With the successful application of nanopore sensing for the sequencing of nucleic acids, research in this area has shifted to the analysis of other biomolecules. Of particular interest is the use of nanopore sensing for the identification of protein analytes. However, this necessitates destabilization of the analytes during data acquisition through the use of agents such as guanidinium chloride (GdmCl), which can also disrupt the structure of the nanopores used for sensing. It is therefore essential that the nanopores used for protein identification can withstand such conditions.
To facilitate protein identification using nanopore sensing, we improved the stability of bryoporin (Bry) pores by introducing halogenated non-canonical amino acids (ncAAs) at the interfaces of the pore protomers using a genetic code expansion. Bry variants containing either m-Cl-tyrosine (ClY) or p-pentafluorosulfanyl phenylalanine (SF5) were produced. Their ability to oligomerize and their stability in GdmCl were tested by native PAGE and nano-differential scanning fluorimetry (nanoDSF), respectively. The pores of the most stable variants were then isolated and characterized by cryo-EM and single channel recordings on planar lipid bilayers.
Despite the nanoDSF results indicating high stability of the SF5-containing pores, we could not isolate them for further analysis due to their poor solubility and high aggregation tendency. ClY-containing pores, on the other hand, were generally less stable but easy to produce. Consequently, we used the latter for further analysis. We performed single channel recordings, where pores of the most stable ClY-containing variant could be used in buffers containing up to 3 M GdmCl, corresponding to a twofold increase in respect to the Bry variants without ncAAs. These promising results give us confidence for achieving our next goal of translocation and identification of full-length protein analytes.