Abstract: The disulfide bond is one of the most common post-translational modifications in proteins, of which determination is essential to the comprehensive understanding of protein structures. Disulfide bond analysis has gone through great improvement due to the development of matrix-assisted laser desorption/ionization mass spectrometry (MALDI MS), especially in terms of speed and sensitivity. In general, the characterization of disulfide-containing peptides is achieved by the reduction of disulfide bonds followed by alkylation. The identification of disulfide/cysteine-containing peptides in digests of proteins is essential to structure elucidation of a protein. In order to present a deep understanding of some phenomena occurring in MALDI MS of disulfide/cysteine-containing proteins, the current work systematically investigated effects of co-crystal size and matrix on MALDI-In Source Decay (ISD) fragmentation of disulfide-containing proteins. Imaging experiments were performed to evaluate the influence of laser shot location on the fragmentation of human insulin and insulin glargine, which were selected as model compounds. The spectrum that was recorded on very finely distributed crystals of FA spot does not exhibit fragments. While a characteristic ‘triplet’ ions with a mass separation of 32 u generated by both symmetric and nonsymmetric cleavages of the disulfide bonds was observed on large crystals. Probably for thermodynamic reasons, the tryptic peptide was subjected to the cleavage of even number of chemical bonds gave rise to radical recombination without reductive ISD. Among several matrices tested including ferulic acid (FA), α-cyano-4-hydroxycinnamic acid (CHCA), sinapinic acid (SA), 2,5-dihydroxybenzoic acid (DHB), 3-aminoquinolin (AQ), and 2,4,6-trihydroxy acetophenone (THAP), FA was shown to be a versatile matrix allowing one to induce or prevent ISD according to the location of laser shots. CHCA and SA were found to promote ISD of disulfide-containing proteins, in a location dependent manner. However, unlike in CHCA (or SA), ISD was not systematically observed on all crystals for FA, suggesting differences between the crystallization processes of CHCA (or SA) and FA. Minor fragments were observed when using DHB and AQ as matrices. As for THAP, no fragmentation was observed probably because its three OH-groups in meta-position relative to each other resulted in the nonoccurrence of redox reaction. The studies provide insights into the experimental conditions required for determination of disulfide-containing protein by MALDI MS and are helpful for mass spectrum interpretation, opening the way to more rational studies of disulfide/cysteine-containing proteins by MALDI mass spectrometry.