Detection of four biogenic amines by liquid chromatography based on aptamer signal replacement combined with cyclic amplification

Abstract

Biogenic amines (BAs) represent a class of potentially harmful substances in foods and medicines. Their content is thus an important indicator of proper hygiene in food preparation, and purity of medicines. It is of great practical significance to establish accurate and sensitive detection of BAs in food and drugs. In this study, a high performance liquid chromatography (HPLC) method was developed for the simultaneous detection of multiple BAs in fish, pork and antibiotics based on aptamer signal replacement and cyclic amplification strategy. First, non-fluorescent targets were converted into fluorescent nucleic acid probes using a two-step replacement process. Subsequently, a large number of nucleic acid probes with different lengths and base sequences were generated using a double-stranded specific nuclease-assisted signal amplification strategy. Finally, various BAs in real samples were accurately identified using an HPLC platform. The influence of base sequence and nucleic acid probe length on separation via HPLC was studied to improve discrimination among fluorescent signals. Four different sequences were selected as tails to the DNA probe, and their retention times increased in turn. Experimental conditions, including column temperature, flow rate, gradient elution process, reaction temperature, and incubation time, were optimized by orthogonal experiments to further improve signal separation efficiency. Specifically, the methanol gradient was changed from 10% to 20% during 0-20 min, 35 ℃ of column temperature and 1.0 mL/min of flow rate were chosen as the HPLC conditions. The final resolution of chromatographic peaks was 3.44, 3.59 and 2.37, indicating complete separation between peaks. Optimal incubation time for BA capture by aptamer was 120 min, and optimal dosage of duplex specific nuclease (DSN) and Mg2+were 0.9 U and 30 mmol/L. The optimal pH, incubation temperature, and DSN incubation time were 7.0, 40 ℃ and 210 min, respectively. The proposed method exhibited high sensitivity towards BAs, with a linear range of 1 pmol/L-1 μmol/L, and the limits of detection of tyramine, histamine, spermine, and tryptamine were 0.25, 0.21, 0.27 and 0.19 pmol/L, respectively. The feasibility of this method was verified, and contrast experiments indicated that it could achieve highly selective detection of four BAs in one run. The applicability of this integrated method was also investigated for the detection of real samples (gentamycin sulfate, fish and pork). To assess the matrix effect, each BA with different concentrations were spiked into real fish and pork samples. Relative recoveries and relative standard deviations (RSDs) ranged from 101.2% to 104.5% and from 1.5% to 4.3%, respectively. The above detection results for real samples showed that this method could accurately capture, separate, and identify BAs in complex matrix samples. This strategy can effectively improve analyte selectivity and reduce the matrix effect. This assay is thus expected to provide a new approach for food and drug analyses.