Cyanobacteria are photosynthetic bacteria encountered in various aquatic environments. Some of them are able to produce powerful toxins called cyanotoxins. Among cyanotoxins, microcystins (MCs) constitute a group of closely related cyclic heptapeptides. Their sequences are made up of classical amino acids as well as post- translational modified ones. Interestingly, in vivo metabolism of microcystins seems to be greatly dependent on various minor structural differences and particularly those of the seventh amino acid, which can be either dehydroalanine (or a derivative), dehydroaminobutyric acid (or a derivative), serine or alanine. As a consequence, microcystins have been classified on the basis of the nature of this singular amino acid. A major difficulty in the classification of such toxins is that some of them share the same molecular masses and the same molecular formulas. Consequently, a simple mass measurement is not sufficient to determine the structure and the class of a toxin of interest. Heavy and expensive techniques are used to classify them, such as multi-dimensional nuclear magnetic resonance and amino acid analysis. In this work, a new matrix-assisted laser desorption/ionization time-of-flight method leading to an easy classification of MCs is proposed. The methodology relies on the reductive properties of the matrix 1,5-diaminonaphtalene (1,5-DAN) which appears to be able to selectively reduce the double carbon-carbon bond belonging to the seventh amino acid. Moreover, the yield of reduction seems to be influenced by the degree of substitution of this double bond, allowing a discrimination between dehydroalanine and dehydroaminobutyric acid. This selective reduction was confirmed by the study of three synthetic peptides by mass spectrometry and tandem mass spectrometry. According to these results, the use of reductive matrices seems to be promising in the study of microcystins and in their classification. More generally, 1,5-DAN allows the selective reduction of double carbon-carbon bonds. This property could also be employed in the characterization of others types of compound displaying double bonds (petrochemistry, metabolomics....).