A cross-sectional study of dairy farms in western Türkiye has found that bovine mastitis, one of the costliest infectious diseases in dairy production, can carry some of the most clinically feared antimicrobial resistance phenotypes known to medicine. Between May 2025 and January 2026, researchers screened 504 cows across 15 privately owned dairy farms in Aydın and İzmir provinces and collected 450 quarter-milk samples that met clinical or California Mastitis Test-based mastitis criteria. Bacterial culture yielded 160 non-duplicate gram-negative isolates, including 60 Escherichia coli, 68 Klebsiella pneumoniae, and 32 Acinetobacter baumannii strains, all of which were analyzed within a single comparative resistance-severity framework.
The isolates were sorted into five resistance-defined groups: ESBL-producing E. coli (n = 40), carbapenem-resistant E. coli (n = 20), ESBL-producing K. pneumoniae (n = 35), carbapenem-resistant K. pneumoniae (n = 33), and carbapenem-resistant A. baumannii (n = 32). Extended-spectrum beta-lactamases (ESBLs) are major mechanisms of beta-lactam resistance in these pathogens, while carbapenemases are enzymes that hydrolyze carbapenems, a class of beta-lactam antibiotics. The World Health Organization’s 2024 bacterial priority pathogens list places carbapenem-resistant A. baumannii and carbapenem-resistant Enterobacterales among its critical-priority pathogens, which is why finding these phenotypes in milk samples from dairy cows draws attention far beyond veterinary medicine.
The most striking pattern in the dataset was the relationship between resistance severity and clinical presentation. Overall, 122 of the 160 isolates (76.3%) came from clinical mastitis cases, and 38 (23.8%) from subclinical cases. ESBL-producing E. coli was the only group found more often in subclinical mastitis, with 27 of 40 isolates (67.5%) recovered from animals showing no visible udder or milk abnormalities but testing positive on the California Mastitis Test. By contrast, the carbapenem-resistant groups were concentrated almost entirely in clinical cases: 19 of 20 carbapenem-resistant E. coli isolates (95.0%), all 33 carbapenem-resistant K. pneumoniae isolates, and all 32 carbapenem-resistant A. baumannii isolates came from clinical mastitis. When the isolates were pooled into ESBL-associated and carbapenemase-associated categories, the difference was highly significant, with 84 of 85 carbapenemase-associated isolates (98.8%) recovered from clinical mastitis and only one from a subclinical case.
Resistance burden also climbed in a stepwise fashion across the groups. The mean number of resistant antimicrobial classes rose from 4.45 ± 1.38 in ESBL-producing E. coli to 5.54 ± 1.40 in ESBL-producing K. pneumoniae, 6.70 ± 0.92 in carbapenem-resistant E. coli, 7.09 ± 0.95 in carbapenem-resistant K. pneumoniae, and 8.00 ± 0.00 in carbapenem-resistant A. baumannii, where every isolate was resistant to all eight antimicrobial classes represented in the tested panel. The proportion of isolates classified as panel-defined extensively drug resistant, meaning resistant to six or more of the eight tested classes, increased from 25.0% in ESBL-producing E. coli to 100.0% in carbapenem-resistant A. baumannii. Because colistin and tigecycline were not included in the antimicrobial panel, the researchers deliberately described these as panel-defined XDR classifications rather than formal pandrug-resistant designations.
Within-species comparisons confirmed that the escalation was not simply an artifact of comparing different bacteria. In E. coli, resistance to imipenem and cefoxitin rose from 0 of 40 isolates (0.0%) in the ESBL-producing group to 20 of 20 (100.0%) in the carbapenem-resistant group for both agents, and the mean resistant-class burden increased from 4.45 ± 1.38 to 6.70 ± 0.92. K. pneumoniae showed the same pattern, with imipenem and cefoxitin resistance rising from 0 of 35 (0.0%) to 33 of 33 (100.0%) and the class burden climbing from 5.54 ± 1.40 to 7.09 ± 0.95. Resistance to cefotaxime and ceftriaxone was complete in every group, indicating a shared background of third-generation cephalosporin resistance. Non-beta-lactam co-resistance followed a similar trajectory: gentamicin resistance increased from 7.5% in ESBL-producing E. coli to 100.0% in carbapenem-resistant A. baumannii, and chloramphenicol resistance reached 100.0% in that same group.
Targeted PCR testing revealed distinct beta-lactamase gene profiles behind these phenotypes. In ESBL-producing E. coli, blaCTX-M was the most frequent ESBL-associated gene, detected in 36 of 40 isolates (90.0%), followed by blaTEM (70.0%) and blaSHV (42.5%). ESBL-producing K. pneumoniae carried blaCTX-M in all 35 isolates, with blaSHV in 88.6% and blaTEM in only 31.4%, meaning the two species showed predominantly blaCTX-M/blaTEM and blaCTX-M/blaSHV profiles, respectively. Among the carbapenem-resistant groups, all 20 carbapenem-resistant E. coli isolates carried blaNDM, with blaOXA-48 in 60.0% and no blaKPC, blaIMP, or blaVIM detected. Carbapenem-resistant K. pneumoniae showed the most diverse profile among the Enterobacterales, with blaNDM in 90.9%, blaOXA-48 in 75.8%, and blaKPC in 57.6% of isolates. Carbapenem-resistant A. baumannii showed the broadest profile of all, with blaNDM, blaOXA-48, blaOXA-51-like, and blaOXA-23 detected in every isolate and blaOXA-24 in 59.4%. Neither blaIMP nor blaVIM was detected in any carbapenem-resistant group, suggesting that carbapenemase-associated resistance in this collection was mainly linked to NDM- and OXA-related mechanisms.
The study also compared how well five phenotypic carbapenemase detection assays performed across the different organisms, and the results varied considerably. The modified carbapenem inactivation method (mCIM), which follows CLSI criteria, showed high positivity in carbapenem-resistant E. coli (95.0%) and K. pneumoniae (90.9%) but dropped to just 53.1% positivity in carbapenem-resistant A. baumannii, with a 95% confidence interval of 36.4–69.1. In contrast, the optimized carbapenem inactivation method (oCIM), adapted specifically for A. baumannii, and the in-house UniCarb assay each reached 96.9% positivity in that group (95% CI: 84.3–99.4), and the meropenem–EDTA test was positive for all 32 isolates. The CarbAcineto NP test showed intermediate positivity of 75.0% in A. baumannii. Overall, oCIM, UniCarb, and MEM-EDTA demonstrated the most consistent positivity and the strongest PCR-supported agreement across the carbapenemase-associated groups, with mCIM missing 15 of 32 PCR-positive A. baumannii isolates. The authors note that these agreement figures describe performance within the tested isolate collection and are not external diagnostic sensitivity estimates.
Phenotypic ESBL confirmation was fully consistent across the ESBL-associated groups: all ESBL-producing E. coli and K. pneumoniae isolates tested positive by both the combined disk test and the double-disk synergy test, and AmpC exclusion testing found no phenotypically detectable AmpC activity, supporting ESBL-mediated rather than AmpC-mediated resistance. Species identification was confirmed by partial 16S rRNA gene sequencing, with A. baumannii assignments additionally supported by detection of the intrinsic blaOXA-51-like marker. Antimicrobial susceptibility testing was performed by disk diffusion on Mueller–Hinton agar using a twelve-agent panel, and the researchers emphasize that these interpretations were used to compare resistance burden across groups rather than as direct bovine mastitis treatment recommendations.
The authors are careful about what these findings do and do not mean. The study was cross-sectional and limited to milk isolates from western Türkiye, so temporal trends, evolutionary pathways, and broader geographic generalization should be interpreted cautiously. Complete twelve-month farm-level antimicrobial consumption records, detailed treatment histories, and mastitis management data were not systematically available, so farm-level antimicrobial use pressure could not be assessed. No human, environmental, food-chain, or farm-environment isolates were included, meaning the results should not be read as evidence of cross-sector transmission. Whole-genome sequencing was not performed, which limits resolution of clonality, plasmid context, and transmission dynamics, and the PCR panel targeted major beta-lactamase genes rather than the complete resistome. The study was also not designed as an external multicenter validation of the UniCarb assay.
Despite these limitations, the researchers argue that the integration of mastitis origin, resistance burden, phenotypic confirmation, and PCR-based gene detection within a single comparative framework offers a clearer picture of resistance severity than routine susceptibility testing or species-level reporting alone. Bovine mastitis is already recognized as an important veterinary setting for antimicrobial resistance surveillance because antimicrobial therapy is frequently used in mastitis control, and in Türkiye veterinary antimicrobial use is regulated within national animal-health and food-safety frameworks that include e-prescription and veterinary medicine tracking systems, with antibiotic growth promoters in animal feed prohibited. The finding that mastitis-associated gram-negative isolates can carry resistance phenotypes and beta-lactamase genes overlapping with major global AMR priorities supports the authors’ conclusion that bovine mastitis is a relevant veterinary surveillance interface with potential One Health relevance, while future studies incorporating animal, environmental, food-chain, and human comparator isolates would be needed to evaluate actual transmission pathways.
Subject of Research: ESBL- and carbapenemase-associated antimicrobial resistance in gram-negative bovine mastitis pathogens from western Türkiye
Article Title: Comparative phenotypic and genotypic profiling of ESBL- and carbapenemase-associated gram-negative bovine mastitis pathogens in western Türkiye, 2025–2026
Article References: Hassan, F., Türkyılmaz, M. K., & Türkyılmaz, S. (2026). Comparative phenotypic and genotypic profiling of ESBL- and carbapenemase-associated gram-negative bovine mastitis pathogens in western Türkiye, 2025–2026. Animal Diseases, 6(1), Article 55. https://doi.org/10.1186/s44149-026-00271-5
Image Credits: AI Generated
DOI: 10.1186/s44149-026-00271-5
Keywords: bovine mastitis, antimicrobial resistance, ESBL, carbapenemase, Escherichia coli, Klebsiella pneumoniae, Acinetobacter baumannii, blaNDM, blaOXA-48, One Health, dairy cattle, Türkiye
News Source: Kristina Jarvis. (October 28, 2026). Research report: Comparative phenotypic and genotypic profiling of ESBL- and carbapenemase-associated gram-negative bovine mastitis pathogens in western Türkiye, 2025–2026. Scienmag.
