One of the last places you’d expect to find a charged amino acid residue is buried within the hydrophobic interior of a lipid bilayer. And for the most part, this expectation holds true: portions of proteins that span membranes are typically composed of hydrophobic residues. But in some cases, the positively charged residues lysine and arginine find their way into such apparently unfavorable positions, and it remains unclear why they haven’t been selected out of existence.

One hypothesis to explain why such positively charged residues don’t always interfere with a membrane-anchoring domain of a protein’s function is that they “snorkel”. That is, their side chains wriggle into a position in which the tip, where the charge resides, pokes out of the hydrophobic interior of the membrane into a more hospitable environment, surrounded by the polar head groups of the lipids and the aqueous phase. Some lipid head groups are also negatively charged, which would make snorkeling into this region even more favorable for a positively charged amino acid residue like lysine or arginine.

So far, most of the evidence to support snorkeling has come from in vitro or structural studies, but in the February issue of GENETICS, Keskin et al. report in vivo data supporting the snorkeling hypothesis. Using deep mutational scanning of the yeast mitochondrial outer membrane protein Fis1p, the researchers hoped to understand which structural features of the protein’s carboxyl-terminal tail target it to its membrane.

They found that, in certain positions in the protein’s sequence, positively charged residues were far better at promoting mitochondrial membrane targeting than were negatively charged residues. This makes sense if the residues’ side chains are snorkeling, because negatively charged side chains would be repelled by negatively charged lipid head groups. Not only that, but the negatively charged residues aspartate and glutamate have shorter side chains than lysine and arginine, meaning they might not have a long enough hydrophobic stretch to reach the surface of the lipid bilayer, leaving them trapped among the hydrophobic lipid tails.

In addition to providing strong in vivo evidence for amino acid snorkeling, these results suggest that prediction methods designed to identify membrane-anchoring domains might be excessively stringent. If Keskin et al.’s findings are generalizable, allowing positively charged residues at certain positions in such domains might allow us to better predict the structure and function of proteins from their sequences.  


Keskin, A.; Akdoğan, E.; Dunn, C. Evidence for Amino Acid Snorkeling from a High-Resolution, In Vivo Analysis of Fis1 Tail-Anchor Insertion at the Mitochondrial Outer Membrane.
GENETICS, 205(2), 691-705.
DOI: 10.1534/genetics.116.196428

Nicole Haloupek is a freelance science writer and a recent graduate of UC Berkeley's molecular and cell biology PhD program.

View all posts by Nicole Haloupek »