The ability to sequence DNA with nanopores has been a technical coup, but as polymers go, DNA is pretty tame: besides having a relatively uniform secondary structure and charge, it is made up of only four bases. Compare this to proteins and their 20 amino acids, variable charge and hydrophobicity, and considerable secondary and tertiary structure. Despite this complexity, nanopores may soon be useful for assessing protein structural domains, modifications and interactions.

Pulling proteins through nanopores can provide information on structure and modifications. Credit: Debbie Maizels/Nature Publishing Group

Nanopore sensors measure the disruption of ionic current as a single molecule transits the pore; the level of disruption depends on the molecule's sequence and structural features. The same voltage that generates current across the pore causes negatively charged DNA to enter it. To coax weakly charged proteins into the pore, researchers have had to add a string of negatively charged amino acids (Nat. Biotechnol. 31, 247–250, 2013) or a short stretch of DNA (Nat. Nanotechnol. 8, 288–295, 2013) to one end. Optimizing the efficiency of adding the leader will be important for generalizing and scaling up assays.

Threading a leader through the pore causes some proteins to continue unwinding, but others need more force to denature and pull them through. Adding a recognition site for the unfoldase ClpX to the leader made it possible to ratchet target protein through a pore in a sequence-independent fashion (Nat. Biotechnol. 31, 247–250, 2013). For tightly folded proteins, denaturants may need to be tested.

Nanopores have been used to characterize functional aspects of unfolding proteins during translocation. One study used them to discriminate site-specific phosphorylation states of thioredoxin (Nat. Biotechnol. 32, 179–181, 2014). Engineering larger protein pores (J. Am. Chem. Soc. 135, 13456–13463, 2013) may also allow for folded domains or proteins to enter the pore in order to assay protein interactions with other proteins, drugs or substrates.

Ultimately, designing motors that move protein through the pore one amino acid at a time will be a critical step toward single-protein sequencing. It may be impossible to resolve the signature of amino acids individually or in each of their bewildering number of combinations. But recognizing domains and coarse features may be sufficient to identify proteins, characterize unfolding or protein states, or assess interactions. We look forward to seeing nanopores become established tools for studying proteins at the single-molecule level.