Journal of Structural Biology xxx (2006) xxx–xxx www.elsevier.com/locate/yjsbi 1047-8477/$ - see front matter  2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jsb.2006.07.011 ARTICLE IN PRESS Please cite this article as: Gian A. Signorell et al., Controlled 2D crystallization of membrane proteins using methyl--cyclodextrin, Journal of Structural Biology (2006), doi:10.1016/j.jsb.2006.07.011 Controlled 2D crystallization of membrane proteins using methyl--cyclodextrin Gian A. Signorell, Thomas C. Kaufmann, Wanda Kukulski, Andreas Engel, Hervé-W. Rémigy ¤ M. E. Müller Institute for Microscopy at the Biozentrum, University of Basel, Basel, Switzerland Received 9 May 2006; received in revised form 21 July 2006; accepted 22 July 2006 Abstract High-resolution structural data of membrane proteins can be obtained by studying 2D crystals by electron crystallography. Finding the right conditions to produce these crystals is one of the major bottlenecks encountered in 2D crystallography. Many reviews address 2D crystallization techniques in attempts to provide guidelines for crystallographers. Several techniques including new approaches to remove detergent like the biobeads technique and the development of dedicated devices have been described (dialysis and dilution machines). In addition, 2D crystallization at interfaces has been studied, the most prominent method being the 2D crystallization at the lipid monolayer. A new approach based on detergent complexation by cyclodextrins is presented in this paper. To prove the ability of cyclodextrins to remove detergent from ternary mixtures (lipid, detergent and protein) in order to get 2D crystals, this method has been tested with OmpF, a typical -barrel protein, and with SoPIP2;1, a typical -helical protein. Experiments over diVerent time ranges were performed to analyze the kinetic eVects of detergent removal with cyclodextrins on the formation of 2D crystals. The quality of the pro- duced crystals was assessed with negative stain electron microscopy, cryo-electron microscopy and diVraction. Both proteins yielded crys- tals comparable in quality to previous crystallization reports.  2006 Elsevier Inc. All rights reserved. Keywords: 2D Crystallization; Membrane protein; Cyclodextrin; Electron microscopy 1. Introduction Detergents are used to extract proteins from membranes and during subsequent puriWcation to obtain samples suit- able for crystallization. During this process the membrane protein structure and function are aVected by the loss of interaction with the lipids. To reconstitute membrane pro- teins into a lipid bilayer in order to obtain 2D crystals, the detergent must be removed from a protein–lipid–detergent ternary mixture. There are mainly three ways to remove detergent: dialysis, adsorption to biobeads and dilution (Rigaud et al., 1997; Hasler et al., 1998; Remigy et al., 2003). However, all of these methods have their limitations. For dialysis, the rate of detergent removal is closely related to the critical micellar concentration (CMC) of the detergent (Jap et al., 1992; Kuhlbrandt, 1992). For example, dialysis of low CMC detergents (such as Triton X-100 or dodecyl- -D-maltoside (DDM)) takes weeks at room temperature. Low CMC detergents are often necessary to keep the pro- tein in its active form, as they are better substitutes for the bilayer than high CMC detergents. When low CMC deter- gents are required, the use of biobeads oVers the advantage of a fast detergent removal. However, this can be a prob- lem, since the high eYciency of detergent adsorption to bio- beads may result in too fast detergent removal leading to aggregation primarily when small (<20 l) reconstitution volumes are used. Therefore a minimal initial volume of ternary mixture of 100 l is required for removing the deter- gent in a controlled manner. This will lead to compromises * Corresponding author. Fax: +41 61 267 21 09. E-mail address: herve.remigy@unibas.ch (H.-W. Rémigy).