Membrane Proteins 753 ISSN 0100-879X Review

Home , Membrane protein

Brazilian Journal of Medical and Biological Research (2002) 35: 753-766 Reconstitution of membrane proteins 753 ISSN 0100-879X Review

Membrane proteins: functional and structural studies using reconstituted proteoliposomes and 2-D crystals

J.-L. Rigaud Institut Curie, UMR-CNRS 168 and LRC-CEA 8, Paris, France

Abstract

Correspondence Reconstitution of membrane proteins into lipid bilayers is a powerful Key words J.-L. Rigaud tool to analyze functional as well as structural areas of membrane • Membrane proteins Institut Curie protein research. First, the proper incorporation of a purified mem- • Proteoliposomes UMR-CNRS 168 and LRC-CEA 8 brane protein into closed lipid vesicles, to produce proteoliposomes, • 2-D crystals 11 Rue Pierre et Marie Curie, 75231 allows the investigation of transport and/or catalytic properties of any • Electron crystallography Paris, Cedex 05 • Reconstitution France membrane protein without interference by other membrane compo- E-mail: [email protected] nents. Second, the incorporation of a large amount of membrane proteins into lipid bilayers to grow crystals confined to two dimen- Presented at the XXX Annual sions has recently opened a new way to solve their structure at high Meeting of the Brazilian Society resolution using electron crystallography. However, reconstitution of of Biochemistry and Molecular membrane proteins into functional proteoliposomes or 2-D crystalli- Biology, Caxambu, MG, Brazil, zation has been an empirical domain, which has been viewed for a May 19-22, 2001. long time more like “black magic” than science. Nevertheless, in the

Research supported in part by last ten years, important progress has been made in acquiring knowl- CNRS (France)-CNPq (Brazil) edge of lipid-protein-detergent interactions and has permitted to build collaborative programs. upon a set of basic principles that has limited the empirical approach of reconstitution experiments. Reconstitution strategies have been improved and new strategies have been developed, facilitating the success rate of proteoliposome formation and 2-D crystallization. This Received January 14, 2002 review deals with the various strategies available to obtain proteolipo- Accepted May 7, 2002 somes and 2-D crystals from detergent-solubilized proteins. It gives an overview of the methods that have been applied, which may be of help for reconstituting more proteins into lipid bilayers in a form suitable for functional studies at the molecular level and for high-resolution structural analysis.

Introduction biological membranes makes it difficult to study these membrane proteins in situ. There- Screening of the genomes of different fore, purification from the native membrane organisms has demonstrated that about 25% and further reincorporation of the purified of the sequenced genes code for strongly membrane protein into an artificial mem- hydrophobic proteins, which are integrated brane to form proteoliposomes or to produce into cell membranes (1). These results have two-dimensional (2-D) crystals continue to emphasized the importance of membrane be a crucial step in studying the function and proteins in many biological processes essen- structure of these molecules. tial for life. However, the complexity of most First of all, membrane protein reconstitu-

Braz J Med Biol Res 35(7) 2002 754 J.-L. Rigaud

tion into liposomes has played an important on the one hand, and electron crystallogra- role and should continue to be a powerful phy on the other. Thus, one of the limiting tool that can be used to identify and charac- factors in obtaining molecular information terize the mechanisms of action of mem- was related to the lack of reproducible meth- brane proteins. The necessity for reconstitu- ods of reconstitution. Therefore enormous tion is due to the fact that many membrane efforts were required to understand the proteins express their full activity only when mechanisms of reconstitution and for new correctly oriented and inserted into a lipid approaches to be evaluated, refined and ap- bilayer. In particular, reconstitution has plied to available proteins in order to make played a central role in the identification and reconstitution even more important as a tool characterization of the mechanisms of ac- for establishing structure-function relation- tion of membrane proteins with a vectorial ships at the molecular level. transport function. More generally, biochemi- This review will deal with the different cal and biophysical approaches have provid- strategies commonly used to reconstitute pro- ed important information about lipid-protein teoliposomes and 2-D crystals. It will also and protein-protein interactions as well as focus on recent new approaches, which have topological and topographic features of dif- led to promising perspectives for the produc- ferent classes of membrane proteins recon- tion of highly functional proteoliposomes stituted into proteoliposomes (2-5). and more 2-D crystals of membrane proteins Second, structure determination at high amenable to structural analysis. General resolution is actually a difficult challenge for guidelines and rules will be proposed in this membrane proteins and the number of mem- field which has long been seen more as art brane proteins that have been crystallized in than as science. three dimensions (3-D) is still small and far behind that of soluble proteins (6). Reconsti- Reconstitution of proteoliposomes tution of membrane proteins into artificial membranes to form 2-D crystals has opened Strategies a new way to solve their structures inside a native-like environment (7-10). Indeed, elec- From the analysis of the abundant litera- tron crystallography of 2-D crystals has de- ture concerning the insertion of membrane veloped to the point that significant struc- proteins into liposomes, four basic strategies tural atomic models have been built and that have been outlined: mechanical means, many membrane protein structures have been freeze-thaw, organic solvents, and detergents. evaluated at resolutions allowing the sec- Although these reconstitution strategies have ondary structure to be assessed (11). These proven to be very useful to prepare pure recent results have made electron crystallog- phospholipid vesicles (12), the additional raphy an excellent alternative and a comple- insertion of a membrane protein during the mentary discipline to X-ray crystallography reconstitution process has imposed many in membrane protein structural biology. constraints that have seriously hampered the However, the ability to investigate func- applicability of phospholipid vesicles to pro- tional aspects or to solve the structures of teoliposome reconstitution. Indeed, besides membrane proteins through the use of recon- the need for conditions that preserve protein stituted systems has been limited by the fact activity, many criteria have to be considered that methods for producing high quality pro- to fully optimize the reconstitution of a mem- teoliposomes or 2-D crystals have not ad- brane protein: the homogeneity of protein vanced at the same rate as biochemical, bio- insertion, the final orientation of the protein, physical and molecular biology techniques, the morphology and the size of the reconsti-

Braz J Med Biol Res 35(7) 2002 Reconstitution of membrane proteins 755 tuted proteoliposomes, as well as their re- formation as well as understanding of the sidual permeability. Even more limiting is physical behavior of lipid-detergent systems the fact that many methods for preparing permitted the construction of a set of basic pure liposomes have been difficult to apply principles that has limited the number of successfully to proteoliposome reconstitu- experimental variables and thus the empiri- tion because most membrane proteins are cal approach to proteoliposome reconstitu- purified by the use of detergents, which in- tion (5,14,15). terfere with the process of vesicle formation. To allow realistic monitoring of the Thus, the vast majority of membrane pro- mechanisms by which proteins may associ- tein reconstitution procedures are those in- ate with lipids during detergent-mediated volving the use of detergents. Indeed, due to reconstitutions, we have developed a strat- their amphiphilic character, most membrane egy (Figure 2) based on the idea that the proteins require detergents, not only as a process of reconstitution by detergent re- means of disintegrating the structure of na- moval is the reverse of the process of solubi- tive membranes in the initial step of their lization by increasing amounts of detergent solubilization, but also as a means of keep- (5,16). To this end, the first stage in our ing the protein in a non-denaturing environ- strategy was to add detergent to preformed ment during further purification (13). The liposomes using the entire range of detergent standard procedure used to reconstitute a concentrations that cause the transformation purified membrane protein involves co- of lamellar structures to mixed micelles (17- micellization of the protein in an excess of 19). The lipid-detergent mixed amphiphilic phospholipid and appropriate detergent, to structures that are formed during the solubi- form a solution of mixed lipid-protein-deter- lization process are similar to those pre- gent and lipid-detergent micelles. Next, the detergent is removed from the micellar solutions, resulting in the progressive formation of closed lipid bilayers into which the pro- Native membrane teins eventually incorporate. All the deter- Solubilization gent-mediated reconstitutions described in the literature rely on this standard procedure Detergent but differ essentially in the techniques used Lipid Protein to remove the detergent, which can be dialy- Purification sis, gel chromatography, dilution, or hydrophobic adsorption onto polystyrene beads

(Figure 1). Dialysis As shown by the abundant literature, re- Proteoliposome constitutions from detergent micellar mix- Gel chromatography tures have yielded proteoliposomes of dif- Reconstitution Dilution ferent sizes and compositions depending on the nature of the detergent, the particular Polystyrene procedure to remove it, as well as the nature beads of the protein and the lipid composition. 2-D crystals Therefore, each membrane protein responded Figure 1. Detergent-mediated reconstitutions. Most membrane proteins are extracted from differently to the various reconstitution pro- native membranes with solubilizing detergent concentrations. After purification, the solubi- cedures and the approach has long been lized protein is supplemented with an excess of lipids and detergents, leading to a solution of mixed lipid-protein-detergent and lipid-detergent micelles. For proteoliposome or 2-D entirely empirical. In the 90’s, important crystal reconstitution, detergent is removed from these micellar solutions using different knowledge of the mechanisms of liposome strategies.

Braz J Med Biol Res 35(7) 2002 756 J.-L. Rigaud

dicted to form during the micelle-to-lipid gent was removed and the resulting proteo- bilayer transition in a reconstitution process liposomes were characterized in terms of by detergent removal. Thus, in a second protein incorporation and orientation, vesicle stage, adding the protein at each step of the size distribution, and biological activity. solubilization process allowed a “snap-shot” With respect to optimal protein incorpo- of all the situations that might occur in a ration, we have identified three mechanisms reconstitution experiment following deter- by which membrane proteins can associate gent removal from lipid-protein-detergent with lipids to give proteoliposomes (Figure micelles. Finally, in a third stage, the deter- 3). Depending on the nature of the detergent, proteins can be either directly incorporated into detergent-saturated liposomes (octylglu- Step-by-step solubilization coside- and dodecylmaltoside-mediated reconstitutions), transferred from mixed mi- Preformed liposomes celles to detergent-saturated liposomes (Tri- ton X-100-mediated reconstitutions) or par- ticipate in proteoliposome formation during Increasing detergent concentration the micellar-to-lamellar transition (cholate,

chaps, and C12E8-mediated reconstitutions). In addition to providing original information about the mechanisms of lipid-protein association in the presence of detergent, this “step-by-step” reconstitution strategy proved to be a powerful reconstitution procedure, more suitable than the usual methods. This new reconstitution strategy, which

Rsat Rsol is now widely used, has produced proteoliposomes which sustain biological activi- ties comparable to those measured in the Protein addition native membrane. Indeed, it has allowed an easy determination of the optimal conditions under which a protein can associate with lipids in the presence of detergent. In particular, it should be stressed that reconstitution of a membrane protein into detergent- saturated preformed liposomes ensured a Detergent removal unidirectional insertion of the protein in the

Bio-Beads membrane and generally produced the most efficient proteoliposomes (20-22). An additional advantage of our new reconstitution strategy relied on the use of Bio-Beads SM2 Proteoliposomes to remove the detergent. Indeed, the efficient removal of detergent from reconstituted pro- Figure 2. The “step-by-step” reconstitution strategy. The standard procedure for reconstituting membrane proteins into proteoliposomes consists of three stages. 1, Step-by-step teoliposomes was an absolute necessity since addition of detergent to preformed liposomes. A few steps in the solubilization process are residual detergent might inhibit enzymatic schematically drawn. Rsat corresponds to the detergent-to-lipid ratio in liposomes at the activity and/or drastically increase the pas- onset of solubilization, while Rsol corresponds to the detergent-to-lipid ratio in micelles when total solubilization is reached. 2, Addition of the solubilized protein in each well- sive permeability of the liposomes, render- defined step of the solubilization process. 3, Detergent removal. ing transport functions difficult to measure.

Braz J Med Biol Res 35(7) 2002 Reconstitution of membrane proteins 757

Systematic studies using radioactive deter- proteoliposomes (Figure 4A-C). Many cap- gents have demonstrated the efficiency of sids of phages bound to their receptor were polystyrene beads to allow almost complete partially or totally devoid of DNA, and, in- removal of any kind of detergent, making terestingly, always associated with proteoli- this method of detergent removal much more posomes that contained DNA, providing clear efficient than conventional dialysis (23,24). evidence of DNA transfer from the phage

Functional studies

OD An important benefit of the production of Total solubilization “quasi-ideal” proteoliposomes has been to obtain original information about the func- Rsol tion of different membrane proteins. A few notable examples from our laboratory include the demonstration of a back pressure effect of ∆pH on bacteriorhodopsin (25), the Onset of 2+ + evidence of a direct Ca /H coupling in Ca- solubilization

ATPases (26-28), and the analysis of artifi- Rsat + cial ∆µH driving ATP synthesis by F0F1 ATPases (21). It has also been possible to Detergent co-reconstitute two functionally coupled membrane proteins, permitting the analysis of coupling ∆µH+/ATP synthesis in co-reconstituted systems containing F0F1 ATPases and different light-driven proton generators (29-32). More recently, using reconstituted proteoliposomes, we have been able to mimic and analyze at the molecular level the mechanisms of transfer of the phage genome into bacteria during phage infection. In Esche- richia coli, the mere interaction of phage T5 with FhuA, its outer membrane receptor, is sufficient to trigger the release of DNA from the phage capsid. Therefore, by reconstituting the membrane receptor FhuA into liposomes, we have been able to analyze the mechanism of phage infection, namely phage binding and the subsequent transport of its DNA into the proteoliposomes (Figure 4). Figure 3. Mechanisms of protein incorporation in detergent-mediated reconstitutions. The transfer of DNA has been visualized Proteoliposomes were reconstituted by the “step-by-step” strategy described in Figure 2. using cryo-electron microscopy, i.e., samples The upper part of the Figure describes the steps in the lamellar-to-micellar transition when optimal reconstitution was observed. The lamellar-to-micellar transition can be analyzed by frozen in vitreous ice (33). When T5 phages turbidimetry as schematically shown. Depending upon the nature of the detergent, three were added to FhuA-containing proteolipo- mechanisms have been identified: direct protein incorporation into detergent-saturated somes, electron micrographs clearly demon- liposomes at the onset of solubilization (Rsat, mechanism 1), transfer from mixed micelles to detergent-saturated liposomes (Rsol, mechanism 2) or proteoliposome formation by strated that phages were able to bind to their micellar coalescence (mechanism 3). Note that the final orientation of the protein depends receptor incorporated in the membranes of on the mechanism of association.

Braz J Med Biol Res 35(7) 2002 758 J.-L. Rigaud

capsid into the proteoliposomes. More than For a better understanding of the interac- one complete genome (121,000 bp, 40 µm in tion of the phage with its receptors, informa- length) could be transferred into proteolipo- tion has been retrieved from 3-D images of somes of 150 nm in diameter and several phage-proteoliposome complexes using elec- genomes were observed in proteoliposomes tron tomography (34). The principle of elec- of 250 nm in diameter (Figure 4A). tron tomography consists of the determination of the 3-D structure of a single object from a series of 2-D images of this object tilted at various angles (35). The resolution FhuA Phage T5 of the 3-D reconstructions demonstrated un- ambiguously that, after binding to its receptor, the straight fiber of the phage crossed the Phage T5 lipid bilayer and underwent a major conformational change, allowing the transfer of its double-stranded DNA into the proteoliposome (Figure 4C). On the basis of the crystal FhuA FhuA Porines structure of the FhuA receptor (36), we have been able to conclude that FhuA was only TonB

FhuB ExbB ExbB used as a docking site for the phage and that

FhuC the tip of the phage tail acted like a DNA “injection needle” creating a passageway at the periphery of FhuA. Besides the important consequences of our study for the molecular mechanisms of phage infection, the experimental approach using reconstituted proteoliposomes has important applications in many biological con- texts. For example, it has been an attractive model system to study the mechanisms of DNA condensation. To this end, proteoliposomes were reconstituted in the presence of different amounts of spermine, a well-known DNA condensing agent. In the presence of 50 mM spermine-containing liposomes, the DNA strands transferred from the phage capsids appeared condensed into character- istic toroidal structures that occupied a vol- Figure 4. Use of FhuA-containing proteoliposomes to study the mechanism of phage ume smaller than the internal volume of the infection. I, The proteoliposome strategy. 1, Purification of FhuA, the receptor for phage T5 proteoliposome (Figure 4B). Increasing the on the outer membrane of Escherichia coli. 2, Reconstitution of FhuA into proteoliposomes. 3, Binding of T5 and DNA release into proteoliposomes. II, Panels A, B, and C illustrate cryo- number of DNA strands transferred into the electron micrographs demonstrating the DNA transfer from phage T5 (A) into FhuA-contain- proteoliposomes did not change the number ing proteoliposomes in the absence (B) or presence of 50 mM spermine (C). DNA is of toroids in the liposomes but increased visualized in electron-dense material either in full and partially empty icosahedral capsids or inside proteoliposomes in which DNA transfer has occurred. Note that, as opposed to what significantly the size of the toroid formed is observed in the absence of spermine, DNA strands transferred into spermine-containing (37). From another point of view, the ap- proteoliposomes appear to be condensed into a single toroidal structure. Panel D, Isosurface proach could be important for gene therapy representation of the 3-D reconstruction of the phage bound to the proteoliposome as determined by electron tomography. The top of the vesicle has been cropped to reveal the applications, since the DNA-containing pro- tip of the phage visible inside the vesicle. teoliposomes produced in this way could

Braz J Med Biol Res 35(7) 2002 Reconstitution of membrane proteins 759 serve as alternative vehicles for the transfer knowledge, combined with the information of foreign genomes into eukaryotic cells. from sequence analysis, frequently provided significant insight into the structure of the Reconstitution of 2-D crystals protein. However, as for 3-D crystallization, the production of ordered 2-D crystals re- Elucidation of the structure of membrane mains rare and many proteins continue to proteins is a major challenge. Indeed, less resist efforts. Except for few examples of 2- than 30 original structures of membrane pro- D crystals of membrane proteins in native teins have been solved to atomic resolution, membranes, 2-D crystals have to be recon- a number which lags far behind that of soluble stituted from solubilized purified proteins. proteins, with several thousand accurately determined structures. 2-D crystallization by detergent removal in While conventional 3-D crystallization bulk solution has provided most of the atomic structures of membrane proteins, successful growth of To date, the most frequently employed 3-D crystals of membrane proteins in deter- strategy for 2-D crystallization relies on the gent micelles is relatively infrequent. This general method of detergent-mediated re- drawback is mainly related to the difficulty constitution of proteoliposomes, but at low in maintaining a crystal lattice through the lipid/protein ratios (1 protein molecule per sole interactions between the hydrophilic 10-50 lipid molecules for 2-D crystals as domains of the proteins, with the hydropho- compared to 1 protein molecule per 2000- bic domains being shielded by the detergent 10,000 lipid molecules for proteoliposomes). micelles. To help with this problem, exten- The strategy consists of starting with the sive studies have been performed to develop purified protein and a suitable combination new strategies. In particular, a novel ap- of lipids, both solubilized in detergent. Next, proach has been introduced by Michel’s the detergent is removed from these micellar group (38), in which monoclonal antibody solutions, resulting in the progressive forma- fragments were specifically bound to pro- tion of lipid bilayers into which the proteins teins, increasing the interactions between incorporate and eventually crystallize. Sev- the hydrophilic domains of membrane pro- eral types of 2-D crystals, all containing a teins. More recently, bi-continuous lipidic continuous lipid bilayer in which proteins cubic phases have been successfully used as have been incorporated, can be produced: matrices for nucleation and further 3-D crys- vesicular crystals, tubular crystals in which tallization of different membrane proteins, reconstituted proteins are helically ordered with the idea that proteins in such membrane on the surface of a cylinder, and planar mimetic systems should be more stable than crystalline sheets which in some instances in micelles (39). can lead to thin 3-D crystals upon the stack- It is noteworthy, that reconstitution of ing of the 2-D arrays (Figure 5). membrane proteins into artificial membranes Different methods have been used to to form crystals confined to two dimensions remove detergent, each presenting advan- has opened a new way to solve their struc- tages and disadvantages (for a recent re- ture (7-10). In several cases, the structures view, see Ref. 10). Dialysis is the most obtained by electron crystallography allowed widely used technique in 2-D crystallization atomic models to be built. Besides these trials. Due to the necessity to scale down the examples, many other 2-D crystals have been amount of material, microdialysis devices analyzed at resolution which allowed the have been used in the form of small com- secondary structure to be clearly solved. Such partments (50-100 µl) dialyzed against large

Braz J Med Biol Res 35(7) 2002 760 J.-L. Rigaud

buffer volumes. The dialysis method has solubilized in various detergents and, impor- been successfully applied to many mem- tantly, some of these 2-D crystals have been brane proteins, but may not be well suited useful for high-resolution structural analysis for detergents with low critical micellar con- (36,41-43). centrations, that require a long time for di- The self-assembling process leading to alysis. densely packed vesicles and 2-D crystals The method of detergent removal by hy- depends upon a large matrix of variables. In drophobic adsorption onto polystyrene beads, addition to the composition of the solution previously shown to be very efficient for (salt, pH, additives such as glycerol) and proteoliposome reconstitution (24), has been temperature, variables of primary importance recently demonstrated to be a powerful alter- are related to protein-protein, lipid-protein, native to conventional dialysis for 2-D crys- detergent-protein and detergent-lipid inter- tallization trials (40). Using radioactive de- actions. As opposed to proteoliposome re- tergents, the method has been calibrated pre- constitution, the rules for 2-D crystallization cisely in terms of the adsorptive capacity of by detergent removal are far from being the beads and the rates of detergent removal. generalized and very few physicochemical The mechanisms underlying detergent ad- parameters have been determined as essen- sorption onto beads have been analyzed and tial for growing 2-D crystals of membrane general rules for the use of polystyrene beads proteins (10,44,45). have been proposed. Sufficient reproduc- ibility can be now obtained with experience Induced 2-D crystallization and careful handling. An important benefit of this new strategy has been to produce new The previous reconstitution procedure im- 2-D crystals of several membrane proteins plies that, upon detergent removal, proteins must be incorporated into the bilayer but also find homogeneous optimal interactions Planar Detergent at this transition. These two simultaneous Lipid processes appear crucial to the outcome of Protein 2-D crystallization trials and may explain the difficulty for many proteins to crystallize. Planar stacking For those membrane proteins difficult to crystallize, one can consider bilayer formation separately from crystallization and im- prove the strategy of “induced 2-D crystallization”. In this case, reconstitutions should be first performed at high lipid-to-protein Vesicular ratios, leading to proteoliposomes with pro- Detergent teins homogeneously incorporated and rather removal closely packed. Then, in a second step, crystallization could be induced by physical treat- Tubular ment such as lipid temperature transition (46) or chemical agents including vanadate that cross-links P-type ATPases (47) or phos- 2-D crystals pholipase A2 that decreases the lipid-to-protein ratio in the reconstituted proteolipo- Figure 5. 2-D crystallization of membrane proteins by detergent removal. Following detergent removal from a lipid-protein-detergent micellar solution (lipid-to-protein ratios below 1 some (48). w/w), different 2-D crystals can be produced. We have developed this approach to in-

Braz J Med Biol Res 35(7) 2002 Reconstitution of membrane proteins 761 duce 2-D crystallization of Ca-ATPase after are overproduced and purified using recom- reconstitution into proteoliposomes (47). Ca- binant proteins containing a stretch of con- ATPase-dense proteoliposomes, produced tinuous histidine residues (His-tag), a nickel- through the "step-by-step" reconstitution chelating lipid has been employed as a gen- strategy, were demonstrated to be ideal to eral adaptor molecule that will link any His- induce further crystallization by the addition tag protein; ii) reconstitution of a bilayer of vanadate, leading to very long ordered around the pre-bound proteins by detergent crystalline tubes. These tubular crystals were removal using polystyrene Bio-Beads in the produced by very few preparations of native solution below the lipid layer; iii) diffusion sarcoplasmic reticulum vesicles, and the un- of the lipid-protein complexes and further 2- known parameters that control the growth of D crystallization. long tubes from spherical vesicles could be This innovative strategy opens a new easily analyzed in our reconstituted system. promising field for membrane protein struc- In particular, we found that a critical step to ture determination since: i) it may increase make extended tubular crystals from small the chances of success to produce 2-D crys- proteoliposomes was the inclusion of specif- tals of proteins difficult to crystallize by ic lipids that destabilize bilayers. conventional methods; ii) it permits structure analysis with smaller amounts of mate- 2-D crystallization of membrane proteins on rial. Indeed, since this strategy induces pro- functionalized lipid layers tein concentration at the interface, this leads to the use of less than 1 µg protein as com- Considerable interest exists in develop- pared to 25-50 µg protein required by the ing new strategies to increase the success classical bulk method. This will be of par- rate of 2-D crystallization. The technique of Figure 6. 2-D crystallization of 2-D crystallization on lipid layers is based on A membrane proteins on the lipid a specific interaction between the protein layer at the air-water interface. The three stages in the process and specific ligands coupled to lipid mol- of 2-D crystallization on lipid lay- ecules incorporated into a planar lipid film at ers. A, Binding of protein-lipid- the air-water interface. This method has detergent micelles to the functionalized lipid layer spread at proven to be a successful approach to 2-D the air-water interface. B, Re- crystallization of soluble proteins, providing constitution of proteins into a li- important structural information (49,50). The pid bilayer upon detergent re- question as to whether this method was also moval. C, 2-D crystallization. applicable to membrane proteins was a challenge. B As the result of a careful study, we have recently been able to produce on planar lipidic templates the first 2-D crystals of two radically different membrane proteins, FhuA and F0F1 ATPase, demonstrating the feasi- bility of this new strategy for membrane proteins (51,52). A model for the mechan- C isms of 2-D crystallization of membrane proteins on a lipid layer has been proposed, which include three steps (Figure 6): i) binding of protein-lipid-detergent micelles to the lipid layer. Since many membrane proteins

Braz J Med Biol Res 35(7) 2002 762 J.-L. Rigaud

ticular help to laboratories studying mem- meter, digitized on a microdensitometer and brane proteins which are currently in the computer processed. The digitized images early stages of developing procedures for are Fourier transformed and phases and am- overexpression and large-scale production; plitudes of structure factors selected and iii) although a nickel-chelating lipid has been averaged. In the course of image processing, employed as a general adaptor that can link crystal defects (lattice distortion) and image any His-tag protein at the surface, the strat- defects (lack of focus and astigmatism) are egy has been extended to other types of corrected. The inverse Fourier transform then specific molecular recognition such as elec- generates a projection map which can be trostatic interactions between proteins and represented as a topographic map in which lipids (52). Also it will be important to ex- the hills would reflect higher electron densi- tend the strategy to other types of specific ties. To determine the 3-D structure, electron molecular recognition at the surface between diffraction patterns and electron images of proteins and lipids using other lipid-coupled 2-D crystals are recorded with the specimen physiological ligands. at various tilt angles. By combining the structure factors which contribute to these projec- Structural analysis tions in 3-D, a 3-D data set is compiled. The inverse transform of this data set is the 3-D Electron crystallography. The method of map of the membrane protein. choice for analyzing the results of 2-D crys- Atomic force microscopy. As a comple- tallization is electron microscopy. The quick- mentary technique, atomic force microscopy est specimen preparation is negative staining (AFM) has become an important tool to which yields excellent contrast but does not investigate transmembrane proteins under preserve high-resolution details. Once the physiological conditions. AFM measures the optimal conditions of crystallization have heights of biological samples to obtain topo- been determined by negative staining, cryo- graphic maps of protein structures protrud- electron microscopy is preferred because it ing from the membrane at a lateral resolution preserves the native structure of the protein of ~8 Å and a vertical resolution of ~1 Å to high resolution. One drop of the sample is (54). Specifically, high-resolution topo- deposited onto a carbon-coated grid, blotted graphic maps before and after proteolytic with a filter paper and rapidly frozen in a cleavage of distinct protein domains have liquid ethane bath (53). The frozen samples permitted the localization of these domains need to be maintained at -170ºC throughout and the assignment of the sidedness of the observation under the microscope to pre- whole transmembrane molecule. This ap- serve them against dehydration in the vacuum proach was nicely illustrated by a recent of the microscope and against radiation dam- study on the light-harvesting complex 2 age. Finally, the best method to assay the (LH2) of Rubrivivax gelatinosus, the acces- degree of order is electron diffraction, which sory antenna proteins in the bacterial photo- is generally used to assess high-resolution synthetic apparatus built of αß-heterodimers structure analysis of good quality 2-D crystals. (55). AFM has been used to measure the Once the 2-D crystals have been pro- surface topography of LH2 in a reconstituted duced, the structure of the protein can be membrane and to locate the C-terminal hy- determined by electron crystallography and drophobic extension of 21-amino acid resi- image processing to a resolution that de- dues in the α-subunit. High-resolution topo- pends on the quality of the 2-D crystal (11). graphic maps revealed a nonameric organi- The best electron micrographs are selected zation of the αß-heterodimers, regularly by optical diffraction with a laser diffracto- packed as cylindrical complexes incorpo-

Braz J Med Biol Res 35(7) 2002 Reconstitution of membrane proteins 763 rated into the membrane in both orientations conformational changes by the addition of (Figure 7). Native LH2 showed one surface substrate molecules to the scanning buffer. which protruded by ~5 Å and one which Excellent reviews from Engel’s group (54- protruded by ~14 Å from the membrane. 57) are available that illustrate various appli- Topographic maps of samples reconstituted cations of AFM. with thermolysin-digested LH2 have revealed a height reduction of the strongly protruding Conclusions surface to ~9 Å, while weakly protruding rings remained unchanged. These results Reconstitution of membrane proteins into demonstrated the extrinsic localization of liposomes has played an important role and the C-terminal domain of the α-subunit of R. should continue to be a potentially powerful gelatinosus and allowed the periplasmic sur- tool to be used to identify the mechanism of face to be assigned. action of membrane proteins. As shown in Finally, the AFM has developed to the this review, it appears that reconstitution no point that it permits to unfold membrane longer is “black magic” and the prospects of proteins, to detect flexible and stable extrin- achieving optimal reconstitution of proteoli- sic domains and to monitor function-related posomes are obviously good, with the use of

Figure 7. High-resolution atomic force microscopy topographic maps of the LH2 complex from Rubrivivax gelatinosus. a, To- pographic maps of the strongly (~14 Å) protruding surface of the native LH2. b, Topographic maps of the weakly (~5 Å) protruding surface of the native LH2. c, Topographic maps of the strongly (~9 Å) protruding surface of the digested LH2. d, Topographic maps of the weakly (~5 Å) protruding surface of the digested LH2. In- sets show average images of LH2 complexes calculated from the raw data shown in each corresponding panel. Pro- teolysis only influenced the topography of the strongly protruding surface, allowing to lo- calize the C-terminal position of the α-subunit.

Braz J Med Biol Res 35(7) 2002 764 J.-L. Rigaud

reliable methods. teins, which have been identified in the se- Despite significant progress in the last quencing of the genomes of different organ- few years, 3-D crystallization of membrane isms. Powerful methods of gene overexpres- proteins has proved to be difficult and slow, sion are now available which will permit the and alternative approaches to structure de- production of new interesting membrane pro- termination have become essential. Electron teins. crystallography, associated with 2-D crystallization, is a technique that has produced Acknowledgments many structures and is definitely a viable alternative strategy to X-ray crystallogra- I would like to thank Prof. A. Pereira and phy. However, for electron crystallography the Brazilian Society of Biochemistry and to become even more important, enormous Molecular Biology for their kind invitation efforts are required: i) for a deeper under- to attend the meeting in Caxambu, which standing of the 2-D crystallogenesis process, gives me the great opportunity to establish and ii) for the development of new approaches contact with many laboratories in Brazil. to be evaluated, refined and applied to avail- Special thanks are due to D. Lévy, O. Lam- able proteins. bert, M. Chami, S. Scheuring and B. Pitard The future of membrane protein recon- who have been directly involved in many of stitution appears bright in the light of the the reconstitution studies described in this steadily increasing number of membrane pro- review.

References

1. Jones DT (1998). Do transmembrane pro- & Walz T (1998). 2D crystallization of 348. tein superfolds exist? FEBS Letters, 423: membrane proteins: rationales and ex- 16. Rigaud J-L & Lévy D (2002). Reconstitu- 281-285. amples. Journal of Structural Biology, 121: tion of membrane proteins into proteo- 2. Racker E (1982). Reconstitution of mem- 162-171. liposomes. Methods in Enzymology : Li- brane processes. Methods in Enzymol- 10. Rigaud JL, Chami M, Lambert O, Lévy D posomes, Part B (in press). ogy, 55: 699-711. & Ranck J (2000). Use of detergents in 2D 17. Lévy D, Gulick A, Seigneuret M & Rigaud 3. Eytan GD (1981). Use of liposomes for crystallization of membrane proteins. Bio- J-L (1990). Phospholipid vesicle solubiliza- reconstitution of biological functions. Bio- chimica et Biophysica Acta, 1508: 112- tion and reconstitution by detergent. Sym- chimica et Biophysica Acta, 694: 185-202. 128. metrical analysis of the two processes 4. Cornelius F (1991). Functional reconstitu- 11. Walz T & Grigorieff N (1998). Electron using octaethylene glycol mono-N-dode- tion of the sodium pump. Kinetics of ex- crystallography of two-dimensional crys- cyl ether. Biochemistry, 29: 9480-9488. change reactions performed by reconsti- tals of membrane proteins. Journal of 18. Paternostre MT, Roux M & Rigaud J-L tuted Na/K-ATPase. Biochimica et Biophy- Structural Biology, 121: 142-161. (1988). Mechanisms of membrane pro- sica Acta, 1071: 19-66. 12. Szoka F & Papahadjopoulos D (1980). tein insertion into liposomes during re- 5. Rigaud JL, Pitard B & Lévy D (1995). Re- Comparative properties and methods of constitution procedures involving the use constitution of membrane proteins into preparation of lipid vesicles. Annual Re- of detergents. 1. Solubilization of large liposomes: application to energy-trans- view of Biophysics and Bioengineering, 9: unilamellar liposomes (prepared by re- ducing membrane proteins. Biochimica et 467-508. verse-phase evaporation) by Triton X-100, Biophysica Acta, 1231: 223-246. 13. Lemaire M, Champeil P & Moller JV octyl glucoside and sodium cholate. Bio- 6. Ostermeier C & Michel H (1997). Crystal- (2000). Interaction of membrane proteins chemistry, 27: 2668-2677. lization of membrane proteins. Current and lipids with solubilizing detergents. 19. Lambert O, Lévy D, Ranck JL, Leblanc G Opinion in Structural Biology, 7: 697-701. Biochimica et Biophysica Acta, 1508: 86- & Rigaud J-L (1998). A new gel-like phase 7. Kuhlbrandt W (1992). Two-dimensional 111. in dodecylmaltoside-lipid mixtures: impli- crystallization of membrane proteins. 14. Lichtenberg D (1985). Characterization of cations in solubilization and reconstitution Quarterly Reviews in Biophysics, 25: 1- the solubilization of lipid bilayers by sur- studies. Biophysical Journal, 74: 918-930. 49. factants. Biochimica et Biophysica Acta, 20. Rigaud J-L, Paternostre MT & Bluzat A 8. Jap BK, Zulauf M, Scheybani T, Hefti A, 821: 470-478. (1988). Mechanisms of membrane pro- Baumeister W, Aebi U & Engel A (1992). 15. Silvius JR (1992). Solubilization and func- tein insertion into liposomes during re- 2D crystallization: from art to science. tional reconstitution of biomembrane constitution procedures involving the use Ultramicroscopy, 46: 45-84. components. Annual Review of Biophys- of detergents. 2. Incorporation of the light- 9. Hasler L, Heymann JB, Engel A, Kistler J ics and Biomolecular Structure, 21: 323- driven proton pump bacteriorhodopsin.

Braz J Med Biol Res 35(7) 2002 Reconstitution of membrane proteins 765

Biochemistry, 27: 2677-2688. from bacillus PS3 co-reconstituted into li- Rigaud J-L (1998). Two-dimensional crys- 21. Richard P, Rigaud J-L & Gräber P (1990). posomes with bacteriorhodopsin. Rela- tallization of Ca2+ATPase by total deter- Reconstitution of CF0F1 into liposomes tionships between proton motive force gent removal. Biophysical Journal, 75: using a new reconstitution procedure. and ATP synthesis. European Journal of 1319-1329. European Journal of Biochemistry, 193: Biochemistry, 235: 779-788. 43. Saint N, Lacapère JJ, Ghazi A, Martinac B 921-930. 31. Richard P, Pitard B & Rigaud J-L (1995). & Rigaud J-L (1998). A hexameric trans- 22. Lévy D, Gulik A, Bluzat A & Rigaud J-L ATP synthesis by the F0F1 ATPase from membrane pore revealed by 2D crystalli- (1992). Reconstitution of the sarcoplas- the thermophilic bacillus PS3 co-reconsti- zation of the large mechanosensitive ion mic reticulum Ca2+ATPase: mechanisms tuted with bacteriorhodopsin into lipo- channel (MscL) of E. coli. Journal of Bio- of membrane protein insertion into liposomes. Evidence for stimulation of ATP logical Chemistry, 273: 14667-14670. somes during reconstitution procedures synthesis by ATP bound to a non-catalytic 44. Dolder M, Engel A & Zulauf M (1996). The involving the use of detergents. Biochimi- site. Journal of Biological Chemistry, 270: micelle to vesicle transition of lipids and ca et Biophysica Acta, 1107: 283-298. 21571-21578. detergents in the presence of a mem- 23. Rigaud J-L, Lévy D, Mosser G & Lambert 32. Steinberg-Yfrach G, Rigaud J-L, Moore brane protein: towards a rationale for 2D O (1998). Detergent removal by Bio- LA, Gust D & Moore TA (1998). Light- crystallization. FEBS Letters, 382: 203- Beads. Applications to membrane protein driven, F0F1 ATP synthase-catalyzed syn- 208. reconstitution and 2D crystallization. Eu- thesis of ATP by an artificial photosyn- 45. Engel A, Hoenger A, Hefti A, Henn C, ropean Biophysics Journal, 27: 305-319. thetic membrane. Nature, 392: 479-482. Ford RC, Kisler J & Zulauf M (1992). As- 24. Lévy D, Bluzat A, Seigneuret M & Rigaud 33. Lambert O, Plançon L, Rigaud J-L & sembly of 2D membrane protein crystals: J-L (1990). A systematic study of lipo- Letellier L (1998). Protein-mediated DNA dynamics, crystal order and fidelity of some and proteoliposome reconstitution transfer into liposomes. Molecular Micro- structure analysis by electron microscopy. involving Bio-Bead-mediated Triton X-100 biology, 30: 761-765. Journal of Structural Biology, 109: 219- removal. Biochimica et Biophysica Acta, 34. Boehm J, Lambert O, Letellier L, 234. 1025: 179-190. Baumeister W & Rigaud J-L (2001). FhuA- 46. Cherry RJ, Muller R, Henderson R & Heyn 25. Seigneuret M & Rigaud J-L (1986). Analy- mediated phage genome transfer into li- MP (1997). Temperature-dependent ag- sis of passive and light-driven ion move- posomes: a cryo-electron tomography gregation of bacteriorhodopsin in dipalmi- ments in large bacteriorhodopsin lipo- study. Current Biology, 11: 1168-1175. toyl and dimyristoylphosphatidylcholine. somes reconstituted by reverse-phase 35. Baumeister W, Grimm R & Walz J (1999). Journal of Molecular Biology, 121: 283- evaporation. 2. Influence of passive per- Electron tomography of molecules and 298. meability and back-pressure effects upon cells. Trends in Cell Biology, 9: 81-85. 47. Young HS, Rigaud J-L, Lacapère JJ, light-induced proton uptake. Biochemis- 36. Lambert O, Moeck G, Lévy D, Plançon L, Reddy LG & Stokes DL (1997). How to try, 25: 6723-6729. Letellier L & Rigaud J-L (1999). A pro- make tubular crystals by reconstitution of 26. Lévy D, Seigneuret M, Bluzat A & Rigaud jected structure of FhuA, a ligand-gated detergent-solubilized Ca-ATPase. Bio- J-L (1990). Evidence for proton counter- channel of E. coli outer membrane pro- physical Journal, 72: 2545-2558. transport by the sarcoplasmic reticulum tein. Journal of Structural Biology, 126: 48. Mannella C (1984). Phospholipase- Ca2+ATPase during calcium transport in 145-155. induced crystallization of channels in mi- reconstituted proteoliposomes with “low” 37. Lambert O, Letellier L, Gelbart WM & tochondrial outer membranes. Science, ionic permeability. Journal of Biological Rigaud J-L (2000). DNA delivery by phage 224: 165-166. Chemistry, 265: 19524-19533. as a strategy for encapsulating toroidal 49. Kornberg RD & Darst SA (1991). Two- 27. Hao L, Rigaud J-L & Inesi G (1994). Ca2+/ condensates of arbitrary size into lipo- dimensional crystals of proteins on lipid H+ countertransport and electrogenicity somes. Proceedings of the National Acad- layers. Current Opinion in Structural Biol- in proteoliposomes containing erythro- emy of Sciences, USA, 97: 7248-7253. ogy, 1: 642-646. cyte plasma membrane, Ca2+ATPase and 38. Ostermeier C, Iwata J & Michel H (1996). 50. Chiu W, Avila-Sakar AJ & Schmid MF exogenous lipids. Journal of Biological Cytochrome c oxidase. Current Opinion in (1997). Electron crystallography of macro- Chemistry, 269: 14268-14275. Structural Biology, 6: 460-466. molecular periodic arrays on phospholipid 28. Salvador J, Inesi G, Rigaud J-L & Mata 39. Nollert P, Qiu H, Gaffrey M, Rosenbusch monolayers. Advances in Biophysics, 34: AM (1998). Ca2+ transport by reconsti- JP & Landau EM (2001). Molecular 161-172. tuted synaptosomal ATPase is associated mechanisms for the crystallization of 51. Lévy D, Mosser G, Lambert O, Moeck G, with H+ countertransport and net charge bacteriorhodopsin in lipidic cubic phases. Bald D & Rigaud J-L (1999). 2D crystalliza- displacement. Journal of Biological Chem- FEBS Letters, 504: 179-186. tion on lipid layer: a successful approach istry, 273: 18230-18234. 40. Rigaud J-L, Mosser G, Lacapère JJ, Lévy for membrane proteins. Journal of Struc- 29. Pitard B, Richard P, Dunach M, Girault G D, Oloffson A & Ranck JL (1997). Bio- tural Biology, 127: 44-52. & Rigaud J-L (1996). ATP synthesis by the Beads: an efficient strategy for 2D crystal- 52. Lévy D, Chami M & Rigaud J-L (2001). F0F1 ATP synthase from bacillus PS3 re- lization of membrane proteins. Journal of Two dimensional crystallization of mem- constituted into liposomes with bacterio- Structural Biology, 118: 226-235. brane proteins: the lipid layer strategy. rhodopsin. Factors defining the optimal 41. Mosser G, Breyton C, Oloffson A, Popot FEBS Letters, 504: 187-193. reconstitution of ATP synthases with bac- JL & Rigaud J-L (1997). A 8A projection 53. Dubochet J, Adrian M, Chang JJ, Homo teriorhodopsin. European Journal of Bio- map of cytochrome b6f complex from JC, Lepault J, McDonall AW & Schultz P chemistry, 235: 769-778. Chlamydomonas reinhartdii. Journal of (1988). Cryo-electron microscopy of vitri- 30. Pitard B, Richard P & Rigaud J-L (1996). Biological Chemistry, 272: 20263-20268. fied specimen. Quarterly Reviews of Bio- ATP synthesis by the F0F1 ATP synthase 42. Lacapère JJ, Stokes D, Olofsson A & physics, 21: 129-228.

Braz J Med Biol Res 35(7) 2002 766 J.-L. Rigaud

54. Stahlberg H, Fotiadis D, Scheuring S, 55. Scheuring S, Reiss Husson F, Engel A, force microscope. Nature Structural Biol- Remigny H, Braun T, Mitsuoka Y, Rigaud J-L & Ranck JL (2001). High reso- ogy, 7: 715-718. Fujiyoshi Y & Engel A (2001). Two-dimen- lution AFM topographs of Rubrivivax ge- 57. Heymann JB, Müller DJ, Mitsuoka K & sional crystals: a powerful approach to latinosus light-harvesting complex LH2. Engel A (1997). Electron and atomic force assess structure, function and dynamics EMBO Journal, 20: 3029-3035. microscopy of membrane proteins. Cur- of membrane proteins. FEBS Letters, 56. Engel A & Muller DJ (2000). Observing rent Opinion in Structural Biology, 7: 543- 504: 166-172. single molecules at work with the atomic 549.

Braz J Med Biol Res 35(7) 2002