Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) for detection and identification of albumin phosphylation by organophosphorus pesticides and G- and V-type nerve agents

Abstract

Toxic organophosphorus compounds (OPC), e.g., pesticides and nerve agents (NA), are known to phosphylate distinct endogenous proteins in vivo and in vitro. OPC adducts of butyrylcholinesterase and albumin are considered to be valuable biomarkers for retrospective verification of OPC exposure. Therefore, we have detected and identified novel adducts of human serum albumin (HSA) by means of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Pure albumin and plasma were incubated with numerous pesticides and NA of the V- and G-type in different molar ratios. Samples were prepared either by sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by in-gel enzymatic cleavage using endoproteinase Glu-C (Glu-C) or by combining highly albumin-selective affinity extraction with ultrafiltration followed by reduction, carbamidomethylation, and enzymatic cleavage (Glu-C) prior to MALDI-TOF MS analysis. Characteristic mass shifts for phosphylation revealed tyrosine adducts at Y(411) (Y(401)KFQNALLVRY(411)TKKVPQVSTPTLVE(425)), Y(148) and Y(150) (I(142)ARRHPY(148)FY(150)APE(153), single and double labeled), and Y(161) (L(154)LFFAKRY(161)KAAFTE(167)) produced by original NA (tabun, sarin, soman, cyclosarin, VX, Chinese VX, and Russian VX) as well as by chlorpyrifos-oxon, diisopropyl fluorophosphate (DFP), paraoxon-ethyl (POE), and profenofos. MALDI-MS/MS of the single-labeled I(142)-E(153) peptide demonstrated that Y(150) was phosphylated with preference to Y(148). Aged albumin adducts were not detected. The procedure described was reproducible and feasible for detection of adducts at the most reactive Y(411)-residue (S/N ≥ 3) when at least 1% of total albumin was labeled. This was achieved by incubating plasma with molar HSA/OPC ratios ranging from approximately 1:0.03 (all G-type NA, DFP, and POE) to 1:3 (V-type NA, profenofos). Relative signal intensity of the Y(411) adduct correlated well with the spotted relative molar amount underlining the usefulness for quantitative adduct determination. In conclusion, the current analytical design exhibits potential as a verification tool for high-dose exposure.