Screening for antimicrobial resistant bacteria in cooled stallion semen and post-insemination uterine lavage fluid

Abstract

Cooled semen contains antibiotics to control bacterial growth during storage, which may contribute to the development of antimicrobial resistance (AMR) and introduce resistant bacteria into the uterus during artificial insemination (AI). Our aim was to evaluate the presence of AMR in cooled semen and post-insemination uterine lavage fluid. In Part 1, 98 commercial domestic or international cooled semen shipments were cultured after storage ranging from 4h to overnight. Seven samples were cultured both before and after cooled storage with or without antibiotic-containing extender, within 4h and 30 h after semen collection. In Part 2, culture samples were obtained from 10 cooled AI doses that were used for insemination of 10 mares. Uterine lavage fluid samples were collected at 16 to 25 h after insemination. Samples were cultured on sheep blood agars. After enrichment, samples were also cultured on selective media for detection of extended spectrum beta-lactamase (ESBL) producers, carbapenemase-resistant Enterobacteriaceae (CARBA), methicillin-resistant Staphylococcus aureus (MRSA), and vancomycin-resistant enterococci (VRE). Bacterial species were identified with MALDI-TOF mass spectrometry. Antibiotic susceptibility of selected pure cultures was determined using the disk diffusion method. In Part 1, 23% of semen samples showed bacterial growth on blood agars. Growth on selective media was observed in 21% (MRSA; none were identified as S. aureus), 17% (ESBL), 14% (CARBA), and 9% (VRE, including Enterococcus faecalis) of samples. Only 1/7 of the samples stored without extender showed no growth on any medium. In Part 2, most mare and semen samples (9/10) had no bacterial growth. Bacterial growth on selective media was detected in two semen samples: one with growth of clinically non-significant bacterial species on all selective media, and one with non-significant growth on MRSA media. These semen doses were not associated with bacterial growth in corresponding uterine lavage samples. Bacterial growth was observed in two mare samples: one with non-significant growth on MRSA media and one with Enterobacter cloacae -growth on ESBL media. The corresponding semen samples had no bacterial growth. All Enterobacteriaceae-strains isolated from the positive mare sample on ESBL media were resistant to third generation cephalosporins, with three strains producing AmpC-cephalosporinases and one strain producing both ESBL and AmpC-enzymes. Resistant strains, such as ESBL and AmpC-producing Enterobacter cloacae and VRE may present health risks, including zoonotic spread of AMR. These ESBL and AmpC-enzymes are often plasmid-encoded, facilitating horizontal gene transfer among bacteria, potentially leading to spread of AMR in gram-negative bacteria in horses, complicating infection treatment and posing public health risks due to close human-horse interactions. In conclusion, while bacteria, including resistant strains, may be present in cooled stallion semen, the same species are not necessarily found in the uteri of inseminated mares, as it is likely that uterine defense mechanisms in normal mares clear bacteria effectively.