The ability to predict the structure of a protein from its amino-acid sequence is a long-sought-after, but elusive, goal. However, Hessa et al. have brought us one step closer to the dream with their recent work on the sequence requirements of protein transmembrane helices.

Membrane proteins are vitally important to the integrity and functions of a cell, and to successfully integrate into the complex environment of the membrane, a protein must have a similarly complex structure, the nature of which is compatible with the membrane environment. Most membrane proteins have a bundle of tightly packed α-helices that span the width of the membrane and that are predominantly hydrophobic in character.

To fold successfully inside the cell such membrane proteins must insert into, and fold within, the endoplasmic reticulum (ER) membrane as they are being translated. This insertion is mediated by the Sec61-translocon protein complex that provides a channel through which the translating protein is conducted. It can translocate polypeptides into the aqueous environment of the ER lumen or can move them laterally into the lipid environment of the ER membrane. But what factors determine whether the translocon inserts a polypeptide into the membrane or pushes it out into the lumen?

In Nature, Hessa et al. have begun to answer this question by carrying out extensive quantitative studies of translocon-mediated protein insertion. They systematically designed a large set of transmembrane sequences and measured the efficiency of their membrane integration by Sec61. Analysis of these data was used to assess the contributions of individual amino acids to the efficiency of membrane insertion, and a thermodynamic scale was established that measured the energy that was required for the membrane insertion of each amino acid. In fact, the authors found that these biological measurements closely correlated with previously established biophysical measurements for inserting individual amino acids into membrane environments.

By carefully manipulating the amino-acid sequences of their experimental transmembrane segments, Hessa et al. showed that the overall hydrophobicity of the segment was not the only determinant of membrane insertion. The position of particular residues within the segment had an important role in its membrane insertion, and specific residues 'preferred' to lie in specific regions of the segment depending on the compatibility of their physical properties with the membrane environment. Manipulating the ability of the transmembrane segment to form a helical structure by inserting proline residues at various positions also showed that α-helix formation is vital for proper membrane insertion.

Taken together, the results indicated that the translocon recognizes transmembrane helices by allowing the direct interaction of the transmembrane segment and the surrounding membrane lipids. In fact, the authors suggest that the translocon has a dynamic structure that allows the translocating peptide to 'sample' the translocon–membrane interface, thereby ensuring that the physical properties of the polypeptide help to direct its fate. The authors are hopeful that, with more work on other factors that could influence membrane insertion, their results might be used in the future to help predict the membrane-insertion efficiency of natural polypeptide segments.