Beta roll peptide motifs are composed of repetitive calcium binding nonamers. These motifs are found in several proteases, lipases and haemolysins, and are defined by the consensus sequence: GGXGXDXUX, where U is an aliphatic amino acid and X is any amino acid. Repeating peptide scaffolds have been increasingly utilized as a stable and modular interface for biomolecular recognition of target molecules. These repeat peptides can be concatenated to increase the surface area available to design biomolecular interaction, making possible the design of proteins that can bind targets of varying size. Unlike ankyrin repeats, tetratricopeptide repeat proteins and leucine-rich repeats, beta roll folding is triggered by calcium binding. When in its folded state, two variable amino acids of every other nonamer form an interface on one side of the beta roll. Thus, the binding of calcium controls the assembly of the interface designed to bind its target, and in the absence of calcium, the target can be released from the beta roll.

Here we present the engineering and characterization of beta roll motifs from a serralysin (Serratia marcescens) and adenylate cyclase (Bordetella pertussis). The functional folding of the beta roll peptide was determined using circular dichroism (CD) spectroscopy, dynamic light scattering (DLS) and fluorescence resonance energy transfer (FRET). CD spectroscopy indicated that the naked beta roll domain (without flanking sequences) was unable to fold. However, when flanked by its natural N- and C- terminal caps, unstructured residues formed calcium induced beta sheet secondary structure. Similarly, DLS experiments indicated a decrease in the hydrodynamic diameter of beta roll peptides upon calcium binding. We have developed a sensitive FRET-based assay that may be able to determine more subtle conformational change not observed through traditional CD spectroscopy. FRET experiments also revealed a calcium dependent decrease in fluorophore separation. In an effort to understand the roles of regions flanking the beta-roll, both the N- and C- terminal capping groups were truncated in several logical positions based on secondary structure prediction models. We found that the C-terminal flanking residues were sufficient for calcium induced beta roll folding. The N-terminal flanking region, however, was not sufficient for folding without the addition of ~20 amino acids on the C-terminal end of the beta roll. Currently, we are investigating the thermodynamics and kinetics of calcium induced folding to gain greater insight into the way the flanking regions influence the allosteric behavior of beta roll peptides. These insights, in combination with consensus design and refinement, will lead to the development of a modular and expandable allosteric scaffold on which to design biomolecular recognition.

Extended Abstract Status: Not Uploaded