Identifying associations between management practices and antimicrobial resistances of sentinel bacteria recovered from bulk tank milk on dairy farms.

There is increasing emphasis on the need to reduce antimicrobial use (AMU) on dairy farms to reduce the emergence of resistant bacteria which could compromise animal health and impact human medicine. In addition to AMU, the role of farm management is an area of growing interest and represents an alternative route for possible interventions. The aim of this study was to evaluate the impact of farm management practices and AMU on resistances of sentinel bacteria in bulk milk. Dairy farms from two, geographically separate locations within the British Isles were recruited as part of two study groups. Farm management data from study group 1 (n = 125) and study group 2 (n = 16) were collected by means of a face-to-face questionnaire with farmers carried out during farm visits. For study group 2, additional data on AMU was collated from veterinary medicine sales records. Sentinel bacterial species (Enterococcus spp. and E. coli), which have been reported to be of value in antimicrobial resistance (AMR) studies, were isolated from bulk tank milk to monitor antimicrobial susceptibilities by means of minimum inhibitory concentrations (MICs). MIC data for both groups was used to generate an overall "score" for each farm. For both groups, this overall farm mean MIC was used as the outcome variable to evaluate the impact of farm management and AMU. This was achieved through use of elastic net modelling, a regularised regression method which also featured a bootstrapping procedure to produce robust models. Inference of models was based on covariate stabilities and bootstrapped P-values to identify farm management and AMU practices that have significant effects on MICs of sentinel bacteria. Practices which were found to be of importance with respect to Enterococcus spp. included management of slurry, external entry of livestock to the dairy herd, use of bedding materials and conditioners, cubicle cleaning routines and antibiotic practices, including use of ß-lactams and fluoroquinolones. Practices deemed to be of importance for E. coli MICs included cubicle and bedding management practices, teat preparation routines at milking and the milking procedure itself. We conclude that a variety of routine farm management practices are associated with MICs of sentinel bacteria in bulk milk. Amendment of these practices offers additional possible routes of intervention, alongside alterations to AMU, to mitigate the emergence and dissemination of AMR on dairy farms.

Mass spectrometry and machine learning for the accurate diagnosis of benzylpenicillin and multidrug resistance of Staphylococcus aureus in bovine mastitis.

Staphylococcus aureus is a serious human and animal pathogen threat exhibiting extraordinary capacity for acquiring new antibiotic resistance traits in the pathogen population worldwide. The development of fast, affordable and effective diagnostic solutions capable of discriminating between antibiotic-resistant and susceptible S. aureus strains would be of huge benefit for effective disease detection and treatment. Here we develop a diagnostics solution that uses Matrix-Assisted Laser Desorption/Ionisation-Time of Flight Mass Spectrometry (MALDI-TOF) and machine learning, to identify signature profiles of antibiotic resistance to either multidrug or benzylpenicillin in S. aureus isolates. Using ten different supervised learning techniques, we have analysed a set of 82 S. aureus isolates collected from 67 cows diagnosed with bovine mastitis across 24 farms. For the multidrug phenotyping analysis, LDA, linear SVM, RBF SVM, logistic regression, naïve Bayes, MLP neural network and QDA had Cohen's kappa values over 85.00%. For the benzylpenicillin phenotyping analysis, RBF SVM, MLP neural network, naïve Bayes, logistic regression, linear SVM, QDA, LDA, and random forests had Cohen's kappa values over 85.00%. For the benzylpenicillin the diagnostic systems achieved up to (mean result ± standard deviation over 30 runs on the test set): accuracy = 97.54% ± 1.91%, sensitivity = 99.93% ± 0.25%, specificity = 95.04% ± 3.83%, and Cohen's kappa = 95.04% ± 3.83%. Moreover, the diagnostic platform complemented by a protein-protein network and 3D structural protein information framework allowed the identification of five molecular determinants underlying the susceptible and resistant profiles. Four proteins were able to classify multidrug-resistant and susceptible strains with 96.81% ± 0.43% accuracy. Five proteins, including the previous four, were able to classify benzylpenicillin resistant and susceptible strains with 97.54% ± 1.91% accuracy. Our approach may open up new avenues for the development of a fast, affordable and effective day-to-day diagnostic solution, which would offer new opportunities for targeting resistant bacteria.

Prediction of Streptococcus uberis clinical mastitis treatment success in dairy herds by means of mass spectrometry and machine-learning.

Streptococcus uberis is one of the leading pathogens causing mastitis worldwide. Identification of S. uberis strains that fail to respond to treatment with antibiotics is essential for better decision making and treatment selection. We demonstrate that the combination of supervised machine learning and matrix-assisted laser desorption ionization/time of flight (MALDI-TOF) mass spectrometry can discriminate strains of S. uberis causing clinical mastitis that are likely to be responsive or unresponsive to treatment. Diagnostics prediction systems trained on 90 individuals from 26 different farms achieved up to 86.2% and 71.5% in terms of accuracy and Cohen's kappa. The performance was further increased by adding metadata (parity, somatic cell count of previous lactation and count of positive mastitis cases) to encoded MALDI-TOF spectra, which increased accuracy and Cohen's kappa to 92.2% and 84.1% respectively. A computational framework integrating protein-protein networks and structural protein information to the machine learning results unveiled the molecular determinants underlying the responsive and unresponsive phenotypes.

A Pilot Study to Investigate the Feasibility of a Multiple Locus Variable Number Tandem Repeat Analysis to Understand the Epidemiology of Dichelobacter nodosus in Ovine Footrot.

Dichelobacter nodosus is the essential pathogen in ovine footrot, an important cause of lameness in sheep that reduces productivity and welfare. The aim of this study was to investigate the feasibility of using multiple locus variable number tandem repeat analysis (MLVA) developed to investigate isolates to understand the molecular epidemiology of Dichelobacter nodosus in ovine footrot by investigation of communities of strains. MLVA sensitivity was improved by optimizing PCR conditions to 100% specificity for D. nodosus. The improved MLVA scheme was used to investigate non-cultured DNA purified from swabs (swab DNA) and cultured DNA from isolates (isolate DNA) from 152 foot and 38 gingival swab samples from 10 sheep sampled on four occasions in a longitudinal study. Isolate DNA was obtained from 6/152 (3.9%) feet and 5/6 yielded complete MLVA profiles, three strains were detected. Two of the three isolate strains were also detected in isolate DNA from 2 gingival crevice cultures. Complete MLVA profiles were obtained from swab DNA from 39 (25.7%) feet. There were 22 D. nodosus community types that were comprised of 7 single strain and 15 multi-strain communities. Six community types were detected more than once and three of these were detected on the same four sheep and the same two feet over time. There were a minimum of 17 and a maximum of 25 strain types of D. nodosus in the study. The three isolate strain types were also the most frequently detected strain types in swab DNA. We conclude that the MLVA from swab DNA detects the same strains as culture, is much more sensitive and can be used to describe and differentiate communities and strains on sheep, feet and over time. It is therefore a sensitive molecular tool to study D. nodosus strains directly from DNA without culture.

Sites of persistence of Fusobacterium necrophorum and Dichelobacter nodosus: a paradigm shift in understanding the epidemiology of footrot in sheep.

Sites of persistence of bacterial pathogens contribute to disease dynamics of bacterial diseases. Footrot is a globally important bacterial disease that reduces health and productivity of sheep. It is caused by Dichelobacter nodosus, a pathogen apparently highly specialised for feet, while Fusobacterium necrophorum, a secondary pathogen in footrot is reportedly ubiquitous on pasture. Two prospective longitudinal studies were conducted to investigate the persistence of D. nodosus and F. necrophorum in sheep feet, mouths and faeces, and in soil. Molecular tools were used to detect species, strains and communities. In contrast to the existing paradigm, F. necrophorum persisted on footrot diseased feet, and in mouths and faeces; different strains were detected in feet and mouths. D. nodosus persisted in soil and on diseased, but not healthy, feet; similar strains were detected on both healthy and diseased feet of diseased sheep. We conclude that D. nodosus and F. necrophorum depend on sheep for persistence but use different strategies to persist and spread between sheep within and between flocks. Elimination of F. necrophorum would be challenging due to faecal shedding. In contrast D. nodosus could be eliminated if all footrot-affected sheep were removed and fade out of D. nodosus occurred in the environment before re-infection of a foot.