membrane proteins: functional and structural studies using ... · reconstitution of membrane...

753

Braz J Med Biol Res 35(7) 2002

Reconstitution of membrane proteins

Membrane proteins: functional andstructural studies using reconstitutedproteoliposomes and 2-D crystals

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

Abstract

Reconstitution of membrane proteins into lipid bilayers is a powerfultool to analyze functional as well as structural areas of membraneprotein research. First, the proper incorporation of a purified mem-brane protein into closed lipid vesicles, to produce proteoliposomes,allows the investigation of transport and/or catalytic properties of anymembrane protein without interference by other membrane compo-nents. Second, the incorporation of a large amount of membraneproteins into lipid bilayers to grow crystals confined to two dimen-sions has recently opened a new way to solve their structure at highresolution using electron crystallography. However, reconstitution ofmembrane proteins into functional proteoliposomes or 2-D crystalli-zation has been an empirical domain, which has been viewed for along time more like “black magic” than science. Nevertheless, in thelast ten years, important progress has been made in acquiring knowl-edge of lipid-protein-detergent interactions and has permitted to buildupon a set of basic principles that has limited the empirical approachof reconstitution experiments. Reconstitution strategies have beenimproved and new strategies have been developed, facilitating thesuccess rate of proteoliposome formation and 2-D crystallization. Thisreview deals with the various strategies available to obtain proteolipo-somes and 2-D crystals from detergent-solubilized proteins. It gives anoverview of the methods that have been applied, which may be of helpfor reconstituting more proteins into lipid bilayers in a form suitablefor functional studies at the molecular level and for high-resolutionstructural analysis.

CorrespondenceJ.-L. Rigaud

Institut Curie

UMR-CNRS 168 and LRC-CEA 8

11 Rue Pierre et Marie Curie, 75231

Paris, Cedex 05

France

E-mail: [email protected]

Presented at the XXX Annual

Meeting of the Brazilian Society

of Biochemistry and Molecular

Biology, Caxambu, MG, Brazil,

May 19-22, 2001.

Research supported in part by

CNRS (France)-CNPq (Brazil)

collaborative programs.

Received January 14, 2002

Accepted May 7, 2002

Key words• Membrane proteins• Proteoliposomes• 2-D crystals• Electron crystallography• Reconstitution

Introduction

Screening of the genomes of differentorganisms has demonstrated that about 25%of the sequenced genes code for stronglyhydrophobic proteins, which are integratedinto cell membranes (1). These results haveemphasized the importance of membraneproteins in many biological processes essen-tial for life. However, the complexity of most

biological membranes makes it difficult tostudy these membrane proteins in situ. There-fore, purification from the native membraneand further reincorporation of the purifiedmembrane protein into an artificial mem-brane to form proteoliposomes or to producetwo-dimensional (2-D) crystals continue tobe a crucial step in studying the function andstructure of these molecules.

First of all, membrane protein reconstitu-

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

754


J.-L. Rigaud

tion into liposomes has played an importantrole and should continue to be a powerfultool that can be used to identify and charac-terize the mechanisms of action of mem-brane proteins. The necessity for reconstitu-tion is due to the fact that many membraneproteins express their full activity only whencorrectly oriented and inserted into a lipidbilayer. In particular, reconstitution hasplayed a central role in the identification andcharacterization of the mechanisms of ac-tion of membrane proteins with a vectorialtransport function. More generally, biochemi-cal and biophysical approaches have provid-ed important information about lipid-proteinand protein-protein interactions as well astopological and topographic features of dif-ferent classes of membrane proteins recon-stituted into proteoliposomes (2-5).

Second, structure determination at highresolution is actually a difficult challenge formembrane proteins and the number of mem-brane proteins that have been crystallized inthree dimensions (3-D) is still small and farbehind that of soluble proteins (6). Reconsti-tution of membrane proteins into artificialmembranes to form 2-D crystals has openeda new way to solve their structures inside anative-like environment (7-10). Indeed, elec-tron crystallography of 2-D crystals has de-veloped to the point that significant struc-tural atomic models have been built and thatmany membrane protein structures have beenevaluated at resolutions allowing the sec-ondary structure to be assessed (11). Theserecent results have made electron crystallog-raphy an excellent alternative and a comple-mentary discipline to X-ray crystallographyin membrane protein structural biology.

However, the ability to investigate func-tional aspects or to solve the structures ofmembrane proteins through the use of recon-stituted systems has been limited by the factthat methods for producing high quality pro-teoliposomes or 2-D crystals have not ad-vanced at the same rate as biochemical, bio-physical and molecular biology techniques,

on the one hand, and electron crystallogra-phy on the other. Thus, one of the limitingfactors in obtaining molecular informationwas related to the lack of reproducible meth-ods of reconstitution. Therefore enormousefforts were required to understand themechanisms of reconstitution and for newapproaches to be evaluated, refined and ap-plied to available proteins in order to makereconstitution even more important as a toolfor establishing structure-function relation-ships at the molecular level.

This review will deal with the differentstrategies commonly used to reconstitute pro-teoliposomes and 2-D crystals. It will alsofocus on recent new approaches, which haveled to promising perspectives for the produc-tion of highly functional proteoliposomesand more 2-D crystals of membrane proteinsamenable to structural analysis. Generalguidelines and rules will be proposed in thisfield which has long been seen more as artthan as science.

Reconstitution of proteoliposomes

Strategies

From the analysis of the abundant litera-ture concerning the insertion of membraneproteins into liposomes, four basic strategieshave been outlined: mechanical means,freeze-thaw, organic solvents, and detergents.Although these reconstitution strategies haveproven to be very useful to prepare purephospholipid vesicles (12), the additionalinsertion of a membrane protein during thereconstitution process has imposed manyconstraints that have seriously hampered theapplicability of phospholipid vesicles to pro-teoliposome reconstitution. Indeed, besidesthe need for conditions that preserve proteinactivity, many criteria have to be consideredto fully optimize the reconstitution of a mem-brane protein: the homogeneity of proteininsertion, the final orientation of the protein,the morphology and the size of the reconsti-

755



tuted proteoliposomes, as well as their re-sidual permeability. Even more limiting isthe fact that many methods for preparingpure liposomes have been difficult to applysuccessfully to proteoliposome reconstitu-tion because most membrane proteins arepurified by the use of detergents, which in-terfere with the process of vesicle formation.

Thus, the vast majority of membrane pro-tein reconstitution procedures are those in-volving the use of detergents. Indeed, due totheir amphiphilic character, most membraneproteins require detergents, not only as ameans of disintegrating the structure of na-tive membranes in the initial step of theirsolubilization, but also as a means of keep-ing the protein in a non-denaturing environ-ment during further purification (13). Thestandard procedure used to reconstitute apurified membrane protein involves co-micellization of the protein in an excess ofphospholipid and appropriate detergent, toform a solution of mixed lipid-protein-deter-gent and lipid-detergent micelles. Next, thedetergent is removed from the micellar solu-tions, resulting in the progressive formationof closed lipid bilayers into which the pro-teins eventually incorporate. All the deter-gent-mediated reconstitutions described inthe literature rely on this standard procedurebut differ essentially in the techniques usedto remove the detergent, which can be dialy-sis, gel chromatography, dilution, or hydro-phobic adsorption onto polystyrene beads(Figure 1).

As shown by the abundant literature, re-constitutions from detergent micellar mix-tures have yielded proteoliposomes of dif-ferent sizes and compositions depending onthe nature of the detergent, the particularprocedure to remove it, as well as the natureof the protein and the lipid composition.Therefore, each membrane protein respondeddifferently to the various reconstitution pro-cedures and the approach has long beenentirely empirical. In the 90’s, importantknowledge of the mechanisms of liposome

formation as well as understanding of thephysical behavior of lipid-detergent systemspermitted the construction of a set of basicprinciples that has limited the number ofexperimental variables and thus the empiri-cal approach to proteoliposome reconstitu-tion (5,14,15).

To allow realistic monitoring of themechanisms by which proteins may associ-ate with lipids during detergent-mediatedreconstitutions, we have developed a strat-egy (Figure 2) based on the idea that theprocess of reconstitution by detergent re-moval is the reverse of the process of solubi-lization by increasing amounts of detergent(5,16). To this end, the first stage in ourstrategy was to add detergent to preformedliposomes using the entire range of detergentconcentrations that cause the transformationof lamellar structures to mixed micelles (17-19). The lipid-detergent mixed amphiphilicstructures that are formed during the solubi-lization process are similar to those pre-

Figure 1. Detergent-mediated reconstitutions. Most membrane proteins are extracted fromnative membranes with solubilizing detergent concentrations. After purification, the solubi-lized protein is supplemented with an excess of lipids and detergents, leading to a solutionof mixed lipid-protein-detergent and lipid-detergent micelles. For proteoliposome or 2-Dcrystal reconstitution, detergent is removed from these micellar solutions using differentstrategies.

Proteoliposome

Native membrane

Solubilization

Purification

Dialysis

Gelchromatography

Dilution

Polystyrenebeads

Reconstitution

DetergentLipidProtein

2-D crystals

756


J.-L. Rigaud

gent was removed and the resulting proteo-liposomes were characterized in terms ofprotein incorporation and orientation, vesiclesize distribution, and biological activity.

With respect to optimal protein incorpo-ration, we have identified three mechanismsby which membrane proteins can associatewith lipids to give proteoliposomes (Figure3). Depending on the nature of the detergent,proteins can be either directly incorporatedinto detergent-saturated liposomes (octylglu-coside- and dodecylmaltoside-mediated re-constitutions), transferred from mixed mi-celles to detergent-saturated liposomes (Tri-ton X-100-mediated reconstitutions) or par-ticipate in proteoliposome formation duringthe micellar-to-lamellar transition (cholate,chaps, and C12E8-mediated reconstitutions).

In addition to providing original infor-mation about the mechanisms of lipid-pro-tein association in the presence of detergent,this “step-by-step” reconstitution strategyproved to be a powerful reconstitution pro-cedure, more suitable than the usual meth-ods. This new reconstitution strategy, whichis now widely used, has produced proteo-liposomes which sustain biological activi-ties comparable to those measured in thenative membrane. Indeed, it has allowed aneasy determination of the optimal conditionsunder which a protein can associate withlipids in the presence of detergent. In par-ticular, it should be stressed that reconstitu-tion of a membrane protein into detergent-saturated preformed liposomes ensured aunidirectional insertion of the protein in themembrane and generally produced the mostefficient proteoliposomes (20-22). An addi-tional advantage of our new reconstitutionstrategy relied on the use of Bio-Beads SM2to remove the detergent. Indeed, the efficientremoval of detergent from reconstituted pro-teoliposomes was an absolute necessity sinceresidual detergent might inhibit enzymaticactivity and/or drastically increase the pas-sive permeability of the liposomes, render-ing transport functions difficult to measure.

Figure 2. The “step-by-step” reconstitution strategy. The standard procedure for reconsti-tuting membrane proteins into proteoliposomes consists of three stages. 1, Step-by-stepaddition of detergent to preformed liposomes. A few steps in the solubilization process areschematically drawn. Rsat corresponds to the detergent-to-lipid ratio in liposomes at theonset of solubilization, while Rsol corresponds to the detergent-to-lipid ratio in micelleswhen total solubilization is reached. 2, Addition of the solubilized protein in each well-defined step of the solubilization process. 3, Detergent removal.

dicted to form during the micelle-to-lipidbilayer transition in a reconstitution processby detergent removal. Thus, in a secondstage, adding the protein at each step of thesolubilization process allowed a “snap-shot”of all the situations that might occur in areconstitution experiment following deter-gent removal from lipid-protein-detergentmicelles. Finally, in a third stage, the deter-

Step-by-step solubilization

Preformed liposomes

Increasing detergent concentration

Rsat Rsol

Protein addition

Detergent removal

Bio-Beads

Proteoliposomes

757



Systematic studies using radioactive deter-gents have demonstrated the efficiency ofpolystyrene beads to allow almost completeremoval of any kind of detergent, makingthis method of detergent removal much moreefficient than conventional dialysis (23,24).

Functional studies

An important benefit of the production of“quasi-ideal” proteoliposomes has been toobtain original information about the func-tion of different membrane proteins. A fewnotable examples from our laboratory in-clude the demonstration of a back pressureeffect of ∆pH on bacteriorhodopsin (25), theevidence of a direct Ca2+/H+ coupling in Ca-ATPases (26-28), and the analysis of artifi-cial ∆µH+ driving ATP synthesis by F0F1

ATPases (21). It has also been possible toco-reconstitute two functionally coupledmembrane proteins, permitting the analysisof coupling ∆µH+/ATP synthesis in co-re-constituted systems containing F0F1 ATPasesand different light-driven proton generators(29-32).

More recently, using reconstituted pro-teoliposomes, we have been able to mimicand analyze at the molecular level themechanisms of transfer of the phage genomeinto bacteria during phage infection. In Esche-richia coli, the mere interaction of phage T5with FhuA, its outer membrane receptor, issufficient to trigger the release of DNA fromthe phage capsid. Therefore, by reconstitut-ing the membrane receptor FhuA into lipo-somes, we have been able to analyze themechanism of phage infection, namely phagebinding and the subsequent transport of itsDNA into the proteoliposomes (Figure 4).

The transfer of DNA has been visualizedusing cryo-electron microscopy, i.e., samplesfrozen in vitreous ice (33). When T5 phageswere added to FhuA-containing proteolipo-somes, electron micrographs clearly demon-strated that phages were able to bind to theirreceptor incorporated in the membranes of

OD Totalsolubilization

Rsol

Onset ofsolubilizationRsat

Detergent

Figure 3. Mechanisms of protein incorporation in detergent-mediated reconstitutions.Proteoliposomes were reconstituted by the “step-by-step” strategy described in Figure 2.The upper part of the Figure describes the steps in the lamellar-to-micellar transition whenoptimal reconstitution was observed. The lamellar-to-micellar transition can be analyzed byturbidimetry as schematically shown. Depending upon the nature of the detergent, threemechanisms have been identified: direct protein incorporation into detergent-saturatedliposomes at the onset of solubilization (Rsat, mechanism 1), transfer from mixed micellesto detergent-saturated liposomes (Rsol, mechanism 2) or proteoliposome formation bymicellar coalescence (mechanism 3). Note that the final orientation of the protein dependson the mechanism of association.

proteoliposomes (Figure 4A-C). Many cap-sids of phages bound to their receptor werepartially or totally devoid of DNA, and, in-terestingly, always associated with proteoli-posomes that contained DNA, providing clearevidence of DNA transfer from the phage

758


J.-L. Rigaud

For a better understanding of the interac-tion of the phage with its receptors, informa-tion has been retrieved from 3-D images ofphage-proteoliposome complexes using elec-tron tomography (34). The principle of elec-tron tomography consists of the determina-tion of the 3-D structure of a single objectfrom a series of 2-D images of this objecttilted at various angles (35). The resolutionof the 3-D reconstructions demonstrated un-ambiguously that, after binding to its recep-tor, the straight fiber of the phage crossed thelipid bilayer and underwent a major confor-mational change, allowing the transfer of itsdouble-stranded DNA into the proteolipo-some (Figure 4C). On the basis of the crystalstructure of the FhuA receptor (36), we havebeen able to conclude that FhuA was onlyused as a docking site for the phage and thatthe tip of the phage tail acted like a DNA“injection needle” creating a passageway atthe periphery of FhuA.

Besides the important consequences ofour study for the molecular mechanisms ofphage infection, the experimental approachusing reconstituted proteoliposomes has im-portant applications in many biological con-texts. For example, it has been an attractivemodel system to study the mechanisms ofDNA condensation. To this end, proteolipo-somes were reconstituted in the presence ofdifferent amounts of spermine, a well-knownDNA condensing agent. In the presence of50 mM spermine-containing liposomes, theDNA strands transferred from the phagecapsids appeared condensed into character-istic toroidal structures that occupied a vol-ume smaller than the internal volume of theproteoliposome (Figure 4B). Increasing thenumber of DNA strands transferred into theproteoliposomes did not change the numberof toroids in the liposomes but increasedsignificantly the size of the toroid formed(37). From another point of view, the ap-proach could be important for gene therapyapplications, since the DNA-containing pro-teoliposomes produced in this way could

Figure 4. Use of FhuA-containing proteoliposomes to study the mechanism of phageinfection. I, The proteoliposome strategy. 1, Purification of FhuA, the receptor for phage T5on 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-electron micrographs demonstrating the DNA transfer from phage T5 (A) into FhuA-contain-ing proteoliposomes in the absence (B) or presence of 50 mM spermine (C). DNA isvisualized in electron-dense material either in full and partially empty icosahedral capsids orinside proteoliposomes in which DNA transfer has occurred. Note that, as opposed to whatis observed in the absence of spermine, DNA strands transferred into spermine-containingproteoliposomes appear to be condensed into a single toroidal structure. Panel D, Isosurfacerepresentation of the 3-D reconstruction of the phage bound to the proteoliposome asdetermined by electron tomography. The top of the vesicle has been cropped to reveal thetip of the phage visible inside the vesicle.

capsid into the proteoliposomes. More thanone complete genome (121,000 bp, 40 µm inlength) could be transferred into proteolipo-somes of 150 nm in diameter and severalgenomes were observed in proteoliposomesof 250 nm in diameter (Figure 4A).

Phage T5FhuA

Phage T5

PorinesFhuA

TonB

ExbB ExbBFhuB

FhuC

FhuA

759



serve as alternative vehicles for the transferof foreign genomes into eukaryotic cells.

Reconstitution of 2-D crystals

Elucidation of the structure of membraneproteins is a major challenge. Indeed, lessthan 30 original structures of membrane pro-teins have been solved to atomic resolution,a number which lags far behind that of solubleproteins, with several thousand accuratelydetermined structures.

While conventional 3-D crystallizationhas provided most of the atomic structures ofmembrane proteins, successful growth of3-D crystals of membrane proteins in deter-gent micelles is relatively infrequent. Thisdrawback is mainly related to the difficultyin maintaining a crystal lattice through thesole interactions between the hydrophilicdomains of the proteins, with the hydropho-bic domains being shielded by the detergentmicelles. To help with this problem, exten-sive studies have been performed to developnew strategies. In particular, a novel ap-proach has been introduced by Michel’sgroup (38), in which monoclonal antibodyfragments were specifically bound to pro-teins, increasing the interactions betweenthe hydrophilic domains of membrane pro-teins. More recently, bi-continuous lipidiccubic phases have been successfully used asmatrices for nucleation and further 3-D crys-tallization of different membrane proteins,with the idea that proteins in such membranemimetic systems should be more stable thanin micelles (39).

It is noteworthy, that reconstitution ofmembrane proteins into artificial membranesto form crystals confined to two dimensionshas opened a new way to solve their struc-ture (7-10). In several cases, the structuresobtained by electron crystallography allowedatomic models to be built. Besides theseexamples, many other 2-D crystals have beenanalyzed at resolution which allowed thesecondary structure to be clearly solved. Such

knowledge, combined with the informationfrom sequence analysis, frequently providedsignificant insight into the structure of theprotein. However, as for 3-D crystallization,the production of ordered 2-D crystals re-mains rare and many proteins continue toresist efforts. Except for few examples of 2-D crystals of membrane proteins in nativemembranes, 2-D crystals have to be recon-stituted from solubilized purified proteins.

2-D crystallization by detergent removal inbulk solution

To date, the most frequently employedstrategy for 2-D crystallization relies on thegeneral method of detergent-mediated re-constitution of proteoliposomes, but at lowlipid/protein ratios (1 protein molecule per10-50 lipid molecules for 2-D crystals ascompared to 1 protein molecule per 2000-10,000 lipid molecules for proteoliposomes).The strategy consists of starting with thepurified protein and a suitable combinationof lipids, both solubilized in detergent. Next,the detergent is removed from these micellarsolutions, resulting in the progressive forma-tion of lipid bilayers into which the proteinsincorporate and eventually crystallize. Sev-eral types of 2-D crystals, all containing acontinuous lipid bilayer in which proteinshave been incorporated, can be produced:vesicular crystals, tubular crystals in whichreconstituted proteins are helically orderedon the surface of a cylinder, and planarcrystalline sheets which in some instancescan lead to thin 3-D crystals upon the stack-ing of the 2-D arrays (Figure 5).

Different methods have been used toremove detergent, each presenting advan-tages and disadvantages (for a recent re-view, see Ref. 10). Dialysis is the mostwidely used technique in 2-D crystallizationtrials. Due to the necessity to scale down theamount of material, microdialysis deviceshave been used in the form of small com-partments (50-100 µl) dialyzed against large

760


J.-L. Rigaud

solubilized in various detergents and, impor-tantly, some of these 2-D crystals have beenuseful for high-resolution structural analysis(36,41-43).

The self-assembling process leading todensely packed vesicles and 2-D crystalsdepends upon a large matrix of variables. Inaddition to the composition of the solution(salt, pH, additives such as glycerol) andtemperature, variables of primary importanceare related to protein-protein, lipid-protein,detergent-protein and detergent-lipid inter-actions. As opposed to proteoliposome re-constitution, the rules for 2-D crystallizationby detergent removal are far from beinggeneralized and very few physicochemicalparameters have been determined as essen-tial for growing 2-D crystals of membraneproteins (10,44,45).

Induced 2-D crystallization

The previous reconstitution procedure im-plies that, upon detergent removal, proteinsmust be incorporated into the bilayer butalso find homogeneous optimal interactionsat this transition. These two simultaneousprocesses appear crucial to the outcome of2-D crystallization trials and may explain thedifficulty for many proteins to crystallize.For those membrane proteins difficult tocrystallize, one can consider bilayer forma-tion separately from crystallization and im-prove the strategy of “induced 2-D crystalli-zation”. In this case, reconstitutions shouldbe first performed at high lipid-to-proteinratios, leading to proteoliposomes with pro-teins homogeneously incorporated and ratherclosely packed. Then, in a second step, crys-tallization could be induced by physical treat-ment such as lipid temperature transition(46) or chemical agents including vanadatethat cross-links P-type ATPases (47) or phos-pholipase A2 that decreases the lipid-to-pro-tein ratio in the reconstituted proteolipo-some (48).

We have developed this approach to in-

Planar

Vesicular

Planarstacking

Tubular

Detergentremoval

DetergentLipid

Protein

Figure 5. 2-D crystallization of membrane proteins by detergent removal. Following deter-gent removal from a lipid-protein-detergent micellar solution (lipid-to-protein ratios below 1w/w), different 2-D crystals can be produced.

2-D crystals

buffer volumes. The dialysis method hasbeen successfully applied to many mem-brane proteins, but may not be well suitedfor detergents with low critical micellar con-centrations, that require a long time for di-alysis.

The method of detergent removal by hy-drophobic adsorption onto polystyrene beads,previously shown to be very efficient forproteoliposome reconstitution (24), has beenrecently demonstrated to be a powerful alter-native to conventional dialysis for 2-D crys-tallization trials (40). Using radioactive de-tergents, the method has been calibrated pre-cisely in terms of the adsorptive capacity ofthe beads and the rates of detergent removal.The mechanisms underlying detergent ad-sorption onto beads have been analyzed andgeneral rules for the use of polystyrene beadshave been proposed. Sufficient reproduc-ibility can be now obtained with experienceand careful handling. An important benefitof this new strategy has been to produce new2-D crystals of several membrane proteins

761



duce 2-D crystallization of Ca-ATPase afterreconstitution into proteoliposomes (47). Ca-ATPase-dense proteoliposomes, producedthrough the "step-by-step" reconstitutionstrategy, were demonstrated to be ideal toinduce further crystallization by the additionof vanadate, leading to very long orderedcrystalline tubes. These tubular crystals wereproduced by very few preparations of nativesarcoplasmic reticulum vesicles, and the un-known parameters that control the growth oflong tubes from spherical vesicles could beeasily analyzed in our reconstituted system.In particular, we found that a critical step tomake extended tubular crystals from smallproteoliposomes was the inclusion of specif-ic lipids that destabilize bilayers.

2-D crystallization of membrane proteins onfunctionalized lipid layers

Considerable interest exists in develop-ing new strategies to increase the successrate of 2-D crystallization. The technique of2-D crystallization on lipid layers is based ona specific interaction between the proteinand specific ligands coupled to lipid mol-ecules incorporated into a planar lipid film atthe air-water interface. This method hasproven to be a successful approach to 2-Dcrystallization of soluble proteins, providingimportant structural information (49,50). Thequestion as to whether this method was alsoapplicable to membrane proteins was a chal-lenge.

As the result of a careful study, we haverecently been able to produce on planar lip-idic templates the first 2-D crystals of tworadically different membrane proteins, FhuAand F0F1 ATPase, demonstrating the feasi-bility of this new strategy for membraneproteins (51,52). A model for the mechan-isms of 2-D crystallization of membrane pro-teins on a lipid layer has been proposed,which include three steps (Figure 6): i) bind-ing of protein-lipid-detergent micelles to thelipid layer. Since many membrane proteins

are overproduced and purified using recom-binant proteins containing a stretch of con-tinuous histidine residues (His-tag), a nickel-chelating lipid has been employed as a gen-eral adaptor molecule that will link any His-tag protein; ii) reconstitution of a bilayeraround the pre-bound proteins by detergentremoval using polystyrene Bio-Beads in thesolution below the lipid layer; iii) diffusionof the lipid-protein complexes and further 2-D crystallization.

This innovative strategy opens a newpromising field for membrane protein struc-ture determination since: i) it may increasethe chances of success to produce 2-D crys-tals of proteins difficult to crystallize byconventional methods; ii) it permits struc-ture analysis with smaller amounts of mate-rial. Indeed, since this strategy induces pro-tein concentration at the interface, this leadsto the use of less than 1 µg protein as com-pared to 25-50 µg protein required by theclassical bulk method. This will be of par-

A

B

C

Figure 6. 2-D crystallization ofmembrane proteins on the lipidlayer at the air-water interface.The three stages in the processof 2-D crystallization on lipid lay-ers. A, Binding of protein-lipid-detergent micelles to the func-tionalized lipid layer spread atthe air-water interface. B, Re-constitution of proteins into a li-pid bilayer upon detergent re-moval. C, 2-D crystallization.

762


J.-L. Rigaud

ticular help to laboratories studying mem-brane proteins which are currently in theearly stages of developing procedures foroverexpression and large-scale production;iii) although a nickel-chelating lipid has beenemployed as a general adaptor that can linkany His-tag protein at the surface, the strat-egy has been extended to other types ofspecific molecular recognition such as elec-trostatic interactions between proteins andlipids (52). Also it will be important to ex-tend the strategy to other types of specificmolecular recognition at the surface betweenproteins and lipids using other lipid-coupledphysiological ligands.

Structural analysis

Electron crystallography. The method ofchoice for analyzing the results of 2-D crys-tallization is electron microscopy. The quick-est specimen preparation is negative stainingwhich yields excellent contrast but does notpreserve high-resolution details. Once theoptimal conditions of crystallization havebeen determined by negative staining, cryo-electron microscopy is preferred because itpreserves the native structure of the proteinto high resolution. One drop of the sample isdeposited onto a carbon-coated grid, blottedwith a filter paper and rapidly frozen in aliquid ethane bath (53). The frozen samplesneed to be maintained at -170ºC throughoutobservation under the microscope to pre-serve them against dehydration in the vacuumof the microscope and against radiation dam-age. Finally, the best method to assay thedegree of order is electron diffraction, whichis generally used to assess high-resolutionstructure analysis of good quality 2-D crystals.

Once the 2-D crystals have been pro-duced, the structure of the protein can bedetermined by electron crystallography andimage processing to a resolution that de-pends on the quality of the 2-D crystal (11).The best electron micrographs are selectedby optical diffraction with a laser diffracto-

meter, digitized on a microdensitometer andcomputer processed. The digitized imagesare Fourier transformed and phases and am-plitudes of structure factors selected andaveraged. In the course of image processing,crystal defects (lattice distortion) and imagedefects (lack of focus and astigmatism) arecorrected. The inverse Fourier transform thengenerates a projection map which can berepresented as a topographic map in whichthe hills would reflect higher electron densi-ties. To determine the 3-D structure, electrondiffraction patterns and electron images of2-D crystals are recorded with the specimenat various tilt angles. By combining the struc-ture factors which contribute to these projec-tions in 3-D, a 3-D data set is compiled. Theinverse transform of this data set is the 3-Dmap of the membrane protein.

Atomic force microscopy. As a comple-mentary technique, atomic force microscopy(AFM) has become an important tool toinvestigate transmembrane proteins underphysiological conditions. AFM measures theheights of biological samples to obtain topo-graphic maps of protein structures protrud-ing from the membrane at a lateral resolutionof ~8 Å and a vertical resolution of ~1 Å(54). Specifically, high-resolution topo-graphic maps before and after proteolyticcleavage of distinct protein domains havepermitted the localization of these domainsand the assignment of the sidedness of thewhole transmembrane molecule. This ap-proach was nicely illustrated by a recentstudy on the light-harvesting complex 2(LH2) of Rubrivivax gelatinosus, the acces-sory antenna proteins in the bacterial photo-synthetic apparatus built of αß-heterodimers(55). AFM has been used to measure thesurface topography of LH2 in a reconstitutedmembrane and to locate the C-terminal hy-drophobic extension of 21-amino acid resi-dues in the α-subunit. High-resolution topo-graphic maps revealed a nonameric organi-zation of the αß-heterodimers, regularlypacked as cylindrical complexes incorpo-

763



rated into the membrane in both orientations(Figure 7). Native LH2 showed one surfacewhich protruded by ~5 Å and one whichprotruded by ~14 Å from the membrane.Topographic maps of samples reconstitutedwith thermolysin-digested LH2 have revealeda height reduction of the strongly protrudingsurface to ~9 Å, while weakly protrudingrings remained unchanged. These resultsdemonstrated the extrinsic localization ofthe C-terminal domain of the α-subunit of R.gelatinosus and allowed the periplasmic sur-face to be assigned.

Finally, the AFM has developed to thepoint that it permits to unfold membraneproteins, to detect flexible and stable extrin-sic domains and to monitor function-related

conformational changes by the addition ofsubstrate molecules to the scanning buffer.Excellent reviews from Engel’s group (54-57) are available that illustrate various appli-cations of AFM.

Conclusions

Reconstitution of membrane proteins intoliposomes has played an important role andshould continue to be a potentially powerfultool to be used to identify the mechanism ofaction of membrane proteins. As shown inthis review, it appears that reconstitution nolonger is “black magic” and the prospects ofachieving optimal reconstitution of proteoli-posomes are obviously good, with the use of

Figure 7. High-resolution atomicforce microscopy topographicmaps of the LH2 complex fromRubrivivax gelatinosus. a, To-pographic maps of the strongly(~14 Å) protruding surface ofthe native LH2. b, Topographicmaps of the weakly (~5 Å) pro-truding surface of the nativeLH2. c, Topographic maps ofthe strongly (~9 Å) protrudingsurface of the digested LH2. d,Topographic maps of theweakly (~5 Å) protruding sur-face of the digested LH2. In-sets show average images ofLH2 complexes calculatedfrom the raw data shown ineach corresponding panel. Pro-teolysis only influenced the to-pography of the strongly pro-truding surface, allowing to lo-calize the C-terminal position ofthe α-subunit.

764


J.-L. Rigaud

reliable methods.Despite significant progress in the last

few years, 3-D crystallization of membraneproteins has proved to be difficult and slow,and alternative approaches to structure de-termination have become essential. Electroncrystallography, associated with 2-D crys-tallization, is a technique that has producedmany structures and is definitely a viablealternative strategy to X-ray crystallogra-phy. However, for electron crystallographyto become even more important, enormousefforts are required: i) for a deeper under-standing of the 2-D crystallogenesis process,and ii) for the development of new approachesto be evaluated, refined and applied to avail-able proteins.

The future of membrane protein recon-stitution appears bright in the light of thesteadily increasing number of membrane pro-

teins, which have been identified in the se-quencing of the genomes of different organ-isms. Powerful methods of gene overexpres-sion are now available which will permit theproduction of new interesting membrane pro-teins.

Acknowledgments

I would like to thank Prof. A. Pereira andthe Brazilian Society of Biochemistry andMolecular Biology for their kind invitationto attend the meeting in Caxambu, whichgives me the great opportunity to establishcontact with many laboratories in Brazil.Special thanks are due to D. Lévy, O. Lam-bert, M. Chami, S. Scheuring and B. Pitardwho have been directly involved in many ofthe reconstitution studies described in thisreview.

References

1. Jones DT (1998). Do transmembrane pro-tein superfolds exist? FEBS Letters, 423:281-285.

2. Racker E (1982). Reconstitution of mem-brane processes. Methods in Enzymol-ogy, 55: 699-711.

3. Eytan GD (1981). Use of liposomes forreconstitution of biological functions. Bio-chimica et Biophysica Acta, 694: 185-202.

4. Cornelius F (1991). Functional reconstitu-tion of the sodium pump. Kinetics of ex-change reactions performed by reconsti-tuted Na/K-ATPase. Biochimica et Biophy-sica Acta, 1071: 19-66.

5. Rigaud JL, Pitard B & Lévy D (1995). Re-constitution of membrane proteins intoliposomes: application to energy-trans-ducing membrane proteins. Biochimica etBiophysica Acta, 1231: 223-246.

6. Ostermeier C & Michel H (1997). Crystal-lization of membrane proteins. CurrentOpinion in Structural Biology, 7: 697-701.

7. Kuhlbrandt W (1992). Two-dimensionalcrystallization of membrane proteins.Quarterly Reviews in Biophysics, 25: 1-49.

8. Jap BK, Zulauf M, Scheybani T, Hefti A,Baumeister W, Aebi U & Engel A (1992).2D crystallization: from art to science.Ultramicroscopy, 46: 45-84.

9. Hasler L, Heymann JB, Engel A, Kistler J

& Walz T (1998). 2D crystallization ofmembrane proteins: rationales and ex-amples. Journal of Structural Biology, 121:162-171.

10. Rigaud JL, Chami M, Lambert O, Lévy D& Ranck J (2000). Use of detergents in 2Dcrystallization of membrane proteins. Bio-chimica et Biophysica Acta, 1508: 112-128.

11. Walz T & Grigorieff N (1998). Electroncrystallography of two-dimensional crys-tals of membrane proteins. Journal ofStructural Biology, 121: 142-161.

12. Szoka F & Papahadjopoulos D (1980).Comparative properties and methods ofpreparation of lipid vesicles. Annual Re-view of Biophysics and Bioengineering, 9:467-508.

13. Lemaire M, Champeil P & Moller JV(2000). Interaction of membrane proteinsand lipids with solubilizing detergents.Biochimica et Biophysica Acta, 1508: 86-111.

14. Lichtenberg D (1985). Characterization ofthe solubilization of lipid bilayers by sur-factants. Biochimica et Biophysica Acta,821: 470-478.

15. Silvius JR (1992). Solubilization and func-tional reconstitution of biomembranecomponents. Annual Review of Biophys-ics and Biomolecular Structure, 21: 323-

348.16. Rigaud J-L & Lévy D (2002). Reconstitu-

tion of membrane proteins into proteo-liposomes. Methods in Enzymology : Li-posomes, Part B (in press).

17. Lévy D, Gulick A, Seigneuret M & RigaudJ-L (1990). Phospholipid vesicle solubiliza-tion and reconstitution by detergent. Sym-metrical analysis of the two processesusing octaethylene glycol mono-N-dode-cyl ether. Biochemistry, 29: 9480-9488.

18. Paternostre MT, Roux M & Rigaud J-L(1988). Mechanisms of membrane pro-tein insertion into liposomes during re-constitution procedures involving the useof detergents. 1. Solubilization of largeunilamellar liposomes (prepared by re-verse-phase evaporation) by Triton X-100,octyl glucoside and sodium cholate. Bio-chemistry, 27: 2668-2677.

19. Lambert O, Lévy D, Ranck JL, Leblanc G& Rigaud J-L (1998). A new gel-like phasein dodecylmaltoside-lipid mixtures: impli-cations in solubilization and reconstitutionstudies. Biophysical Journal, 74: 918-930.

20. Rigaud J-L, Paternostre MT & Bluzat A(1988). Mechanisms of membrane pro-tein insertion into liposomes during re-constitution procedures involving the useof detergents. 2. Incorporation of the light-driven proton pump bacteriorhodopsin.

765



Biochemistry, 27: 2677-2688.21. Richard P, Rigaud J-L & Gräber P (1990).

Reconstitution of CF0F1 into liposomesusing a new reconstitution procedure.European Journal of Biochemistry, 193:921-930.

22. Lévy D, Gulik A, Bluzat A & Rigaud J-L(1992). Reconstitution of the sarcoplas-mic reticulum Ca2+ATPase: mechanismsof membrane protein insertion into lipo-somes during reconstitution proceduresinvolving the use of detergents. Biochimi-ca et Biophysica Acta, 1107: 283-298.

23. Rigaud J-L, Lévy D, Mosser G & LambertO (1998). Detergent removal by Bio-Beads. Applications to membrane proteinreconstitution and 2D crystallization. Eu-ropean Biophysics Journal, 27: 305-319.

24. Lévy D, Bluzat A, Seigneuret M & RigaudJ-L (1990). A systematic study of lipo-some and proteoliposome reconstitutioninvolving Bio-Bead-mediated Triton X-100removal. Biochimica et Biophysica Acta,1025: 179-190.

25. Seigneuret M & Rigaud J-L (1986). Analy-sis of passive and light-driven ion move-ments in large bacteriorhodopsin lipo-somes reconstituted by reverse-phaseevaporation. 2. Influence of passive per-meability and back-pressure effects uponlight-induced proton uptake. Biochemis-try, 25: 6723-6729.

26. Lévy D, Seigneuret M, Bluzat A & RigaudJ-L (1990). Evidence for proton counter-transport by the sarcoplasmic reticulumCa2+ATPase during calcium transport inreconstituted proteoliposomes with “low”ionic permeability. Journal of BiologicalChemistry, 265: 19524-19533.

27. Hao L, Rigaud J-L & Inesi G (1994). Ca2+/H+ countertransport and electrogenicityin proteoliposomes containing erythro-cyte plasma membrane, Ca2+ATPase andexogenous lipids. Journal of BiologicalChemistry, 269: 14268-14275.

28. Salvador J, Inesi G, Rigaud J-L & MataAM (1998). Ca2+ transport by reconsti-tuted synaptosomal ATPase is associatedwith H+ countertransport and net chargedisplacement. Journal of Biological Chem-istry, 273: 18230-18234.

29. Pitard B, Richard P, Dunach M, Girault G& Rigaud J-L (1996). ATP synthesis by theF0F1 ATP synthase from bacillus PS3 re-constituted into liposomes with bacterio-rhodopsin. Factors defining the optimalreconstitution of ATP synthases with bac-teriorhodopsin. European Journal of Bio-chemistry, 235: 769-778.

30. Pitard B, Richard P & Rigaud J-L (1996).ATP synthesis by the F0F1 ATP synthase

from bacillus PS3 co-reconstituted into li-posomes with bacteriorhodopsin. Rela-tionships between proton motive forceand ATP synthesis. European Journal ofBiochemistry, 235: 779-788.

31. Richard P, Pitard B & Rigaud J-L (1995).ATP synthesis by the F0F1 ATPase fromthe thermophilic bacillus PS3 co-reconsti-tuted with bacteriorhodopsin into lipo-somes. Evidence for stimulation of ATPsynthesis by ATP bound to a non-catalyticsite. Journal of Biological Chemistry, 270:21571-21578.

32. Steinberg-Yfrach G, Rigaud J-L, MooreLA, Gust D & Moore TA (1998). Light-driven, F0F1 ATP synthase-catalyzed syn-thesis of ATP by an artificial photosyn-thetic membrane. Nature, 392: 479-482.

33. Lambert O, Plançon L, Rigaud J-L &Letellier L (1998). Protein-mediated DNAtransfer into liposomes. Molecular Micro-biology, 30: 761-765.

34. Boehm J, Lambert O, Letellier L,Baumeister W & Rigaud J-L (2001). FhuA-mediated phage genome transfer into li-posomes: a cryo-electron tomographystudy. Current Biology, 11: 1168-1175.

35. Baumeister W, Grimm R & Walz J (1999).Electron tomography of molecules andcells. Trends in Cell Biology, 9: 81-85.

36. Lambert O, Moeck G, Lévy D, Plançon L,Letellier L & Rigaud J-L (1999). A pro-jected structure of FhuA, a ligand-gatedchannel of E. coli outer membrane pro-tein. Journal of Structural Biology, 126:145-155.

37. Lambert O, Letellier L, Gelbart WM &Rigaud J-L (2000). DNA delivery by phageas a strategy for encapsulating toroidalcondensates of arbitrary size into lipo-somes. Proceedings of the National Acad-emy of Sciences, USA, 97: 7248-7253.

38. Ostermeier C, Iwata J & Michel H (1996).Cytochrome c oxidase. Current Opinion inStructural Biology, 6: 460-466.

39. Nollert P, Qiu H, Gaffrey M, RosenbuschJP & Landau EM (2001). Molecularmechanisms for the crystallization ofbacteriorhodopsin in lipidic cubic phases.FEBS Letters, 504: 179-186.

40. Rigaud J-L, Mosser G, Lacapère JJ, LévyD, Oloffson A & Ranck JL (1997). Bio-Beads: an efficient strategy for 2D crystal-lization of membrane proteins. Journal ofStructural Biology, 118: 226-235.

41. Mosser G, Breyton C, Oloffson A, PopotJL & Rigaud J-L (1997). A 8A projectionmap of cytochrome b6f complex fromChlamydomonas reinhartdii. Journal ofBiological Chemistry, 272: 20263-20268.

42. Lacapère JJ, Stokes D, Olofsson A &

Rigaud J-L (1998). Two-dimensional crys-tallization of Ca2+ATPase by total deter-gent removal. Biophysical Journal, 75:1319-1329.

43. Saint N, Lacapère JJ, Ghazi A, Martinac B& Rigaud J-L (1998). A hexameric trans-membrane pore revealed by 2D crystalli-zation of the large mechanosensitive ionchannel (MscL) of E. coli. Journal of Bio-logical Chemistry, 273: 14667-14670.

44. Dolder M, Engel A & Zulauf M (1996). Themicelle to vesicle transition of lipids anddetergents in the presence of a mem-brane protein: towards a rationale for 2Dcrystallization. FEBS Letters, 382: 203-208.

45. Engel A, Hoenger A, Hefti A, Henn C,Ford RC, Kisler J & Zulauf M (1992). As-sembly of 2D membrane protein crystals:dynamics, crystal order and fidelity ofstructure analysis by electron microscopy.Journal of Structural Biology, 109: 219-234.

46. Cherry RJ, Muller R, Henderson R & HeynMP (1997). Temperature-dependent ag-gregation of bacteriorhodopsin in dipalmi-toyl and dimyristoylphosphatidylcholine.Journal of Molecular Biology, 121: 283-298.

47. Young HS, Rigaud J-L, Lacapère JJ,Reddy LG & Stokes DL (1997). How tomake tubular crystals by reconstitution ofdetergent-solubilized Ca-ATPase. Bio-physical Journal, 72: 2545-2558.

48. Mannella C (1984). Phospholipase-induced crystallization of channels in mi-tochondrial outer membranes. Science,224: 165-166.

49. Kornberg RD & Darst SA (1991). Two-dimensional crystals of proteins on lipidlayers. Current Opinion in Structural Biol-ogy, 1: 642-646.

50. Chiu W, Avila-Sakar AJ & Schmid MF(1997). Electron crystallography of macro-molecular periodic arrays on phospholipidmonolayers. Advances in Biophysics, 34:161-172.

51. Lévy D, Mosser G, Lambert O, Moeck G,Bald D & Rigaud J-L (1999). 2D crystalliza-tion on lipid layer: a successful approachfor membrane proteins. Journal of Struc-tural Biology, 127: 44-52.

52. Lévy D, Chami M & Rigaud J-L (2001).Two dimensional crystallization of mem-brane proteins: the lipid layer strategy.FEBS Letters, 504: 187-193.

53. Dubochet J, Adrian M, Chang JJ, HomoJC, Lepault J, McDonall AW & Schultz P(1988). Cryo-electron microscopy of vitri-fied specimen. Quarterly Reviews of Bio-physics, 21: 129-228.

766


J.-L. Rigaud

54. Stahlberg H, Fotiadis D, Scheuring S,Remigny H, Braun T, Mitsuoka Y,Fujiyoshi Y & Engel A (2001). Two-dimen-sional crystals: a powerful approach toassess structure, function and dynamicsof membrane proteins. FEBS Letters,504: 166-172.

55. Scheuring S, Reiss Husson F, Engel A,Rigaud J-L & Ranck JL (2001). High reso-lution AFM topographs of Rubrivivax ge-latinosus light-harvesting complex LH2.EMBO Journal, 20: 3029-3035.

56. Engel A & Muller DJ (2000). Observingsingle molecules at work with the atomic

force microscope. Nature Structural Biol-ogy, 7: 715-718.

57. Heymann JB, Müller DJ, Mitsuoka K &Engel A (1997). Electron and atomic forcemicroscopy of membrane proteins. Cur-rent Opinion in Structural Biology, 7: 543-549.

membrane proteins: functional and structural studies using ... · reconstitution of membrane...

Documents