Unusual features of the c-ring of F1FO ATP synthases.

Membrane integral ATP synthases produce adenosine triphosphate, the universal "energy currency" of most organisms. However, important details of proton driven energy conversion are still unknown. We present the first high-resolution structure (2.3 Å) of the in meso crystallized c-ring of 14 subunits from spinach chloroplasts. The structure reveals molecular mechanisms of intersubunit contacts in the c14-ring, and it shows additional electron densities inside the c-ring which form circles parallel to the membrane plane. Similar densities were found in all known high-resolution structures of c-rings of F1FO ATP synthases from archaea and bacteria to eukaryotes. The densities might originate from isoprenoid quinones (such as coenzyme Q in mitochondria and plastoquinone in chloroplasts) that is consistent with differential UV-Vis spectroscopy of the c-ring samples, unusually large distance between polar/apolar interfaces inside the c-ring and universality among different species. Although additional experiments are required to verify this hypothesis, coenzyme Q and its analogues known as electron carriers of bioenergetic chains may be universal cofactors of ATP synthases, stabilizing c-ring and prevent ion leakage through it.

From protons to OXPHOS supercomplexes and Alzheimer's disease: structure-dynamics-function relationships of energy-transducing membranes.

By the elucidation of high-resolution structures the view of the bioenergetic processes has become more precise. But in the face of these fundamental advances, many problems are still unresolved. We have examined a variety of aspects of energy-transducing membranes from large protein complexes down to the level of protons and functional relevant picosecond protein dynamics. Based on the central role of the ATP synthase for supplying the biological fuel ATP, one main emphasis was put on this protein complex from both chloroplast and mitochondria. In particular the stoichiometry of protons required for the synthesis of one ATP molecule and the supramolecular organisation of ATP synthases were examined. Since formation of supercomplexes also concerns other complexes of the respiratory chain, our work was directed to unravel this kind of organisation, e.g. of the OXPHOS supercomplex I(1)III(2)IV(1), in terms of structure and function. Not only the large protein complexes or supercomplexes work as key players for biological energy conversion, but also small components as quinones which facilitate the transfer of electrons and protons. Therefore, their location in the membrane profile was determined by neutron diffraction. Physico-chemical features of the path of protons from the generators of the electrochemical gradient to the ATP synthase, as well as of their interaction with the membrane surface, could be elucidated by time-resolved absorption spectroscopy in combination with optical pH indicators. Diseases such as Alzheimer's dementia (AD) are triggered by perturbation of membranes and bioenergetics as demonstrated by our neutron scattering studies.

ATP synthase: constrained stoichiometry of the transmembrane rotor.

Recent structural data suggest that the number of identical subunits (c or III) assembled into the cation-powered rotor of F1F0 ATP synthase depends on the biological origin. Atomic force microscopy allowed individual subunits of the cylindrical transmembrane rotors from spinach chloroplast and from Ilyobacter tartaricus ATP synthase to be directly visualized in their native-like environment. Occasionally, individual rotors exhibit structural gaps of the size of one or more subunits. Complete rotors and arch-shaped fragments of incomplete rotors revealed the same diameter within one ATP synthase species. These results suggest the rotor diameter and stoichiometry to be determined by the shape of the subunits and their nearest neighbor interactions.

Microsecond exchange of internal water molecules in bacteriorhodopsin.

The proton-conducting pathway of bacteriorhodopsin (BR) contains at least nine internal water molecules that are thought to be key players in the proton translocation mechanism. Here, we report the results of a multinuclear (1H, 2H, 17O) magnetic relaxation dispersion (MRD) study with the primary goal of determining the rate of exchange of these internal water molecules with bulk water. This rate is of interest in current attempts to elucidate the molecular details of the proton translocation mechanism. The relevance of water exchange kinetics is underscored by recent crystallographic findings of substantial variations in the number and locations of internal water molecules during the photocycle. Moreover, internal water exchange is believed to be governed by conformational fluctuations in the protein and can therefore provide information about the thermal accessibility of functionally important conformational substates. The present 2H and 17O MRD data show that at least seven water molecules, or more if they are orientationally disordered, in BR have residence times (inverse exchange rate constant) in the range 0.1-10 micros at 277 K. At least five of these water molecules have residence times in the more restrictive range 0.1-0.5 micros. These results show that most or all of the deeply buried water molecules in BR exchange on a time-scale that is short compared to the rate-limiting step in the photocycle. The MRD measurements were performed on BR solubilized in micelles of octyl glucoside. From the MRD data, the rotational correlation time of detergent-solubilized BR was determined to 35 ns at 300 K, consistent with a monomeric protein in complex with about 150 detergent molecules. The solubilized protein was found to be stable in the dark for at least eight months at 277 K.

Activity and stability of a thermostable alpha-amylase compared to its mesophilic homologue: mechanisms of thermal adaptation.

To elucidate how enzymes adapt to extreme environmental conditions, a comparative study with a thermostable alpha-amylase from Bacillus licheniformis (BLA) and its mesophilic homologue from Bacillus amyloliquefaciens (BAA) was performed. We measured conformational stability, catalytic activity, and conformational fluctuations on the picosecond time scale for both enzymes as a function of temperature. The objective of this study is to analyze how these properties are related to each other. BLA shows its maximal catalytic activity at about 90-95 degrees C and a strongly reduced activity (only 20% of the maximum) at room temperature. Although B. licheniformis itself is a mesophilic organism, BLA shows an activity profile typical for a thermophilic enzyme. In contrast to this, BAA exhibits its maximal activity at about 80 degrees C but with a level of about 60% activity at room temperature. In both cases the unfolding temperatures T(m) are only 6 degrees C (BAA, T(m) = 86 degrees C) and 10 degrees C (BLA, T(m) = 103 degrees C), respectively, higher than the temperatures for maximal activity. In contrast to many previous studies on other thermophilic-mesophilic pairs, in this study a higher structural flexibility of the thermostable BLA was measured as compared to the mesophilic BAA. The findings of this study neither indicate a proportionality between the observed dynamics and the catalytic activity nor support the idea of more "rigid" thermostable proteins, as often proposed in the concept of "corresponding states".

Dynamics of the proton transfer reaction on the cytoplasmic surface of bacteriorhodopsin.

The cytoplasmic surface of bacteriorhodopsin is characterized by a group of carboxylates that function as a proton attractive domain [Checover, S., Nachliel, E., Dencher, N. A., and Gutman, M. (1997) Biochemistry 36, 13919-13928]. To identify these carboxylates, we selectively mutated them into cysteine residues and monitored the effects of the dynamics of proton transfer between the bulk and the surface of the protein. The measurements were carried out without attachment of a pH-sensor to the cysteine residue, thus avoiding any structural perturbation and change in the surface charge caused by the attachment of a reporter group, and the protein was in its BR state. The purple membranes were suspended in an unbuffered solution of pyranine (8-hydroxypyrene-1,3,6-trisulfonate) and exposed to a train of 1000 laser pulses (2.1 mJ/pulse, lambda = 355 nm, at 10 Hz). The excitation of the dye ejected the hydroxyl's proton, and a few nanoseconds later, a pair of free protons and ground-state pyranine anion was formed. The experimental observation was the dynamics of the relaxation of the system to the prepulse state. The observed signals were reconstructed by a numeric method that replicates the chemical reactions proceeding in the perturbed space. The detailed reconstruction of the measured signal assigned the various proton-binding sites with rate constants for proton binding and proton exchange and the pK values. Comparison of the results obtained by the various mutants indicates that the dominant proton-binding cluster of the wild-type protein consists of D104, E161, and E234. The replacement of D104 or E161 with cysteine lowered the proton binding capacity of the cluster to approximately 60% of that of the native protein. The replacement of E234 with cysteine disrupted the structure of the cluster, causing the two remaining carboxylates to function as isolated residues that do not interact with each other. The possibility of proton transfer between monomers is discussed.

Water and bacteriorhodopsin: structure, dynamics, and function.

A wealth of information has been gathered during the past decades that water molecules do play an important role in the structure, dynamics, and function of bacteriorhodopsin (bR) and purple membrane. Light-induced structural alterations in bR as detected by X-ray and neutron diffraction at low and high resolution are discussed in relationship to the mechanism of proton pumping. The analysis of high resolution intermediate structures revealed photon-induced rearrangements of water molecules and hydrogen bonds concomitant with conformational changes in the chromophore and the protein. These observations led to an understanding of key features of the pumping mechanism, especially the vectoriality and the different modes of proton translocation in the proton release and uptake domain of bR. In addition, water molecules influence the function of bR via equilibrium fluctuations, which must occur with adequate amplitude so that energy barriers between conformational states can be overcome.

Reversible loss of crystallinity on photobleaching purple membrane in the presence of hydroxylamine.

Structural changes of purple membrane during photobleaching in the presence of hydroxylamine were monitored using atomic force microscopy (AFM). The process of bleaching was associated with the disassembly of the purple membrane crystal into smaller crystals. Imaging steps of the photobleaching progress showed that disassembly proceeds until the sample is fully bleached and its crystallinity is almost lost. As revealed from high resolution AFM topographs, the loss of crystallinity was initiated by loss of lattice forming contact between the individual bacteriorhodopsin trimers. The bacteriorhodopsin molecules, however, remained assembled into trimers during the entire photobleaching process. Regeneration of the photobleached sample into intact purple membrane resulted in the reassembly of the bacteriorhodopsin trimers into the trigonal lattice of purple membrane. The data provide novel insights into factors triggering purple membrane formation and structure.

Dye removal, catalytic activity and 2D crystallization of chloroplast H(+)-ATP synthase purified by blue native electrophoresis.

The proton-ATP synthase of thylakoid membranes from spinach chloroplasts (CF(O)F(1)) and its subcomplexes CF(O) and CF(1) were isolated by blue native electrophoresis (BN-PAGE) [Neff, D. and Dencher, N.A. (1999) Biochem. Biophys. Res. Commun. 259, 569-575] and subsequently electroeluted from the gel. A method was developed to remove most of the dye Coomassie G-250 (CBG) using gel filtration, a prerequisite for many biophysical investigations. The dye was removed from the electroeluted CF(O)F(1), CF(O) or CF(1) and exchanged with the detergent CHAPS. ATP hydrolysis activity of CF(1) and ATP synthesis activity of reconstituted CF(O)F(1) were determined before and after dye removal. The secondary structure of CF(O) was studied by CD spectroscopy in the presence and the absence of the dye. CBG neither abolishes the catalytic activity of the isolated CF(O)F(1) and CF(1) nor affects the subunit composition and the high alpha-helical content of CF(O). In crystallization attempts, 2D arrays of CF(O)F(1) and of CF(O) before and after dye removal were obtained. In the aggregates of CF(O), circular structures with a mean diameter of 6.7 nm were observed. Our results indicate that the combination of BN-PAGE and dye removal by gel filtration is a suitable approach to obtain catalytically active protein complexes for further functional and structural characterization.

Dye-ligand chromatographic purification of intact multisubunit membrane protein complexes: application to the chloroplast H+-FoF1-ATP synthase.

n-Dodecyl-beta-D-maltoside was used as a detergent to solubilize the ammonium sulphate precipitate of chloroplast F(O)F(1)-ATP synthase, which was purified further by dye-ligand chromatography. Upon reconstitution of the purified protein complex into phosphatidylcholine/phosphatidic acid liposomes, ATP synthesis, driven by an artificial DeltapH/Deltapsi, was observed. The highest activity was achieved with ATP synthase solubilized in n-dodecyl-beta-D-maltoside followed by chromatography with Red 120 dye. The optimal dye for purification with CHAPS was Green 5. All known subunits were present in the monodisperse proton-translocating ATP synthase preparation obtained from chloroplasts.

Detergent effect on anion exchange perfusion chromatography and gel filtration of intact chloroplast H(+)-ATP synthase.

To gain a pure enzyme preparation for functional and crystallization studies, an additional purification step in the isolation of the chloroplast ATP synthase (CF(0)F(1)) has been introduced. By applying gel filtration or anion exchange perfusion chromatography in presence of the detergents CHAPS and n-dodecyl-beta-d-maltoside, respectively, Rubisco and other contaminants were separated from CF(0)F(1). The purity and activity depended on the chromatographic method and the detergent employed. The highest purity and activity were achieved by anion exchange chromatography for the detergent dodecyl-maltoside and by gel filtration for the detergent CHAPS. The detergent Triton X-100, which is frequently used to solubilize CF(0)F(1), was found to be inadequate to stabilize the ATP synthase during chromatography.

Purification of multisubunit membrane protein complexes: isolation of chloroplast FoF1-ATP synthase, CFo and CF1 by blue native electrophoresis.

The proton-ATP synthase of thylakoid membranes from chloroplasts (CFoF1) is composed of two parts with different structural and functional properties: the membrane-integral, proton-conducting complex CFo and the hydrophilic part, CF1 which catalyze the formation of adenosine-5'-triphosphate (ATP). To date it is difficult to isolate functional CFoF1 from thylakoids in high purity and yield. Blue native polyacrylamide gel electrophoresis (BN-PAGE) was therefore successfully employed to isolate CFoF1 in a one-step procedure from thylakoid membranes. Using a cathode buffer with low Coomassie Blue G-250 (CBG) concentration (0.002%), CFoF1 remains intact and can be obtained in high purity from solubilized, prepurified ATP synthase. Using BN-PAGE and a cathode buffer with 0.02% CBG, the ATP synthase bifurcates, and we were able to isolate both parts, CFo and CF1, separately. CFoF1, CFo, and CF1, respectively, were electroeluted nearly quantitatively electroeluted from the gel. BN-PAGE is a generally applicable method for the isolation and characterization of multisubunit membrane protein complexes in their native structure. However, the combination of neutral detergents and the negatively charged dye CBG seems to mimic properties of mild ionic detergents. This effect can lead to dissociation of labile subunits and subcomplexes, especially when delipidated membrane protein complexes are applied to BN-PAGE. By variation of the initial electrophoresis conditions, i.e., dye concentration in the cathode buffer, amount of lipid and detergent, BN-PAGE can be used for the isolation of either intact complexes or of subcomplexes.

X-ray diffraction from a single layer of purple membrane at the air/water interface.

X-ray diffraction patterns have been recorded from a single layer of purple membrane ( approximately 50 A thickness) at the air/water interface in a Langmuir trough. Grazing-incidence X-ray diffraction is demonstrated to be a promising method for obtaining structural information on membrane proteins under physiological conditions. The method is so sensitive that diffraction can be measured from samples with only 10(13) protein molecules in the beam. Diffraction from hexagonal crystals of purple membrane with a lattice constant of 61. 3 A was observed up to the order {h,k}={4,3}, corresponding to a resolution of approximately 9 A. The work reported here is a first step towards a new way of protein crystallography using grazing-incidence X-ray diffraction at the air/water interface.

Molecular motions and hydration of purple membranes and disk membranes studied by neutron scattering.

Fast stochastic equilibrium fluctuations (time scale: 10(-10)-10(-13) seconds) in purple membranes (MP) and in disk membranes (DM) have been measured with quasielastic incoherent neutron scattering. The comparison of predominantly stochastic motions occurring in purple membranes and in disk membranes revealed qualitatively similar dynamical behaviour. Models of internal motions within restricted volumes have been shown to be useful to fit the spectra from both samples. From fits using these models we found "amplitudes" 15 to 20% larger for motions in DM samples compared to PM samples. This indicates a higher internal flexibility of the DM. Because the dynamical behaviour is very sensitive to the hydration of the protein-lipid complex, we also performed neutron diffraction experiments to determine lamellar spacings as a measure of level of hydration and as a function of temperature. From these studies the interaction of solvent molecules with the surface of the protein-lipid complex appears to be qualitatively similar for both types of membranes.

Function and picosecond dynamics of bacteriorhodopsin in purple membrane at different lipidation and hydration.

By neutron scattering experiments and time-resolved absorption spectroscopy we have investigated picosecond equilibrium fluctuations and the kinetics of the photocycle of bacteriorhodopsin (BR) in the purple membrane (PM). Natural PM samples composed of 75% BR (w/w) and 25% lipid (w/w) as well as delipidated PM having only 5% lipid (w/w) were measured at different levels of hydration. We observed a reduced 'flexibility', due to a diminished weight of stochastic large-amplitude motions occurring in the delipidated PM as compared to the natural PM. This effect is more pronounced for wet samples, indicating the importance of lipid hydration for protein dynamics. The reduced flexibility is accompanied by significantly larger time constants describing the decay of the M-intermediate. Therefore, a correlation between the dynamical behavior of the protein-lipid complex and BR function emerges.

Evidence for charge-controlled conformational changes in the photocycle of bacteriorhodopsin.

The existence of two different M-state structures in the photocycle of the bacteriorhodopsin mutant ASP38ARG was proved. At pH 6.7 (0 to -6 degreesC) a spectroscopic M intermediate (M1) that does not differ significantly in its tertiary structure from the light-adapted ground state accumulates under illumination. At pH > 9 another state (M2), characterized by additional pronounced changes in the Fourier transform infrared difference spectrum in the region of the amide I and II bands, accumulates. The M2 intermediate trapped at pH 9.6 displays the same changes in the x-ray diffraction intensities under continuous illumination as previously described for x-ray experiments with the mutant ASP96ASN. These observations indicate that in this mutant the altered charge distribution at neutral pH controls the tertiary structural changes that seem to be necessary for proton translocation.

Dehydration of biological membranes by cooling: an investigation on the purple membrane.

The lamellar spacing dl of purple membrane (PM) multilayer systems was investigated with neutron diffraction as a function of temperature and of the level of hydration. The observed large T-dependent variations of dl indicate that PM is partially dehydrated when cooled below a "hydration water freezing point". This phenomenon is reversible, but a hysteresis is observed when PM is rehydrated upon reheating. The hydration water remaining bound to the membrane below about 240 K is non-freezing. Its amount was found to be hnf=0.24(+/-0.02) g 2H2O/g BR for all samples equilibrated at room temperature in the presence of 2H2O vapour at >/=84% r.h. It is evident, that the dehydration/rehydration behaviour of PM is strongly correlated with the temperature-dependent behaviour of the dynamical structure factor. Above the well-known "dynamical transition" announcing the onset of localized diffusive molecular motions between 190 K and 230 K, a second dynamical transition is caused by the temperature-induced rehydration of the PM starting near 255 K. This is also correlated with the deviation from a pure Arrhenius law of the rate-limiting process in the photocycle, known to occur upon cooling beyond the ice point into the same temperature region. Our results suggest that the phenomenon of dehydration and rehydration induced by cooling and reheating, respectively, is a general property of biological membranes.

Structure and hydration of the M-state of the bacteriorhodopsin mutant D96N studied by neutron diffraction.

Neutron diffraction from oriented purple membrane fragments at various hydration levels, coupled with H2O/2H2O exchange, was used to compare the structure and hydration of the light-adapted initial state (B-state) and the M photointermediate of bacteriorhodopsin mutant D96N. Diffraction patterns were recorded at 86%, 75% and 57% relative humidity (r.h.). Structural changes observed at 86% and 75% r.h. are absent at 57% r.h., showing that they are uncoupled from the deprotonation of the Schiff base during formation of the M-state. In a current model, the M-state consists of two substates, M1 and M2. Our data suggest that the state trapped at 57% r.h. is M1 and that M2 is trapped at the higher r.h. values. The observed structural changes are, therefore, associated with the M1-->M2 transition, which can only take place at higher r.h. The difference Fourier projections of exchangeable hydrogen atoms and water molecules in the membrane plane are very similar for the B and M-states at 75% and 86% r.h. This shows that contrary to certain models, the structural changes in the M-state are not correlated with major hydration changes in the proton channel projection.