Computational prediction of signal peptides is one of the most important steps in genetic engineering experiments. The periplasmic expression cause the reducing in the inherent destructive behavior of Bofurin I against its host and also reducing its susceptibility to proteolytic degradation. In order to predict the best signal peptides for expression of Buforin I in E. coli, 103 signal sequences were retired from signal peptide databases. Since the purpose of this study was to introduce the optimal signal peptides for periplasmic expression, first, sub-cellular localization site of signal peptides was analyzed. Then, n, h, and c regions of signal peptide, signal peptide probability and physico-chemical features were investigated. Base on the results, MalE, hofQ, papK, ugpB, zraP, and sfmC were introduced as the best signal peptides. For increasing the half-life of mRNA and the increasing the stability of the mRNA against exonuclease activity, secondary structures of mRNA including Shine-Dalgarno, untranslated region of ompA, start codon, signal peptide and sequences of Buforin I were analyzed. Based on the total free energy pilot evaluated and mRNA conformations, papK seemed more appropriate than the rest of the signal peptides. The obtained result of this study can be used for design the periplasmic expression constructs.

Keywords