Pitfalls and Solutions in Mass Spectrometry-Based Identification of Protein Glycation

Emerging evidence suggests that advanced glycation end-products (AGEs) such as N^ε-(carboxymethyl)lysine (CML) and N^ε-(carboxymethyl)lysine (CEL) may play important roles in certain human diseases. Reliable analytical methods are needed for their characterizations and measurements. Pitfalls have been reported for applications of LC–MS/MS to identify various types of post-translational modifications, but not yet for the case of AGEs. Here, we showed that in the absence of manual inspection, cysteine alkylation with 2-iodoacetamide (IAA) can result in false-positive/ambiguous identifications of CML >20%. They were attributed to offsite alkylation together with incorrect monoisotopic peak assignment (pitfall 1) or together with deamidation (pitfall 2). For pitfall 1, false-positive identifications can be alleviated using a peptide mass error tolerance ≤5 ppm during the database search. Pitfall 2 results in ambiguous modification assignments, which may be overcome by using other alkylation reagents. According to calculations of theoretical mass shifts, the use of other common alkylation reagents (iodoacetic acid, 2-chloroacetamide, and acrylamide) should face similar pitfalls. The use of acrylamide can result in false-positive identifications of CEL instead of CML. Subsequently, we showed that compared to IAA, the use of N-isopropylacrylamide (NIPAM) as an alkylation reagent achieved similar levels of proteome coverage, while reducing the offsite alkylation reactions at lysine by more than five times. Furthermore, false-positive/ambiguous identifications of CML due to the two types of pitfalls were absent when using NIPAM. NIPAM alkylation results in a unique mass shift that allows reliable identifications of CML and most likely other AGEs, such as CEL.