The messy process of guiding proteins into membranes

O ne of the keys to predicting the threedimensional structure of a membrane protein from its sequence of amino acid residues is to understand how structures called translocons guide the protein to its final folded state. Translocons are generally thought of as channels that allow proteins to cross cell membranes. In eukaryotes, it is thought that newly-formed secreted proteins pass through the Sec61 translocon as they emerge from the ribosome. New membrane proteins are thought to follow a similar path, except that the hydrophobic transmembrane helices in these proteins are diverted sideways so that they become embedded in the cell membrane. This ‘sequential-insertion’ scheme seems logical in the context of what we know about the structure of translocons (Rapoport et al., 2004; Cymer et al., 2015), but is it correct? We cannot answer this question because we do not have experimental methods that can follow, residue-by-residue, the insertion and folding of the protein chains as they pass from the ribosome and into the membrane. The alternative is to simulate the process. However, a newly-formed protein chain elongates at a rate of about one residue every 50–100 milliseconds, which is orders of magnitude faster than can be modeled using standard molecular dynamics simulation methods. Now, in eLife, Reid van Lehn, Bin Zhang and Thomas Miller of the California Institute of Technology report a simplified approach that allows insertion and folding to be simulated on biological time scales (Van Lehn et al., 2015). Their results suggest that the membrane protein insertion/folding process is more complicated than commonly depicted in the sequential-insertion scheme. Van Lehn et al. modeled a protein called EmrE that sits in the inner membrane of Escherichia coli bacteria and is able to transport a wide range of antibiotic drugs out of the cell. This helps to make the bacteria resistant to these treatments. EmrE is a homodimer, and each monomer has four transmembrane helices (Chen et al., 2007). EmrE is unusual in that the two monomers are oriented in opposite directions (Figure 1A): this is known as dual topology. The topology (orientation) of membrane proteins is largely determined by the positive-inside rule (von Heijne, 1986). This rule suggests that if the connecting loops that join the transmembrane regions of the protein are rich in lysine and arginine residues, then these loops tend to orient inward, toward the cytoplasm of the cell. This is known as the K+R bias. EmrE, which is encoded in a single gene, has a weak K+R bias, and this means that the monomers can be inserted into the membrane in one of two opposite orientations (Rapp et al., 2006, 2007). In 2010, researchers at Stockholm University reported, based on extensive mutation studies, Copyright White. This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.