Bovine Mastitis: Establishing Bacterial Diversity, Associated Risk Factors, and Antimicrobial Resistance Profiles of the Isolates in Embu and Kajiado Counties, Kenya

Abstract

Bovine mastitis is one of the most important global diseases of cattle in which it adversely affects animal and human health, quality and quantity of milk, and the economics of almost every country. Despite the global challenge of bovine mastitis, studies on prevalence, bacterial diversity of mastitis-causing pathogens, risk factors and antibiotic resistance profiles of the isolate in dairy cows in Kenya remain limited. This cross-sectional study was undertaken in Embu and Kajiado counties of Kenya with the following objectives; (1) To determine the prevalence of clinical and subclinical mastitis in dairy cows, (2) To isolate and characterize the bacterial communities from clinical and subclinical mastitic cow milk using culture and 16S rRNA metagenomics analysis, (3) To establish the phenotypic and genotypic antimicrobial susceptibility profiles of the isolates (4) To determine the risk factors associated with subclinical mastitis in dairy cows. The study was conducted among 395 randomly selected dairy cows from 154 smallholder farms. In each of the farms, a semi-structured questionnaire was used to collect data on mastitis management practices and cow level risk factors associated with mastitis. A total of 1574 milk samples were aseptically collected from each mammary quarter of the 395 cows. Six quarters were blocked and hence did not produce any milk. The milk was initially checked for clinical mastitis and screened for subclinical mastitis using the California Mastitis Test (CMT) before being analyzed for bacterial infection using standard bacterial culture methods. Sixty-six (66) mastitic milk samples based on their culture results were selected and further analyzed using 16S rRNA metagenomics analysis to further understand their bacterial diversity. Additionally, phenotypic antimicrobial susceptibility profiles for Staphylococcus species (n=183), Streptococcus species (n=22) Escherichia coli (n=12), and Pseudomonas aeruginosa (n=19) were determined against 10 antimicrobial drugs using the

disc diffusion method. Investigation of seven resistance genes to the various antimicrobial drugs was further done on the 183 Staphylococcus isolates using polymerase chain reaction (PCR) amplification and partial sequencing. Overall, the farm-level, cow-level and quarter-level prevalence of mastitis were at 76.6% (118/154), 80.0% (316/395) and 67.8% (1068/1574) respectively. Of the mastitic cows, 8.5% (27/316) were clinical and 91.4% (289/316) had subclinical mastitis. On culture, eight genera of bacteria were identified where Coagulase-negative Staphylococcus (CNS) at 42.8% (435/1016) were the most prevalent bacteria. Twenty percent, (217/1068) of the mastitic milk samples yielded no bacterial growth on aerobic culture-based methods. Failure to milk mastitic cow last (p=0.04) and previous history of mastitis (p=0.03) were significantly associated with subclinical mastitis. Alpha and beta diversity comparison showed that there were no significant differences in bacterial number and diversities in mastitic milk from quarters based on the region, clinical/subclinical status and culture growth status. Genera level analysis using 16S rRNA metagenomics analysis revealed that 11 genera dominated by Pseudomonas (2.6%-83.8%) were shared among the three categories. An increased relative abundance of some phyla and genera which could not be identified using standard culture methods such as Chlamydiae, Mycoplasma and Solibacillus in culture-negative mastitic milk were also reported. Overall, antimicrobial resistance (AMR) among the staphylococci ranged between 3.5% (8/183) for fluoroquinolones and 66.1% (121/183) for ampicillin. Strikingly, 25.0% (23/91) of S. aureus and 10.8% (10/92) of the Coagulase-negative Staphylococcus (CNS) isolates, were methicillin-resistant staphylococci (MRS) phenotypically. Among the Streptococcus species AMR ranged between 31.8% (7/22) for ampicillin and zero for fluoroquinolones, E. coli showed the highest

phenotypic resistance to ampicillin at 75.0% (9/12) while no resistance to fluoroquinolones. Pseudomonas aeruginosa showed the highest phenotypic resistance in cefaclor 89.5% (17/19) while lower resistance was reported in ciprofloxacin 5.3% (1/19). Unexpectedly 13.6% (3/22) vancomycin-resistant streptococci and 21.0% (4/19) carbapenem-resistant P. aeruginosa isolates were reported in this study. Higher multidrug resistance (MDR) was present in 66.0% (8/12), 78.9% (15/19) of the E. coli and P. aeruginosa isolates respectively. The most common antimicrobial resistant genes in S. aureus and CNS was blaZ at 44.3% (35/79) and 75.3% (55/73) respectively. This study shows a high prevalence of subclinical mastitis both at the farm and animal levels. Both the clinical and subclinical mastitis were predominantly associated with Coagulase-negative Staphylococcus and Pseudomonas species based on culture and 16S rRNA metagenomics analysis respectively. There was an increased relative abundance of some bacterial phyla and genera which could not be identified using standard culture-based methods in culture-negative mastitic milk implying the usefulness of using more sensitive techniques in the diagnosis of mastitis. There is a need to improve on management of the dairy farms through culling of cows with a previous history of mastitis, use of individual udder drying towels, and milking mastitic cows last as control measures for mastitis. The bacterial isolates revealed high resistance to beta-lactams with high blaZ genes being detected in staphylococci signifying a public health concern and a challenge to bovine mastitis therapy hence the need to prevent the emergence and control the spread of AMR in dairy farms. This is the first study to report on the presence, methicillin-resistant Coagulase-negative Staphylococcus, carbapenem-resistant P. aeruginosa and vancomycin-resistant Streptococcus in cow mastitic milk from Kenyan dairy farms and therefore further monitoring is recommended.