The 29-amino-acid-long peptide sequence computationally designed in the Saven Lab, named BNDL1, forms homotetrameric, antiparallel, α-helical bundle monomers. These bundles can be linked into rigid rods, utilizing the thiol-maleimide reaction to efficiently link bundles into long polymers of bundles.This project aims to use the computationally designed variants of the 29-residue long sequence to generate and probe hypotheses regarding structure formation. Studies to explore polymeric rod formation and its sensitivity to sequences with varying hydrophobic content at the interface will allow us to investigate the interactions between adjacent bundles. Variants (“mutants”) of the BNDL1 sequence have already been identified: I4A, I4V, E28M. Molecular simulations suggest that changing the 28th residue (glutamic acid) to a hydrophobic methionine residue is expected to stabilize the structure. This is because the change is expected to protect the bundles from solvation by water molecules that disrupt peptide assembly. In contrast, changing the interior hydrophobic amino acid (isoleucine) to a smaller one (alanine or valine) is conjectured to reduce the content of rod-like structure because the cavity between monomers in the homotetrameric bundle is wider in the mutant, allowing for more water molecules to enter this space. We hypothesize that, in their cross-linked polymeric form, the designed variants will exhibit distinct structure and assembly properties. 

I conducted experimental characterization of the I4A variant. Two 25mg samples of the I4A peptide, one with an extra cysteine at the N-terminus and the other with a maleimide modification at the N-terminus, were purchased from GenScript. However, the I4A-Mal peptide had problems during purification according to Genscript. Therefore, we were sent the purified I4A-Cys peptide and crude I4A-Mal peptide, which we attempted to purify on our own with size exclusion chromatography. I used MALDI-TOF MS to verify the identity of the peptides and circular dichroism (CD) spectroscopy to study the secondary structure at various temperatures, melting temperature, and reversibility of thermal denaturing. Our data shows that the I4A mutant of the BNDL1 peptide forms homotetrameric, antiparallel, α-helical bundles as hypothesized, and the mutant also exhibits a melting temperature of around 66C, lower than that of BNDL1.

Rod formation has not been successfully executed yet due to complications in the purification of the I4A-maleimide peptide. As alternatives to the original plan, I reattempted bundle linkage through reacting 1,4-bismaleimidobutane (BMB) cross linkers with I4A-Cysteine bundles. The delivery of the peptide calibration standard that is necessary for detecting the 7.5kDa I4A-Cys dimer connected with a BMB linker got delayed. Therefore for the meantime I ran the peptide program that could detect up to 4kDa molecules. Although the resulting MALDI spectrum suggests the presence of the dimer, the quantification will have to be redone with the protein program. Furthermore, since MALDI would not be able to distinguish between intra-bundle and inter-bundle linkages, a more comprehensive method would have to be developed to verify that the desired inter-bundle linkage occurred.