Quantitative detection of ricin in beverages using trypsin/Glu-C tandem digestion coupled with ultra-high-pressure liquid chromatography-tandem mass spectrometry

Abstract

The toxic protein of ricin has drawn wide attention in recent years as a potential bioterrorism agent due to its high toxicity and wide availability. For the verification of the potential anti-terrorism activities, it is urgent for the quantification of ricin in food-related matrices. Here, a novel strategy of trypsin/Glu-C tandem digestion was introduced for quantitative detection of ricin marker peptides in several beverage matrices using isotope-labeled internal standard (IS)-mass spectrometry. The ricin in beverages was captured and enriched by biotinylated anti-ricin polyclonal antibodies conjugated to streptavidin magnetic beads. The purified ricin was cleaved using the developed trypsin/Glu-C tandem digestion method and then quantitatively detected by ultra-high-pressure liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) with isotope-labeled T7A and TG11B selected as IS. The use of trypsin/Glu-C digestion allows shorter peptides, which are more suitable for MS detection, to be obtained than the use of single trypsin digestion. Under the optimized tandem digestion condition, except for T7A in the A-chain, two resulting specific peptides of TG13A, TG28A from the A-chain and two of TG11B, TG33B from the B-chain were chosen as novel marker peptides with high MS response. The uniqueness of the selected marker peptides allows for unambiguous identification of ricin among its homologous proteins in a single run. The MS response of the four novel marker peptides is increased by more than 10 times compared with that of individual corresponding tryptic peptides. Both the marker peptides of A-chain T7A and B-chain TG11B were selected as quantitative peptides based on the highest MS response among the marker peptides from their individual chains. The limit of detection (LOD) of ricin is 0.1 ng/mL in PBS and 0.5 ng/mL in either milk or orange juice. The linear range of calibration curves for ricin were 0.5-300 ng/mL in PBS, 1.0-400 ng/mL in milk, and 1.0-250 ng/mL in orange juice. The method accuracy ranged between 82.6 and 101.8% for PBS, 88.9-105.2% for milk, and 95.3-118.7% for orange juice. The intra-day and inter-day precision had relative standard deviations (%RSD) of 0.3-9.4%, 0.7-8.9%, and 0.2-6.9% in the three matrices respectively. Furthermore, whether T7A or TG11B is used as a quantitative peptide, the quantitative results of ricin are consistent. This study provides not only a practical method for the absolute quantification of ricin in beverage matrices but also a new strategy for the investigation of illegal use of ricin in chemical weapon verification tasks such as OPCW biotoxin sample analysis exercises.