Determination of Residues of 4 Nitrofuran Metabolites in Freshwater Fish by Ultra-High Performance Liquid Chromatography-Tandem Mass Spectrometry Based on Information Entropy Theory

Based on the information entropy theory, the recovery comprehensive score M of the 4 nitrofuran metabolites, including 5-morpholinomethyl-3-amino-2-oxazolidinone (AMOZ), 3-amino-2-oxazolidinone (AOZ), 1-aminohydantoin hydrochloride (AHD) and semicarbazide hydrochloride (SEM), was used as the evaluation index to optimize the derivatization conditions for the detection of nitrofuran metabolites residues in freshwater fish by single factor and response surface tests. The freshwater fish sample was hydrolyzed by hydrochloric acid solution, and internal standards solution was added. Then, 350 μL of 0.05 mol·L^-1 ortho nitrobenzaldehyde (derivatization reagent) solution with methanol as the medium was added, and derivatization reaction was made by oscillation at 46℃ under the light proof condition for 2.5 h. The acidity of the solution obtained was adjust to pH 7.4 ± 0.2, and extraction was made twice by ethyl acetate. The extract was collected, and blown to dryness by nitrogen. n-Hexane was added for impurity removal, and 50% (volume fraction) acetonitrile solution was added for redissolving the residue. After centrifugation, the lower phase was taken for filtration, and the filtrate was analyzed by ultra-high performance liquid chromatography-tandem mass spectrometry. In chromatographic analysis, Agilent Zorbax RRHD Eclipse Plus C₁₈ chromatographic column (100 mm×2.1 mm, 1.8 μm) was used for separation, and the mixed solution composed of 0.1% (volume fraction) formic acid solution and acetonitrile at volume ratio of 1:1 was used for isocratic elution. In mass spectrometry analysis, atmospheric pressure spray ion source was used for ionization, and multiple reaction monitoring (MRM) mode was used for detection. In quantitative analysis, matrix matching and internal standard methods were used to reduce matrix interference and improve the sensitivity of the method. It was shown that factors of derivatization reagent volume, derivatization temperature, and derivatization time, as well as the interaction factor of derivatization reagent volume and derivatization temperature, had a significant effect on the recovery comprehensive score M (P<0.01). Under the optimal derivatization conditions, the test value of the quadratic polynomial model (M_test=1.020 4) and the prediction value (M_prediction=0.999 2) were basically consistent. The linear ranges of the working curves for the 4 nitrofuran metabolites were found in the same range of 0.02-50.00 μg·L^-1, with detection limits (3S/N) in the range of 0.02-0.19 μg·kg^-1. Test for recovery was made according to the standard addition method, giving recoveries in the range of 85.1%-116%, and RSDs (n=5) of the determined values ranged from 0.64% to 9.8%. The proposed method was used for the analysis of 55 batches of actual samples, and the detection rates of AOZ, AMOZ, and SEM were 1.82%, 1.82%, and 3.64%, respectively. The highest detection amounts of AOZ and AMOZ were 0.24, 0.13 μg·kg^-1, and the detection amount of SEM was below the lower limit of determination.