Determination of the number of detergent molecules associated with the reaction center protein isolated from the photosynthetic bacterium Rhodopseudomonas viridis. Effects of the amphiphilic molecule 1,2,3-heptanetriol

Refined definition of the critical micelle concentration and application to alkyl maltosides used in membrane protein research

The critical micelle concentration (CMC) of nonionic detergents is defined as the breaking point in the monomer concentration as a function of the total detergent concentration, identified by setting the third derivate of this function to zero. Combined with a mass action model for micelle formation, this definition yields analytic formulae for the concentration ratio of monomers to total detergent at the CMC and the relationship between the CMC and the free energy of micellization g mic. The theoretical breaking point is shown to coincide with the breaking point of the experimental titration curve, if the fluorescence enhancement of 8-anilino-1-naphthalene-sulfonic acid (ANS) or a similar probe dye is used to monitor micelle formation. Application to a series of n-alkyl-β-d-maltosides with the number of carbon atoms in the alkyl chain ranging from 8 to 12 demonstrates the good performance of a molecular thermodynamic model, in which the free energy of micellization is given by g mic = σΦ + g pack + g st. In this model, σ is a fit parameter with the dimension of surface tension, Φ represents the change in area of hydrophobic molecular surfaces in contact with the aqueous phase, and g pack and g st are contributions, respectively, from alkyl chain packing in the micelle interior and steric repulsion of detergent head groups. The analysis of experimental data from different sources shows that varying experimental conditions such as co-solutes in the aqueous phase can be accounted for by adapting only σ, if the co-solutes do not bind to the detergent to an appreciable extent. The model is considered a good compromise between theory and practicability to be applied in the context of in vitro investigations of membrane proteins.

Lipidic cubic phase serial femtosecond crystallography structure of a photosynthetic reaction centre

Serial crystallography relies upon the growth of microcrystals at high concentration. An LCP crystallization protocol for the Blastochloris viridis photosynthetic reaction centre that utilizes seeding from detergent-grown crystals is reported.

Serial crystallography is a rapidly growing method that can yield structural insights from microcrystals that were previously considered to be too small to be useful in conventional X-ray crystallography. Here, conditions for growing microcrystals of the photosynthetic reaction centre of Blastochloris viridis within a lipidic cubic phase (LCP) crystallization matrix that employ a seeding protocol utilizing detergent-grown crystals with a different crystal packing are described. LCP microcrystals diffracted to 2.25 Å resolution when exposed to XFEL radiation, which is an improvement of 0.15 Å over previous microcrystal forms. Ubiquinone was incorporated into the LCP crystallization media and the resulting electron density within the mobile Q_B pocket is comparable to that of other cofactors within the structure. As such, LCP microcrystallization conditions will facilitate time-resolved diffraction studies of electron-transfer reactions to the mobile quinone, potentially allowing the observation of structural changes associated with the two electron-transfer reactions leading to complete reduction of the ubiquinone ligand.

Visualizing a protein quake with time-resolved X-ray scattering at a free-electron laser

We describe a method to measure ultrafast protein structural changes using time-resolved wide-angle X-ray scattering at an X-ray free-electron laser. We demonstrated this approach using multiphoton excitation of the Blastochloris viridis photosynthetic reaction center, observing an ultrafast global conformational change that arises within picoseconds and precedes the propagation of heat through the protein. This provides direct structural evidence for a 'protein quake': the hypothesis that proteins rapidly dissipate energy through quake-like structural motions.

Crystallization of membrane proteins

The crystallization of membrane proteins is an essential technique for the determination of atomic models of three-dimensional structures by X-ray crystallography. The compositions of solutions of purified membrane proteins are altered, so as to transiently induce supersaturation, a requirement for crystal nucleation and growth. The establishment of the precise optimal crystallization conditions has to be performed individually by a combination of systematic approaches and trial-and-error. These procedures have become more efficient due to the introduction of laboratory automation. Here we describe the crystallization of the dihaem-containing quinol:fumarate reductase (QFR) membrane protein complex and illustrate key factors important in the screening process.

Detergent organisation in solutions and in crystals of membrane proteins

The use of neutron scattering in studying the organisation of detergents in pure micelles, in protein/detergent mixed micelles and in crystals of membrane proteins, is reviewed. Small angle scattering has been used to study the size, shape and composition of pure and mixed protein/detergent micelles as well as the effects of adding small amphiphiles. The technique of contrast variation applied to single crystals is described and its application to the determination of the organization of detergent in single crystals of membrane proteins is discussed. A better understanding of protein/detergent interactions should help in producing crystals of membrane proteins more easily as well as clues to the nature of protein/lipid interactions in vivo.

Influence of Cardiolipin on the Functionality of the QA Site of the Photosynthetic Bacterial Reaction Center

The effect of cardiolipin on the functionality of the Q(A) site of a photosynthetic reaction center (RC) was studied in RCs from the purple non-sulfur bacterium Rhodobacter sphaeroides by means of time-resolved absorbance measurements. The binding of the ubiquinone-10 to the Q(A) site of the RC embedded in cardiolipin or lecithin liposomes has been followed at different temperatures and phospholipid loading. A global fit of the experimental data allowed us to get quite reliable values of the thermodynamic parameters joined to the binding process. The presence of cardiolipin does not affect the affinity of the Q(A) site for ubiquinone but has a marked influence on the rate of P+QA(-) --> PQA electron transfer. The P+QA(-) charge recombination kinetics has been examined in liposomes made of cardiolipin/lecithin mixtures and in detergent (DDAO) micelles doped with cardiolipin. The electron-transfer rate constant increases upon cardiolipin loading. It appears that the main effect of cardiolipin on the electron transfer can be ascribed to a destabilization of the charge-separated state. Results obtained in micelles and vesicles follow the same titration curve when cardiolipin concentration evaluated with respect to the apolar phase is used as a relevant variable. The dependence of the P+QA(-) recombination rate on cardiolipin loading suggests two classes of binding sites. In addition to a high-affinity site (compatible with previous crystallographic studies), a cooperative binding, involving about four cardiolipin molecules, takes place at high cardiolipin loading.

Extinction coefficients and critical solubilisation concentrations of photosystems I and II from Thermosynechococcus elongatus

The absorption properties of chlorophyll a (Chla) in active core complexes of photosystems I (PSI) and II (PSII) isolated in high purity from the thermophilic cyanobacterium Thermosynechococcus elongatus were correlated with those of extracts in 80% acetone to determine effective extinction coefficients of protein-bound Chla and molar extinction coefficients of core complexes and reaction centers (RC). These coefficients allow a quick determination of Chla and protein concentrations from steady-state absorption spectra of intact samples without the need for pigment extraction and protein destruction. In the visible range, epsilon(680)(p) = 57 mM(-1) cm(-1) for trimeric PSI (PSIt) and epsilon(674)(p) = 70 mM(-1) cm(-1) for dimeric (PSIId) and monomeric (PSIIm) PSII (error +/-6%; superscript "p" refers to Chla bound to intact protein, subscripts are the peak maxima in nm). The integral extinction coefficient phi(p) = 2.8 nm microM(-1) cm(-1) for the wavelength interval between 550 and 800 nm and the extinction coefficient epsilon(B)(p) = 14 mM(-1) cm(-1) for the smaller absorption maximum (B = 632 nm for PSI and 627 nm for PSII) were found to be essentially the same for both types of PS. The coefficients of PSIt are shown to remain unaltered when 65% (v/v) of the buffer is replaced with glycerol. Molar extinction coefficients of core complexes were determined using Chla/RC ratios of 96+/-1 for PSI and 35+/-2 for PSII based on X-ray data. In addition, the critical solubilisation concentration of n-dodecyl-beta-d-maltoside (betaDM), necessary to keep the core complexes in solution, was determined by turbidimetric titrations. It was found that at least approximately 500 betaDM molecules per PSIt ( approximately 2 betaDM per Chla) and 190 betaDM molecules per PSIIm ( approximately 5 betaDM per Chla, also for PSIId) in excess of the critical micelle concentration of 0.16 +/- 0.03 mM are necessary for a complete solubilisation of the core complexes.

Two-dimensional crystallogenesis of transmembrane proteins

Two-dimensional crystallogenesis is a crucial step in the long road that leads to the determination of macromolecules structure via electron crystallography. The necessity of having large and highly ordered samples can hold back the resolution of structural works for a long time, and this, despite improvements made in electron microscopes or image processing. Today, finding good conditions for growing two-dimensional crystals still rely on either "biocrystallo-cooks" or on lucky ones. The present review presents the field by first describing the different crystals that one can encounter and the different crystallisation methods used. Then, the effects of different components (such as protein, lipids, detergent, buffer, and temperature) and the different methods (dialysis, hydrophobic adsorption) are discussed. This discussion is punctuated by correspondences made to the world of three-dimensional crystallogenesis. Finally, a guide for setting up 2D crystallogenesis experiments, built on the discussion mentioned before, is proposed to the reader. More than giving recipes, this review is meant to open up the discussions in this field.

Cumulant analysis of charge recombination kinetics in bacterial reaction centers reconstituted into lipid vesicles

The kinetics of charge recombination between the primary photoxidized donor (P(+)) and the secondary reduced quinone acceptor (Q(B)(-)) have been studied in reaction centers (RCs) from the purple photosynthetic bacterium Rhodobacter sphaeroides incorporated into lecithin vesicles containing large ubiquinone pools over the temperature range 275 K </= T </= 307 K. To account for the non-exponential kinetics of P(+) re-reduction observed following a flash, a new approach has been developed, based on the following assumptions: 1) the exchange of quinone between different vesicles is negligible; 2) the exchange of quinone between the Q(B) site of the RC and the quinone pool within each single vesicle is faster than the return of the electron from the primary reduced acceptor Q(A)(-) to P(+); 3) the size polydispersity of proteoliposomes and the distribution of quinone molecules among them result in a quinone concentration distribution function, P(Q). The first and second moments of P(Q) have been evaluated from the size distribution of proteoliposomes probed by quasi-elastic light scattering (mean radius, <R> = (50 +/- 15) nm). Following these premises, we describe the kinetics of P(+)Q(B)(-) recombination with a truncated cumulant expansion and relate it to P(Q) and to the free energy changes for Q(A)(-)Q(B) --> Q(A)Q(B)(-) electron transfer (DeltaG(AB)(o)) and for quinone binding (DeltaG(bind)(o)) at Q(B). The model accounts well for the temperature and quinone dependence of the charge recombination kinetics, yielding DeltaG(AB)(o) = -7.67 +/- 0.05 kJ mol(-1) and DeltaG(bind)(o) = -14.6 +/- 0.6 kJ mol(-1) at 298 K.

X-Ray diffraction analysis of three-dimensional crystals of bovine rhodopsin obtained from mixed micelles

Rhodopsin, a prototypic G protein-coupled receptor responsible for absorption of photons in retinal rod photoreceptor cells, was selectively extracted from bovine rod outer segment membranes, employing mixed micelles of nonyl beta-d-glucoside and heptanetriol. Highly purified rhodopsin was crystallized from solutions containing varying amounts of detergent and amphiphile. The crystals contained ground state rhodopsin molecules as judged by their red color and the linear dichroism originating from the 11-cis-retinal chromophore. However, when exposed to visible light, even at 4 degrees C, rhodopsin was bleached and the crystals decomposed. Reflections in the diffraction pattern were observed out to 3.5-A resolution at 100 K for the most ordered crystals. Diffraction data have been processed to 3.85-A resolution. The symmetry of the diffraction pattern and the systematic absences indicate that the crystals have tetragonal symmetry, space group P4(1)22 or P4(3)22, a = b = 96.51 A, c = 148.55 A. A value of 4.12 A(3)/Da for V(M) was obtained for one monomer in the asymmetric unit (eight molecules per unit cell). Our study is the first characterization of a three-dimensional crystal of a G protein-coupled receptor and may be valuable for future structural studies on related receptors of this important superfamily.

Crystallization of membrane proteins

Five new membrane protein structures have been determined since 1995 using X-ray crystallography: bacterial light-harvesting complex; bacterial and mitochondrial cytochrome c oxidases; mitochondrial bc1 complex; and alpha-hemolysin. These successes are partly based on advances in the crystallization procedures for integral membrane proteins. Variation of the size of the detergent micelle and/or increasing the size of the polar surface of the membrane protein is the most important route to well-ordered membrane protein crystals. The use of bicontinuous lipidic cubic phases also appears to be promising.

Influence of iron-removal procedures on sequential electron transfer in photosynthetic bacterial reaction centers studied by transient EPR spectroscopy

Electron spin polarized electron paramagentic resonance (ESP EPR) spectra were obtained with deuterated iron-removed photosynthetic bacterial reaction centers (RCs) to specifically investigate the effect of the rate of primary charge separation, metal-site occupancy, and H-subunit content on the observed P865+QA- charge-separated state. Fe-removed and Zn-substituted RCs from Rb. sphaeroides R-26 were prepared by refined procedures, and specific electron transfer rates (kQ) from the intermediate acceptor H- to the primary acceptor QA of (200 ps)-1 vs (3-6 ns)-1 were observed. Correlation of the transient EPR and optical results shows that the observed slow kQ rate in Fe-removed RCs is H-subunit-independent, and, in some cases, independent of Fe-site occupancy as Zn2+ substitution does not ensure retention of the native kQ. In addition, shifts in the optical spectrum of P865 and differences in the high-field region of the Q-band ESP spectrum for Fe-removed RCs with slow kQ indicate possible structural changes near P865. The experimental X-band and Q-band spin-polarized EPR spectra for deuterated Fe-removed RCs where kQ is at least 15-fold slower at room temperature than the (200 ps)-1 rate observed for native Fe-containing RCs have different relative amplitudes and small g-value shifts compared to the spectra of Zn-RCs which have a kQ unchanged from native RCs. These differences reflect the trends in polarization predicted from the sequential electron transfer polarization (SETP) model [Morris et al. (1995) J. Phys. Chem. 99, 3854-3866; Tang et al. (1996) Chem. Phys. Lett. 253, 293-298]. Thus, SETP modeling of these highly resolved ESP spectra obtained with well-characterized proteins will provide definitive information about any light-induced structural changes of P865, H, and QA that occur upon formation of the P865+QA- charge-separated state.

The interaction of quinone and detergent with reaction centers of purple bacteria. I. Slow quinone exchange between reaction center micelles and pure detergent micelles

The kinetics of light-induced electron transfer in reaction centers (RCs) from the purple photosynthetic bacterium Rhodobacter sphaeroides were studied in the presence of the detergent lauryldimethylamine-N-oxide (LDAO). After the light-induced electron transfer from the primary donor (P) to the acceptor quinone complex, the dark re-reduction of P+ reflects recombination from the reduced acceptor quinones, QA- or QB-. The secondary quinone, QB, which is loosely bound to the RC, determines the rate of this process. Electron transfer to QB slows down the return of the electron to P+, giving rise to a slow phase of the recovery kinetics with time tau P approximately 1 s, whereas charge recombination in RCs lacking QB generates a fast phase with time tau AP approximately 0.1 s. The amount of quinone bound to RC micelles can be reduced by increasing the detergent concentration. The characteristic time of the slow component of P+ dark relaxation, observed at low quinone content per RC micelle (at high detergent concentration), is about 1.2-1.5 s, in sharp contrast to expectations from previous models, according to which the time of the slow component should approach the time of the fast component (about 0.1 s) when the quinone concentration approaches zero. To account for this large discrepancy, a new quantitative approach has been developed to analyze the kinetics of electron transfer in isolated RCs with the following key features: 1) The exchange of quinone between different micelles (RC and detergent micelles) occurs more slowly than electron transfer from QB- to P+; 2) The exchange of quinone between the detergent "phase" and the QB binding site within the same RC micelle is much faster than electron transfer between QA- and P+; 3) The time of the slow component of P+ dark relaxation is determined by (n) > or = 1, the average number of quinones in RC micelles, calculated only for those RC micelles that have at least one quinone per RC (in excess of QA). An analytical function is derived that relates the time of the slow component of P+ relaxation, tau P, and the relative amplitude of the slow phase. This provides a useful means of determining the true equilibrium constant of electron transfer between QA and QB (LAB), and the association equilibrium constant of quinone binding at the QB site (KQ+). We found that LAB = 22 +/- 3 and KQ = 0.6 +/- 0.2 at pH 7.5. The analysis shows that saturation of the QB binding site in detergent-solubilized RCs is difficult to achieve with hydrophobic quinones. This has important implications for the interpretation of apparent dependencies of QB function on environmental parameters (e.g. pH) and on mutational alterations. The model accounts for the effects of detergent and quinone concentration on electron transfer in the acceptor quinone complex, and the conclusions are of general significance for the study of quinone-binding membrane proteins in detergent solutions.

Two distinct conformations of the primary electron donor in reaction centers from Rhodobacter sphaeroides revealed by ENDOR/TRIPLE-spectroscopy

The effect of solubilization of photosynthetic reaction centers (RCs) from Rhodobacter sphaeroides with different detergents on the electronic structure of the oxidized primary donor, P.+, is investigated. Electron paramagnetic resonance spectroscopy and related multiple resonance techniques (ENDOR/Special TRIPLE) show that two distinct conformations of P.+ can be obtained, depending on the detergent properties, the detergent/RC ratio, and the temperature. The two states correspond to different positions of the long-wavelength Qy-band of the neutral state, P (lambda1 = 866 nm and lambda2 = 850 nm at room temperature) and therefore are called P866.+ and P850.+, respectively. P866.+ is found in chromatophores and in RCs solubilized with nonionic detergents and bile salts. P850.+ is induced by zwitterionic and ionic detergents with aliphatic hydrophobic chains. The TRIPLE resonance spectra reveal that both states coexist in the range lambda2 < lambda(max) < lambda1. The main property of the detergent that determines the ability to induce P850.+ is the polarity of the head group. A simple phenomenological model is presented that relates the standard Gibbs free energy difference between the two conformations to the detergent/RC ratio and the temperature. Of special interest is the observation that the widely used detergent LDAO can induce P850.+ upon freezing the RCs without cryoprotectants. The spectroscopic properties of the two states are compared and their possible roles in RC function are discussed.

The association of different detergents with the photosynthetic reaction center protein of Rhodobacter sphaeroides R26 and the effects on its photochemistry

Detergent-free reaction centers from Rhodobacter sphaeroides R26 were used to study the solubilization of reaction centers in various detergents and their effects on reaction center photochemistry. 500 +/- 100 n-octyl-beta-D-glucopyranoside or 51 +/- 5 Triton X-100 molecules were associated with one reaction center. For N.N-alkylamine N-oxide detergents with chain lengths in the range from 8-13 carbon atoms, the number of detergent molecules associated with the reaction centers increased with decreasing chain length. The amount of detergent molecules associated with the reaction centers decreased almost tenfold if the pH was increased from pH 6 to pH 10. The addition of 5% 1,2,3-heptanetriol to various detergents lowered the detergent/reaction center ratio by a factor of two compared to that for the pure detergent. The detergent concentration at which solubilization of the reaction center occurs was close to the critical micelle concentration for all detergents studied. The absorption band at 865 nm of the primary donor in the reaction center shifts to 846 nm when detergent was removed from the reaction center; upon resolubilization with various detergents, this band shifts back to 865 nm. In 80-90% of the detergent-free reaction centers, the secondary electron transfer from QA to QB was inhibited: this electron transfer was restored after re-addition of detergent.

The crystal structure of the light-harvesting complex II (B800-850) from Rhodospirillum molischianum

BACKGROUND

The light-harvesting complexes II (LH-2s) are integral membrane proteins that form ring-like structures, oligomers of alpha beta-heterodimers, in the photosynthetic membranes of purple bacteria. They contain a large number of chromophores organized optimally for light absorption and rapid light energy migration. Recently, the structure of the nonameric LH-2 of Rhodopseudomonas acidophila has been determined; we report here the crystal structure of the octameric LH-2 from Rhodospirillum molischianum. The unveiling of similarities and differences in the architecture of these proteins may provide valuable insight into the efficient energy transfer mechanisms of bacterial photosynthesis.

RESULTS

The crystal structure of LH-2 from Rs. molischianum has been determined by molecular replacement at 2.4 A resolution using X-ray diffraction. The crystal structure displays two concentric cylinders of sixteen membrane-spanning helical subunits, containing two rings of bacteriochlorophyll-a (BChl-a) molecules. One ring comprises sixteen B850 BChl-as perpendicular to the membrane plane and the other eight B800 BChl-as that are nearly parallel to the membrane plane; eight membrane-spanning lycopenes (the major carotenoid in this complex) stretch out between the B800 and B850 BChl-as. The B800 BChl-as exhibit a different ligation from that of Rps. acidophila (aspartate is the Mg ligand as opposed to formyl-methionine in Rps. acidophila).

CONCLUSIONS

The light-harvesting complexes from different bacteria assume various ring sizes. In LH-2 of Rs. molischianum, the Qy transition dipole moments of neighbouring B850 and B800 BChl-as are nearly parallel to each other, that is, they are optimally aligned for Föster exciton transfer. Dexter energy transfer between these chlorophylls is also possible through interactions mediated by lycopenes and B850 BChl-a phytyl tails; the B800 BChl-a and one of the two B850 BChl-as associated with each heterodimeric unit are in van der Waals distance to a lycopene, such that singlet and triplet energy transfer between lycopene and the BChl-as can occur by the Dexter mechanism. The ring structure of the B850 BChl-as is optimal for light energy transfer in that it samples all spatial absorption and emission characteristics and places all oscillator strength into energetically low lying, thermally accessible exciton states.

Fv fragment-mediated crystallization of the membrane protein bacterial cytochrome c oxidase

Crystallization of membrane proteins, a prerequisite for their X-ray crystallographic analysis, remains difficult. Here, we demonstrate that the crystallization of the cytochrome c oxidase from Paracoccus denitrificans can be mediated by co-crystallization with an antibody Fv fragment. The crystals obtained contain all four subunits of this membrane protein complex and the Fv fragment. The approach of co-crystallizing membrane proteins with antibody fragments should be useful in obtaining well-ordered crystals of membrane proteins in general.

Synthesis of phospholipid-based detergents and their effects on the Ca(2+)-ATPase of sarcoplasmic reticulum

Phospholipid molecules containing a cholate or hemisuccinylcholate moiety at the 2-position were synthesized as possible detergents for solubilization of membrane proteins. 1-Myristoleoyl-2-cholyl-sn-glycero-3- phosphatidylcholine ((C14:1,cholyl)PC) was found to solubilize sarcoplasmic reticulum vesicles at concentrations above its cmc of ca. 4 micrograms/ml. Effects of (C14:1,cholyl)PC on the activity of the Ca(2+)-ATPase of sarcoplasmic reticulum were complex, as for other detergents. High concentrations (0.2 mg/ml) of (C14:1,cholyl)PC were able to displace phospholipids from the lipid/protein interface of the ATPase. Although under these conditions the activity of the ATPase was low, the ATPase was not denatured since activity could be regained by displacement of (C14:1,cholyl)PC with the detergent C12E8. 1-oleoyl-2-cholyl-sn-glycero-3-phosphatidylcholine ((C18:1,cholyl)PC) was found to have a very low water solubility, but this could be increased by the introduction of a hemisuccinyl group to give 1-oleoyl-2-(3 alpha-hemisuccinyl)cholyl-sn-glycero-3-phosphatidylcholine ((C18:1,cholylCOOH)PC). This was able to solubilize and delipidate the Ca(2+)-ATPase; the ATPase was stable in (C18:1,cholylCOOH)PC for long periods of time.

Asymmetric binding of the primary acceptor quinone in reaction centers of the photosynthetic bacterium Rhodobacter sphaeroides R26, probed with Q-band (35 GHz) EPR spectroscopy

The reaction center (RC)-bound primary acceptor quinone QA of the photosynthetic bacterium Rhodobacter sphaeroides R26 functions as a one-electron gate. The radical anion QA.- is proposed to have an asymmetric electron distribution, induced by the protein environment. We replace the native ubiquinone-10 (UQ10) with specifically 13C-labelled UQ10, and use Q-band (35 GHz) EPR spectroscopy to investigate this phenomenon in closer detail. The direct observation of the 13C-hyperfine splitting of the gz-component of UQ10A.- in the RC and in frozen isopropanol shows that the electron spin distribution is symmetric in the isopropanol glass, and asymmetric in the RC. Our results allow qualitative assessment of the spin and charge distribution for QA.- in the RC. The carbonyl oxygen of the semiquinone anion nearest to the S = 2 Fe(2+)-ion and QB is shown to acquire the highest (negative) charge density.