroughs and Kate Sawford, have upskilled my epidemiology knowledge and understanding, from a Programme perspective. A special thank-you to Alan Harvey, the ICP Manager throughout the outbreak and Emma Bramley for retrieving data from the MPI database for me. Penelope Bilton, Mosgiel, ‘adopted’ me early in my study and enthusiastically taught and assisted me with statistics/epidemiology, R, and modelling work. Winston Mason, epidemiologist at VetEnt, also assisted me later in the study. The advice and experience of both proved invaluable. My veterinary partners at the Veterinary Centre, Hamish Newton (Oamaru) and Ryan Luckman (Waimate) have been sounding boards for ideas and very tolerant teachers of the mechanics of spreadsheets. Kirsty McMurtrie, Veterinary Centre Oamaru has been a cornerstone of help with the workings of Word, design of Tables and final formatting. Cristina Gans, Palmerston North, for the inclusion of her histopathology and IHC work, from her Master’s Pathology thesis. Massey University library staff, especially Kim Baxter and Jeff Phillips have been very generous with their time, promptly answering a lot of queries. Early in this study, Larry Fox, USA, willingly shared his wealth of experience in this field. And finally, but certainly not least, to my wife Clare and four adult children who have kept a keen eye on progress – thank you. iii Table of Contents Chapter 1 Introduction ........................................................................................................................... 1 1.1 Introduction .................................................................................................................................. 1 1.2 Format of Thesis ........................................................................................................................... 3 Chapter 2 Literature Review ................................................................................................................... 4 2.1 Introduction .................................................................................................................................. 4 2.2 Mycoplasma bovis – The Pathogen .............................................................................................. 4 2.2.1 Microbiology ......................................................................................................................... 4 2.2.2 Mycoplasma Species ............................................................................................................. 5 2.2.3 The Use of Molecular Techniques for Typing Mycoplasma Strains ...................................... 6 2.3 Epidemiology ................................................................................................................................ 8 2.3.1 Pathogenicity ......................................................................................................................... 8 2.3.2 Transmission ........................................................................................................................ 12 2.3.3 Colonisation ......................................................................................................................... 14 2.3.4 Shedding .............................................................................................................................. 15 2.3.5 Prevalence ........................................................................................................................... 16 2.3.6 Incidence ............................................................................................................................. 19 2.3.7 Herd-Level Risk Factors ....................................................................................................... 19 2.4 Clinical Manifestations of Mycoplasma bovis ............................................................................ 21 2.4.1 Mastitis ................................................................................................................................ 21 2.4.2 Otitis Media in Calves .......................................................................................................... 24 2.4.3 Genital Disorders ................................................................................................................. 25 2.4.4 Pneumonia .......................................................................................................................... 25 2.4.5 Keratoconjunctivitis ............................................................................................................ 26 2.4.6 Arthritis................................................................................................................................ 26 2.5 Diagnostic Techniques ................................................................................................................ 27 2.5.1 Microbiology ....................................................................................................................... 27 2.5.2 Molecular Diagnostics ......................................................................................................... 27 2.5.3 Immunodiagnostics ............................................................................................................. 29 2.5.4 Immunohistochemistry ....................................................................................................... 30 2.6 Sensitivity and Specificity of Tests .............................................................................................. 31 2.7 Pathology of Mycoplasma bovis Mastitis ................................................................................... 32 iv 2.7.1 Gross Lesions ....................................................................................................................... 32 2.7.2 Histopathology .................................................................................................................... 32 2.8 Control and Eradication .............................................................................................................. 33 2.8.1 Test and Segregate/Test and Slaughter .............................................................................. 33 2.8.2 Vaccination .......................................................................................................................... 34 2.8.3 Antimicrobial Therapy ......................................................................................................... 35 2.9 Economic Cost of M. bovis .......................................................................................................... 36 Chapter 3 Materials and Methods ........................................................................................................ 37 3.1 Background to Study................................................................................................................... 37 3.2 Enterprise Background ............................................................................................................... 38 3.2.1 Outline of Farming Operation ............................................................................................. 38 3.2.2 Aspects of Enterprise History .............................................................................................. 41 3.3 Samples Collected from Farms and Selection of Animals for Testing ........................................ 42 3.3.1 Herd-level Sampling ............................................................................................................ 42 3.3.2 Animal-level Sampling ......................................................................................................... 42 3.3.3 Quarter-level Sampling ....................................................................................................... 43 3.3.4 Agreement Study ................................................................................................................. 43 3.3.5 Pathology samples at slaughter .......................................................................................... 44 3.4 Definitions Used in Case Study ................................................................................................... 47 3.4.1 Definition of Outbreak of M. bovis Clinical Mastitis ........................................................... 47 3.4.2 Terms for M. bovis Mastitis ................................................................................................. 47 3.4.3 Case Definitions ................................................................................................................... 47 3.4.4 Date of Detection of M. bovis on Study Farms ................................................................... 49 3.5 Further Aspects of the Enterprise .............................................................................................. 52 3.5.1 Young Stock Movements ..................................................................................................... 52 3.5.2 Cow Movement to Farm 4 in Winter 2018 ......................................................................... 53 3.6 Antimicrobial Therapies Used for Mastitis Treatments ............................................................. 53 3.7 Methodology of Sample Collection ............................................................................................ 53 3.7.1 Herd-level ............................................................................................................................ 53 3.7.2 Animal-level ......................................................................................................................... 54 3.7.3 Quarter Sampling ................................................................................................................ 57 3.7.4 Pathology Samples of Mammary Glands from M. bovis CM Cows at Slaughter ................ 57 3.8 Laboratory Analysis .................................................................................................................... 58 v 3.8.1 DNA Extraction and M. bovis qPCR ..................................................................................... 58 3.8.2 Confirmatory Conventional PCR Testing and Sequencing at AHL ....................................... 58 3.8.3 Other Methodologies (see Appendix Document 3 Chapter 3, 6.1.3) ................................. 59 3.9 Data Sources ............................................................................................................................... 59 3.9.1 The Programme Database ................................................................................................... 59 3.9.2 The Programme Exotic Disease Investigative Report (EDIR) ............................................... 59 3.9.3 Infovet (Zoetis New Zealand Ltd, Auckland, NZ) ................................................................. 59 3.9.4 MINDA®, LIC, Hamilton, NZ ................................................................................................. 60 3.9.5 Vision VPT™ Covetrus Software Services, Australia ............................................................ 60 3.9.6 Farmer Records ................................................................................................................... 60 3.10 Statistical Analysis and Data Management .............................................................................. 60 3.10.1 R Core Team (2018) ........................................................................................................... 60 3.10.2 Bulk Tank Milk ELISA Linear Regression ............................................................................ 60 3.10.3 Seroconversion and Seroprevalence of Farms 1 to 4........................................................ 61 3.10.4 Data Management for Modeling of Individual M. bovis CM Cow ELISA SP% Over Time .. 62 3.10.5 Incidence of Clinical Mastitis ............................................................................................. 63 3.10.6 Analysis of Age of M. bovis CM cows ................................................................................ 64 3.10.7 Agreement Study Between Serum ELISA SP% and Tonsil Swab qPCR in Individual Cows 64 3.11 Prospective Cohort Study on Farm 1 ........................................................................................ 67 3.12 Ethics approval ......................................................................................................................... 67 Chapter 4 Results .................................................................................................................................. 68 4.1 Herd-level Results ....................................................................................................................... 70 4.1.1 Bulk Tank Milk PCR .............................................................................................................. 70 4.1.2 Bulk Tank Milk ELISA Across the Four Study Farms ............................................................. 72 4.1.3 Bulk Tank Milk ELISA Linear Regression .............................................................................. 72 4.1.4 Distribution of SP% Over Time for Study Farms 1-4 Aggregated by Farm Sampling Date .. 74 4.1.5 Estimated True Infection Prevalence .................................................................................. 77 4.1.6 Comparison Over Time of BTM ELISA and Estimated True Infection Prevalence ............... 79 4.2 Animal-level Results ................................................................................................................... 80 4.2.1 Individual Cow Composite Milk Sampled (CMS) at Programme Surveillance Visits ........... 80 4.2.2 Vet Visits to Study Farms 1 and 2 for Collection of Samples from Farmer-selected CM Cows ............................................................................................................................................. 81 4.2.3 Mycoplasma bovis CM Individual Cow SP% Over Time ...................................................... 82 4.2.4 Incidence of Clinical Mastitis ............................................................................................... 85 vi 4.2.5 Age of M. bovis CM Cows and CM Cows for Farms 1 and 2 ................................................ 87 4.2.6 Subclinical Intramammary Infection (IMI) .......................................................................... 90 4.2.7 Calf Surveillance .................................................................................................................. 91 4.3 Quarter-level Results .................................................................................................................. 92 4.3.1 Quarter Milk Samples from CM Cows Delivered to the Clinic from Farm 1, 2 and 3.......... 92 4.3.2 Other Pathogens Present in Milk with M. bovis .................................................................. 92 4.3.3 Clinical Presentation of M. bovis CM During Outbreak ....................................................... 94 4.4 Agreement Study Between Serum ELISA SP% and Tonsil Swab in Both Clinical and Non-clinical Surveillance Slaughter Cows ............................................................................................................. 99 4.4.1 Two by Two Tables for Analysis of Palatine Tonsil Swabs for M. bovis qPCR and M. bovis Serum ELISA .................................................................................................................................. 99 4.4.2 Agreement Between Serum ELISA and Palatine Tonsil qPCR on Farms 1 and 2 ............... 102 4.5 Morphological Patterns of Mammary Gland Lesions in Dairy Cows with M. bovis Clinical Mastitis ........................................................................................................................................... 103 4.5.1 Gross Pathology ................................................................................................................. 103 4.5.2 Histopathology .................................................................................................................. 106 4.5.3 Immunohistochemistry ..................................................................................................... 106 Chapter 5 Discussion ........................................................................................................................... 109 5.1 Introductory Comments and Key Results ................................................................................. 109 5.2 Herd-level Investigations .......................................................................................................... 111 5.2.1 Bulk Tank Milk qPCR and Bulk Tank Milk ELISA ................................................................. 111 5.2.2 Individual Cow Serum ELISA Testing (aggregated from individual up to herd-level, by farm sampling date) ............................................................................................................................ 112 5.2.3 Estimation of True Prevalence .......................................................................................... 113 5.2.4 Association Between BTM ELISA and Estimated True Infection Prevalence .................... 114 5.2.5 Bulk Milk Somatic Cell Count ............................................................................................ 114 5.3 Animal-level Investigations ....................................................................................................... 115 5.3.1 Mycoplasma bovis CM Cases and SP% Values .................................................................. 115 5.3.2 Longevity of anti-M. bovis Antibodies ............................................................................... 116 5.3.3 Incidence of Clinical Mastitis ............................................................................................. 117 5.3.4 Age Incidence of M. bovis CM Cows ................................................................................. 118 5.3.5 Pattern of the CM Outbreak.............................................................................................. 119 5.3.6 Subclinical IMI (M. bovis IMI) ............................................................................................ 120 5.3.7 Calf Surveillance Testing .................................................................................................... 120 vii 5.4 Quarter-level Investigations ..................................................................................................... 121 5.4.1 Other Pathogens Present in Milk of M. bovis CM Cows ................................................... 121 5.4.2 Clinical Aspects of the North Otago M. bovis CM Outbreak ............................................. 122 5.4.3 Antimicrobial Therapy for M. bovis CM ............................................................................ 124 5.5 Part 2 Agreement Study ........................................................................................................... 126 5.6 Part 3 Pathological Findings...................................................................................................... 127 5.7 The Diagnosis ............................................................................................................................ 129 5.8 Limitations ................................................................................................................................ 130 5.9 Future Research ........................................................................................................................ 131 Chapter 6 Appendices ......................................................................................................................... 132 6.1 Appendices for Chapter 3 Materials and Methods .................................................................. 132 6.1.1 Document 1 Antimicrobial Therapy Used for Mastitis Treatments .................................. 132 6.1.2 Document 2 Laboratory Analysis Methodologies (Mycoplasma bovis) ............................ 134 6.1.3 Document 3 Other Laboratory Analysis ............................................................................ 137 6.1.4 Document 4 Methodology of Prospective Cohort study .................................................. 139 6.2 Appendices for Chapter 4 Results ............................................................................................ 140 6.2.1 Document 1 BMSCC Curves for Farms 1-4 ........................................................................ 140 6.2.2 Document 2 Age and Farm Origin ..................................................................................... 142 6.2.3 Document 3 Apparent and Estimated True Prevalence .................................................... 143 6.2.4 Document 4 Historical Information About Cows Milk Sampled at Surveillance ............... 144 6.2.5 Document 5 Model for SP Values...................................................................................... 145 6.2.6 Document 6 Pathological Findings .................................................................................... 147 Chapter 7 References .......................................................................................................................... 149 viii List of Figures Figure 3-1 Map of North Otago showing location of study farms ........................................................ 38 Figure 3-2 Diagrammatic representation of animal movements and sampling times in the dairy farming enterprise Including four dairy units (ID 1 – 4 on left of graphic) ........................................... 51 Figure 3-3 Visualization of the palatine tonsillar crypts, via PVC pipe in an adult dairy cow .............. 55 Figure 3-4 Exposed nasopharynx and palantine tonsil swab collection at slaughter .......................... 56 Figure 4-1 Bulk tank milk (BTM)1 Mycoplasma bovis PCR test results across 4 study farms sampled between 30.8.2018 and 4.3.2019 ......................................................................................................... 71 Figure 4-2 Bulk tank milk (BTM)1 Mycoplasma bovis ELISA across 4 study farms sampled between 30.8.2018 and 4.3.2019 ........................................................................................................................ 73 Figure 4-3 Distribution of individual cow M. bovis SP% for Farms 1-4 aggregated from individual up to herd-level, by farm sampling date .................................................................................................... 75 Figure 4-4 True Prevalence for study Farms 1 – 4 for Serum ELISA surveillance testing of cows ....... 78 Figure 4-5 Bulk tank milk ELISA (black dots) and estimated true infection prevalence (red dots) from serum sampling over time for study Farms 1-4 .................................................................................... 79 Figure 4-6 Farm 1 and Farm 2 plot of individual cows SP% over time ................................................. 82 Figure 4-7 Smoothed relationship between M. bovis serum ELISA SP% and time from first diagnosis of clinical mastitis from which M. bovis was also detected from a general additive model of data from Farms 1 and 2 ............................................................................................................................... 84 Figure 4-8 Farm 1 Stacked bar plot of CM cases per week at cow level, and outcome of individual cow milk M. bovis qPCR testing for cases sampled .............................................................................. 86 Figure 4-9 Farm 2 Stacked bar plot of CM cases per week at cow level, and outcome of individual cow milk M. bovis qPCR testing for cases sampled .............................................................................. 86 Figure 4-10 Seasonal incidence until depopulation by farm and age group of clinical mastitis (in red), M. bovis clinical mastitis (in green) and national level (in blue)1 ......................................................... 89 Figure 4-11 Farm 1. Two cows presented with light quarters. ............................................................ 96 Figure 4-12 Mycoplasma bovis was detected in all samples. (a), (b) and (e) Farm 1. (c) and (d) Farm 2 .............................................................................................................................................................. 97 Figure 4-13 Mammary glands of cows affected with M. bovis clinical mastitis - exudates. .............. 103 Figure 4-14 Mammary glands of cows affected with M. bovis clinical mastitis – tissue fibrosis....... 104 Figure 4-15 Mammary glands of cows affected with M. bovis clinical mastitis – caseous necrosis . 105 Figure 4-16 Mammary glands of cows affected with M. bovis clinical mastitis – cystic dilatation ... 105 Figure 4-17 Mammary glands of cows affected with M. bovis clinical mastitis – histopathology .... 106 Figure 4-18 Immunohistochemistry for M. bovis antigen ................................................................. 107 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876667 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876670 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876670 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876674 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876675 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876675 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876675 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876676 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876676 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876677 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876677 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876679 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876680 file://///kevins-hp/Data/Masters%202019/Copy%20M%20bovis%208.12.20/Formatting%20Long%20Document/19.7.21%20Using%20Draft%202%20FC.docx%23_Toc77876680 ix Figure 6-1 BMSCC Farms 1-4 2018/19 (red), 2017/18 (blue) ............................................................ 140 x List of Tables Table 2-1 Prevalence of Mycoplasma bovis .......................................................................................... 18 Table 3-1 Key Performance Indicators and test results for the dairy farming enterprise described in this study ............................................................................................................................................... 40 Table 3-2 Summary of herd-, animal-, quarter- level testing of Programme surveillance and vet diagnostic samples and tests taken during outbreak for Farms 1-4 on-farm and at slaughter. Unshaded is on-farm Programme surveillance ..................................................................................... 45 Table 3-3 Dates of vet visits to study Farms 1 and 2 for sampling and management of outbreak, and dates of slaughter of Mycoplasma bovis CM cows (number of cows) ................................................. 46 Table 3-4 Diagnostic test cut points with programme estimates of performances ............................ 46 Table 4-1 Bulk tank milk (BTM) Mycoplasma bovis qPCR test results from study Farms 1 – 4 sampled between 30.8.2018 and 4.3.2019 ......................................................................................................... 70 Table 4-2 Bulk tank milk (BTM) Mycoplasma bovis ELISA test results from study Farms 1 – 4 sampled 30.8.2018 and 4.3.2019 ........................................................................................................................ 72 Table 4-3 Regression output of bulk tank milk (BTM) ELISA and Farm................................................ 72 Table 4-4 Count and proportion of individual cow Mycoplasma bovis serum ELISA test positive results for on farm surveillance visits and slaughter surveillance for study Farms 1-4 ........................ 76 Table 4-5 Age group prevalence of Mycoplasma bovis serum ELISA test results for on farm surveillance Visit 1 for study Farms 1-4. ............................................................................................... 76 Table 4-6 Trends in true prevalence for study Farms 1-4 .................................................................... 77 Table 4-7 Individual cow milk samples for qPCR at Programme surveillance visits for study Farms 1- 4, proportion test detected. ................................................................................................................. 80 Table 4-8 Count and proportion results for sampling carried out at vet visits of farmer-selected clinical mastitis (CM) cows on study Farms 1 and 2 ............................................................................. 81 Table 4-9 Results of the final general additive model of the association of SP% with estimated time since diagnosis for combined data for Farm 1 and Farm 2 ................................................................... 83 Table 4-10 Cumulative Incidence (herd level) and Incidence rate (cow level) of clinical mastitis on study Farms 1-4 ..................................................................................................................................... 85 Table 4-11 Counts of disease incidence for Farm 1 ............................................................................. 87 Table 4-12 Counts of disease incidence for Farm 2 ............................................................................. 87 Table 4-13 Standardised residuals for Farm 1 ..................................................................................... 87 Table 4-14 Standardised residuals for Farm 2 ..................................................................................... 88 Table 4-15 Visit dates for on-farm surveillance calf nasal swabs for M. bovis qPCR, and slaughter surveillance for palatine tonsil swabs qPCR and M. bovis serum ELISA ............................................... 91 xi Table 4-16 Proportion of farmer-selected and sampled quarter milk samples (QMS) from cows with clinical mastitis (both apparent treatment failures and apparent treatment success) delivered to the clinic from study Farms 1-3 that had M. bovis detected by qPCR tests ............................................... 92 Table 4-17 Other pathogens present in QMS M. bovis positive CM on a gland level, for study Farms 1 and 2 ..................................................................................................................................................... 93 Table 4-18 Number of quarters recorded as treated at first treatment of clinical mastitis for Farms 1 and 2 ..................................................................................................................................................... 94 Table 4-19 Spread to a second quarter in M. bovis CM diagnosed cows ............................................ 95 Table 4-20 Spread of clinical mastitis in M. bovis diagnosed cows to 3 or 4 quarters ........................ 95 Table 4-21 Light (agalactic) quarters and gross udder findings in M. bovis CM cows ......................... 96 Table 4-22 The number of mastitis treatments for M. bovis CM cows on Farms 1 and 2 ................... 98 Table 4-23 Observed Agreement and Gwet AC1 Value for Clinicals and Surveillance Slaughter Cows ............................................................................................................................................................ 102 Table 4-24 Description of gross lesions, Histological lesion characterization, Special stains and Immunohistochemistry from 55 M. bovis clinical mastitis quarters at slaughter. ............................. 108 Table 6-1 Age and Farm origin at Surveillance Blood Test 1 Farm 4 .................................................. 142 Table 6-2 Age and Farm origin at Surveillance Blood Test 2 Farm 4 ................................................. 142 Table 6-3 Apparent and Estimated True Prevalence of Farms 1-4 for Programme Surveillance Testing ............................................................................................................................................................ 143 Table 6-4 Historical information about cows milk sampled at surveillance ...................................... 144 Table 6-5 Results of the final general additive model of the association of SP% with estimated time since diagnosis, for Farm 1 and Farm 2 .............................................................................................. 146 Table 6-6 Description of gross morphological lesions, lesion characterisation, special stains & IHC results from 55 M. bovis Clinical Mastitis quarters from 14 dairy cows at slaughter ........................ 147 xii Glossary List of Abbreviations Ab – Antibody ACR – Automatic cup removers AHL – Animal Health Laboratory AI – Artificial insemination AFLP – Amplified fragment length polymorphism AFO – Acid fast organism AP – Apparent prevalence ATF – Apparent treatment failure ATS – Apparent treatment success BHV-1 – Bovine herpes virus 1 BLAST - Basic Local Alignment Search Tool BMSCC – Bulk milk somatic cell count BMT – Bulk milk testing BPI-3 – Bovine Parainfluenza-3 virus BRDC – Bovine respiratory disease complex BRSv – Bovine respiratory syncytial virus BTM – Bulk tank milk BVDV – Bovine diarrhoea virus cfu – colony forming units CM – Clinical Mastitis CMI – Cell mediated immunity CMS – Composite milk sample CNS – Coagulase Negative Staphylococcus CT – Cycle threshold DDGE – Denaturing gradient gel electrophoresis DNA – Deoxyribonucleic acid EBL – Embryonic bovine lung edf – Estimated degrees of freedom xiii ELISA – Enzyme-linked immunosorbent assay GAM – Generalised additive model HE – Haematoxylin and eosin ICR – In calf rate Ig - Immunoglobulin IHA – Indirect Haemagglutination IHC – Immunohistochemistry i.m - Intramuscular IMI – Intramammary infection IP – Incubation period ISCC – Individual somatic cell count KPI – Key performance indicators LCA – Latent class analysis LIC – Livestock Improvement Corporation LID – Lifetime identification LOSM – Lower order sharemilker MALDI-TOF - matrix-assisted laser desorption/ionisation – time of flight mass spectrometry MbAD – Mycoplasma bovis-associated disease MIC – Minimum inhibitory concentration MLST – Multi-locus sequence typing MLVA – Multiple locus variable number tandem repeat analysis MPI – Ministry for Primary Industries NADH - Nicotinamide Adenine Dinucleotide Hydrogen NAIT – National animal identification and tracing NCBI – National Centre for biotechnology information NEB – Negative energy balance NEFA – Non-esterified fatty acid NGS – Next generation sequencing NO – North Otago NZ – New Zealand OAD – Once a day milking xiv ODC – Optical density coefficient PBMC – Peripheral blood mononuclear cells PCR – Polymerase chain reaction PFGE – Pulse-field gel electrophoresis PI – Persistently infected PM – post-mortem QMS – Quarter milk sample qPCR – Real-time PCR RAPD – Random amplified polymorphic DNA RFID – Radio-frequency identification RMT – Rapid mastitis test ROS – Reactive oxygen species RNS – Reactive nitrogen species rRNA – Ribosomal RNA s.c – Subcutaneous SCM – Subclinical mastitis S/P – Sample to positive Se – Sensitivity Sp – Specificity spp. - Species ST – Sequence type TAD – Twice a day milking TMR – Total Mixed Ration URT – Upper respiratory tract Vsps – Variable surface lipoproteins WGS – Whole genome sequence WHP – Withholding period WHT – Withholding time 1 Chapter 1 Introduction 1.1 Introduction Mycoplasma bovis is a highly contagious pathogen of cattle and is known to cause significant economic losses and animal welfare impacts worldwide (Nicholas, 2011). There are numerous clinical manifestations of M. bovis in a herd, including mastitis, arthritis and pneumonia in cattle of all ages (Maunsell et al., 2011; Nicholas & Ayling, 2003), and otitis media in young calves (Maunsell et al., 2012). The presence of asymptomatic carrier animals in an infected herd, together with the poor diagnostic test sensitivity of samples from individual animals, make M. bovis elimination from a herd very difficult (Gille et al., 2018; Maunsell et al., 2011). Additionally, there are still large gaps in our understanding of the epidemiology and pathophysiology of M. bovis disease (Calcutt et al., 2018) that make control of this pathogen difficult (Fox et al., 2005). The clinical signs of mycoplasmal mastitis and specifically M. bovis clinical mastitis (CM) are considered nonspecific. The classical presentation is a mastitis which progresses to a chronic multi-quarter mastitis, with a marked drop in milk production and milk quality, but systemic clinical signs are rarely reported. Clinical cases are unresponsiveness to antimicrobial therapy, and the mastitis may become purulent in nature and progress to agalactia (Nicholas et al., 2016; Pothmann et al., 2015; Radaelli et al., 2011). Additionally, the epidemiological characteristics of M. bovis infection at the herd-level are reported to be related to management factors such as herd size, introduction of stock to a farm and animal replacement policies (McCluskey et al., 2003; Punyapornwithaya et al., 2010). However, the main body of work on the clinical manifestations of M. bovis has been reported in European farming system (Petersen et al., 2018; Vähänikkilä et al., 2019), which are quite dissimilar to those in New Zealand (NZ) which may therefore impact clinical presentation. The July 2017 detection of M. bovis in a NZ South Island dairy farm prompted a national response. In NZ, the Ministry for Primary Industries (Manatu Ahu Matua) (MPI), is the competent authority which responds to incursions of Unwanted and New Organisms, in accordance with the Biosecurity Act 1993 (Government, 1993). In May 2018, the NZ Government, together with the dairy and beef industries, made the decision to eradicate M. bovis from NZ (Ministry for Primary Industries, 2019b; New Zealand Government, 2018). At the time of undertaking this research, there was only one brief report of M. bovis clinical disease in a NZ dairy herd, from the index case (Hay, 2018). Additionally, the management effects on the epidemiology of M. bovis and the use of different diagnostic tests in other countries limit the relevance of the descriptions of outbreaks in other countries to the NZ farmer and veterinarian. Therefore, there 2 is a need to describe an outbreak of M. bovis CM in the NZ setting. This thesis investigates an outbreak of M. bovis CM on a multi-farm dairy operation and reports, in detail, herd-, cow- and quarter-level epidemiological findings, gross and microscopic lesions in infected mammary glands, as well as considering the agreement of two diagnostic tests used in the situation. The three tenets of a veterinary diagnosis of a disease are firstly identify, the specific cause, secondly identify the abnormality of structure or function that the causative agent has produced and which is detrimental to normal body structure or function, and thirdly, identify the clinical manifestation of the causative agent (Radostits et al., 2000). Hazelton et al. (2018) considers a clinical diagnosis of M. bovis- associated disease (MbAD) to be based on the clinical presentation with confirmation by microbiological culture or polymerase chain reaction (PCR) of the milk, or other tissues. Given the clinical signs of infection associated with M. bovis are non-specific, Wawegama and Browning (2017) suggest a tentative diagnosis is reached when clinical disease is present, with a combination of clinical signs, postmortem (PM) findings, histopathology and immunohistochemistry (IHC). However, further testing is required to determine the prevalence of disease at the herd level. González and Wilson (2003) reminded the veterinary practitioner to keep an open mind diagnosing intramammary infections (IMI) on dairy farms, as the variations in the pathogenicity of different strains of M. bovis, of animal susceptibility, and of farming practices on the dairy farm are key determinants for the presentation of mycoplasmal mastitis. This descriptive study includes both case and prospective cohort study methods to describe key epidemiological, clinical and pathological findings which enabled the diagnosis of M. bovis CM to be made in NZ, across a multi-farm dairy operation, with a single hypothesized infection source and date. While the findings in this case study will offer an understanding of the use of a range of diagnostic tests, applied both at a herd- and cow-level, it is imperative for the NZ dairy industry that the diagnosis of M. bovis CM at a cow-level can be made by veterinarians in the field. This would allow the implementation of effective control measures in a timely way to limit further spread of infection within and between farms. The three objectives of this thesis are to: 1. Provide an epidemiological description of the diagnostic test findings and clinical manifestations of a case study of an outbreak of M. bovis CM in NZ. This will be reported on a herd-, cow-, and quarter-level with the expectation to provide NZ-based data for a dairy clinician to make a diagnosis of M. bovis CM. This will be reported as the “Outbreak Investigation”. 2. Provide a comparison on how well the serum enzyme-linked immunosorbent assay (ELISA) sample to positive (SP) ratio agreed with tonsil swab real-time PCR (qPCR) in both clinical and non- 3 clinical cows. The tonsil qPCR, as a parallel diagnostic test, could be used as part of an eradication program, or wider surveillance program, to confirm the clinical diagnosis of M. bovis. This will be reported as the “Agreement Study”. 3. Present the patterns of the gross lesions found in cases of M. bovis CM. This description will enable the dairy clinician to use gross pathology (and associated tests) as additional diagnostic tools to recognize M. bovis-associated lesions in the mammary gland, as different morphological patterns are currently unreported. This will be reported as “Pathological Findings – Morphological patterns of mammary gland lesions in dairy cows with M. bovis CM”. 1.2 Format of Thesis Chapter 2 Literature Review summarises the scientific literature of M. bovis. This review will primarily detail our current knowledge of the epidemiology and clinical manifestations of M. bovis, the diagnostic tests used in the diagnosis of M. bovis, the pathology of M. bovis CM and finally describe the control and eradication strategies for M. bovis. Chapter 3 Material and Methods outlines the background to the case study, collection of samples, laboratory analysis, data sources used and statistical analysis of data. Chapter 4 Results reports the findings of the three main study objectives. Chapter 5 Discussion considers the findings of this study in the context of other relevant literature and draws conclusions on each of the three objectives. These conclusions will aid the veterinary diagnosis of M. bovis CM. The limitations of this study, and their possible impact, will also be highlighted. Important directions for future research are considered. 4 Chapter 2 Literature Review 2.1 Introduction This review details M. bovis as a pathogen, followed by the epidemiology of M. bovis, specifically pathogenicity, transmission, colonisation, shedding, prevalence, and risk factors. The international literature describing the clinical manifestations of M. bovis, with emphasis on M. bovis mastitis and associated pathology, together with the diagnostic tests used in the diagnosis of M. bovis are examined. Finally, the control and eradication of M. bovis is considered. 2.2 Mycoplasma bovis – The Pathogen 2.2.1 Microbiology The first recorded CM outbreak in a dairy herd due to the species now called M. bovis was in Connecticut, USA in 1961 (Hale et al., 1962). The bacterium has had several reclassifications over the decades, from Mycoplasma agalactiae var bovis, to Mycoplasma bovimastitidis in 1967 (Jain et al., 1967) then Mycoplasma agalactiae subspecies bovis in 1970 (Freundt & Edward, 1971) and finally in 1976 after further microbiological observations, these strains were named as a new species, M. bovis (Askaa, 1976). Seminal work on M. bovis mastitis which was reported during this period of taxonomic change (Jain et al., 1969; Jasper & Al Aubaidi, 1974) and mainly reported on experimentally induced disease, still offers some of the more detailed reports of disease progression. The Genus Mycoplasma belong to the Class Mollicutes and are a class of bacteria distinguished by the absence of a cell wall. Mycoplasma bovis is small and pleomorphic; with a small genome size of 1,080 kilobase pairs (kbps) and low G+C ratio of 27.8-32.9 mol%. The bacterium has complex nutritional requirements, lacks the tricarboxylic acid cycle (TCA), relies on the host for external sources of lipids, amino acids, nucleic acid precursors and is unable to ferment glucose (Hermann R, 1992; Khan et al., 2005). Mycoplasma bovis can survive at 4oC for nearly 2 months in sponges and milk, and over 2 weeks in water; however at higher temperatures, survival drops considerably (Ruffo et al., 1974; Ruffo et al., 1969). Without a cell wall, the cell membrane is directly exposed to the host environment and the bacterium is vulnerable to osmotic shock (Bürki et al., 2015; Kumar et al., 2014) and resistant to antimicrobials of the beta-lactam family, which inhibit the synthesis of peptidoglycan (Rosenbusch, 1994). 5 2.2.2 Mycoplasma Species There are over 100 species of mycoplasmas (Razin et al., 1998). Mycoplasma bovis is the most common cause of mycoplasmal mastitis in cows (Ayling et al., 2004; Bushnell, 1984; González & Wilson, 2003). Mycoplasma alkalescens, Mycoplasma canadense, Mycoplasma californicum, and Mycoplasma bovigenitalium together with M. bovis are considered the more common Mycoplasma spp. which cause mastitis (González & Wilson, 2003). Other species associated with mastitis outbreaks include Mycoplasma bovine group 7, now named Mycoplasma leachii (Hum et al., 2000), Mycoplasma bovirhinis (Hirose et al., 2001), and Mycoplasma dispar (Jasper, 1981a). Mycoplasma canis, Mycoplasma gallinarium and Mycoplasma bovoculi have been isolated from milk samples, but their role in bovine mastitis is not clear (Ayling et al., 2004). Differentiation of species by culture can be challenging as they possess similar morphology, cultural and biochemical characteristics (Fox et al., 2005; Kumar et al., 2014; Parker et al., 2018). While M. bovis and Mycoplasma agalactiae share a number of related proteins, epitopes, and antigenic determinants, they can be diagnostically differentiated by DNA-based detection methods more easily, as the PCR assays target different genomic regions (Bashiruddin et al., 2005). Two species of mycoplasma have previously been reported in NZ dairy cattle, where the key clinical syndrome has been outbreaks of unresponsive mastitis (Pharo, 2018). In 1969, a member of the bovine mycoplasma group 8, now renamed M. alkalescens was diagnosed in Northland (Brookbanks E et al., 1969), and in 1983 in the Waikato, M. dispar was diagnosed in an outbreak of dry cow mastitis (Hodges R et al., 1983). Both of these species have been described as infrequent and sporadic cause of mastitis, with low transmission rates (Rosenbusch, 2005). Several other mycoplasmal diseases have been reported in cattle in NZ. Diseases include regenerative anaemia and haemoglobuinuria in a cow where haemotropic mycoplasmas (M. wenyonii, Candidatus M. haemobos) were detected by PCR (McFadden et al., 2016), and polyarthritis in a cow where Mycoplasma mycoides mycoides large colony (MmmLC) was demonstrated on both PCR and culture of the joint tissues (Johnstone & King, 2003). 6 2.2.3 The Use of Molecular Techniques for Typing Mycoplasma Strains Bacteria can be classified firstly into genus e.g. Mycoplasma, then secondly, species e.g. M. bovis, and the third level of classification is strain classification e.g. Strain PG45, Strain M590 (Aebi et al., 2012). While the bacteriological terms strain and clone have specific scientific definitions, some scientific papers, do not define them well (Dijkshoorn et al., 2000), with some strain studies for M. bovis interchanging the terms (Aebi et al., 2012). Strain typing has many uses especially for epidemiological analysis. These include firstly, tracing the region or country of origin of an M. bovis outbreak, especially where movement of imported animals are suspected (Ayling et al., 2004), and secondly use in epidemiological studies of M. bovis and M. californicum for between-herd, within-herd and also within-cow studies (Hata et al., 2014). Thirdly, the study of certain aspects of mycoplasma diseases i.e. improving the ability to classify and characterize mycoplasma strains, assessing the genetic diversity of mycoplasmal populations, strain virulence, disease outcomes and importantly, added to our knowledge of internal dissemination of M. bovis within the animal (Biddle et al., 2005). Isolates of M. bovis can be genetically characterized using a number of different methods, the details of which are outside the scope of this review. The six common methods are random amplified polymorphic DNA (RAPD), pulse-field gel electrophoresis (PFGE), amplified fragment length polymorphism (AFLP) (McAuliffe et al., 2004), insertion sequence fingerprinting (IS) (Miles et al., 2005), multiple locus variable number tandem repeat analysis (MLVA) (Pinho et al., 2012), and multi- locus sequence typing (MLST) (Rosales et al., 2015). Seminal work by Biddle and co-workers, showed in a mastitis outbreak on one farm, that PFGE patterns of mammary gland isolates were often (44/70) identical to isolates collected from multiple other body sites (Biddle et al., 2005). This finding suggested that there is potential for haematogenous spread of mycoplasmas (Biddle et al., 2005). An Idaho, USA study looking at M. bovis mastitis strains also using the PGFE method, found while four strains of M. bovis were isolated from different body sites, only one strain, Strain O, caused CM. It was proposed that a virulence factor present in Strain O enabled it to cause CM (Punyapornwithaya et al., 2010). This finding conflicts with other reports, where it was considered that all strains, regardless where in the body they were isolated from, can cause CM (Pfützner & Sachse, 1996). Danish work (Kusiluka, Kokotovic, et al., 2000) looked at over 40 field isolates of M. bovis over a 17- year period and found significant genetic homogeneity. They employed the ALFP method and their findings were consistent with the American work of Biddle et al. (2005). In one cow, they found indistinguishable genetic patterns from M. bovis isolates from the nose, lung, and milk, which 7 suggested the pathogen spread internally. This study also demonstrated the AFLP technique could be used for both discrimination of M. bovis strains and genomic fingerprinting. In contrast, an investigation into the prevalence of M. bovis in pneumonic lungs in Danish cattle, using PFGE, found the 11 M. bovis field isolates from nine different farms, showed different profiles, except for two isolates from the same farm (Kusiluka, Ojeniyi, et al., 2000). An increase in M. bovis mastitis outbreaks in Switzerland initiated a large strain study, with over 1400 samples collected from 19 herds, to determine if one clone or strain had caused these outbreaks. Insertion sequencing techniques showed M. bovis strains diverged between herds, and strains were mostly herd specific. The outbreaks were deemed to be caused by numerous strains and not the introduction of a sole new strain or clone (Aebi et al., 2012). Multiple locus variable number tandem repeat analysis and MLST, a powerful DNA-typing tool for evaluating intraspecies genetic relatedness, are also used for genotyping isolates of M. bovis (Sulyok et al., 2014). A large number of M. bovis isolates, 137 collected from 12 countries, from both clinically infected and healthy cows, were analyzed by MLST. The isolates fell into two population clusters that were distinct. This finding was in agreement with the hypothesis that geographical independent evolution of M. bovis occurs when it is introduced into a new country (Rosales et al., 2015). The MLST method was used to analyze the lineage of a large number of Swiss and Austrian M. bovis isolates. Two distinct lineages were described, one for isolates collected since 2007, and the other prior to 2007. Further work is needed to understand if infection with isolates in the 2007 and beyond cluster, lineage 1, leads to more severe outbreaks of mastitis compared to infections with isolates in the prior to 2007 lineage, lineage 2, as there has been an emergence of severe M. bovis-associated mastitis cases seen in both countries (Bürki et al., 2016). The MLST method has been used in the M. bovis outbreak in NZ. This method involves PCR amplification followed by DNA sequencing, and measures variations in the DNA sequence of a set of housekeeping genes. Strains are then characterized by unique allelic profiles, which are assigned as a sequence type (ST). Phylogenetic models output based on the M. bovis isolates collected in NZ, which compare STs of different clonal complexes, suggest the isolates of M. bovis found in NZ all originated from a single strain (Biosecurity New Zealand, 2018b). It has therefore been hypothesized that there has been an incursion of only a single M. bovis strain into NZ. 8 2.3 Epidemiology The epidemiology of M. bovis, including the pathogenicity, transmission, colonisation (or carriage), shedding, prevalence, incidence rates, and risk factors of M. bovis is reviewed. While the literature offers consensus is some areas, conjecture and opposing views are raised in others. However, there is agreement that there are still areas of the epidemiology of M. bovis that require further investigation. 2.3.1 Pathogenicity The understanding of the pathogenic characteristics of M. bovis and their role in the pathogenesis of disease is still limited, but does appear to be multifactorial (Bürki et al., 2015). Mycoplasma bovis has microbial characteristics that facilitate it to both colonise and persist on a number of host mucosal surfaces, and then adapt to the hosts’ immune response to persist and cause disease (Lysnyansky et al., 1999). The virulence factors of M. bovis to be reviewed are antigenic variation, which includes the presence of variable surface lipoproteins (Vsps) and chromosomal rearrangement, adherence to and internalization into host cells, immunomodulatory characteristics, the production of secondary metabolites, biofilm production and synergistic infections with other pathogens. These virulence factors allow the bacterium to elude defence mechanisms of the host and hence persist within host cells (Bürki et al., 2015). Mycoplasma bovis can also colonise several host sites without fulminant disease developing. 2.3.1.1 Antigenic Variation Variation in the expression of the cell surface antigens is thought to be a form of adaptive variation by the bacteria to the host. An important source of these surface antigens are exposed membrane proteins, which are a group of immuno-dominant Vsps. These Vsps are one of the best-studied pathogenetic mechanisms of M. bovis. This family of proteins undergoes high-frequency phase (turning expression ON and OFF) and size variation, which results in the cell gaining and/or losing surface antigens (Bürki et al., 2015). As an example, in the type strain PG45 of M. bovis, the vsp-locus, which is an organised chromosomal cluster, is comprised of a family of 13 distinctive, single-copy vsp genes. The deduced proteins are identified as VspA to VspO (Behrens et al., 1994; Lysnyansky et al., 1999). Features like these allow M. bovis to evade immune surveillance and facilitates chronic infections (Buchenau et al., 2010). Apart from antigenic variation, Vsps have also been associated with other pathogenicity features of M. bovis, including enhancing colonisation and epithelial cell adhesion (Lysnyansky et al., 2016; Sachse et al., 1996; Sachse et al., 1993). These Vsps also have the ability to induce the expression of both up- and down-regulating cytokines e.g. IFN-γ in leukocytes and 9 lymphocytes e.g. CD4+ (helper) and CD8+ (cytotoxic) T cells (Bush & Rosenbusch, 2003; Kauf et al., 2007; Rosati et al., 1999). Behrens et al. (1996) demonstrated a Vsp-unrelated immunodominant membrane protein, labelled pMB67, was also involved in M. bovis surface antigenic variation. Antigenic variation is also achieved through chromosomal rearrangement, including deletions, duplications, and insertions. This high-frequency size variation, in specific repetitive blocks, within a locus, may result in a number of size variants of each Vsp (Behrens et al., 1994; Lysnyansky et al., 1996). Variable surface lipoproteins offer M. bovis an immense capacity to vary its surface antigens, which presents a challenge to developing effective vaccines and antimicrobials (Maunsell et al., 2011). Further investigation is needed to determine whether these complex mechanisms of antigenic variation are utilized by M. bovis to avoid the humoral host responses in natural infections. 2.3.1.2 Adherence to Host Cell Adhesion to the host cell in the target tissue, which is facilitated by cell–surface adhesins of the bacterium, is a prerequisite for colonisation and infection (Razin, 1999; Rottem, 2003). The mycoplasma membrane adhesins are known to have direct contact with the host cell (Sachse et al., 1996), aid mycoplasmal survival and are considered important virulence factors of M. bovis. Several adhesins and cytoadherence-related proteins have been identified including P26, a surface-located protein that adheres to embryonic bovine lung (EBL) cells (Sachse et al., 1996); a family of Vsps including VspA, VspB, VspE and VspF (Sachse et al., 2000) and a new Vsp protein (Thomas et al., 2005). Mycoplasma bovis NADH oxidase has also been found to behave as an adhesin (Zhao et al., 2017). Other in vitro studies have identified a plasminogen-binding protein, α-enolase (Song et al., 2012), and VpmaX (Zou et al., 2013) as adhesins. In an in vitro study investigating adherence rates to various host cell lines, the only non-pathogenic strain studied showed lower adherence rates compared to three CM isolates (Thomas et al., 2003). Researchers proposed that merging of the host membrane to these adhesins allows an intracellular exchange of components (Razin et al., 1998). Adhesion to the host cell is advantageous for the bacterium, as it can access specific nutrient requirements, including amino acids, lipids and precursors of nucleic acids (Calcutt et al., 2018; Fox, 2012). It is considered that both a fusion of the cell membrane of the mycoplasma bacterium and the host cell, and penetration of mycoplasmal tip organelle into the host cell are potential virulence factors (Razin, 1999). Contagious mastitis pathogens, unlike non-contagious mastitis pathogens, have the characteristic of being able to adhere to mammary gland epithelial surfaces (Frost et al., 1977). An in vitro study showed the ability of M. bovis to adhere to a number of differing host cell lines was not correlated to 10 the pathological background of the isolate, regardless of whether it was from a case of pneumonia, arthritis or mastitis (Thomas et al., 2003). It has not yet been proven if M. bovis has the ability to adhere to epithelial cells in the mammary gland (Fox, 2012). 2.3.1.3 Internalisation into Host Cells The ability of M. bovis to invade and then survive within the host cell gives it protection against the host’s own immune response (Fox, 2012). Mycoplasmaemia has been seen in calves that have been infected with M. bovis (D. Adegboye et al., 1995). The pathogen was reported in neutrophils and macrophages, and also in hepatocytes and epithelial cells of bile ducts (Bürki et al., 2015). In other in vitro work, M. bovis was shown to invade both bovine peripheral blood mononuclear cells (PBMC), including both T and B cells, and erythrocytes (van der Merwe et al., 2010). The internalization in host cells of M. bovis could contribute to the dissemination of the bacterium to multiple organ systems by the lymphatic or haematogenous routes. This finding is consistent with the ability of M. bovis to spread to multiple body sites of diseased cattle (Biddle et al., 2005; Jain et al., 1969). This internalisation is also considered to afford M. bovis protection against antimicrobial therapy (Bürki et al., 2015; Fox, 2012). Mycoplasma bovis being able to invade phagocytes aids its survival and allows the pathogen to persist in the animal (Bürki et al., 2015). The mechanism that allows M. bovis to survive phagocytosis has not been totally elucidated (Kleinschmidt et al., 2013). 2.3.1.4 Immunomodulatory Characteristics Another virulence factor of M. bovis is its’ ability to modulate the host immune system to enhance its survival (Fox, 2012). The membrane proteins of M. bovis, including both the Vsps and the Vsp- unrelated pMB67, are critical in this modulation as they interact with the host’s immune system. While immunomodulatory mechanisms are complex, controversial and not fully understood (van der Merwe et al., 2010; Vanden Bush & Rosenbusch, 2002), there does seem to be consensus that the bacterium can secrete a peptide that inhibits proliferation of host lymphocytes (Vanden Bush & Rosenbusch, 2004). Mycoplasma bovis may both stimulate (Razin et al., 1998; van der Merwe et al., 2010) and suppress the immune system of the host (Mulongo, Prysliak, Scruten, et al., 2013). The mechanisms of immune stimulation include the immune response being upregulated by the induction of cytokines, complement being activated, macrophages or T cells (Bush & Rosenbusch, 2003; Jungi et al., 1996; Kauf et al., 2007). Whereas immune suppression takes place by expression of anti-inflammatory cytokines (e.g. IL-10), or suppression of pro-inflammatory cytokines e.g. IFN-γ and TNF-α (Mulongo, Prysliak, Scruten, et al., 2013). Host immune suppression is also achieved by inhibiting lymphocyte proliferation via a lympho-inhibitory peptide (Vanden Bush & Rosenbusch, 2004), suppressing 11 lymphocyte response to phytohaemagglutinin (Thomas et al., 1990), and the in vitro ability of M. bovis to induce apoptotic death of bovine lymphocytes (Vanden Bush & Rosenbusch, 2002). However, a delay in apoptosis in M. bovis-infected bovine monocytes was noted in another study (Mulongo, Prysliak, Scruten, et al., 2013). The host immune response to M. bovis is hampered by the bacterium binding to neutrophils and inhibiting the oxidative burst (Thomas et al., 1991). Nevertheless, the bacterium can cause immunomodulation of both the cell-mediated and humoral responses (Fox, 2012). Host immune response modulation is consistent with a protracted survival and systemic dissemination of the bacterium in infected cattle (Mulongo, Prysliak, & Perez-Casal, 2013). 2.3.1.5 Secondary Metabolites and Biofilm Production Like other Mycoplasma spp., M. bovis produces secondary metabolites that are involved in the pathogenesis of M. bovis disease (Hames et al., 2009). Secondary metabolites such as H2O2 (hydrogen peroxide) and superoxide radicals are known to damage host cells and lead to cell death, lipid peroxidation or ciliary action inhibition (Hames et al., 2009). The enzyme NADH oxidase, an M. bovis adhesin, also generates H2O2 (Khan et al., 2005). For many Mycoplasma spp, including M. bovis, H2O2 generation is considered as an important virulence factor (Maunsell et al., 2011; Schott et al., 2014). Host tissue damage sees the recruitment and stimulation of phagocytes, both macrophages and neutrophils, which release lysosomal enzymes, reactive oxygen and nitrogen species (ROS and RNS) (Beckman & Koppenol, 1996; Fligger et al., 1999; Hermeyer et al., 2011). The superoxide anion with nitric oxide or alternatively nitrite with H2O2 forms peroxynitrite and subsequently causes nitrative injury (Sugiura & Ichinose, 2011). Mycoplasmal H2O2 together with ROS/RNS from macrophages may cause oxidative and nitrative injury, which result in the characteristic caseonecrotic lung lesions seen in cases of M. bovis pneumonia (Hermeyer et al., 2011; Schott et al., 2014). Caseonecrotic lesions have recently been reported in M. bovis CM (Radaelli et al., 2011) with a similar pathogenesis is believed to result in these lesions suggested. Some strains of M. bovis produce biofilms (McAuliffe et al., 2006). Biofilms are communities of sessile micro-organisms attached to a surface, often surrounded by an extracellular polysaccharide matrix (McAuliffe et al., 2006). Their production contributes to bacterial persistence in both the environment and inside the host, which may lead to disease chronicity (Bürki et al., 2015). Biofilm production also aids survival of the bacteria against both environmental stressors and host defences (Mah & O'Toole, 2001). It has been shown that strains of M. bovis that produce biofilms, have better survival in the environment. The biofilms aid in the prevention of desiccation and enable the bacteria to survive in 12 hotter temperatures, although in one study, the percentage of M. bovis bacteria surviving in a biofilm, at 30 hours, was only 0.01% of the initial inoculated dose (McAuliffe et al., 2006). Mycoplasma bovis also survives in bedding sand, from infected dairies for at least eight months (Justice-Allen et al., 2010). An increasing appreciation of biofilms in bovine mastitis is emerging (Gomes et al., 2016). While this work was focused on bacterial mastitis pathogens other than M. bovis, biofilms were shown to be of importance in pathogenicity. Biofilms maybe biologically important in recurrent infections, antimicrobial response, and in host defence mechanisms. 2.3.1.6 Co-infections with Other Pathogens Co-infections with other bacteria and viruses play a role in the development of the Bovine Respiratory Disease Complex (BRDC) (Bürki et al., 2015; Maunsell et al., 2011), see 2.4.4 Pneumonia. In an investigation into chronic pneumonia in Canadian feedlot cattle, a synergism between bovine viral diarrhea virus (BVDV), and its ability to cause immunosuppression, and the pneumonia and arthritis caused by M. bovis was proposed in the pathogenesis of these mycoplasma syndromes (Gagea et al., 2006; Shahriar et al., 2002). There has not been any specific discussion in the literature on co-infection and potential synergistic effects with other pathogens for M. bovis mastitis. 2.3.2 Transmission While there is a link between transmission (movement of infection from an infected animal to a susceptible or naive animal within an infected population), colonisation (presence of a bacteria on a body or mucosal surface, without causing disease in that animal) and shedding (discharge of an infectious agent into the environment, by excretion, secretion, exhalation or open wounds) in the epidemiology of M. bovis mastitis, these will be reviewed separately. Traditionally Mycoplasma spp. including M. bovis, have been considered highly contagious pathogens (González & Wilson, 2003) known to colonise mucosal surfaces in cattle, including the nose, eyes, ears, mammary gland, respiratory tract, prepuce, vagina and tonsils (Fox et al., 2005; Maunsell et al., 2012). Mycoplasma spp. are transmitted via the secretions from these mucosal surfaces. The literature has focussed on the premise that M. bovis mastitis, while contagious in nature, is mostly transmitted during the milking routine, from the udder, which is the reservoir for infection. This transmission may be via fomites, milkers’ hands and intramammary syringes of antimicrobial preparations. As opposed to environmental mastitis, where the primary reservoir for the pathogen (e.g. Streptococcus uberis), is the environment and not the infected udder (Smith et al., 1985), stringent hygiene practices at milking have played a major role in controlling the more traditional 13 contagious mastitis bacteria (e.g. Staphylococcus aureus and Streptococcus agalactiae). While indirect transmission of M. bovis from udder to udder during the milking routine is recognised as a major route of transmission (Fox, 2012), researchers found that despite high hygiene standards, post milking teat disinfectants, attention to detail with milking routine and shed maintenance, the prevalence of M. bovis mastitis on infected farms still increased (Enger et al., 2015). Similarly Punyapornwithaya et al. (2012) showed it was not possible to associate elimination of mycoplasmas from a dairy herd with any milk hygiene or control practices. The transmission of Mycoplasma spp. in dairy cows on an infected dairy platform, where CM is present may occur via two mechanisms (Fox, 2012). Firstly, via large droplets and short-range aerosols due to the presence of M. bovis in respiratory secretions and milk. This is direct transmission or transmission via nose-to-nose contact, where the bacteria may be shed through an external mucosal surface of an infected or colonised animal to a naive animal (Calcutt et al., 2018; Maunsell et al., 2011). Secondly, indirect transmission via milking equipment and other fomites (Fox, 2012), or feeding of infected colostrum/milk to calves (Maunsell & Donovan, 2009). Dissemination of M. bovis within an infected animal can also occur where M. bovis spreads haematogenously from an infected organ, body system or mucosal surface, to the mammary gland or in reverse, with initial infection of the udder followed by spread to other systems or mucosal surfaces. Isolates from the respiratory and urogenital systems, as well as the mucosal surfaces of the ear and eye, have been of the same strain type as isolates found in the mammary gland (Biddle et al., 2005). In an infected herd, multiple potential transmission routes could play a role in the transmission of M. bovis. Other potential transmission routes are noted in the literature. Vertical transmission of M. bovis mastitis has been suggested following a case of mastitis in pre-pubertal heifers, where these heifers were infected with the same strain as their infected dams and other herd mates (Fox et al., 2008). Mycoplasma spp. can survive in the environment for some months at varying temperatures and on different materials (Justice-Allen et al., 2010; Wilson et al., 2011). However, further research is necessary to define the role of the environment as a reservoir for Mycoplasma mastitis (Fox, 2012). Seminal vesiculitis in bulls has been experimentally induced in M. bovis studies (Ruhnke, 1994) and Mycoplasma spp. have been isolated from the semen of bulls (Ungureanu et al., 1986). Recent outbreaks of M. bovis mastitis in two naive Finnish dairy herds where the introduction of M. bovis infection was via semen used for artificial insemination (AI) is strongly suspected (Haapala et al., 2018). The practice of feeding colostrum or contaminated milk from cows with M. bovis mastitis has been linked to calves becoming infected (Maunsell et al., 2012). The M. bovis nasal prevalence and colonisation in calves fed infected colostrum or contaminated milk is higher than in calves fed clean milk (Bennett & Jasper, 1977b). 14 2.3.3 Colonisation The ability of M. bovis to colonise numerous mucosal surfaces has been well documented (Fox et al., 2005; Hazelton et al., 2018; Punyapornwithaya et al., 2010). Similarly, haematogenous spread of M. bovis from a foci of M. bovis arthritis and/or bronchopneumonia to allow mammary gland colonisation, potentially leading to M. bovis CM, has been postulated (Pfützner & Sachse, 1996; Punyapornwithaya et al., 2010). Early work suggested that the upper respiratory tract (URT) is the initial colonisation site for naturally infected calves (Bennett & Jasper, 1977b; Brys & Pfützner, 1989). In an experimental study, the colonisation of the URT was demonstrated by oral inoculation with M. bovis in calves which were slaughtered 14 days later. The tonsils, both palatine and pharyngeal, had high microbial loading at slaughter, while the nasal passages (deep nasal swabs) did not. Otitis media was also clinically diagnosed in these calves. Recent NZ work (Buckle et al., 2020), further supports the colonisation of palatine tonsils in naturally infected calves, where from 51 palatine tonsils that were swabbed at slaughter, 92.7% (95% CI 82.4-98.0%) were qPCR positive for M. bovis, compared to 12.7% (95% CI 5.3-24.5%) for swabs from mainstem bronchi. These calves originated from a M. bovis-infected herd. The mucosal surfaces of the nasal cavities, eyes, ears, mammary gland, respiratory tract, vagina, prepuce, and tonsils are known colonisation sites for M. bovis (Biddle et al., 2005; Hazelton et al., 2018; Maunsell et al., 2012). The easily accessible sites, namely the nasal cavity, which is sampled routinely in young animals under eight months of age (Bennett & Jasper, 1977b), eyes, ear, milk and ear (Fox et al., 2005; Maunsell et al., 2012), are usually used in experimental or clinical diagnosis where isolation of the bacteria is sought to confirm the colonisation and/or infection status of the cow/calf or herd. Deep nasopharyngeal swabs have been used for investigation into respiratory disease of mixed aetiology in calves (Godinho et al., 2007). A colonised or infected animal may be an asymptomatic subclinical carrier or clinical case. To aid in M. bovis diagnosis, or support control of M. bovis in asymptomatic cows, palatine tonsil swabs can also be taken from live cattle in the field by a clinician. There is no published literature on the colonisation of the palatine tonsil in the M. bovis CM cow or clinically asymptomatic cow in an infected dairy herd. Two important studies have further elucidated our understanding of M. bovis colonisation. Their study objectives, study populations and testing methodologies were slightly different. Firstly, an Idaho study (Punyapornwithaya et al., 2010) in a herd that had experienced M. bovis mastitis and was evaluating the association between M. bovis mastitis and colonisation at different body sites in asymptomatic carriers. The ears, eyes, nose, vulvovaginal tract, and milk of asymptomatic carriers were swabbed for 15 culture four times over a year. Mycoplasma spp. isolated were speciated and fingerprinted, using PFGE. This work showed the nasal mucosa was the most likely site to be colonised with M. bovis cultured from 21% (18/84) cows, 3.6 % (3/84) of cows’ eyes were culture positive, and no isolations from ears. If M. bovis was isolated from a cow at one site at one time point, sampling of this site was never repeated during subsequent sampling, nor did M. bovis isolation from a body site preceded mastitis. Also, in the initial stages of this outbreak of M. bovis mastitis colonisation of different body sites with the outbreak strain was common, but the prevalence of colonisation decreased over time. Secondly, Australian work Hazelton et al. (2018) swabbed the accessible mucosal surfaces of the eye, nose and vagina of 16 cows that developed M. bovis CM one to two weeks prior to sample collection. The M. bovis CM had been diagnosed by milk qPCR and 15/16 cows were serum ELISA positive on serum (using Bio-X kit with an optical density coefficient (ODC%) cut point of 37%, and diagnostic sensitivity of 93.8%). From the three mucosal sites, M. bovis was only detected in three (18.8%) vaginal swabs, with no detection from ear or nose swabs (Hazelton et al., 2018). This low prevalence of M. bovis on mucosal surfaces suggests that colonisation of these sites may be sporadic, given the analytical sensitivity of M. bovis qPCR is high. 2.3.4 Shedding The shedding of M. bovis from mucosal surfaces of infected cows is known to be intermittent and inconsistent (Biddle et al., 2003; Hazelton et al., 2018; Wilson et al., 2009). This intermittent shedding is commonly described with chronic and subclinical mastitis (SCM) cases (Bushnell, 1984; Gonzalez et al., 1992; Jasper, 1981a). The reasons for this intermittent shedding are not fully understood, but hypothesized reasons are the stress status of the animal and the time course of the infection (Calcutt et al., 2018). Shedding of M. bovis from infected animals may last from a few weeks to several months (Punyapornwithaya et al., 2010) or possibly years (Bayoumi et al., 1988). Prolific shedding (>106 cfu/ml by milk culture) of Mycoplasma spp. is often seen in cases of mycoplasma CM (Biddle et al., 2003). The limit of detection for M. bovis in milk culture is reported as 102 cfu/ml (Cai et al., 2005) to 103 cfu/ml (Parker et al., 2018) while 102 cfu/ml is the accepted PCR limit for detection of M. bovis in milk samples (Clothier et al., 2010; Rossetti et al., 2010). An American study that followed 10 infected cows for a 28-day period, assessing the frequency of shedding with chronic mycoplasma intramammary infection (IMI), found variable shedding (Biddle et al., 2003), where Mycoplasma spp. were isolated, by culture, from 71% of composite milk samples (CMS). At a cow level, 10% of the time cows shed between 102 and 104 cfu/ml, 1% of the time between 104 and 105 cfu/ml, and 60% of the time >105 cfu/ml. In another study, a cow with chronic mastitis from M. bovis infection was reported to not shed for a 56- 16 day period (González, 1999). These findings in part explain why intermittent shedding and asymptomatic carriage can hinder consistent detection of M. bovis, especially at an animal level (Hazelton et al., 2018), which increases the risk of a misdiagnosis (Biddle et al., 2003). Also, asymptomatic infected carrier cows are able to shed mycoplasma bacteria in milk or nasal secretions for months to years with no clinical signs (Calcutt et al., 2018; Nicholas et al., 2008). More research is needed into the shedding of M. bovis in mastitis. 2.3.5 Prevalence There are many Mycoplasma spp. and M. bovis prevalence studies in the literature. The 2018 National Mastitis Council Research Committee Report (Lopez-Benavides et al., 2018) reported that while IMI and mastitis denote different entities, these terms have been used incorrectly interchangeably over many years. Three states of M. bovis infection can be considered. Firstly, M. bovis seropositivity in milk or blood, with M. bovis not detected in the milk, by culture or qPCR, does not reflect an IMI or mastitis but rather a systemic response to previous M. bovis infection. Secondly, M. bovis detected in milk but no mammary gland change indicative of mastitis is a M. bovis subclinical mastitis (SCM), and an IMI present. Thirdly, M. bovis detected in milk and mammary gland changes indicative of mastitis is a M. bovis CM, and an IMI present. Studies report differences in the herd-level prevalence of M. bovis infection and within-herd prevalence of M. bovis infection. Together with M. bovis subclinical IMI from a small number of studies, they are presented in Table 1. (Different methodologies have been used, and these have been reported under Diagnostic test used). Herd-level Prevalence of M. bovis Estimated M. bovis herd-level prevalence varied between countries, from 0.9% in Australia to 5.4% in Greece. Most studies involved sample collection from a region (i.e. state or province) as opposed to sample collection from across the country. While all estimates are derived from testing of BTM samples, different studies used M. bovis PCR, culture, or ELISA. Some studies reported the herd-level prevalence of Mycoplasma spp. (Fox, 2012). There is limited herd-level seroprevalence work published, though a Swiss study (Burnens et al., 1999) found that 47% dairy herds in the canton of Jura, Switzerland were BTM M. bovis ELISA positive. Mycoplasma bovis Subclinical Mastitis The study of M. bovis SCM using culture methodology has historically been considered both costly and challenging (Fox, 2012). More recently qPCR technology has allowed a more extensive investigation of SCM. The literature offers a small dataset of studies to evaluate M. bovis SCM. Two Estonian studies (Timonen et al., 2017; Timonen et al., 2020) that looked at M. bovis IMIs presented very different 17 results. One study of a 600-cow M. bovis-infected herd had a within-herd prevalence of M. bovis IMI, using milk qPCR, of 17.2% (Timonen et al., 2017). A more recent repeated cross-sectional study of four herds with endemic M. bovis infection, where all cows were milk qPCR tested three times over a six-month period, had a much lower prevalence (Timonen et al., 2020). The two studies were not compared, and no explanation given as to the marked difference. Another recent Australian study (Hazelton et al., 2020) demonstrated a very low (0.0 to 0.2%) apparent prevalence (AP) herd-level of M. bovis IMIs, across a study in four herds with clinical M. bovis cases. The samples were collected after CM cases were removed from the herds. More work on the prevalence of M. bovis IMI is needed to fully explain and quantify this problem. These studies suggest that while most farms have a low prevalence of M. bovis SCM, occasional farms may have a high prevalence. Mycoplasma bovis Clinical Mastitis Estimation of the cow-level prevalence of M. bovis CM has been more difficult (Fox, 2012), with several studies presenting prevalences only at a Mycoplasma spp. level (Bradley et al., 2007; Hertl et al., 2011). Nevertheless, in North America, M. bovis is considered one of the most economically important mastitis pathogens (Rosengarten & Citti, 1999). In a US study, Brown (Brown et al., 1990) reported that, in herds with M. bovis up to 70% of mastitis cases, were infected with M. bovis. In contrast, two multi-herd studies reported low prevalences of M. bovis CM. Firstly, 19 known M. bovis infected dairy farms in Finland, with a median size of only 61 cows, were followed over a two-year period. Vähänikkilä et al. (2019) reported only a few cases of M. bovis CM. Fifty-one cases were recorded over the 19 farms, with eight farms only experiencing one case of M. bovis CM, with the range of cases/farm being one to eight. Milk qPCR was used on all mastitis cases. Of note, 88% of new clinical cases were reported within eight weeks of the index case. Secondly, 19 case herds were also enrolled in a Swiss study, on farms with confirmed M. bovis infection. Two visits to each farm were made to milk sample CM cows, including cows with multiple quarter CM. Milk qPCR positive cows were only found at the first visit and accounted for 18/742 CM cases (2.4%; 95%-CI: 1.5-3.8%). A large Saudi Arabian study (Al-Abdullah & Fadl, 2006) showed the prevalence of cases of M. bovis CM was highest in heifers in their first lactation, post calving. This finding agreed with early American work (Bayoumi et al., 1988). Contrary to these studies, other American work notes that despite using teat disinfectants post milking as part of the milking routine, M. bovis CM prevalence increased with cow age (Fox, 2012; Zadoks, 2015). 18 Table 2-1 Prevalence of Mycoplasma bovis References 1 (Morton et al., 2014) 2 (Hazelton et al., 2020) 3 (Passchyn et al., 2012) 4 (Filioussis et al., 2007) 5 (Nielsen et al., 2015) 6 (Lysnyansky et al., 2016) 7 (Murai & Higuchi, 2019) 8 (Fox, 2012) 9 (Miranda‐Morales et al., 2008) 10 (Ghazaei, 2006) 11 (Murai et al., 2014) 12 (Timonen et al., 2017) 13 (Timonen et al., 2020) 14 (Hazelton et al., 2020) 19 2.3.6 Incidence Incident rates of M. bovis CM are not widely discussed in the literature. Swiss work (Aebi et al., 2015) reported the incidence rate of M. bovis CM of 0 – 0.1 case per animal year at risk, and clinical pneumonia at 0.1 – 0.6 cases per animal year at risk. Using different indices, American work saw M. bovis CM incidence rates in the milking herd of 0.01 cases per 100 cow-days at risk, and 1.7 cases per 100 cow-days at risk in the hospital pens (Punyapornwithaya et al., 2011). 2.3.7 Herd-Level Risk Factors There are many risk factors for outbreaks of mycoplasmal disease on a dairy farm, especially outbreaks of CM, including increased herd size, purchase and introduction of stock, the seasonal movement of different classes of animals, stress factors and also infected semen used in AI (Aebi et al., 2015; Bayoumi et al., 1988; Murai & Higuchi, 2019; Nicholas et al., 2016). Early work by Thomas et al. (1981) showed there was a positive and significant correlation in herds that had mycoplasma mastitis between herd size and culling percentage. From a large two-year study of 650 herds in the USA, large herd size was shown to be significantly correlated with a BTM culture of M. bovis, where herd size was measured by total milk production i.e. milk production was used as a proxy for cow numbers (Fox et al., 2003). Conclusive evidence for herd size as a risk factor was demonstrated when BTM prevalence of Mycoplasma spp. increased with herd size from 2.1% for herds <100 cows, 3.9% for 100-499 cows, and 21.7% for herds >500 cows (McCluskey et al., 2003). A marked increase in the incidence of mycoplasma mastitis cases was seen in Israel with the expansion of herd size to >600 cows (Lysnyansky et al., 2016). While management practices in large herds may lead to high cow turnover, large numbers of stock movements and purchases increase the risk of introducing an infected animal into a herd (Fox et al., 2003; Pinho et al., 2013). Therefore, it is deemed the large herd size itself, irrespective of management practices, affords greater opportunity for introduction of mycoplasmas and enables infection to be more easily maintained in the herd (Nicholas et al., 2016). Stock introduction from an outside source is the most significant risk factor for mycoplasmal mastitis, including M. bovis (Punyapornwithaya et al., 2010). This introduction may include infected asymptomatic carriers moving into a naive herd or the movement of stock for grazing or shows where there is the potential for mixing of uninfected and infected cattle. In one Swiss study (Burnens et al., 1999) where multivariate analysis was used to identify possible risk factors, animal purchase was the only variable significantly associated with herd serological status. Another retrospective case study, which used logistic regression, identified movement of animals, high mean herd milk production, one 20 brand of milking cups, and other stress factors as herd-level risk factors (Aebi et al., 2015). In a Japanese study, after controlling for herd size, purchase of cows and corporate farming models where frequent movement of stock was common, were risk factors. (Murai & Higuchi, 2019). Asymptomatic carriers may carry Mycoplasma spp. until a stress factor precipitates clinical disease (Fox, 2012). Calving is deemed a common stress factor (Bushnell, 1984; Punyapornwithaya et al., 2010). High stocking density, mouldy feed, high in-barn temperature, and concomitant disease are also noted risk factors (Aebi et al., 2015). Environmental, physical or nutritional stress factors may cause host immunosuppression that precipitates clinical outbreaks (Bayoumi et al., 1988; Boothby et al., 1986; Jasper, 1981a). In an univariable logistic regression model looking at the herd-level presence of M. bovis and associated risk factors, peri-calving conditions such as metritis, hypocalcaemia, clinical ketosis, and abomasal displacements were noted as concomitant diseases (Aebi et al., 2015). Mouldy feed where the fusarium toxin, deoxynivalenol, is present is another possible risk factor. This toxin is able to decrease the immune response, by reducing neutrophil phagocytic function (Fink-Gremmels, 2008). High producing herds are at higher risk of being in a negative energy balance (NEB), especially if poorly fed around calving, and would therefore be more predisposed to infectious disease (Goff, 2006). Two studies (Aebi et al., 2015; Feenstra et al., 1991) demonstrated that M. bovis case herds had higher mean milk production compared to control herds. There are also other risk factors. The dry cow period is a risk period for M. bovis mastitis outbreaks (Bicknell et al., 1983; Otter et al., 2015). The exact mechanism of dry cow mastitis needs further investigation (Nicholas et al., 2016). Production systems where potentially infected waste milk or colostrum is fed to calves increases transmission risk. Pasteurisation of milk being fed to calves is recommended (Foster et al., 2009). The mixing of infected calves with cows and vice versa is a risk factor as direct contact, large droplets, and aerosols are known to spread respiratory disease (Lysnyansky et al., 2016). While seminal vesiculitis in bulls has been experimentally induced in M. bovis studies (Ruhnke, 1994), there has been isolation of Mycoplasma spp. from the semen of bulls (Ungureanu et al., 1986). Artificial insemination (AI) of cows with M. bovis infected semen is considered a route of infection (Pfutzner, 1990; Wrathall et al., 2007) and hence another risk factor. A recent Finnish investigation (Haapala et al., 2018) found that M. bovis-infected semen used for AI in two self-replacing herds was the likely source of infection that caused M. bovis CM outbreaks. Mycoplasma bovis can persist in frozen semen for years (Pfutzner, 1990). Embryo transfer (ET) is also a potential risk factor (Bielanski et al., 2000). 21 2.4 Clinical Manifestations of Mycoplasma bovis There are numerous clinical manifestations associated with M. bovis infection. Mastitis, arthritis, tenosynovitis, pneumonia, and reproductive syndromes are seen in adult cattle, including dairy, beef and feedlot cattle (Maunsell et al., 2011; Pfützner & Sachse, 1996). While the most common clinical syndromes observed in young calves aged two to six weeks old are pneumonia (Nicholas et al., 2002), arthritis (Maunsell & Donovan, 2009), and otitis media (Maunsell et al., 2012), less common presentations include tenosynovitis (Adegboye et al., 1996), decubital abscesses, meningitis (Stipkovits et al., 1993), and keratoconjunctivitis (Levisohn et al., 2004). For both experimental and natural infections that are M. bovis-associated, there is variability in disease expression. Epidemiologically, the maintenance and dissemination of M. bovis in populations of cattle does not seem to rely on clinical disease (Maunsell et al., 2011). 2.4.1 Mastitis 2.4.1.1 Clinical Mastitis Mycoplasma bovis is considered to be a contagious mastitis pathogen (Royster & Wagner, 2015; USDA, 2008). Mycoplasma bovis mastitis at the herd level can vary from endemic subclinical IMI, to mild CM, to outbreaks of severe clinical disease. Chronic infections can also occur (Maunsell et al., 2011; Timonen et al., 2020). Mastitis outbreaks may be diagnosed concurrently with arthritis (Wentink et al., 1987; Wilson et al., 2007) and/or pneumonia caused by M. bovis (Petersen et al., 2018). Outbreaks of CM vary in time to resolution. Outbreaks have varied in duration, from two months, to a year, to several years (Bayoumi et al., 1988; Fox et al., 2003; Jasper, 1981b; Punyapornwithaya et al., 2012). In contrast, in some herds with clinical disease attributable to M. bovis