- Email: [email protected]
Membrane protein crystallization from lipidic phases
Available online at www.sciencedirect.com
Membrane protein crystallization from lipidic phases Linda C Johansson1, Annemarie B Wo¨hri2, Gergely Katona1, Sven Engstro¨m2 and Richard Neutze1 Membrane protein structural biology is enjoying a steady acceleration in the rate of success. Nevertheless, numerous membrane protein targets are resistant to the traditional approach of directly crystallizing detergent solubilized and purified protein and the ‘niche market’ of lipidic phase crystallization is emerging as a powerful complement. These approaches, including lipidic cubic phase, lipidic sponge phase, and bicelle crystallization methods, all immerse purified membrane protein within a lipid rich matrix before crystallization. This environment is hypothesized to contribute to the protein’s long-term structural stability and thereby favor crystallization. Spectacular recent successes include the highresolution structures of the b2-adrenergic G-protein-coupled receptor, the A2A adenosine G-protein-coupled receptor, and the mitochondrial voltage dependent anion channel. In combination with technical innovations aiming to popularize these methods, lipidic phase crystallization approaches can be expected to deliver an increasing scientific impact as the field develops. Addresses 1 Department of Chemistry, Biochemistry and Biophysics, University of Gothenburg, SE-405 30 Gothenburg, Sweden 2 Department of Chemical and Biological Engineering, Chalmers University of Technology, SE-41296 Gothenburg, Sweden Corresponding author: Neutze, Richard ([email protected])
Current Opinion in Structural Biology 2009, 19:372–378 This review comes from a themed issue on Membranes Edited by Declan Doyle and Graham Shipley Available online 4th July 2009 0959-440X/$ – see front matter # 2009 Elsevier Ltd. All rights reserved. DOI 10.1016/j.sbi.2009.05.006
Introduction Membrane proteins are fickle entities and repeatedly resist even the most determined efforts to overproduce, purify, and crystallize them for structural studies. Despite this, the field of membrane protein structural biology is undergoing a period of rapid expansion with accelerating rates of successful structural determination, in part due to the incorporation of technical innovations developed for high-throughput structural biology [1]. Detergent based crystallization protocols, whereby detergent solubilized and purified membrane protein is crystallized using the vapor-diffusion method essentially as if it were a soluble Current Opinion in Structural Biology 2009, 19:372–378
protein, are by far the most popular and successful approach developed to date [2]. Despite this, it has almost become a truism that the improvement of initial membrane protein crystals is painstakingly slow and frequently the quality of crystals can plateau at a point unsuitable for structural determination. In other cases no crystal leads whatsoever emerge despite success in recovering pure, stable protein. When faced with such disappointments, it is attractive to explore other crystallization protocols. A celebrated approach to improve the chances of recovering high quality crystals is to enhance the crystal contacts by enlarging the soluble domains through the addition of proteins specific antibodies [3], although this typically falls outside the capabilities of most laboratories. A variation of this theme is to increase the membrane protein’s soluble domains by protein engineering [4] but this may potentially introduce other complications. Another approach, the crystallization of membrane proteins in lipidic environments, explicitly recognizes and addresses the fact that membrane proteins are most stable in lipid bilayers. Detergent solubilization, the extraction of membrane proteins from their native membrane, often causes structural lipids and weakly bound subunits to be lost and potentially impairs protein integrity. As a consequence, membrane proteins frequently display reduced activity and poor stability in detergent solution and this role of lipids in membrane protein structural biology has been extensively discussed [5]. Alternatively, as originally conceived and demonstrated by Landau and Rosenbusch [6], purified membrane proteins can be reconstituted into lipidic bilayer environments before crystallization. By reintroducing proteins into a lipidic bilayer it was hypothesized that enhanced protein stability would be achieved, thus aiding the crystallization process. In this review we focus on recent progress using lipidic phase environments as vehicles aiding membrane protein crystallization. We first sketch the underlying ideas of lipidic phases and describe their application to the methods of lipidic cubic phase (LCP) crystallization [6], its offspring lipidic sponge phase (LSP) crystallization [7], and the conceptually related approach of bicelle crystallization [8]. We emphasize the recent structural results to emerge using each of these methods and highlight efforts to increase their popularity. In closing we speculate upon the future role likely to be played by lipidic phase crystallization protocols within the broader discipline of membrane protein structural biology. www.sciencedirect.com
Lipidic phase crystallisation Johansson et al. 373
Lipidic phases
Figure 1
Amphipathic lipid molecules consist of both hydrophobic and hydrophilic moieties. When exposed to aqueous environments they spontaneously form welldefined structures so as to maximize energetically favorable interactions of the hydrophilic head group with water and to minimize the exposure of hydrophobic tail group. The lipidic phases that emerge from the optimal enthalpy/entropy balance depend upon the specific geometry of the lipid itself, its concentration, its temperature, as well as the presence of other additives including ions and amphiphiles. The spontaneous formation of detergent micelles and unilamellar vesicles are used daily by membrane protein biochemists, but other less familiar lipidic aggregates are also easily accessed including reverse micelles and various other lipidic bilayer structures. Lipidic bilayer phases encompass vesicles, lamellar phases (stacked bilayers) as well as the bicontinuous cubic and sponge phases, which can be traversed in any direction along either hydrophilic or hydophobic paths. These mesophases [9] are more ordered than a liquid but less ordered than a solid. LCPs are semi-solid in texture and their long-range order can be visualized using small-angle X-ray scattering which reveals scattering peaks corresponding to the characteristic lattice parameters of their cubic packing [7]. Sponge phases, by comparison, can be thought of as swollen cubic phases with aqueous pores up to three times larger than those of a cubic phase [10] and lower long-range order. Another characteristic distinguishing the lipidic cubic and sponge phases is that the latter are liquid in nature. Lamellar phases also have low viscosity, but they are anisotropic and are therefore birefringent, whereas the isotropic cubic and sponge phases are non-birefringent. Thus cross-polarized microscopy provides a useful and rapid diagnostic distinguishing the lamellar and sponge phases. Bicelles can be thought of as solubilized lipidic bilayer disks. They are formed by the addition of detergent (or a short-chain lipid) to a long-chain lipid [8,11]. These mixtures spontaneously form disc-shaped aggregates of lipids and detergents, with the long-chain phospholipid forming a central planar bilayer which is surrounded by a rim of detergents that protect the bilayer from water. The physical diameter of the bicelle is controlled by the ratio of long-chain to short-chain amphiphiles. Moreover, membrane proteins reconstituted into bicelles maintain functionality [12].
Lipidic cubic phase crystallization Monoolein has been the lipid of choice for membrane protein crystallization from LCPs [6], although some successes have also been reported using closely related lipids such as monovaccenin [13]. When mixed with water, monoolein spontaneously swells to form several www.sciencedirect.com
Phase diagram of the PEG:monoolein:water system illustrating the presence of LCP (dark blue), LSP (light blue) and lamellar (gray) phases. Sponge phases do not form in the absence of PEG (triangle base) but require approximately 30% water content (dashed line). The PEG (25– 60%) and monoolein (10–45%) concentrations can vary over a broader range [10]. Phase diagrams recovered from PEGs of variable length are rather similar.
mesophases. At room temperature the Pn3m cubic phase is recovered at approximately 40% water concentration with the (larger cell) Ia3d cubic phase emerging at slightly lower water content [9]. The traditional starting point for any LCP crystallization experiment has been to incorporate membrane protein into either of these semi-solid cubic phases by mixing a buffer containing purified protein with monoolein in an approximate ratio of 2:3 by volume. As with any other batch crystallization experiment, this protein-containing LCP is then dispensed into tubes [6] or crystallization plates [14], overlaid with a crystallization agent and sealed. Dehydration of the cubic phase drives a phase-transition to a lamellar phase (Figure 1) and it is believed that this phase-transition is a key ingredient in successfully recovering crystals, a hypothesis which receives support from the observation of birefringence in the immediate vicinity of crystals grown in the LCP [15]. Bacteriorhodopsin was the first membrane protein to be successfully crystallized using LCP crystallization [6] and improvements in the crystal quality shortly later yielded high-resolution crystal structures of bacteriorhodopsin in its resting conformation [16–18]. LCP crystals of bacteriorhodopsin grow as stacked layers of 2D crystals and are functionally active, such that these crystals continue to diffract after illumination. Thus several conformational changes associated with the photo-cycle of bacteriorhodopsin have been convincingly demonstrated using light illuminated LCP crystals [19–23] and have contributed to Current Opinion in Structural Biology 2009, 19:372–378
374 Membranes
a uniquely detailed picture of the structural mechanism of a membrane protein in action [24].
Figure 2
Three other archaeal rhodopsins have also been successfully crystallized using the LCP technique and highresolution crystal structures of halorhodopsin [25], N. pharaonis sensory rhodopsin II alone [26] and in complex with its integral membrane transducer protein [13], and Anabaena sensory rhodopsin [27] all emerged using this technique. Two criticisms, that LCP crystallization was a niche technique successful for the archaeal rhodopsins but unsuitable for the crystallization of large membrane protein complexes or colorless proteins, were countered by the successful LCP crystallization of the R. sphaeroides reaction center complex [28], the colorless cobalamin transporter BtuB [29] and outer membrane adhension protein OpcA [30], respectively. Moreover, as with bacteriorhodopsin, the LCP grown crystals of halorhodopsin [31], sensory rhodopsin II [32], sensory rhodopsin II in complex with its transducer [33], and the photosynthetic reaction center [34] are all functionally active and have yielded convincing difference electron density changes following illumination. To the wider community, however, the scientific impact of the above successes has recently been dwarfed by the LCP crystallization and structure determination to 2.4 A˚ resolution of a modified form of the human b2-adrenergic G-protein-coupled receptor [35] and the 2.6-A˚ resolution structure of the human A2A adenosine G-proteincoupled receptor [36]. These high-resolution structures of ligand-activated G-protein-coupled receptors have generated considerable excitement within the structural biology community [37]. Moreover, the challenge associated with crystallizing highly flexible membrane proteins is widely appreciated, and crystal structures of G-proteincoupled receptors have become something of a Holy Grail within the field. Thus these structures provide dramatic support for the original hypothesis of Landau and Rosenbusch [6] that by maintaining membrane proteins in a lipid-bilayer environment during crystallization, protein stability will be improved and enhance the chances of success.
Lipidic sponge phase crystallization Membrane protein crystallization in LSP emerged as a serendipitous offshoot of the LCP technique. Visual inspection of the LCP crystallization setups of the R. sphaeroides reaction center [28] indicated that the semisolid LCP liquefied before crystal formation, and smallangle X-ray scattering studies revealed that this liquefied cubic phase was in fact a LSP [7]. Since it is easier to pipette a liquid rather than a semi-solid phase, we sought to grow crystals of the reaction center using a traditional hanging-drop vapor-diffusion experiment with the liquid sponge phase added directly to a drop containing concentrated protein. This approach was successful without Current Opinion in Structural Biology 2009, 19:372–378
Crystals of the R. sphaeroides reaction center grown in (a) the LCP and (b) close-up of setup. (c) Hanging-drop crystallization experiment for the R. sphaeroides reaction center grown in a LSP and (d) close-up view of a crystal recovered in the drop illustrated in (c). Although the morphology of the two crystal forms appears to be quite different, these crystals are isomorphous. (e) Hanging-drop crystallization experiment for the reaction center of Bl. viridis grown in a LSP and (f) close-up view of a crystal recovered in the drop illustrated in (e).
any significant change to the crystallization conditions [7] and larger crystals with improved diffraction were recovered from optimized LSP setups. Curiously, while their morphology appeared to be different to that observed when using the LCP setups (Figure 2), both crystal forms were isomorphous. In the LSP crystallization of the R. sphaeroides reaction center, Jeffamine M-600 was the reagent driving the formation of the monoolein sponge phase. Jeffamine, however, has not proven to be a common reagent used during the successful crystallization of membrane proteins [2], whereas polyethyleneglycols of various lengths (PEGs) have yielded the largest number of successes [2]. Somewhat fortuitously, monoolein will also spontaneously form a stable sponge phase when mixed with PEGs at concentrations close to those successful for membrane protein crystallization [10] (Figure 1). To further explore the possibility of crystallizing membrane proteins within a LSP environment, we developed a 48condition LSP screen with a strong bias toward PEG www.sciencedirect.com
Lipidic phase crystallisation Johansson et al. 375
conditions and a small number of Jeffamine M-600 conditions. In our hands, using hanging-drop vapor-diffusion experiments, this screen yielded crystallization leads for eight membrane proteins [38] and the LSP structure of the Bl. viridis reaction center to 1.85 A˚ has been determined (Wo¨hri et al., under review). In an independent development, Cherezov et al. crystallized a bacterial light harvesting complex II using LCPs swollen by the addition of 20% PPO to form a LSP [39]. Similarly, an earlier LCP crystal form of the bacterial cobalamin transporter BtuB [29] was improved by swelling the LCP by the addition of 10–12% MPD to form an LSP during crystallization [40]. Thus LSP structures of four bacterial membrane proteins have been solved to date (Table 1, Figure 3). Although the pioneering saturated-salt LCP crystallization of bacteriorhodopsin [6] did not proceed via the sponge phase, another successful condition containing 27% PEG 2000 has been reported [41] which lies near the sweet spot for sponge phase formation [10] (Figure 1). Likewise, the LCP crystallization conditions of the b2adrenergic G-protein-coupled receptor (30–35% PEG 400 in [35] and 28% PEG 400 in [42]), the A2A adenosine G-protein-coupled receptor (28–32% PEG 400 in [36]), and the outer membrane protein OpcA [30] (18% PEG 400) all appear tantalizingly close to that required to form a LSP. It is possible that one or more of these LCP crystallization setups may have passed via the sponge phase before crystal growth, in analogy with that observed during the LCP crystallization of the R. sphaeroides reaction center [7], and it would be of interest to consider the possibility of reproducing these crystals starting directly from the LSP.
Lipidic bicelle crystallization This crystallization method was conceived by Faham and Bowie [8] as a compromise between the detergent solubilized crystallization of membrane proteins and the LCP crystallization approach, elegantly unifying the convenience of working with detergents with the potential benefits of reconstituting membrane proteins into lipidic bilayers before crystallization. Bicelles formed by mixing the long-tail lipids dimyristoyl phosphatidylcholine (DMPC) or ditridecanoyl phosphatidylcholine (DTPC) with the detergent CHAPSO or nonyl maltoside have proven to be successful media for aiding the crystallization of membrane proteins, with a final lipid content of approximately 8% in the crystallization drop. In practical terms, bicelle crystallization proceeds as with any detergent crystallization protocol with an additional step of mixing purified and concentrated membrane protein (on ice) with bicelle formulations before setting up drops. Thus any commercially available or home-grown screen is compatible with the bicelle crystallization approach and the frequently arduous process of crystal optimization proceeds exactly as with the more familiar detergent based protocols. As with the LCP crystallization, bacteriorhodopsin provided the first example of a successful bicelle crystallization experiment [8], with other crystal forms later being found under different bicelle crystallization conditions [11]. The history of success for bicelles crystallization has also followed a path uncannily similar to that of the LCP technique, with three recent major breakthroughs. Xanorhodopsin, a member of the bacterial rhodopsin family but containing two pigments, was crystallized and its structure solved to 1.9 A˚ resolution [43] using bicelle crystallization; the voltage gated
Table 1 Summary of deposited structures solved using LCP, LSP, and bicelle crystallization Protein Bacteriorhodopsin Halorhodopsin Sensory rhodopsin II Sensory rhodopsin II-transducer complex Reaction center Anabaena Sensory Rhodopsin Engineered human b2-adrenergic receptor OpcA outer membrane adhesin Human A2A adenosine receptor Reaction center Light harvesting complex II BtuB Reaction center Bacteriorhodopsin b2-Adrenergic G-protein-coupled receptor Voltage-dependent anion channel Xantorhodopsin a b
Resolution (A˚)
Source
PDB entry
Method
Year
First lipidic phase structure
2.5 1.8 2.1 1.94 2.35 2.0 2.4 1.95 2.6 2.2 2.45 1.95 1.95 2.0 3.4/3.7 2.3 1.9
H. salinarum H. salinarum N. pharaonis N. pharaonis R. sphaeroides Anabaena Homo sapiens N. meningitidis Homo sapiens R. sphaeroides Rps. acidophila E. coli Bl. viridis H. salinarum Homo sapiens Mus musculus S. ruber
1AP9 1E12 1H68 1H2S 1OGV 1XIO 2RH1 2VDF 3EML 2GNU 2FKW 2GUF 2WJM 1KME 2R4R 3EMN 3DDL
LCP LCP LCP LCP LCP a LCP LCP a LCP a LCP a LSP LSP LSP LSP Bicelle Bicelle Bicelle Bicelle
1997 2000 2001 2002 2003 2004 2007 2007 2008 2006 2006 2006 2009 2002 2007 2008 2008
Pebay-Peyroula et al. Kolbe et al. Royant et al. Gordeliy et al. Katona et al. Vogeley et al. Cherezov et al. Cherezov et al. Jaakola et al. Wadsten et al. Cherezov et al. Cherezov et al. Wo¨hri et al. b Faham et al. Rasmussen et al. Ujwal et al. Luecke et al.
Originally reported as LCP crystallization but likely to have proceeded via LSP. Under review .
www.sciencedirect.com
Current Opinion in Structural Biology 2009, 19:372–378
376 Membranes
Figure 3
Growth in the number of deposited membrane protein structures solved using lipidic phase crystallization methods. Structures solved using LCP crystallization are shown in dark blue; structures solved using LSP crystallization are show in light blue; and structures solved using bicelle crystallization are shown in dark green. The LCP and bicelle structures of bacteriorhodopsin, and the LCP and LSP structures of the R. sphaeroides reaction center, are all counted. Mutant structures or alternative conformations of a previously counted structure are not included within this summary. Relevant citations and PDB entries are given in Table 1.
anion channel (VDAC1) a regulated eukaryotic b-barrel channel, was solved to 2.3 A˚ resolution using this method [44]; and the structure of human b2-adrenergic G-protein-coupled receptor in complex with a Fab fragment was solved to 3.4 A˚ resolution using bicelle crystallization [45]. As with similar recent successes using LCP crystallization, these recent outstanding structural results have greatly raised the profile of the bicelle crystallization method within the structural biology community.
Popularization Thirteen years have passed since the first demonstration of membrane protein crystallization within a lipidic phase environment [6]. A number of practical disadvantages associated with crystallization using the semi-solid LCP have been addressed by moving to submicroliter crystallization drops [14] and developing a crystallization plate and an automated dispensing device suited to LCP setups [46]. Moreover, a LCP start-up kit and crystallization screen have recently been marketed by Jena Bioscience (http://www.jenabioscience.com/) and Emerald BioSystems (http://www.emeraldbiosystems.com/), a LSP crystallization screen has been marketed by Molecular Dimensions Ltd. (http://www.moleculardimensions.com/), Qiagen (http://www1.qiagen.com/) have developed a crystallization plate with a thin layer of monoolein pre-dispersed, and a bicelle kit containing lipids and detergents is marketed by Anatrace (http://www. anatrace.com/). Current Opinion in Structural Biology 2009, 19:372–378
Adopting any new practice necessarily requires that a learning curve be surmounted, yet vapor diffusion crystallization experiments with lipidic sponge and bicelle phases proceed almost exactly as with standard detergent crystallization protocols once the LSP or bicelle preparations have been mastered. Moreover, the results from crystallization drops are as easily interpreted (Figure 2c–f) and the techniques are amenable to automation using standard crystallization robots. Thus both methods could easily be applied even at an early stage in a crystallization project in parallel with the more widespread initial crystal screening of detergent solubilized membrane protein.
Conclusions Figure 3 illustrates the growth of membrane protein structures solved using LCP, LSP, and bicelle crystallization techniques. This list of membrane protein structures (Table 1) includes 7 TM bacterial (rhodopsins) and eukaryotic (GPCRs) membrane proteins; 9, 11, and 18 TM membrane protein complexes (SRII:Htr, reaction centers, and LH II complexes); and 10, 19, and 22 stranded b-barrel structure (OpcA, VDAC1, and BtuB). All crystal forms grow as stacked layers of 2D crystals and frequently display significant protein:protein interactions within the plan of the membrane, characteristic of Type I membrane protein crystals [47]. The last two years mark a golden period for the development and wider acceptance of lipidic phase crystallization techniques. Nothing speaks louder to the www.sciencedirect.com
Lipidic phase crystallisation Johansson et al. 377
scientific community than high-resolution structures of new eukaryotic membrane proteins and, with several recent high-profile successes [35,36,44,45], these methods must now be said to have come of age. Consequently, we anticipate a more widespread pursuit of these approaches and look forward to further novel structural insights into fold and function emerging from the crystallization of membrane proteins in lipidic phases.
Conflict of interest
13. Gordeliy VI, Labahn J, Moukhametzianov R, Efremov R, Granzin J, Schlesinger R, Buldt G, Savopol T, Scheidig AJ, Klare JP et al.: Molecular basis of transmembrane signalling by sensory rhodopsin II-transducer complex. Nature 2002, 419:484-487. 14. Nollert P, Navarro J, Landau EM: Crystallization of membrane proteins in cubo. Methods Enzymol 2002, 343:183-199. 15. Nollert P, Qiu H, Caffrey M, Rosenbusch JP, Landau EM: Molecular mechanism for the crystallization of bacteriorhodopsin in lipidic cubic phases. FEBS Lett 2001, 504:179-186.
The authors declare no conflict of interest arising from this work.
16. Pebay-Peyroula E, Rummel G, Rosenbusch JP, Landau EM: X-ray structure of bacteriorhodopsin at 2.5 A˚ from microcrystals grown in lipidic cubic phases. Science 1997, 277:1676-1681.
Acknowledgements
17. Belrhali H, Nollert P, Royant A, Menzel C, Rosenbusch JP, Landau EM, Pebay-Peyroula E: Protein, lipid and water organization in bacteriorhodopsin crystals: a molecular view of the purple membrane at 1.9 A˚ resolution. Structure 1999, 7:909-917.
This work was supported by the Swedish Science Research Council (VR), the European Commission (E-MEP).
References and recommended reading Papers of particular interest published within the period of review have been highlighted as: of special interest of outstanding interest 1.
Carpenter EP, Beis K, Cameron AD, Iwata S: Overcoming the challenges of membrane protein crystallography. Curr Opin Struct Biol 2008, 18:581-586.
18. Luecke H, Schobert B, Richter HT, Cartailler JP, Lanyi JK: Structure of bacteriorhodopsin at 1.55 A˚ resolution. J Mol Biol 1999, 291:899-911. 19. Edman K, Nollert P, Royant A, Belrhali H, Pebay-Peyroula E, Hajdu J, Neutze R, Landau EM: High-resolution X-ray structure of an early intermediate in the bacteriorhodopsin photocycle. Nature 1999, 401:822-826. 20. Luecke H, Schobert B, Richter HT, Cartailler JP, Lanyi JK: Structural changes in bacteriorhodopsin during ion transport at 2 A˚ resolution. Science 1999, 286:255-261.
2. Newstead S, Ferrandon S, Iwata S: Rationalizing alpha-helical membrane protein crystallization. Protein Sci 2008, 17:466-472. Comprehensive summary of successful membrane protein crystallization conditions.
21. Royant A, Edman K, Ursby T, Pebay-Peyroula E, Landau EM, Neutze R: Helix deformation is coupled to vectorial proton transport in the photocycle of bacteriorhodopsin. Nature 2000, 406:645-648.
3.
22. Luecke H, Schobert B, Cartailler JP, Richter HT, Rosengarth A, Needleman R, Lanyi JK: Coupling photoisomerization of retinal to directional transport in bacteriorhodopsin. J Mol Biol 2000, 300:1237-1255.
Hunte C, Michel H: Crystallization of membrane proteins mediated by antibody fragments. Curr Opin Struct Biol 2002, 12:503-508.
4.
Rosenbaum DM, Cherezov V, Hanson MA, Rasmussen SG, Thian FS, Kobilka TS, Choi HJ, Yao XJ, Weis WI, Stevens RC et al.: GPCR engineering yields high-resolution structural insights into b2-adrenergic receptor function. Science 2007, 318:12661273. High-resolution LCP crystal structure of a GPCR. 5.
Hunte C, Richers S: Lipids and membrane protein structures. Curr Opin Struct Biol 2008, 18:406-411.
6.
Landau EM, Rosenbusch JP: Lipidic cubic phases: a novel concept for the crystallization of membrane proteins. Proc Natl Acad Sci U S A 1996, 93:14532-14535.
7.
8.
9.
Wadsten P, Wo¨hri AB, Snijder A, Katona G, Gardiner AT, Cogdell RJ, Neutze R, Engstro¨m S: Lipidic sponge phase crystallization of membrane proteins. J Mol Biol 2006, 364:44-53. Faham S, Bowie JU: Bicelle crystallization: a new method for crystallizing membrane proteins yields a monomeric bacteriorhodopsin structure. J Mol Biol 2002, 316:1-6. Caffrey M: A lipid’s eye view of membrane protein crystallization in mesophases. Curr Opin Struct Biol 2000, 10:486-497.
10. Bender J, Jarvoll P, Nyden M, Engstrom S: Structure and dynamics of a sponge phase in the methyl deltaaminolevulinate/monoolein/water/propylene glycol system. J Colloid Interface Sci 2008, 317:577-584. 11. Faham S, Boulting GL, Massey EA, Yohannan S, Yang D, Bowie JU: Crystallization of bacteriorhodopsin from bicelle formulations at room temperature. Protein Sci 2005, 14:836-840. 12. Czerski L, Sanders CR: Functionality of a membrane protein in bicelles. Anal Biochem 2000, 284:327-333. www.sciencedirect.com
23. Sass HJ, Buldt G, Gessenich R, Hehn D, Neff D, Schlesinger R, Berendzen J, Ormos P: Structural alterations for proton translocation in the M state of wild-type bacteriorhodopsin. Nature 2000, 406:649-653. 24. Neutze R, Pebay-Peyroula E, Edman K, Royant A, Navarro J, Landau EM: Bacteriorhodopsin: a high-resolution structural view of vectorial proton transport. Biochim Biophys Acta 2002, 1565:144-167. 25. Kolbe M, Besir H, Essen LO, Oesterhelt D: Structure of the lightdriven chloride pump halorhodopsin at 1.8 A˚ resolution. Science 2000, 288:1390-1396. 26. Royant A, Nollert P, Edman K, Neutze R, Landau EM, PebayPeyroula E, Navarro J: X-ray structure of sensory rhodopsin II at 2.1 A˚ resolution. Proc Natl Acad Sci U S A 2001, 98:10131-10136. 27. Vogeley L, Sineshchekov OA, Trivedi VD, Sasaki J, Spudich JL, Luecke H: Anabaena sensory rhodopsin: a photochromic color sensor at 2.0 A˚. Science 2004, 306:1390-1393. 28. Katona G, Andreasson U, Landau EM, Andreasson LE, Neutze R: Lipidic cubic phase crystal structure of the photosynthetic reaction centre from Rhodobacter sphaeroides at 2.35 A˚ resolution. J Mol Biol 2003, 331:681-692. 29. Misquitta LV, Misquitta Y, Cherezov V, Slattery O, Mohan JM, Hart D, Zhalnina M, Cramer WA, Caffrey M: Membrane protein crystallization in lipidic mesophases with tailored bilayers. Structure 2004, 12:2113-2124. 30. Cherezov V, Liu W, Derrick JP, Luan B, Aksimentiev A, Katritch V, Caffrey M: In meso crystal structure and docking simulations suggest an alternative proteoglycan binding site in the OpcA outer membrane adhesin. Proteins 2008, 71:24-34. Current Opinion in Structural Biology 2009, 19:372–378
378 Membranes
LSP structure of a bacterial outer membrane protein. 31. Gmelin W, Zeth K, Efremov R, Heberle J, Tittor J, Oesterhelt D: The crystal structure of the L1 intermediate of halorhodopsin at 1.9 angstroms resolution. Photochem Photobiol 2007, 83:369-377. Trapped low-temperature structure of an early halorhodopsin intermediate. 32. Edman K, Royant A, Nollert P, Maxwell CA, Pebay-Peyroula E, Navarro J, Neutze R, Landau EM: Early structural rearrangements in the photocycle of an integral membrane sensory receptor. Structure 2002, 10:473-482. 33. Moukhametzianov R, Klare JP, Efremov R, Baeken C, Goppner A, Labahn J, Engelhard M, Buldt G, Gordeliy VI: Development of the signal in sensory rhodopsin and its transfer to the cognate transducer. Nature 2006, 440:115-119. 34. Katona G, Snijder A, Gourdon P, Andreasson U, Hansson O, Andreasson LE, Neutze R: Conformational regulation of charge recombination reactions in a photosynthetic bacterial reaction center. Nat Struct Mol Biol 2005, 12:630-631. 35. Cherezov V, Rosenbaum DM, Hanson MA, Rasmussen SG, Thian FS, Kobilka TS, Choi HJ, Kuhn P, Weis WI, Kobilka BK et al.: High-resolution crystal structure of an engineered human b2adrenergic G protein-coupled receptor. Science 2007, 318:1258-1265. High-resolution LCP crystal structure of a GPCR. 36. Jaakola VP, Griffith MT, Hanson MA, Cherezov V, Chien EY, Lane JR, Ijzerman AP, Stevens RC: The 2.6 angstrom crystal structure of a human A2A adenosine receptor bound to an antagonist. Science 2008, 322:1211-1217. High-resolution LCP crystal structure of a GPCR. 37. Weis WI, Kobilka BK: Structural insights into G-protein-coupled receptor activation. Curr Opin Struct Biol 2008, 18:734-740. 38. Wo¨hri AB, Johansson LC, Wadsten-Hindrichsen P, Wahlgren WY, Fischer G, Horsefield R, Katona G, Nyblom M, Oberg F, Young G et al.: A lipidic-sponge phase screen for membrane protein crystallization. Structure 2008, 16:1003-1009. Development and validation of a lipidic sponge phase crystallization screen.
Current Opinion in Structural Biology 2009, 19:372–378
39. Cherezov V, Clogston J, Papiz MZ, Caffrey M: Room to move: crystallizing membrane proteins in swollen lipidic mesophases. J Mol Biol 2006, 357:1605-1618. 40. Cherezov V, Yamashita E, Liu W, Zhalnina M, Cramer WA, Caffrey M: In Meso structure of the cobalamin transporter, BtuB, at 1.95 A˚ resolution. J Mol Biol 2006, 364:716-734. 41. Nollert P: Lipidic cubic phases as matrices for membrane protein crystallization. Methods 2004, 34:348-353. 42. Hanson MA, Cherezov V, Griffith MT, Roth CB, Jaakola VP, Chien EY, Velasquez J, Kuhn P, Stevens RC: A specific cholesterol binding site is established by the 2.8 A˚ structure of the human b2-adrenergic receptor. Structure 2008, 16:897-905. High-resolution LCP crystal structure of a GPCR bound with bound ligand. 43. Luecke H, Schobert B, Stagno J, Imasheva ES, Wang JM, Balashov SP, Lanyi JK: Crystallographic structure of xanthorhodopsin, the light-driven proton pump with a dual chromophore. Proc Natl Acad Sci U S A 2008, 105:16561-16565. High-resolution bicelle crystal structure a bacterial rhodopsin with bound antenna. 44. Ujwal R, Cascio D, Colletier JP, Faham S, Zhang J, Toro L, Ping P, Abramson J: The crystal structure of mouse VDAC1 at 2.3 A˚ resolution reveals mechanistic insights into metabolite gating. Proc Natl Acad Sci U S A 2008, 105:17742-17747. High-resolution bicelle structure of a mitochondrial voltage sensitive anion channel. 45. Rasmussen SG, Choi HJ, Rosenbaum DM, Kobilka TS, Thian FS, Edwards PC, Burghammer M, Ratnala VR, Sanishvili R, Fischetti RF et al.: Crystal structure of the human b2 adrenergic G-protein-coupled receptor. Nature 2007, 450:383-387. Medium-resolution bicelle crystal structure of a GPCR. 46. Cherezov V, Peddi A, Muthusubramaniam L, Zheng YF, Caffrey M: A robotic system for crystallizing membrane and soluble proteins in lipidic mesophases. Acta Crystallogr D: Biol Crystallogr 2004, 60:1795-1807. 47. Ostermeier C, Michel H: Crystallization of membrane proteins. Curr Opin Struct Biol 1997, 7:697-701.
www.sciencedirect.com
Membrane protein crystallization from lipidic phases Linda C Johansson1, Annemarie B Wo¨hri2, Gergely Katona1, Sven Engstro¨m2 and Richard Neutze1 Membrane protein structural biology is enjoying a steady acceleration in the rate of success. Nevertheless, numerous membrane protein targets are resistant to the traditional approach of directly crystallizing detergent solubilized and purified protein and the ‘niche market’ of lipidic phase crystallization is emerging as a powerful complement. These approaches, including lipidic cubic phase, lipidic sponge phase, and bicelle crystallization methods, all immerse purified membrane protein within a lipid rich matrix before crystallization. This environment is hypothesized to contribute to the protein’s long-term structural stability and thereby favor crystallization. Spectacular recent successes include the highresolution structures of the b2-adrenergic G-protein-coupled receptor, the A2A adenosine G-protein-coupled receptor, and the mitochondrial voltage dependent anion channel. In combination with technical innovations aiming to popularize these methods, lipidic phase crystallization approaches can be expected to deliver an increasing scientific impact as the field develops. Addresses 1 Department of Chemistry, Biochemistry and Biophysics, University of Gothenburg, SE-405 30 Gothenburg, Sweden 2 Department of Chemical and Biological Engineering, Chalmers University of Technology, SE-41296 Gothenburg, Sweden Corresponding author: Neutze, Richard ([email protected])
Current Opinion in Structural Biology 2009, 19:372–378 This review comes from a themed issue on Membranes Edited by Declan Doyle and Graham Shipley Available online 4th July 2009 0959-440X/$ – see front matter # 2009 Elsevier Ltd. All rights reserved. DOI 10.1016/j.sbi.2009.05.006
Introduction Membrane proteins are fickle entities and repeatedly resist even the most determined efforts to overproduce, purify, and crystallize them for structural studies. Despite this, the field of membrane protein structural biology is undergoing a period of rapid expansion with accelerating rates of successful structural determination, in part due to the incorporation of technical innovations developed for high-throughput structural biology [1]. Detergent based crystallization protocols, whereby detergent solubilized and purified membrane protein is crystallized using the vapor-diffusion method essentially as if it were a soluble Current Opinion in Structural Biology 2009, 19:372–378
protein, are by far the most popular and successful approach developed to date [2]. Despite this, it has almost become a truism that the improvement of initial membrane protein crystals is painstakingly slow and frequently the quality of crystals can plateau at a point unsuitable for structural determination. In other cases no crystal leads whatsoever emerge despite success in recovering pure, stable protein. When faced with such disappointments, it is attractive to explore other crystallization protocols. A celebrated approach to improve the chances of recovering high quality crystals is to enhance the crystal contacts by enlarging the soluble domains through the addition of proteins specific antibodies [3], although this typically falls outside the capabilities of most laboratories. A variation of this theme is to increase the membrane protein’s soluble domains by protein engineering [4] but this may potentially introduce other complications. Another approach, the crystallization of membrane proteins in lipidic environments, explicitly recognizes and addresses the fact that membrane proteins are most stable in lipid bilayers. Detergent solubilization, the extraction of membrane proteins from their native membrane, often causes structural lipids and weakly bound subunits to be lost and potentially impairs protein integrity. As a consequence, membrane proteins frequently display reduced activity and poor stability in detergent solution and this role of lipids in membrane protein structural biology has been extensively discussed [5]. Alternatively, as originally conceived and demonstrated by Landau and Rosenbusch [6], purified membrane proteins can be reconstituted into lipidic bilayer environments before crystallization. By reintroducing proteins into a lipidic bilayer it was hypothesized that enhanced protein stability would be achieved, thus aiding the crystallization process. In this review we focus on recent progress using lipidic phase environments as vehicles aiding membrane protein crystallization. We first sketch the underlying ideas of lipidic phases and describe their application to the methods of lipidic cubic phase (LCP) crystallization [6], its offspring lipidic sponge phase (LSP) crystallization [7], and the conceptually related approach of bicelle crystallization [8]. We emphasize the recent structural results to emerge using each of these methods and highlight efforts to increase their popularity. In closing we speculate upon the future role likely to be played by lipidic phase crystallization protocols within the broader discipline of membrane protein structural biology. www.sciencedirect.com
Lipidic phase crystallisation Johansson et al. 373
Lipidic phases
Figure 1
Amphipathic lipid molecules consist of both hydrophobic and hydrophilic moieties. When exposed to aqueous environments they spontaneously form welldefined structures so as to maximize energetically favorable interactions of the hydrophilic head group with water and to minimize the exposure of hydrophobic tail group. The lipidic phases that emerge from the optimal enthalpy/entropy balance depend upon the specific geometry of the lipid itself, its concentration, its temperature, as well as the presence of other additives including ions and amphiphiles. The spontaneous formation of detergent micelles and unilamellar vesicles are used daily by membrane protein biochemists, but other less familiar lipidic aggregates are also easily accessed including reverse micelles and various other lipidic bilayer structures. Lipidic bilayer phases encompass vesicles, lamellar phases (stacked bilayers) as well as the bicontinuous cubic and sponge phases, which can be traversed in any direction along either hydrophilic or hydophobic paths. These mesophases [9] are more ordered than a liquid but less ordered than a solid. LCPs are semi-solid in texture and their long-range order can be visualized using small-angle X-ray scattering which reveals scattering peaks corresponding to the characteristic lattice parameters of their cubic packing [7]. Sponge phases, by comparison, can be thought of as swollen cubic phases with aqueous pores up to three times larger than those of a cubic phase [10] and lower long-range order. Another characteristic distinguishing the lipidic cubic and sponge phases is that the latter are liquid in nature. Lamellar phases also have low viscosity, but they are anisotropic and are therefore birefringent, whereas the isotropic cubic and sponge phases are non-birefringent. Thus cross-polarized microscopy provides a useful and rapid diagnostic distinguishing the lamellar and sponge phases. Bicelles can be thought of as solubilized lipidic bilayer disks. They are formed by the addition of detergent (or a short-chain lipid) to a long-chain lipid [8,11]. These mixtures spontaneously form disc-shaped aggregates of lipids and detergents, with the long-chain phospholipid forming a central planar bilayer which is surrounded by a rim of detergents that protect the bilayer from water. The physical diameter of the bicelle is controlled by the ratio of long-chain to short-chain amphiphiles. Moreover, membrane proteins reconstituted into bicelles maintain functionality [12].
Lipidic cubic phase crystallization Monoolein has been the lipid of choice for membrane protein crystallization from LCPs [6], although some successes have also been reported using closely related lipids such as monovaccenin [13]. When mixed with water, monoolein spontaneously swells to form several www.sciencedirect.com
Phase diagram of the PEG:monoolein:water system illustrating the presence of LCP (dark blue), LSP (light blue) and lamellar (gray) phases. Sponge phases do not form in the absence of PEG (triangle base) but require approximately 30% water content (dashed line). The PEG (25– 60%) and monoolein (10–45%) concentrations can vary over a broader range [10]. Phase diagrams recovered from PEGs of variable length are rather similar.
mesophases. At room temperature the Pn3m cubic phase is recovered at approximately 40% water concentration with the (larger cell) Ia3d cubic phase emerging at slightly lower water content [9]. The traditional starting point for any LCP crystallization experiment has been to incorporate membrane protein into either of these semi-solid cubic phases by mixing a buffer containing purified protein with monoolein in an approximate ratio of 2:3 by volume. As with any other batch crystallization experiment, this protein-containing LCP is then dispensed into tubes [6] or crystallization plates [14], overlaid with a crystallization agent and sealed. Dehydration of the cubic phase drives a phase-transition to a lamellar phase (Figure 1) and it is believed that this phase-transition is a key ingredient in successfully recovering crystals, a hypothesis which receives support from the observation of birefringence in the immediate vicinity of crystals grown in the LCP [15]. Bacteriorhodopsin was the first membrane protein to be successfully crystallized using LCP crystallization [6] and improvements in the crystal quality shortly later yielded high-resolution crystal structures of bacteriorhodopsin in its resting conformation [16–18]. LCP crystals of bacteriorhodopsin grow as stacked layers of 2D crystals and are functionally active, such that these crystals continue to diffract after illumination. Thus several conformational changes associated with the photo-cycle of bacteriorhodopsin have been convincingly demonstrated using light illuminated LCP crystals [19–23] and have contributed to Current Opinion in Structural Biology 2009, 19:372–378
374 Membranes
a uniquely detailed picture of the structural mechanism of a membrane protein in action [24].
Figure 2
Three other archaeal rhodopsins have also been successfully crystallized using the LCP technique and highresolution crystal structures of halorhodopsin [25], N. pharaonis sensory rhodopsin II alone [26] and in complex with its integral membrane transducer protein [13], and Anabaena sensory rhodopsin [27] all emerged using this technique. Two criticisms, that LCP crystallization was a niche technique successful for the archaeal rhodopsins but unsuitable for the crystallization of large membrane protein complexes or colorless proteins, were countered by the successful LCP crystallization of the R. sphaeroides reaction center complex [28], the colorless cobalamin transporter BtuB [29] and outer membrane adhension protein OpcA [30], respectively. Moreover, as with bacteriorhodopsin, the LCP grown crystals of halorhodopsin [31], sensory rhodopsin II [32], sensory rhodopsin II in complex with its transducer [33], and the photosynthetic reaction center [34] are all functionally active and have yielded convincing difference electron density changes following illumination. To the wider community, however, the scientific impact of the above successes has recently been dwarfed by the LCP crystallization and structure determination to 2.4 A˚ resolution of a modified form of the human b2-adrenergic G-protein-coupled receptor [35] and the 2.6-A˚ resolution structure of the human A2A adenosine G-proteincoupled receptor [36]. These high-resolution structures of ligand-activated G-protein-coupled receptors have generated considerable excitement within the structural biology community [37]. Moreover, the challenge associated with crystallizing highly flexible membrane proteins is widely appreciated, and crystal structures of G-proteincoupled receptors have become something of a Holy Grail within the field. Thus these structures provide dramatic support for the original hypothesis of Landau and Rosenbusch [6] that by maintaining membrane proteins in a lipid-bilayer environment during crystallization, protein stability will be improved and enhance the chances of success.
Lipidic sponge phase crystallization Membrane protein crystallization in LSP emerged as a serendipitous offshoot of the LCP technique. Visual inspection of the LCP crystallization setups of the R. sphaeroides reaction center [28] indicated that the semisolid LCP liquefied before crystal formation, and smallangle X-ray scattering studies revealed that this liquefied cubic phase was in fact a LSP [7]. Since it is easier to pipette a liquid rather than a semi-solid phase, we sought to grow crystals of the reaction center using a traditional hanging-drop vapor-diffusion experiment with the liquid sponge phase added directly to a drop containing concentrated protein. This approach was successful without Current Opinion in Structural Biology 2009, 19:372–378
Crystals of the R. sphaeroides reaction center grown in (a) the LCP and (b) close-up of setup. (c) Hanging-drop crystallization experiment for the R. sphaeroides reaction center grown in a LSP and (d) close-up view of a crystal recovered in the drop illustrated in (c). Although the morphology of the two crystal forms appears to be quite different, these crystals are isomorphous. (e) Hanging-drop crystallization experiment for the reaction center of Bl. viridis grown in a LSP and (f) close-up view of a crystal recovered in the drop illustrated in (e).
any significant change to the crystallization conditions [7] and larger crystals with improved diffraction were recovered from optimized LSP setups. Curiously, while their morphology appeared to be different to that observed when using the LCP setups (Figure 2), both crystal forms were isomorphous. In the LSP crystallization of the R. sphaeroides reaction center, Jeffamine M-600 was the reagent driving the formation of the monoolein sponge phase. Jeffamine, however, has not proven to be a common reagent used during the successful crystallization of membrane proteins [2], whereas polyethyleneglycols of various lengths (PEGs) have yielded the largest number of successes [2]. Somewhat fortuitously, monoolein will also spontaneously form a stable sponge phase when mixed with PEGs at concentrations close to those successful for membrane protein crystallization [10] (Figure 1). To further explore the possibility of crystallizing membrane proteins within a LSP environment, we developed a 48condition LSP screen with a strong bias toward PEG www.sciencedirect.com
Lipidic phase crystallisation Johansson et al. 375
conditions and a small number of Jeffamine M-600 conditions. In our hands, using hanging-drop vapor-diffusion experiments, this screen yielded crystallization leads for eight membrane proteins [38] and the LSP structure of the Bl. viridis reaction center to 1.85 A˚ has been determined (Wo¨hri et al., under review). In an independent development, Cherezov et al. crystallized a bacterial light harvesting complex II using LCPs swollen by the addition of 20% PPO to form a LSP [39]. Similarly, an earlier LCP crystal form of the bacterial cobalamin transporter BtuB [29] was improved by swelling the LCP by the addition of 10–12% MPD to form an LSP during crystallization [40]. Thus LSP structures of four bacterial membrane proteins have been solved to date (Table 1, Figure 3). Although the pioneering saturated-salt LCP crystallization of bacteriorhodopsin [6] did not proceed via the sponge phase, another successful condition containing 27% PEG 2000 has been reported [41] which lies near the sweet spot for sponge phase formation [10] (Figure 1). Likewise, the LCP crystallization conditions of the b2adrenergic G-protein-coupled receptor (30–35% PEG 400 in [35] and 28% PEG 400 in [42]), the A2A adenosine G-protein-coupled receptor (28–32% PEG 400 in [36]), and the outer membrane protein OpcA [30] (18% PEG 400) all appear tantalizingly close to that required to form a LSP. It is possible that one or more of these LCP crystallization setups may have passed via the sponge phase before crystal growth, in analogy with that observed during the LCP crystallization of the R. sphaeroides reaction center [7], and it would be of interest to consider the possibility of reproducing these crystals starting directly from the LSP.
Lipidic bicelle crystallization This crystallization method was conceived by Faham and Bowie [8] as a compromise between the detergent solubilized crystallization of membrane proteins and the LCP crystallization approach, elegantly unifying the convenience of working with detergents with the potential benefits of reconstituting membrane proteins into lipidic bilayers before crystallization. Bicelles formed by mixing the long-tail lipids dimyristoyl phosphatidylcholine (DMPC) or ditridecanoyl phosphatidylcholine (DTPC) with the detergent CHAPSO or nonyl maltoside have proven to be successful media for aiding the crystallization of membrane proteins, with a final lipid content of approximately 8% in the crystallization drop. In practical terms, bicelle crystallization proceeds as with any detergent crystallization protocol with an additional step of mixing purified and concentrated membrane protein (on ice) with bicelle formulations before setting up drops. Thus any commercially available or home-grown screen is compatible with the bicelle crystallization approach and the frequently arduous process of crystal optimization proceeds exactly as with the more familiar detergent based protocols. As with the LCP crystallization, bacteriorhodopsin provided the first example of a successful bicelle crystallization experiment [8], with other crystal forms later being found under different bicelle crystallization conditions [11]. The history of success for bicelles crystallization has also followed a path uncannily similar to that of the LCP technique, with three recent major breakthroughs. Xanorhodopsin, a member of the bacterial rhodopsin family but containing two pigments, was crystallized and its structure solved to 1.9 A˚ resolution [43] using bicelle crystallization; the voltage gated
Table 1 Summary of deposited structures solved using LCP, LSP, and bicelle crystallization Protein Bacteriorhodopsin Halorhodopsin Sensory rhodopsin II Sensory rhodopsin II-transducer complex Reaction center Anabaena Sensory Rhodopsin Engineered human b2-adrenergic receptor OpcA outer membrane adhesin Human A2A adenosine receptor Reaction center Light harvesting complex II BtuB Reaction center Bacteriorhodopsin b2-Adrenergic G-protein-coupled receptor Voltage-dependent anion channel Xantorhodopsin a b
Resolution (A˚)
Source
PDB entry
Method
Year
First lipidic phase structure
2.5 1.8 2.1 1.94 2.35 2.0 2.4 1.95 2.6 2.2 2.45 1.95 1.95 2.0 3.4/3.7 2.3 1.9
H. salinarum H. salinarum N. pharaonis N. pharaonis R. sphaeroides Anabaena Homo sapiens N. meningitidis Homo sapiens R. sphaeroides Rps. acidophila E. coli Bl. viridis H. salinarum Homo sapiens Mus musculus S. ruber
1AP9 1E12 1H68 1H2S 1OGV 1XIO 2RH1 2VDF 3EML 2GNU 2FKW 2GUF 2WJM 1KME 2R4R 3EMN 3DDL
LCP LCP LCP LCP LCP a LCP LCP a LCP a LCP a LSP LSP LSP LSP Bicelle Bicelle Bicelle Bicelle
1997 2000 2001 2002 2003 2004 2007 2007 2008 2006 2006 2006 2009 2002 2007 2008 2008
Pebay-Peyroula et al. Kolbe et al. Royant et al. Gordeliy et al. Katona et al. Vogeley et al. Cherezov et al. Cherezov et al. Jaakola et al. Wadsten et al. Cherezov et al. Cherezov et al. Wo¨hri et al. b Faham et al. Rasmussen et al. Ujwal et al. Luecke et al.
Originally reported as LCP crystallization but likely to have proceeded via LSP. Under review .
www.sciencedirect.com
Current Opinion in Structural Biology 2009, 19:372–378
376 Membranes
Figure 3
Growth in the number of deposited membrane protein structures solved using lipidic phase crystallization methods. Structures solved using LCP crystallization are shown in dark blue; structures solved using LSP crystallization are show in light blue; and structures solved using bicelle crystallization are shown in dark green. The LCP and bicelle structures of bacteriorhodopsin, and the LCP and LSP structures of the R. sphaeroides reaction center, are all counted. Mutant structures or alternative conformations of a previously counted structure are not included within this summary. Relevant citations and PDB entries are given in Table 1.
anion channel (VDAC1) a regulated eukaryotic b-barrel channel, was solved to 2.3 A˚ resolution using this method [44]; and the structure of human b2-adrenergic G-protein-coupled receptor in complex with a Fab fragment was solved to 3.4 A˚ resolution using bicelle crystallization [45]. As with similar recent successes using LCP crystallization, these recent outstanding structural results have greatly raised the profile of the bicelle crystallization method within the structural biology community.
Popularization Thirteen years have passed since the first demonstration of membrane protein crystallization within a lipidic phase environment [6]. A number of practical disadvantages associated with crystallization using the semi-solid LCP have been addressed by moving to submicroliter crystallization drops [14] and developing a crystallization plate and an automated dispensing device suited to LCP setups [46]. Moreover, a LCP start-up kit and crystallization screen have recently been marketed by Jena Bioscience (http://www.jenabioscience.com/) and Emerald BioSystems (http://www.emeraldbiosystems.com/), a LSP crystallization screen has been marketed by Molecular Dimensions Ltd. (http://www.moleculardimensions.com/), Qiagen (http://www1.qiagen.com/) have developed a crystallization plate with a thin layer of monoolein pre-dispersed, and a bicelle kit containing lipids and detergents is marketed by Anatrace (http://www. anatrace.com/). Current Opinion in Structural Biology 2009, 19:372–378
Adopting any new practice necessarily requires that a learning curve be surmounted, yet vapor diffusion crystallization experiments with lipidic sponge and bicelle phases proceed almost exactly as with standard detergent crystallization protocols once the LSP or bicelle preparations have been mastered. Moreover, the results from crystallization drops are as easily interpreted (Figure 2c–f) and the techniques are amenable to automation using standard crystallization robots. Thus both methods could easily be applied even at an early stage in a crystallization project in parallel with the more widespread initial crystal screening of detergent solubilized membrane protein.
Conclusions Figure 3 illustrates the growth of membrane protein structures solved using LCP, LSP, and bicelle crystallization techniques. This list of membrane protein structures (Table 1) includes 7 TM bacterial (rhodopsins) and eukaryotic (GPCRs) membrane proteins; 9, 11, and 18 TM membrane protein complexes (SRII:Htr, reaction centers, and LH II complexes); and 10, 19, and 22 stranded b-barrel structure (OpcA, VDAC1, and BtuB). All crystal forms grow as stacked layers of 2D crystals and frequently display significant protein:protein interactions within the plan of the membrane, characteristic of Type I membrane protein crystals [47]. The last two years mark a golden period for the development and wider acceptance of lipidic phase crystallization techniques. Nothing speaks louder to the www.sciencedirect.com
Lipidic phase crystallisation Johansson et al. 377
scientific community than high-resolution structures of new eukaryotic membrane proteins and, with several recent high-profile successes [35,36,44,45], these methods must now be said to have come of age. Consequently, we anticipate a more widespread pursuit of these approaches and look forward to further novel structural insights into fold and function emerging from the crystallization of membrane proteins in lipidic phases.
Conflict of interest
13. Gordeliy VI, Labahn J, Moukhametzianov R, Efremov R, Granzin J, Schlesinger R, Buldt G, Savopol T, Scheidig AJ, Klare JP et al.: Molecular basis of transmembrane signalling by sensory rhodopsin II-transducer complex. Nature 2002, 419:484-487. 14. Nollert P, Navarro J, Landau EM: Crystallization of membrane proteins in cubo. Methods Enzymol 2002, 343:183-199. 15. Nollert P, Qiu H, Caffrey M, Rosenbusch JP, Landau EM: Molecular mechanism for the crystallization of bacteriorhodopsin in lipidic cubic phases. FEBS Lett 2001, 504:179-186.
The authors declare no conflict of interest arising from this work.
16. Pebay-Peyroula E, Rummel G, Rosenbusch JP, Landau EM: X-ray structure of bacteriorhodopsin at 2.5 A˚ from microcrystals grown in lipidic cubic phases. Science 1997, 277:1676-1681.
Acknowledgements
17. Belrhali H, Nollert P, Royant A, Menzel C, Rosenbusch JP, Landau EM, Pebay-Peyroula E: Protein, lipid and water organization in bacteriorhodopsin crystals: a molecular view of the purple membrane at 1.9 A˚ resolution. Structure 1999, 7:909-917.
This work was supported by the Swedish Science Research Council (VR), the European Commission (E-MEP).
References and recommended reading Papers of particular interest published within the period of review have been highlighted as: of special interest of outstanding interest 1.
Carpenter EP, Beis K, Cameron AD, Iwata S: Overcoming the challenges of membrane protein crystallography. Curr Opin Struct Biol 2008, 18:581-586.
18. Luecke H, Schobert B, Richter HT, Cartailler JP, Lanyi JK: Structure of bacteriorhodopsin at 1.55 A˚ resolution. J Mol Biol 1999, 291:899-911. 19. Edman K, Nollert P, Royant A, Belrhali H, Pebay-Peyroula E, Hajdu J, Neutze R, Landau EM: High-resolution X-ray structure of an early intermediate in the bacteriorhodopsin photocycle. Nature 1999, 401:822-826. 20. Luecke H, Schobert B, Richter HT, Cartailler JP, Lanyi JK: Structural changes in bacteriorhodopsin during ion transport at 2 A˚ resolution. Science 1999, 286:255-261.
2. Newstead S, Ferrandon S, Iwata S: Rationalizing alpha-helical membrane protein crystallization. Protein Sci 2008, 17:466-472. Comprehensive summary of successful membrane protein crystallization conditions.
21. Royant A, Edman K, Ursby T, Pebay-Peyroula E, Landau EM, Neutze R: Helix deformation is coupled to vectorial proton transport in the photocycle of bacteriorhodopsin. Nature 2000, 406:645-648.
3.
22. Luecke H, Schobert B, Cartailler JP, Richter HT, Rosengarth A, Needleman R, Lanyi JK: Coupling photoisomerization of retinal to directional transport in bacteriorhodopsin. J Mol Biol 2000, 300:1237-1255.
Hunte C, Michel H: Crystallization of membrane proteins mediated by antibody fragments. Curr Opin Struct Biol 2002, 12:503-508.
4.
Rosenbaum DM, Cherezov V, Hanson MA, Rasmussen SG, Thian FS, Kobilka TS, Choi HJ, Yao XJ, Weis WI, Stevens RC et al.: GPCR engineering yields high-resolution structural insights into b2-adrenergic receptor function. Science 2007, 318:12661273. High-resolution LCP crystal structure of a GPCR. 5.
Hunte C, Richers S: Lipids and membrane protein structures. Curr Opin Struct Biol 2008, 18:406-411.
6.
Landau EM, Rosenbusch JP: Lipidic cubic phases: a novel concept for the crystallization of membrane proteins. Proc Natl Acad Sci U S A 1996, 93:14532-14535.
7.
8.
9.
Wadsten P, Wo¨hri AB, Snijder A, Katona G, Gardiner AT, Cogdell RJ, Neutze R, Engstro¨m S: Lipidic sponge phase crystallization of membrane proteins. J Mol Biol 2006, 364:44-53. Faham S, Bowie JU: Bicelle crystallization: a new method for crystallizing membrane proteins yields a monomeric bacteriorhodopsin structure. J Mol Biol 2002, 316:1-6. Caffrey M: A lipid’s eye view of membrane protein crystallization in mesophases. Curr Opin Struct Biol 2000, 10:486-497.
10. Bender J, Jarvoll P, Nyden M, Engstrom S: Structure and dynamics of a sponge phase in the methyl deltaaminolevulinate/monoolein/water/propylene glycol system. J Colloid Interface Sci 2008, 317:577-584. 11. Faham S, Boulting GL, Massey EA, Yohannan S, Yang D, Bowie JU: Crystallization of bacteriorhodopsin from bicelle formulations at room temperature. Protein Sci 2005, 14:836-840. 12. Czerski L, Sanders CR: Functionality of a membrane protein in bicelles. Anal Biochem 2000, 284:327-333. www.sciencedirect.com
23. Sass HJ, Buldt G, Gessenich R, Hehn D, Neff D, Schlesinger R, Berendzen J, Ormos P: Structural alterations for proton translocation in the M state of wild-type bacteriorhodopsin. Nature 2000, 406:649-653. 24. Neutze R, Pebay-Peyroula E, Edman K, Royant A, Navarro J, Landau EM: Bacteriorhodopsin: a high-resolution structural view of vectorial proton transport. Biochim Biophys Acta 2002, 1565:144-167. 25. Kolbe M, Besir H, Essen LO, Oesterhelt D: Structure of the lightdriven chloride pump halorhodopsin at 1.8 A˚ resolution. Science 2000, 288:1390-1396. 26. Royant A, Nollert P, Edman K, Neutze R, Landau EM, PebayPeyroula E, Navarro J: X-ray structure of sensory rhodopsin II at 2.1 A˚ resolution. Proc Natl Acad Sci U S A 2001, 98:10131-10136. 27. Vogeley L, Sineshchekov OA, Trivedi VD, Sasaki J, Spudich JL, Luecke H: Anabaena sensory rhodopsin: a photochromic color sensor at 2.0 A˚. Science 2004, 306:1390-1393. 28. Katona G, Andreasson U, Landau EM, Andreasson LE, Neutze R: Lipidic cubic phase crystal structure of the photosynthetic reaction centre from Rhodobacter sphaeroides at 2.35 A˚ resolution. J Mol Biol 2003, 331:681-692. 29. Misquitta LV, Misquitta Y, Cherezov V, Slattery O, Mohan JM, Hart D, Zhalnina M, Cramer WA, Caffrey M: Membrane protein crystallization in lipidic mesophases with tailored bilayers. Structure 2004, 12:2113-2124. 30. Cherezov V, Liu W, Derrick JP, Luan B, Aksimentiev A, Katritch V, Caffrey M: In meso crystal structure and docking simulations suggest an alternative proteoglycan binding site in the OpcA outer membrane adhesin. Proteins 2008, 71:24-34. Current Opinion in Structural Biology 2009, 19:372–378
378 Membranes
LSP structure of a bacterial outer membrane protein. 31. Gmelin W, Zeth K, Efremov R, Heberle J, Tittor J, Oesterhelt D: The crystal structure of the L1 intermediate of halorhodopsin at 1.9 angstroms resolution. Photochem Photobiol 2007, 83:369-377. Trapped low-temperature structure of an early halorhodopsin intermediate. 32. Edman K, Royant A, Nollert P, Maxwell CA, Pebay-Peyroula E, Navarro J, Neutze R, Landau EM: Early structural rearrangements in the photocycle of an integral membrane sensory receptor. Structure 2002, 10:473-482. 33. Moukhametzianov R, Klare JP, Efremov R, Baeken C, Goppner A, Labahn J, Engelhard M, Buldt G, Gordeliy VI: Development of the signal in sensory rhodopsin and its transfer to the cognate transducer. Nature 2006, 440:115-119. 34. Katona G, Snijder A, Gourdon P, Andreasson U, Hansson O, Andreasson LE, Neutze R: Conformational regulation of charge recombination reactions in a photosynthetic bacterial reaction center. Nat Struct Mol Biol 2005, 12:630-631. 35. Cherezov V, Rosenbaum DM, Hanson MA, Rasmussen SG, Thian FS, Kobilka TS, Choi HJ, Kuhn P, Weis WI, Kobilka BK et al.: High-resolution crystal structure of an engineered human b2adrenergic G protein-coupled receptor. Science 2007, 318:1258-1265. High-resolution LCP crystal structure of a GPCR. 36. Jaakola VP, Griffith MT, Hanson MA, Cherezov V, Chien EY, Lane JR, Ijzerman AP, Stevens RC: The 2.6 angstrom crystal structure of a human A2A adenosine receptor bound to an antagonist. Science 2008, 322:1211-1217. High-resolution LCP crystal structure of a GPCR. 37. Weis WI, Kobilka BK: Structural insights into G-protein-coupled receptor activation. Curr Opin Struct Biol 2008, 18:734-740. 38. Wo¨hri AB, Johansson LC, Wadsten-Hindrichsen P, Wahlgren WY, Fischer G, Horsefield R, Katona G, Nyblom M, Oberg F, Young G et al.: A lipidic-sponge phase screen for membrane protein crystallization. Structure 2008, 16:1003-1009. Development and validation of a lipidic sponge phase crystallization screen.
Current Opinion in Structural Biology 2009, 19:372–378
39. Cherezov V, Clogston J, Papiz MZ, Caffrey M: Room to move: crystallizing membrane proteins in swollen lipidic mesophases. J Mol Biol 2006, 357:1605-1618. 40. Cherezov V, Yamashita E, Liu W, Zhalnina M, Cramer WA, Caffrey M: In Meso structure of the cobalamin transporter, BtuB, at 1.95 A˚ resolution. J Mol Biol 2006, 364:716-734. 41. Nollert P: Lipidic cubic phases as matrices for membrane protein crystallization. Methods 2004, 34:348-353. 42. Hanson MA, Cherezov V, Griffith MT, Roth CB, Jaakola VP, Chien EY, Velasquez J, Kuhn P, Stevens RC: A specific cholesterol binding site is established by the 2.8 A˚ structure of the human b2-adrenergic receptor. Structure 2008, 16:897-905. High-resolution LCP crystal structure of a GPCR bound with bound ligand. 43. Luecke H, Schobert B, Stagno J, Imasheva ES, Wang JM, Balashov SP, Lanyi JK: Crystallographic structure of xanthorhodopsin, the light-driven proton pump with a dual chromophore. Proc Natl Acad Sci U S A 2008, 105:16561-16565. High-resolution bicelle crystal structure a bacterial rhodopsin with bound antenna. 44. Ujwal R, Cascio D, Colletier JP, Faham S, Zhang J, Toro L, Ping P, Abramson J: The crystal structure of mouse VDAC1 at 2.3 A˚ resolution reveals mechanistic insights into metabolite gating. Proc Natl Acad Sci U S A 2008, 105:17742-17747. High-resolution bicelle structure of a mitochondrial voltage sensitive anion channel. 45. Rasmussen SG, Choi HJ, Rosenbaum DM, Kobilka TS, Thian FS, Edwards PC, Burghammer M, Ratnala VR, Sanishvili R, Fischetti RF et al.: Crystal structure of the human b2 adrenergic G-protein-coupled receptor. Nature 2007, 450:383-387. Medium-resolution bicelle crystal structure of a GPCR. 46. Cherezov V, Peddi A, Muthusubramaniam L, Zheng YF, Caffrey M: A robotic system for crystallizing membrane and soluble proteins in lipidic mesophases. Acta Crystallogr D: Biol Crystallogr 2004, 60:1795-1807. 47. Ostermeier C, Michel H: Crystallization of membrane proteins. Curr Opin Struct Biol 1997, 7:697-701.
www.sciencedirect.com