Depletion of Amoxicillin and Its Major Metabolites in Anatolian Water Buffalo Milk After Intramuscular Administration

Antibiotic residues in water buffalo milk are a food-safety concern, yet depletion data are scarce. The purpose of this study was to characterize the depletion profiles of amoxicillin (AMOX) and its two major metabolites, amoxicilloic acid (AMA) and amoxicillin diketopiperazine-2',5'-dione (2,5-DKP), in Anatolian water buffalo milk after a single intramuscular administration and to estimate a milk withdrawal time relative to the EU MRL. We tested the hypothesis that AMOX concentrations would decrease below the EU MRL over successive milkings and that AMA and 2,5-DKP would exhibit depletion kinetics distinct from the parent compound. Five lactating Anatolian water buffaloes received a single intramuscular injection of amoxicillin (15 milligrams per kilogram). Milk was collected at each milking (twice daily) for seven days and analyzed by liquid chromatography-tandem mass spectrometry with quantification limits below the European Union maximum residue limit for amoxicillin in milk (4 micrograms per kilogram). Amoxicillin peaked at the second milking (mean 13.65 micrograms per kilogram), mean concentrations fell below the maximum residue limit from the sixth milking, and they became non-quantifiable from the tenth milking onward. Two major metabolites, amoxicillinic acid and amoxicillin diketopiperazine-2',5'-dione, peaked earlier (2,5-DKP Tmax 12 h) or at higher concentrations (AMA Cmax 32.64 µg/kg vs. AMOX 13.65 µg/kg) and remained detectable up to the thirteenth milking, with longer apparent terminal half-lives (32.0 and 52.8 h) than amoxicillin (23.5 h); the mixed-effects model confirmed different depletion rates among analytes (milking × analyte interaction p = 4.63 × 10^-5). A log-linear withdrawal model applying the EMA 95/95 tolerance limit indicated that the first time point at which the upper tolerance limit fell below the EU MRL was 84.7 h after dosing; rounded up to the next 12 h milking interval, this corresponds to a reported withdrawal period of 96 h (≈8 milkings). These results provide species-specific residue kinetics for amoxicillin in Anatolian buffalo milk and support considering metabolites in monitoring and withdrawal-time decisions.