Despite the ubiquity of helical membrane proteins in nature and their pharmacological importance, the mechanisms guiding their folding remain unclear. We performed kinetic folding and unfolding experiments on 69 mutants (engineered every 2-3 residues throughout the 178-residue transmembrane domain) of GlpG, a membraneembedded rhomboid protease from Escherichia coli. The only clustering of significantly positive ϕ-values occurs at the cytosolic termini of transmembrane helices 1 and 2, which we identify as a compact nucleus. The three loops flanking these helices show a preponderance of negative ϕ-values, which are sometimes taken to be indicative of nonnative interactions in the transition state. Mutations in transmembrane helices 3-6 yielded predominantly ϕ-values near zero, indicating that this part of the protein has denaturedstate-level structure in the transition state. We propose that loops 1-3 undergo conformational rearrangements to position the folding nucleus correctly, which then drives folding of the rest of the domain. A compact N-terminal nucleus is consistent with the vectorial nature of cotranslational membrane insertion found in vivo. The origin of the interactions in the transition state that lead to a large number of negative ϕ-values remains to be elucidated.GlpG | membrane protein | rhomboid | folding | kinetics T he biologically active structure of a protein is encoded in its sequence, and protein-folding studies aim to elucidate how this native state is reached. Great progress has been made in understanding the mechanisms of folding of water-soluble proteins based on comprehensive protein-engineering studies in combination with computational efforts (1, 2) and application of theoretical models (3-5). Much less is known about the folding mechanisms of membrane proteins that present extra challenges such as low expression levels and the need for a membrane-like environment to fold (6-11). In vivo, α-helical membrane proteins insert into the membrane cotranslationally via the signal recognition particle and Sec-translocon complex (12). Transmembrane helices exit one by one or in pairs into the lipid environment through a lateral gate in the translocon. Folding to the native state occurs spontaneously after helices are inserted into the membrane. To mimic this process, most in vitro membrane protein-folding experiments first denature the protein in SDS; renaturation is then achieved by adding excess nonionic surfactants such as dodecyl maltoside (DDM) (13).A complete protein folding mechanism must include descriptions of the denatured state (D), the native state (N), any metastable intermediates, and the transiently populated transition states (TS) that connect them. TS can only be analyzed indirectly using methods based on kinetic experiments, such as Fersht's ϕ-value approach (14, 15). The ϕ-value is the ratio between the energy perturbation to N (from equilibrium measurements or a combination of folding and unfolding kinetics) and the energy perturbation to TS (from kinetic measurements) ca...