Abstract/Summary

Antimicrobial resistance (AMR) is a current and increasing threat to both animal and human health alike, and particularly common where farming is concerned. Of importance is the role that antibiotic treatment of mastitis in cattle plays in this health crisis. Improved diagnostics for bacterial detection and antibiotic resistance detection is one area that can be targeted to improve antibiotic prescribing and treatment regimes, optimising the use of antimicrobials. However, rapid, on-farm tools are needed to effectively make antibiotic-prescribing decisions. On-farm diagnostic tests are already in use, such as the California Mastitis Test (CMT), somatic cell counting (SCC) and infrared thermography. However, these generally rely on host biomarkers for mastitis rather than bacterial culture. This is due to the complex nature of the milk microbiome and slow culture-based methods increasing the time to treatment. New technology offers the promise of rapid and more complex diagnostics for on-farm diagnosis of mastitis and the detection of AMR. Here we aim to review the state-of-the-art and emerging approaches to detecting bovine mastitis, with a particular focus on novel on-farm bacterial and antibiotic resistance detection. The detection and treatment of this disease can prove problematic and often time-consuming, leading to the inappropriate use and overuse of antibiotics in farming. Common diagnostic and AMR detection techniques, although standardised and highly sensitive, are often too laborious and require laboratory time and equipment. Meanwhile, the literature suggests that novel AMR detection systems need to be rapid, accessible, affordable and more information to be useful in aiding antibiotic stewardship. The overarching aim of this project is to address this problem by building on microfluidic technologies, combining traditional laboratory techniques with the use of microcapillary film (MCF) to carry out high-throughput investigations and perform testing that would usually be confined to a laboratory setting, on-farm. Of importance is the ability to carry out direct testing in milk from mastitis-infected cattle. The main objective was to produce a microfluidic test that can be used on-farm to detect bacteria growth and help indicate antibiotic resistance and therefore inform on a course of treatment. We describe a system by which we can detect the growth of bacteria within milk samples using resazurin dye in an assay imaging system. The development of the PiRamid imaging system takes advantage of open-source computing allowing for automated imaging of a range of microbiological experiments including novel MCF and bacteria motility assays and conventional culture, as well as proving useful in chemistry analysis through the imaging of crystal formation. The effects of spoiled milk on a miniaturised Matthew Long ©University of Reading 2025 Page 9 AST system were identified by the comparison of fresh, pasteurised milk and yoghurt as a simulated spoiled milk sample. It was also determined the effect on time to results by a milk sample that may contain both a pathogen of interest and commensal bacteria, indicating the potential for a period of false susceptibility in the presence of a high commensal bacterial load. It was also demonstrated that the resazurin-based assay can be adapted to accommodate the potential antimicrobial properties of resazurin dye, evaluating the use of lower concentrations of resazurin with reduced antimicrobial effect. The fluorescence intensities analysed allow for the detection of bacteria growth at lower concentrations of resazurin dye, with the time to detection of growth for all species more rapid than the standard laboratory culture method. These developments begin to pave the way for high-throughput, on-farm diagnosis of mastitis in cattle, reducing treatment time and eventually aiding in the choice of treatment antibiotics. It is important to understand the challenges relating to the matrix effects of milk as a sample. Milk can become spoiled during storage, therefore a comparison of fresh, pasteurised milk and yoghurt as a simulated spoiled milk sample was made in a miniaturised AST system. Moreover, not only does a milk sample collected from a dairy farm potentially contain a pathogen of interest related to a mastitis infection, but it can also harbour commensal bacteria from the surrounding environment and the cow itself. It was hypothesised the potential for false susceptibility in the presence of a high commensal bacterial load. Investigation was required to evaluation time to results and the performance of this miniaturised AST system containing a simulated milk sample that may contain both a pathogen of interest and commensal bacteria, indicating. Sample dilution reduces the effects of milk on bacteria growth and observed MIC and reduces the interference of commensal bacteria on observed MIC and time to result in this system. It is possible to monitor Gram-negative bacterial growth colourimetrically even in the presence of milk and yoghurt (used to simulate spoiled milk samples), as long as this sample matrix was diluted 1:5 or more in growth medium. Growth detection kinetics using resazurin was not changed by milk at final concentrations of 20% or lower, but a significant delay was seen with yoghurt above 10%. The minimum inhibitory concentration (MIC) for ciprofloxacin and gentamicin was increased in the presence of higher concentrations of milk and yoghurt. When diluted to 1% all observed MIC were within range, indicating dilution may be sufficient to avoid milk matrix interfering with microfluidic AST. Finally, overlap is seen between mastitis-causing bacteria with commensal organisms proving difficulty to avoid false positive culture from the growth of commensals found in healthy milk. It is determined that a susceptible commensal would only be likely to mask the resistance of a pathogen in samples where the commensal cell density significantly exceeds the pathogen cell density. Matthew Long ©University of Reading 2025 Page 10 Where previously the growth and AST of gram-negative bacteria has been validated in microcapillary film (MCF), it was important to understand if it can be used as a portable, microenvironment enabling the growth of Streptococci. Experimentally, 78% of Streptococcus strains were found to be resistant to resazurin dye at a concentration of ≥ 10µg/mL, where previously a concentration of 60µg/mL was shown to provide fluorescent intensity suitable for analysis of bacterial growth kinetics. Where the reduction in resazurin dye concentration may be required to allow Streptococcal growth, the difference in start and endpoint is also reduced, therefore making colourimetric detection unviable for Streptococcal growth. Compared to traditional microtitre plates (MTP) used for BMD, MCF has a decreased path length, decreasing the colour intensity of resazurin dye and decreasing the range between start and endpoint absorbance. Resazurin dye has been shown to be useful colorimetric indicator of the growth of bacteria in MCF. It was determined that haemolysis alone cannot indicate the growth of Streptococcus spp. in MCF where the difference between the absorbance calculated at 0h does not vary enough from the absorbance calculated at 20 h, where growth is present. Fluorometric detection with resazurin dye at 10µg/mL and 5µg/mL saw the difference between the highest fluorescent intensity and the background intensity as acceptable to detecting growth. In this system, it was determined that the maximum time to detection for Streptococci in this system was 9 h at a bacterial concentration of 2.5x104 CFU/mL. All species had detectable growth between 5.5-9 h, considerably faster than the method deployed for traditional AST (> 24 hrs). In AST validation experiments, essential agreement for ampicillin minimum inhibitory concentration (MIC) across all the strains was 100% with all MCF MICs within ±1 doubling dilution of antibiotic. For benzylpenicillin, essential agreement was 80% within ±1 doubling dilution of antibiotic. Future resazurin-based microfluidic assays for the detection of bovine mastitis organisms require lower concentrations of resazurin dye and should potentially utilise fastidious Mueller Hinton broth (MHB-F) to incorporate the detection of Streptococcus spp. growth and AST detection. Rapid AST in mastitis-infected milk still requires optimisation and careful consideration of the problems faced by such a complex media. With an extensive microbial community, we still see effects of false susceptibility, microbial depletion and pH changes that must be considered. Reducing time to testing and therefore time of storage will have a positive effect on the microbial community being tested and will provide the most accurate simulation of the pathogenic bacteria for AST. We have demonstrated reduced time to results sufficiently below common culture methods for AST of mastitis bacteria. Time to results < 10 h have been demonstrated, significantly faster than traditional culture methods of 24 – 48 h. There are still problems with Matthew Long ©University of Reading 2025 Page 11 understanding the bacterial load of a mastitis infection in milk and currently insufficient consistent data on the bacteria inoculum found in milk of a mastitis infection to conclude a threshold for infection. Although the limit of the detection for a novel MCF assay is as low as 103 CFU/mL, complications surrounding the detection of false susceptibility in the presence of high loads of commensal bacteria may still arise. This device can provide a comprehensive screening for AMR in mastitis-infected milk. Moreover, if coupled with a rapid method of microbial identification, will prove a powerful tool for detecting AMR and informing effective antibiotic treatment of mastitis.