The procedure was developed by pipetting 20 lL BAs standard solution or rice wine sample into a 1.5 mL screw-capped vial, 100 lL sodium borate buffer (0.1 M, pH 9.4) and 280 lL derivatization reagent (1.0 103 M) were successively added.
The vial was sealed and heated in a thermostatic water-bath at 55 C for 15 min.
After the derivatization reaction, cooled to room temperature and 100 lL of 50% acetic acid was
added for the stability of the mixture.
Then the diluted solution (0.5 lL) was injected directly into the HPLC–MS/MS system.
2.5. HPLC–MS/MS analysis
An Agilent 1290 series HPLC system coupled with an Agilent 6460 Triple Quadrupole MS/MS system (Agilent, USA) equipped with an Agilent Jet Stream electrospray ionization source (ESI
source) was employed in the process of the HPLC–MS/MS analysis.
HPLC separation was achieved using an Eclipce Plus C18 column (2.1 mm 50 mm, 1.8 lm) made in USA which was kept at 30 C. Eluent A was 5% acetonitrile containing 0.1% formic acid and B was 0.1% formic acid in pure acetonitrile.
The gradient conditions were as follows: 80–10% A from 0 to 6 min and then held for 2 min.
The injection volume was 0.5 lL at a constant flow rate of 0.2 mL/min.
The optimal ESI source conditions were described as follows: capillary voltage +4.0 kV; nebulizer 40 psi; dry gas 11.0 L/min; dry temperature 300 C; Sheath gas temperature 280 C; Sheath gas flow 10 L/min.
Agilent Mass Hunter Data Acquisition, Qualitative Analysis and Quantitative Analysis software were used for method development and data acquisition.
The optimal HPLC–MS/MS method parameters of the corresponding d0-MASC/d3-MASC derivatives are shown in Table 1.
The whole chromatographic windows were divided into three time segments: Segment I is set to waste; Segment II and III are set to source.
Data were collected under positive ionization mode with multiple reactions monitoring (MRM).
2.6. Quantification
The relative quantification method used in this work was established as follows.
Two aliquots of BAs were derivatized by d3-MASC (heavy form) and d0-MASC (light form), respectively.
To construct the calibration curves, d0-MASC and d3-MASC derivatives of standards were mixed in ratios of 1:10–10:1 for all BAs, and analyzed by HPLC–MS/MS.
The linear regression equations were obtained by plotting the experimental peak area ratios of d0-MASC/d3-MASC derivatives (y) against nominal concentration ratios (x).
Slopes of regression equation of calibration curves for the seven derivatives approximated to 1.0, the correlation coefficient values (R2) of above 0.99 were obtained for all the analytes, indicating good correlation of the experimental data with the theoretical ratios (See Supplementary Material Fig. S3).
For real