Bilayer structure and physical dynamics of the cytochrome b5 dimyristoylphosphatidylcholine interaction.

Cytochrome b5 is a microsomal membrane protein which provides reducing potential to delta 5-, delta 6-, and delta 9-fatty acid desaturases through its interaction with cytochrome b5 reductase. Low angle x-ray diffraction has been used to determine the structure of an asymmetrically reconstituted cytochrome b5:DMPC model membrane system. Differential scanning calorimetry and fluorescence anisotropy studies were performed to examine the bilayer physical dynamics of this reconstituted system. These latter studies allow us to constrain structural models to those which are consistent with physical dynamics data. Additionally, because the nonpolar peptide secondary structure remains unclear, we tested the sensitivity of our model to different nonpolar peptide domain configurations. In this modeling approach, the nonpolar peptide moiety was arranged in the membrane to meet such chemically determined criteria as protease susceptibility of carboxyl- and amino-termini, tyrosine availability for pH titration and tryptophan 109 location, et cetera. In these studies, we have obtained a reconstituted cytochrome b5:DMPC bilayer structure at approximately 6.3 A resolution and conclude that the nonpolar peptide does not penetrate beyond the bilayer midplane. Structural correlations with calorimetry, fluorescence anisotropy and acyl chain packing data suggest that asymmetric cytochrome b5 incorporation into the bilayer increases acyl chain order. Additionally, we suggest that the heme peptide:bilayer interaction facilitates a discreet heme peptide orientation which would be dependent upon phospholipid headgroup composition.     

The separate profile structures of the functional calcium pump protein and the phospholipid bilayer within isolated sarcoplasmic reticulum membranes determined by X-ray and neutron diffraction

The detailed profile structure of the isolated sarcoplasmic reticulum membrane was studied utilizing a combination of X-ray and neutron diffraction. The water and lipid profile structures within the sarcoplasmic reticulum membrane were determined at 28 A resolution directly by neutron diffraction and selective deuteration of the water and lipid components. The previously determined electron density profile structure of the sarcoplasmic reticulum membrane at 12 A resolution was subjected to model refinement analysis constrained by the neutron diffraction results, thereby providing unique higher resolution calculated lipid and protein profile structures. It was found that the lipid bilayer profile structure of the isolated sarcoplasmic reticulum membrane is asymmetric, primarily the result of more lipid residing in the inner versus the outer monolayer of the sarcoplasmic reticulum lipid bilayer. The asymmetry in the lipid composition was necessarily coincident with a complimentary asymmetry in the protein mass distribution between the two monolayers in order to preserve the overall cross-sectional area of lipid and protein throughout the lipid bilayer region of the sarcoplasmic reticulum membrane profile structure. Approximately 50% of the mass of the total protein was found to be localized externally to the sarcoplasmic reticulum membrane lipid bilayer protruding from the outer lipid monolayer into the extravesicular medium. The structural features of the protein protrusion appear to be rather variable depending upon the environment of the sarcoplasmic reticulum membrane. This highly asymmetric structural organization of the sarcoplasmic reticulum membrane profile is consistent with its primary function of unidirectional calcium transport.     

Phospholipid asymmetry in the isolated sarcoplasmic reticulum membrane

The total phospholipid content and distribution of phospholipid species between the outer and inner monolayers of the isolated sarcoplasmic reticulum membrane was measured by phospholipase A2 activities and neutron diffraction. Phospholipase measurements showed that specific phospholipid species were asymmetric in their distribution between the outer and inner monolayers of the sarcoplasmic reticulum lipid bilayer; phosphatidylcholine (PC) was distributed 48/52 +/- 2% between the outer and inner monolayer of the sarcoplasmic reticulum bilayer, 69% of the phosphatidyl-ethanolamine (PE) resided mainly in the outer monolayer of the bilayer, 85% of the phosphatidylserine (PS) and 88% of the phosphatidylinositol (PI) were localized predominantly in the inner monolayer. The total phospholipid distribution determined by these measurements was 48/52 +/- 2% for the outer/inner monolayer of the sarcoplasmic reticulum lipid bilayer. Sarcoplasmic reticulum phospholipids were biosynthetically deuterated and exchanged into isolated vesicles with both a specific lecithin and a general exchange protein. Neutron diffraction measurements directly provided lipid distribution profiles for both PC and the total lipid content in the intact sarcoplasmic reticulum membrane. The outer/inner monolayer distribution for PC was 47/53 +/- 1%, in agreement with phospholipase measurements, while that for the total lipid was 46/54 +/- 1%, similar to the phospholipase measurements. These neutron diffraction results regarding the sarcoplasmic reticulum membrane bilayer were used in model calculations for decomposing the electron-density profile structure (10 A resolution) of isolated sarcoplasmic reticulum previously determined by X-ray diffraction into structures for the separate membrane components. These structure studies showed that the protein profile structure within the membrane lipid bilayer was asymmetric, complementary to the asymmetric lipid structure. Thus, the total phospholipid asymmetry obtained by two independent methods was small but consistent with a complementary asymmetric protein structure, and may be related to the highly vectorial functional properties of the calcium pump ATPase protein in the sarcoplasmic reticulum membrane.     

Neutron diffraction analysis of cytochrome b5 reconstituted in deuterated lipid multilayers

Cytochrome b5 was reconstituted with a highly deuterated phospholipid to form ordered multilayers consisting of repeated centrosymmetric pairs of asymmetric lipid-protein bilayers. Lamellar neutron diffraction data were collected to approximately 29 A resolution, and have been interpreted using models for the interaction of the membrane-binding domain of cytochrome b5 with the lipid bilayer. A range of different models was examined, and those in which the protein penetrates well into the bilayer, possibly spanning it, are favored.     

Enzyme and phospholipid asymmetry in liver microsomal membranes

The transverse distribution of enzyme proteins and phospholipids within microsomal membranes was studied by analyzing membrane composition after treatment with proteases and phospholipases. Upon trypsin treatment of closed microsomal vesicles, NADH- and NADPH-cytochrome c reductases as well as cytochrome b5 were solubilized or inactivated, while cytochrome P-450 was partially inactivated. When microsomes were exposed to a concentration of deoxycholate which makes them permeable to macromolecules but does not disrupt the membrane, the detergent alone was sufficient to release four enzymes: nucleoside diphosphatase, esterase, beta-glucuronidase, and a portion of the DT-diaphorase. Introduction of trypsin into the vesicle lumen inactivated glucose-6-phosphatase completely and cytochrome P-450 partially. The rest of this cytochrome, ATPase, AMPase, UDP-glucuronyltransferase, and the remaining 50% of DT-diaphorase activity were not affected by proteolysis from either side of the membrane. Phospholipase A treatment of intact microsomes in the presence of albumin hydrolyzed all of the phosphatidylethanolamine, phosphatidylserine, and 55% of the phosphatidylcholine. From this observation, it was concluded that these lipids are localized in the outer half of the bilayer of the microsomal membrane; Phosphatidylinositol, 45% of the phosphatidylcholine, and sphingomyelin are tentatively assigned to the inner half of this bilayer. It appears that the various enzyme proteins and phospholipids of the microsomal membrane display an asymmetric distribution in the transverse plane.     

Time-resolved x-ray diffraction studies of the sarcoplasmic reticulum membrane during active transport.

X-ray and neutron diffraction studies of oriented multilayers of a highly purified fraction of isolated sarcoplasmic reticulum (SR) have previously provided the separate profile structures of the lipid bilayer and the Ca2+-ATPase molecule within the membrane profile to approximately 10-A resolution. These studies used biosynthetically deuterated SR phospholipids incorporated isomorphously into the isolated SR membranes via phospholipid transfer proteins. Time-resolved x-ray diffraction studies of these oriented SR membrane multilayers have detected significant changes in the membrane profile structure associated with phosphorylation of the Ca2+-ATPase within a single turnover of the Ca2+-transport cycle. These studies used the flash photolysis of caged ATP to effectively synchronize the ensemble of Ca2+-ATPase molecules in the multilayer, synchrotron x-radiation to provide 100-500-ms data collection times, and double-beam spectrophotometry to monitor the Ca2+-transport process directly in the oriented SR membrane multilayer.     

The reaction center profile structure derived from neutron diffraction

Both reaction center protein from the photosynthetic bacteria Rhodopseudomonas sphaeroides and egg phosphatidylcholine can be deuterium labelled; the reaction center protein can be incorporated into the phosphatidylcholine bilayers forming a homogeneous population of unilamellar vesicles. The lipid profile and the reaction center profile within these reconstituted membrane profiles were directly determined to 32 Å resolution using lamellar neutron diffraction from oriented membrane multilayers containing either deuterated or protonated reaction centers, and either deuterated or protonated phosphatidylcholine. The 32 Å resolution reaction center profile shows that the protein spans the membranes, and has an asymmetric mass distribution along the perpendicular to the membrane plane. These results were combined with previously described X-ray diffraction results in order to extend the resolution of the derived reaction center profile to 9 Å.     

Structure determination of lipid bilayers

A method of determining the phases of X-ray reflections from oriented model membrane systems at low resolution is described. The method involves deconvolution and requires that d less than or equal to 2v where v is the width of the head group region within the bilayer and d is the thickness of the bilayer. The method can be used with a single set of X-ray data and applies to lipid bilayers which have a relatively constant density in the hydrocarbon region. Phases for the first five or six orders of phosphatidylethanolamine and lecithin are derived. A refined analysis based upon deconvolution but using information inherent in the Fourier profile is also described.     

Electron density profile of two-dimensionally crystalline membranous cytochrome c oxidase at low resolution.

Unilamellar vesicles of membranous cytochrome c oxidase have been isolated whose distribution of protein in the membrane plane was predominantly crystalline. The vesicles were collapsed via controlled partial dehydration, resulting, at first, in the formation of unoriented, mostly unstacked, membrane pairs. Further controlled partial dehydration resulted in the formation of oriented multilayers of stacks of membrane pairs, retaining the in-plane crystallinity. The above were monitored by electron microscopy and x-ray diffraction. Analysis of the x-ray diffraction from unoriented, unstacked membrane pairs by two independent methods provided the membrane electron density profile to 30 A resolution.     

Membrane fusogenic activity of the Alzheimer's peptide A beta(1-42) demonstrated by small-angle neutron scattering

Amyloid-β peptide (Aβ) is considered a triggering agent of Alzheimer's disease. In relation to a therapeutic treatment of the disease, the interaction of Aβ with the cell membrane has to be elucidated at the molecular level to understand its mechanism of action. In previous works, we had ascertained by neutron diffraction on stacked lipid multilayers that a toxic fragment of Aβ is able to penetrate and perturb the lipid bilayer. Here, the influence of Aβ(1-42), the most abundant Aβ form in senile plaques, on unilamellar lipid vesicles of phospholipids is investigated by small-angle neutron scattering. We have used the recently proposed separated form factor method to fit the data and to obtain information about the vesicle diameter and structure of the lipid bilayer and its change upon peptide administration. The lipid membrane parameters were obtained with different models of the bilayer profile. As a result, we obtained an increase in the vesicle radii, indicating vesicle fusion. This effect was particularly enhanced at pH 7.0 and at a high peptide/lipid ratio. At the same time, a thinning of the lipid bilayer occurred. A fusogenic activity of the peptide may have very important consequences and may contribute to cytotoxicity by destabilizing the cell membrane. The perturbation of the bilayer structure suggests a strong interaction and/or insertion of the peptide into the membrane, although its localization remains beyond the limit of the experimental resolution.     

Spatial organization of redox active centers in the bovine heart ubiquinol-cytochrome c oxidoreductase

We have examined the spatial organization of the redox active centers in the Site II segment of the bovine heart respiratory chain by using reconstituted proteoliposomes of ubiquinol-cytochrome c oxidoreductase (Complex III or cytochrome bc1 complex) and EPR techniques. 1) Mutual spin-spin interactions between intrinsic redox active centers were detected. The spin relaxation of the Rieske iron-sulfur cluster was enhanced by the paramagnetic cytochrome c1 and b566 hemes but not by cytochrome b562. 2) Relative distances of the individual redox active centers to the P-side and N-side surfaces of the reconstituted Complex III proteoliposome were measured by our paramagnetic probe method (Blum, H., Bowyer, J. R., Cusanovich, M. A., Waring, A. J., and Ohnishi, T. (1983) Biochim. Biophys. Acta 748, 418-428). The cytochrome b562 heme was shown to be close to the middle of the phospholipid bilayer, while the Rieske iron-sulfur cluster and cytochrome b566 heme were assigned to be near the P-side surface level of the membrane. This probe method is a low resolution technique from the structural viewpoint; however, it can provide direct and reliable assignment of the topographical locations of redox active centers within the membrane. This is the first direct demonstration of the transmembranous location of the two cytochrome b hemes, although electron transfer between these two hemes crosses only half of the membrane thickness. Our data support the assignment of transmembranous distribution of the redox active centers based on electrochromic measurements (Robertson, D.E., and Dutton, P.L. (1988) Biochim, Biophys. Acta 935, 273-291). The implication of these results on the mechanism of Site II energy coupling is discussed.     

Neutron scattering shows that cytochrome b5 penetrates deeply into the lipid bilayer

Cytochrome b5 was asymmetrically reconstituted into small lipid vesicles made of a highly deuterated phospholipid. Small-angle neutron diffraction patterns were collected in a series of H2O-D2O mixtures from vesicles consisting of lipid and native or trypsinized cytochrome b5. The second moment of the radial distribution of scattering density in the vesicles was derived from these data and was compared to values calculated from three proposed models, which differ by the degree that cytochrome b5 penetrates the lipid bilayer. The model in which the hydrophobic domain of the protein is distributed across the bilayer agreed most closely with the data.     

Peptide model helices in lipid membranes: insertion,positioning, and lipid response on aggregation studied by X-ray scattering

Studying membrane active peptides or protein fragments within the lipid bilayer environment is particularly challenging in the case of synthetically modified, labeled, artificial, or recently discovered native structures. For such samples the localization and orientation of the molecular species or probe within the lipid bilayer environment is the focus of research prior to an evaluation of their dynamic or mechanistic behavior. X-ray scattering is a powerful method to study peptide/lipid interactions in the fluid, fully hydrated state of a lipid bilayer. For one, the lipid response can be revealed by observing membrane thickening and thinning as well as packing in the membrane plane; at the same time, the distinct positions of peptide moieties within lipid membranes can be elucidated at resolutions of up to several angstroms by applying heavy-atom labeling techniques. In this study, we describe a generally applicable X-ray scattering approach that provides robust and quantitative information about peptide insertion and localization as well as peptide/lipid interaction within highly oriented, hydrated multilamellar membrane stacks. To this end, we have studied an artificial, designed β-helical peptide motif in its homodimeric and hairpin variants adopting different states of oligomerization. These peptide lipid complexes were analyzed by grazing incidence diffraction (GID) to monitor changes in the lateral lipid packing and ordering. In addition, we have applied anomalous reflectivity using synchrotron radiation as well as in-house X-ray reflectivity in combination with iodine-labeling in order to determine the electron density distribution ρ(z) along the membrane normal (z axis), and thereby reveal the hydrophobic mismatch situation as well as the position of certain amino acid side chains within the lipid bilayer. In the case of multiple labeling, the latter technique is not only applicable to demonstrate the peptide’s reconstitution but also to generate evidence about the relative peptide orientation with respect to the lipid bilayer.     

Distribution of cytochrome b5 between small and large unilamellar phospholipid vesicles

Cytochrome b5 is an amphipathic integral membrane protein that spontaneously inserts, post-translationally, into intracellular membranes. When added to preformed phospholipid vesicles, it binds in a so-called "loose" or transferable configuration, characterized by the ability of the protein to rapidly equilibrate between vesicles. In a preliminary report we showed that the distribution of cytochrome b5 among a heterogeneous population of small sonicated phosphatidylcholine vesicles (212 to about 350 A in diameter) lies in favor of the smallest vesicles by a factor of at least 20 (Greenhut, S.F. and Roseman, M.A. (1985) J. Biol. Chem. 260, 5883-5886). In the present studies we have attempted to determine the maximal extent to which bilayer curvature can influence the intervesicle distribution of cytochrome b5, by measuring the distribution of the protein between a population of limit-size vesicles 212 A in diameter and a population of large unilamellar vesicles approximately 1000 A in diameter. (The effect of bilayer curvature on the physical properties of the lipids in the large vesicles is considered to be negligible.) The results show that cytochrome b5 favors the small vesicle population by a factor of about 200. This observation suggests that the formation of highly curved regions in biological membranes (or the formation of regions in which the physical state of the lipids is similar to that in small vesicles) may cause the accumulation of certain membrane proteins at those sites. We also observed that a significant fraction (11-20%) of the cytochrome b5, when added directly to the large vesicles, spontaneously inserts into the "tight," physiologically proper configuration. A possible mechanism is discussed.     

The structure of a cytochrome oxidase-lipid model membrane.

The structure of "membranous cytochrome oxidase" has been investigated by X-ray diffraction, optical polarization spectroscopy and EPR spectroscopy. These studies indicate that the cytochrome oxidase molecules are oriented symmetrically in the membrane profile with a significant portion of their mass occurring within the extravesicular surface of the membrane; the oxidase molecultes span the membrane profile; the distribution of the oxidase molecules over the plane of these membranes is non-crystalline; the oxidase molecules contain bundles of alpha-helical polypeptide chain segments where the average orientation of the helices is normal to the membrane plane; and the average heme orientation within the oxidase molecules is such that the normal to the heme plane lies in the plane of the membrane.     

The structure of a cytochrome oxidase-lipid model membrane

The structure of “membranous cytochrome oxidase” has been investigated by X-ray diffraction, optical polarization spectroscopy and EPR spectroscopy. These studies indicate that the cytochrome oxidase molecules are oriented asymmetrically in the membrane profile with a significant portion of their mass occurring within the extravesicular surface of the membrane; the oxidase molecultes span the membrane profile; the distribution of the oxidase molecules over the plane of these membranes is non-crystalline; the oxidase molecules contain bundles of α-helical polypeptide chain segments where the average orientation of the helices is normal to the membrane plane; and the average heme orientation within the oxidase molecules is such that the normal to the heme plane lies in the plane of the membrane.     

Evaluation of lipid exposure of tryptophan residues in membrane peptides and proteins

Fluorescence quenching is used to gain information on the exposure of tryptophan residues to lipid in membrane-bound proteins and peptides. A protocol is developed to calculate this exposure, based on a comparison of quenching efficiency and of a fluorescence lifetime (or quantum yield) measured for a protein and for a model tryptophan-containing compound. Various methods of analysis of depth-dependent quenching are compared and three universal measures of quenching profile are derived. One of the measures, related to the area under profile, is used to estimate quenching efficiency. The method is applied to single tryptophan mutants of a membrane-anchoring nonpolar peptide of cytochrome b(5) and of an outer membrane protein A. Analysis of quenching of the cytochrome's nonpolar peptide by a set of four brominated lipids reveals a temperature-controlled reversible conformational change, resulting in increased exposure of tryptophan to lipid and delocalization of its transverse position. Kinetic quenching profiles and fluorescence binding kinetics reported by Kleinschmidt et al. (Biochemistry (1999) 38, 5006-5016) were analyzed to extract information on the relative exposure of tryptophan residues during folding of an outer membrane protein A. Trp-102, which translocates across the bilayer, was found to be noticeably shielded from the lipid environment throughout the folding event compared to Trp-7, which remains on the cis side. The approach described here provides a new tool for studies of low-resolution structure and conformational transitions in membrane proteins and peptides.     

Intramembrane position of the fluorescent tryptophanyl residue in membrane-bound cytochrome b5

We have developed a method to measure the intramembrane position of the fluorescent tryptophanyl residue in whole cytochrome b5 and the nonpolar membrane binding segment when these molecules are bound to phospholipid vesicles [Koppel, D.E., Fleming, P., & Strittmatter, P. (1979) Biochemistry (preceding paper in this issue)]. The method utilizes excitation energy transfer from the donor tryptophanyl residue in the protein to trinitrophenyl or danysl acceptor groups on the surface of the phospholipid bilayer. It was determined that that single fluorescent tryptophanyl residue in vesicle-bound cytochrome b5 and the nonpolar segment is located approximately 20-22 A below the surface of the bilayer. This position represents a minimum depth of penetration of this portion of the cytochrome in the membrane.     

The reactivities of tyrosine and tryptophan residues in lipid-bound cytochrome b5.

Purified cytochrome b5 from rabbit liver microsomes was bound to liposomes prepared from microsomal lipids. Tyrosyl and tryptophyl side chains of the protein were modified by water-soluble reagents and the reactivities of these amino acid residues in the liposome-bound cytochrome b5 were compared to those of the free protein. At pH 13, 80% of the tyrosines in lipid-free cytochrome b5 ionized immediately, whereas in the lipid-bound protein only 65% ionized within the first minute. In contrast, acetylation with acetylimidazole resulted in the conversion of all 5 tyrosine groups of lipid-free as well as lipid-bound cytochrome b5 into O-acetylated derivatives, which upon treatment with hydroxylamine were completely deacetylated. Reaction with N-bromosuccinimide revealed that only 60% of the 4 tryptophan residues present in cytochrome b5 were accessible to the reagent in the lipid-bound protein, although all tryptophans could be modified in lipid free cytochrome b5. It was concluded that the two tyrosines in the region linking the protein to the membrane are not shielded by lipid bilayer but that of the three tryptophans in the same region one is completely buried in the membrane, whereas the remaining two tryptophans may be both partly exposed to the solvent or alternatively, one may be partially and the other completely exposed.     

Molecular organization (topography) of cytochrome P-450(11)beta in mitochondrial membrane and phospholipid vesicles as studied by trypsinolysis

Cytochrome P-450(11)beta from adrenal cortex is an intrinsic membrane protein embedded in the inner mitochondrial membrane. Topography of the protein inside a phospholipid bilayer was examined using controlled proteolysis of purified cytochrome P-450(11)beta following its integration into artificial liposomes. Inclusion of the protein into phospholipid vesicles led to a marked stabilization of the cytochrome activity. Trypsin treatment of the liposome-integrated cytochrome resulted in the rapid disappearance of the native protein moiety (47 kDa), while a major 34 kDa peptide component was formed. This peptide core retained the heme moiety and part of the cytochrome steroid-11 beta hydroxylase activity. Very similar observations were obtained when inside-out vesicles prepared from isolated adrenocortical mitoplasts were examined with the same approach. It is thus suggested that adrenocortical cytochrome P-450(11)beta is embedded in the inner mitochondrial membrane as well as in artificial liposomes by a major hydrophobic domain associated with the heme moiety while a limited domain remains accessible on the matrix side of the membrane surface. The previous described phosphorylation of the cytochrome P-450(11)beta on a serine residue, by the cAMP-dependent protein kinase is suggested to occur in the protein domain oriented toward the membrane surface, the phosphorylation site being lost under mild proteolytic digestion of the membrane-integrated protein.