Converting a Marginally Hydrophobic Soluble Protein into a Membrane Protein Morten H.H. Nørholm 1 , Fiona Cunningham 2,3 , Charles M. Deber 2,3 and Gunnar von Heijne 1 ⁎ 1 Department of Biochemistry and Biophysics, Center for Biomembrane Research, Stockholm University, SE-106 91 Stockholm, Sweden 2 Division of Molecular Structure and Function, Research Institute, Hospital for Sick Children, Toronto, M5G 1X8 Ontario, Canada 3 Department of Biochemistry, University of Toronto, Toronto, M5S 1A8 Ontario, Canada Received 10 December 2010; received in revised form 13 January 2011; accepted 15 January 2011 Available online 22 January 2011 Edited by J. Bowie Keywords: membrane proteins; membrane insertion; transmembrane helix; protein folding; protein localization δ-Helices are marginally hydrophobic α-helical segments in soluble proteins that exhibit certain sequence characteristics of transmembrane (TM) helices [Cunningham, F., Rath, A., Johnson, R. M. & Deber, C. M. (2009). Distinctions between hydrophobic helices in globular proteins and TM segments as factors in protein sorting. J. Biol. Chem., 284, 5395–402]. In order to better understand the difference between δ-helices and TM helices, we have studied the insertion of five TM-like δ-helices into dog pancreas microsomal membranes. Using model constructs in which an isolated δ- helix is engineered into a bona fide membrane protein, we find that, for two δ-helices originating from secreted proteins, at least three single-nucleotide mutations are necessary to obtain efficient membrane insertion, whereas one mutation is sufficient in a δ-helix from the cytosolic protein P450BM-3. We further find that only when the entire upstream region of the mutated δ-helix in the intact cytochrome P450BM-3 is deleted does a small fraction of the truncated protein insert into microsomes. Our results suggest that upstream portions of the polypeptide, as well as embedded charged residues, protect δ-helices in globular proteins from being recognized by the signal recognition particle–Sec61 endoplasmic-reticulum-targeting machin- ery and that δ-helices in secreted proteins are mutationally more distant from TM helices than δ-helices in cytosolic proteins. © 2011 Elsevier Ltd. All rights reserved. Introduction What distinguishes transmembrane (TM) helices from helices in soluble proteins? Obviously, hydro- phobicity is a major determinant influencing the destination of a given helical segment. Still, marginally hydrophobic protein segments are observed to form both TM helices and parts of globular proteins, 1,2 and hence, additional factors must contribute to ensure their proper localization. A well-studied example is the membrane-embed- ded voltage-sensor domain present in voltage-gated ion channels. 3 Voltage-sensor domains contain an unusual, highly charged TM helix (the S4 helix) and undergo major conformational changes in response to changes in the membrane electric potential. In the Shaker and KAT1 voltage-sensitive potassium channels, membrane insertion of the S4 helix is aided by electrostatic interactions with charged residues in neighboring helices. 4,5 In the opposite case, unusually hydrophobic so- called δ-helices exist in proteins that are not *Corresponding author. E-mail address: gunnar@dbb.su.se. Abbreviations used: TM, transmembrane; ER, endoplasmic reticulum; endoH, endoglycosidase H. doi:10.1016/j.jmb.2011.01.035 J. Mol. Biol. (2011) 407, 171–179 Contents lists available at www.sciencedirect.com Journal of Molecular Biology journal homepage: http://ees.elsevier.com.jmb 0022-2836/$ - see front matter © 2011 Elsevier Ltd. All rights reserved.