In recent years, the discovery of a large family of macrocyclic peptides, the cyclotides, has revealed Natures ingenuity in molecular drug design. The incorporation of a cyclic peptide backbone and a knotted arrangement of disulfide bridges into their structures confers extraordinary chemical, thermal, and enzymatic stability on these biologically active peptides. However, these structural attributes present challenges in the identification of cyclotides. Until now, the sequencing of cyclotides has been slow and inefficient owing to inherent difficulties in the separation of these hydrophobic peptides from plants, the multiple chemical and enzymatic derivatization steps required to make them amenable to mass spectrometric sequencing, and the lack of software tools to efficiently deal with these circular permutants. The current bottleneck slowing the speed of cyclotide sequencing is the requirement for multiple HPLC purification steps before analysis. Here, we have applied proteomic strategies to fast-track the discovery of known, modified and novel sequences. Using four fractions from a previously well-characterized cyclotide-containing plant species, Viola odorata, 11 new sequences, as well as a plethora of known and modified cyclotides, were uncovered. In addition, the methodology was validated through analysis of crude leaf extracts ofOldenlandia affinis and Arabidopsis thaliana. The unambiguous identification of a suite of cyclotides in the Oldenlandia affinis extract provided the ultimate proof-of-concept for this application. Major advances in methodology include the use of optimized LC-MS/MS conditions and design of a custom-built cyclotide database, in which mature cyclotide sequences are excised, replicated and appended, marking a new "era" for cyclotide sequencing.
Read full abstract