A software tool for the analysis of mass spectrometric disulfide mapping experiments

Disulfide cross-linkages stabilize the structure of many proteins (Creighton, 1993). Therefore, knowledge of the disulfide linkages in a protein provides useful, albeit partial, highresolution structural information which can be obtained from experiments that are much less time consuming than the Xray crystallography or NMR spectroscopy experiments necessary for complete structure elucidation. Determination of the disulfide bonds in recombinant or synthetic proteins is also important, since formation of the correct disulfide crosslinkages is an indication of proper folding and function. In addition, protein folding can be studied by determining the disulfide bonds in folding intermediates (Creighton, 1993). The disulfide linkages in a protein can be determined by proteolytically digesting the protein under conditions where the disulfide bonds are stable, separating the proteolytic peptides, and identifying those that contain disulfide linkages. The identification can be performed by amino acid analysis, amino-terminal sequencing, or mass spectrometry. Sometimes, a combination of these techniques is used. The advantages of using mass spectrometry for disulfide mapping (Smith and Zhou, 1990) include its speed, sensitivity, specificity and ability to analyze mixtures. In a typical experiment for determining the disulfide linkage pattern, the protein is digested with a proteolytic enzyme and the reaction mixture analyzed with mass spectrometry before and after reduction of the disulfide bonds. The two mass spectra are compared and the masses of the peptides that contain disulfide bonds are determined. If the mass spectra are complicated, the reduced material can be alkylated and a third mass spectrum acquired to identify the cysteinecontaining peptides. Also, the non-reduced sample can be alkylated to determine the free cysteines in the protein. In cases where the protein of interest is small, contains few disulfide linkages, and can be satisfactorily digested with a protease having high specificity, it is frequently feasible to identify the peptides that are cross-linked simply from the mass(es) of the cross-linked peptides. In cases where more