Modification-tolerant Shotgun Protein Sequencing of a Snake Venom Proteome

Despite the steady accumulation of fully sequenced genomes for model organisms, limited or no sequence information is available for most organisms. Moreover, natural mechanisms of variation such as accelerated mutation and combinatorial recombination in immunoglobulins regularly create novel sequences in the proteomes of model organisms. However, since protein identification via database searching of MS/MS spectra is not applicable in these contexts and de novo sequencing of individual peptides by MS/MS tends to produce inaccurate short sequences, laborious Edman degradation of isolated peptides remains the primary approach to design degenerate PCR primers for cloning the genes of individual proteins. To catalyze the otherwise impractical application of proteomics methodologies in these cases, we have developed Shotgun Protein Sequencing to produce highly accurate protein contigs by acquiring, assembling and simultaneously sequencing multiple MS/MS spectra from overlapping (and possibly modified) peptides generated via multiple proteases of different specificities. By capitalizing on the redundant sequence information in multiple spectra covering the same protein regions, we were able to achieve substantially higher de novo sequencing accuracy than what is achievable through the interpretation of individual spectra. Thus we believe that Shotgun Protein Sequencing opens the door to automated proteomic studies in otherwise neglected organisms that lack sequenced genomes.