Protein Identification with a Single Accurate Mass of a Cysteine-Containing Peptide and Constrained Database Searching

A method for rapid and unambiguous identification of proteins by sequence database searching using the accurate mass of a single peptide and specific sequence constraints is described. Peptide masses were measured using electrospray ionization-Fourier transform ion cyclotron resonance mass spectrometry to an accuracy of 1 ppm. The presence of a cysteine residue within a peptide sequence was used as a database searching constraint to reduce the number of potential database hits. Cysteine-containing peptides were detected within a mixture of peptides by incorporating chlorine into a general alkylating reagent specific for cysteine residues. Secondary search constraints included the specificity of the protease used for protein digestion and the molecular mass of the protein estimated by gel electrophoresis. The natural isotopic distribution of chlorine encoded the cysteine-containing peptide with a distinctive isotopic pattern that allowed automatic screening of mass spectra. The method is demonstrated for a peptide standard and unknown proteins from a yeast lysate using all 6118 possible yeast open reading frames as a database. As judged by calculation of codon bias, low-abundance proteins were identified from the yeast lysate using this new method but not by traditional methods such as tandem mass spectrometry via data-dependent acquisition or mass mapping.