Search

We present a model from which the observed morphology of the inner mitochondrial membrane can be inferred as minimizing the system's free energy. In addition to the usual energetic terms for bending, surface area, and pressure difference, our free energy includes terms for tension that we hypothesize to be exerted by proteins and for an entropic contribution due to many dimensions worth of shapes available at a given energy. We also present measurements of the structural features of mitochondria in HeLa cells and mouse embryonic fibroblasts using three-dimensional electron tomography. Such tomograms reveal that the inner membrane self-assembles into a complex structure that contains both tubular and flat lamellar crista components. This structure, which contains one matrix compartment, is believed to be essential to the proper functioning of mitochondria as the powerhouse of the cell. Interpreting the measurements in terms of the model, we find that tensile forces of ∼20 pN would stabilize a stress-induced coexistence of tubular and flat lamellar cristae phases. The model also predicts a pressure difference of -0.036 ± 0.004 atm (pressure higher in the matrix) and a surface tension equal to 0.09 ± 0.04 pN/nm., (Copyright © 2010 Biophysical Society. Published by Elsevier Inc. All rights reserved.)

Mitochondrial fusion requires the coordinated fusion of the outer and inner membranes. Three large GTPases--OPA1 and the mitofusins Mfn1 and Mfn2--are essential for the fusion of mammalian mitochondria. OPA1 is mutated in dominant optic atrophy, a neurodegenerative disease of the optic nerve. In yeast, the OPA1 ortholog Mgm1 is required for inner membrane fusion in vitro; nevertheless, yeast lacking Mgm1 show neither outer nor inner membrane fusion in vivo, because of the tight coupling between these two processes. We find that outer membrane fusion can be readily visualized in OPA1-null mouse cells in vivo, but these events do not progress to inner membrane fusion. Similar defects are found in cells lacking prohibitins, which are required for proper OPA1 processing. In contrast, double Mfn-null cells show neither outer nor inner membrane fusion. Mitochondria in OPA1-null cells often contain multiple matrix compartments bounded together by a single outer membrane, consistent with uncoupling of outer versus inner membrane fusion. In addition, unlike mitofusins and yeast Mgm1, OPA1 is not required on adjacent mitochondria to mediate membrane fusion. These results indicate that mammalian mitofusins and OPA1 mediate distinct sequential fusion steps that are readily uncoupled, in contrast to the situation in yeast.

Group A Streptococcus (GAS) is a leading human bacterial pathogen capable of producing invasive infections even in previously healthy individuals. As frontline components of host innate defense, macrophages play a key role in control and clearance of GAS infections. We find GAS induces rapid, dose-dependent apoptosis of primary and cultured macrophages and neutrophils. The cell death pathway involves apoptotic caspases, is partly dependent on caspase-1, and requires GAS internalization by the phagocyte. Analysis of GAS virulence factor mutants, heterologous expression, and purified toxin studies identified the pore-forming cytolysin streptolysin O (SLO) as necessary and sufficient for the apoptosis-inducing phenotype. SLO-deficient GAS mutants induced less macrophage apoptosis in vitro and in vivo, allowed macrophage cytokine secretion, and were less virulent in a murine systemic infection model. Ultrastructural evidence of mitochondrial membrane remodeling, coupled with loss of mitochondrial depolarization and cytochrome c release, suggests a direct attack of the toxin initiates the intrinsic apoptosis pathway. A general caspase inhibitor blocked SLO-induced apoptosis and enhanced macrophage killing of GAS. We conclude that accelerated, caspase-dependent macrophage apoptosis induced by the pore-forming cytolysin SLO contributes to GAS immune evasion and virulence.

Three-dimensional light microscopy and three-dimensional electron microscopy (electron tomography) separately provide very powerful tools to study cellular structure and physiology, including the structure and physiology of mitochondria. Fluorescence microscopy allows one to study processes in live cells with specific labels and stains that follow the movement of labeled proteins and changes within cellular compartments but does not have sufficient resolution to define the ultrastructure of intracellular organelles such as mitochondria. Electron microscopy and electron tomography provide the highest resolution currently available to study mitochondrial ultrastructure but cannot follow processes in living cells. We describe the combination of these two techniques in which fluorescence confocal microscopy is used to study structural and physiologic changes in mitochondria within apoptotic HeLa cells to define the apoptotic timeframe. Cells can then be selected at various stages of the apoptotic timeframe for examination at higher resolution by electron microscopy and electron tomography. This is a form of "virtual" 4-dimensional electron microscopy that has revealed interesting structural changes in the mitochondria of HeLa cells during apoptosis. The same techniques can be applied, with modification, to study other dynamic processes within cells in other experimental contexts.

In addition to their role in providing ATP for cellular functions via oxidative phosphorylation, mitochondria also play a critical role in initiating and/or regulating apoptosis through the release of proteins such as cytochrome c from intermembrane and intracristal compartments. The mechanism by which these proteins are able to cross the outer mitochondrial membrane has been a subject of controversy. This paper will review some recent results that demonstrate that inner mitochondrial membrane remodeling does occur during apoptosis in HeLa cells but does not appear to be a requirement for release of cytochrome c from intracristal compartments. Inner membrane remodeling does appear to be related to fragmentation of the mitochondrial matrix, and the form of the remodeling suggests a topological mechanism for inner membrane fission and fusion.

In addition to their role in cellular bioenergetics, mitochondria also initiate common forms of programmed cell death (apoptosis) through the release of proteins such as cytochrome c from the intermembrane and intracristal spaces. The release of these proteins is studied in populations of cells by western blotting mitochondrial and cytoplasmic fractions of cellular extracts, and in single cells by fluorescence microscopy using fluorescent indicators and fusion proteins. However, studying the changes in ultrastructure associated with release of proteins requires the higher resolution provided by transmission electron microscopy. Here, we have used fluorescence microscopy to characterize the state of apoptosis in HeLa cells treated with etoposide followed by electron microscopy and three-dimensional electron microscope tomography of the identical cells to study the sequence of structural changes. We have identified a remodelling of the inner mitochondrial membrane into many separate vesicular matrix compartments that accompanies release of proteins; however, this remodelling is not required for efficient release of cytochrome c. Swelling occurs only late in apoptosis after release of cytochrome c and loss of the mitochondrial membrane potential.

Electron microscope tomography produces three-dimensional reconstructions and has been used to image organelles both isolated and in situ, providing new insight into their structure and function. It is analogous to the various tomographies used in medical imaging. Compared with light microscopy, electron tomography offers an improvement in resolution of 30- to 80-fold and currently ranges from 3 to 8 nm, thus filling the gap between high-resolution structure determinations of isolated macromolecules and larger-scale studies on cells and tissues by light microscopy. Here, we provide an introduction to electron tomography and applications of the method in characterizing organelle architecture that also show its power for suggesting functional significance. Further improvements in labeling modalities, imaging tools, specimen preparation, and reconstruction algorithms promise to increase the quality and breadth of reconstructions by electron tomography and eventually to allow the mapping of the cellular proteomes onto detailed three-dimensional models of cellular structure.

Multiple isotope effects have been measured for the acid-catalyzed hydrolysis of methyl formate in 0.5 M HCl at 20 degrees C. The isotope effects in the present investigation include the carbonyl carbon (13k = 1.028 +/- 0.001), the carbonyl oxygen (18k = 0.9945 +/- 0.0009), the nucleophile oxygen (18k = 0.995 +/- 0.001), and the formyl hydrogen ((D)k = 0.81 +/- 0.02). Determination of the carbonyl carbon, carbonyl oxygen, and formyl hydrogen isotope effects was performed via isotopic analysis of residual substrate. However, determination of the oxygen nucleophile isotope effect required analysis of the oxygen atoms of the product (formic acid), which exchange with the solvent (water) under acid conditions. This necessitated measurement of the rate of exchange of these oxygen atoms under the conditions for hydrolysis (k(ex) = 0.0723 min(-1)) and correction of the raw isotope ratios measured during the nucleophile-O isotope effect experiment. These results, along with the previously reported isotope effect for the leaving oxygen (18k = 1.0009) and the ratio of the rate of hydrolysis to that of exchange of the carbonyl oxygen with water (k(h)/k(ex) = 11.3), give a detailed picture of the transition-state structure for the reaction.

The inner mitochondrial membrane has been shown to have a novel structure that contains tubular components whose radii are of the order of 10 nm as well as comparatively flat regions. The structural organization of mitochondria is important for understanding their functionality. We present a model that can account, thermodynamically, for the observed size of the tubules. The model contains two lipid constituents with different shapes. They are allowed to distribute in such a way that the composition differs on the two sides of the tubular membrane. Our calculations make two predictions: (1) there is a pressure difference of 0.2 atmospheres across the inner membrane as a necessary consequence of the experimentally observed tubule radius of 10 nm, and (2) migration of differently shaped lipids causes concentration variations of the order of 7% between the two sides of the tubular membrane.

In recent years, electron tomography has provided detailed three-dimensional models of mitochondria that have redefined our concept of mitochondrial structure. The models reveal an inner membrane consisting of two components, the inner boundary membrane (IBM) closely apposed to the outer membrane and the cristae membrane that projects into the matrix compartment. These two components are connected by tubular structures of relatively uniform size called crista junctions. The distribution of crista junction sizes and shapes is predicted by a thermodynamic model based upon the energy of membrane bending, but proteins likely also play a role in determining the conformation of the inner membrane. Results of structural studies of mitochondria during apoptosis demonstrate that cytochrome c is released without detectable disruption of the outer membrane or extensive swelling of the mitochondrial matrix, suggesting the formation of an outer membrane pore large enough to allow passage of holo-cytochrome c. The possible compartmentation of inner membrane function between the IBM and the cristae membrane is also discussed.

The use of electron tomography has allowed the three-dimensional membrane topography of the mitochondrion to be better understood. The most striking feature of this topology is the crista junction, a structure that may serve to divide functionally the inner membrane and intermembrane spaces. In situ these junctions seem to have a preferred size and shape independent of the source of the mitochondrion with few exceptions. When mitochondria are isolated and have a condensed matrix the crista junctions enlarge and become nondiscrete. Upon permeation of the inner membrane and subsequent swelling of the matrix space, the uniform circular nature of the crista junction reappears. We examine the distribution of shapes and sizes of crista junctions and suggest a thermodynamic model that explains the distribution based on current theories of bilayer membrane shapes. The theory of spontaneous curvature shows the circular junction to be a thermodynamically stable structure whose size and shape is influenced by the relative volume of the matrix. We conclude that the crista junction exists predominantly as a circular junction, with other shapes as exceptions made possible by specific characteristics of the lipid bilayer.

Electron tomography was used to help redefine the membrane architecture of mitochondria in neurons of the brain. Investigations were conducted on unexplored questions of structural homogeneity between mitochondria in the four intensely studied regions of the brain and in the functionally distinct neuronal sub-compartments. These mitochondria have the majority of cristae composed of both tubular and lamellar segments with the tubes arranged more peripherally and the lamellae more centrally located. Cristae that are entirely tubular were not commonly seen and those that are entirely lamellar were rare. It was determined that cristae connect through narrow, sometimes very long tubular regions to the peripheral surface of the inner membrane. A structurally distinct type of contact site was revealed in brain mitochondria, which we named the bridge contact site. These bridges may play a role in the structural integrity of the outer and inner membrane systems. It was found that the membrane architecture in the various brain regions and neuronal compartments was strikingly uniform, including consistently tubular crista junctions. The functional consequences of this junctional architecture are discussed in relation to the segregation of proteins between the inner boundary membrane and the cristae membranes, and in relation to the model of microcompartmentation of macromolecules inside cristae., (Copyright 2001 Wiley-Liss, Inc.)

In a mutant form of Neurospora crassa, in which sheltered RIP (repeat induced point mutation) was used to deplete Tom19, protein transport through the TOM/TIM pathway is arrested by the addition of p-fluorophenylalanine (FPA). Using intermediate-voltage electron tomography, we have generated three-dimensional reconstructions of 28 FPA-treated mitochondria at four time points (0-32 h) after the addition of FPA. We determined that the cristae surface area and volume were lost in a roughly linear manner. A decrease in mitochondrial volume was not observed until after 16 h of FPA treatment. The inner boundary membrane did not appear to shrink or contract away from the outer membrane. Interestingly, the close apposition of these membranes remained over the entire periphery, even after all of the cristae had disappeared. The different dynamics of the shrinkage of cristae membrane and inner boundary membrane has implications for compartmentalization of electron transport proteins. Two structurally distinct types of contact sites were observed, consistent with recently published work. We determined that the cristae in the untreated (control) mitochondria are all lamellar. The cristae of FPA-treated mitochondria retain the lamellar morphology as they reduce in size and do not adopt tubular shapes. Importantly, the crista junctions exhibit tubular as well as slot-like connections to the inner boundary membrane, persisting until the cristae disappear, indicating that their stability is not dependent on continuous protein import through the complex containing Tom19.

Electron microscopic (EM) tomography is providing important new insights into the internal organization of mitochondria. The standard baffle model for cristae structure, called into question years ago, has now clearly been shown to be inaccurate. Depending on source and conformational state, cristae can vary from simple tubular structures to more complex lamellar structures merging with the inner boundary membrane through tubular structures 28 nm in diameter. The structural information provided by EM tomography has important implications for mitochondrial bioenergetics, biogenesis and the role of mitochondria in apoptosis. The structural paradigm defined by EM tomography is helping in the design of new experimental approaches to mitochondrial function.

Novel applications of microscopy have recently provided new insights into mitochondrial structures. Diverse techniques such as high resolution scanning electron microscopy, transmission electron microscopy, electron microscope tomography and light microscopy have contributed a better understanding of mitochondrial compartmentalization, dynamic networks of mitochondria, intermembrane bridges, segregation of mitochondrial DNA and contacts with the endoplasmic reticulum among other aspects. This review focuses on advances reported in the last five years concerning aspects of mitochondrial substructure or dynamics gained through new techniques, whether they be novel microscope methods or new ways to prepare or label specimens. Sometimes these advances have produced surprising results and more often than not, they have challenged current conceptions of how mitochondria work.

Electron microscope tomography was used to examine the membrane topology of brown adipose tissue (BAT) mitochondria prepared by cryofixation or chemical fixation techniques. These mitochondria contain an uncoupling protein which results in the conversion of energy from electron transport into heat. The three-dimensional reconstructions of BAT mitochondria provided a view of the inner mitochondrial membrane different in important features from descriptions found in the literature. The work reported here provides new insight into BAT mitochondria architecture by identifying crista junctions, including multiple junctions connecting a crista to the same side of the inner boundary membrane, in a class of mitochondria that have no tubular cristae, but only lamellar cristae. Crista junctions were defined previously as the tubular membranes of relatively uniform diameter that connect a crista membrane with the inner boundary membrane. We have also found that the cristae architecture of cryofixed mitochondria, including crista junctions, is similar to that found in chemically fixed mitochondria, suggesting that this architecture is not a fixation artifact. The stacks of lamellar cristae extended through more of the BAT mitochondrial volume than did the cristae we observed in neuronal mitochondria. Hence, the inner membrane surface area was larger in the former. In chemically fixed mitochondria, contact sites were easily visualized because the outer and inner boundary membranes were separated by an 8 nm space. However, in cryofixed mitochondria almost all the outer membrane was observed to be in close contact with the inner boundary membrane.

Electron tomography is an extremely useful method for deriving three-dimensional structure from electron microscope images. The application of this technique to the reconstruction of large, complex structures such as mitochondria is described in conjunction with several tools for segmentation, measurement, classification, and visualization. In addition, the use of massively parallel computers to perform the tomographic reconstruction efficiently using R-weighted backprojection or iterative techniques is described.

Two-dimensional crystals of beef heart mitochondrial cytochrome c oxidase dimers were labeled at Cys-115 of subunit III with a monomaleimide derivative of an undecagold cluster compound. The binding site of the gold cluster compound and hence the site of subunit III were identified by image processing of cryoelectron micrographs of the crystals preserved in a mixture of glucose and uranyl acetate. The shape of the cytochrome oxidase dimer can be approximated as a parallelogram which is 44 by 82 A with an included angle of 80 degrees oriented with its long dimension along the a axis of the crystal. Labeling of subunit III was confirmed by a shift in the mobility of approximately 50% of subunit III molecules upon electrophoresis in polyacrylamide gels in the presence of sodium dodecyl sulfate. Averaged images of undecagold cluster labeled crystals and of unlabeled crystals were calculated; each image represents an average of approximately 17,000 molecules of either labeled or unlabeled cytochrome oxidase. On the basis of a statistical analysis of the differences between the two images, the gold cluster binds along a line 30 degrees from the a axis and 29 A from the center of the dimer. This result is interpreted in the context of other structural studies including the site of cytochrome c binding which Frey and Murray found to be near the a axis and 18 A from the center of the dimer [Frey, T. G., & Murray, J. M. (1994) J. Mol. Biol. 237, 275-297].(ABSTRACT TRUNCATED AT 250 WORDS)

Cytochrome c oxidase was isolated from beef heart mitochondria by detergent extraction yielding two different crystal forms. Extraction with Triton detergents produced vesicular crystals with two-dimensional crystalline arrays of cytochrome c oxidase dimers while extraction with sodium deoxycholate produced crystalline sheets of cytochrome c oxidase monomers. The structures of both crystal forms were determined in two-dimensional projection along an axis normal to the plane of the membrane by cryoelectron microscopy of crystals embedded in vitreous ice (frozen-hydrated). The projection structures of unstained frozen hydrated monomers and of dimers are similar to the structures of the crystals in negative stain. The molecular outline of dimers can be approximated by a parallelogram 44 A by 82 A with an included angle of 80 degrees. Monomers are less regular consisting of two large domains with a smaller domain at one end and a total length of approximately 82 A. Comparison of the two structures reveals the orientation of cytochrome c oxidase monomers within dimers, an orientation which is different from earlier models of monomer-monomer interaction, and suggests a very close interaction between monomers when they associate to form dimers. The crystalline sheets of cytochrome c oxidase monomers bind tightly the small peripheral membrane protein substrate, cytochrome c, and this binding accentuates a tendency of these crystals to stack upon one another. Images of crystals of the cytochrome c oxidase/cytochrome c complex were analyzed by crosscorrelation analysis versus the monomer crystal image. Two types of two-layer crystals have been identified. Both types have one layer rotated by 180 degrees with respect to the other, but they differ in the shifts of origin along crystal axes of the two layers. Difference images formed by subtracting simulated multilayered crystal images (which have no bound cytochrome c) from the complex crystals (cytochrome c oxidase plus cytochrome c) contain one positive difference peak for each cytochrome oxidase monomer within a unit cell. Comparison of the difference peak loci among the different crystal forms is interpreted based upon a consensus cytochrome c binding site in the single layer cytochrome oxidase monomer crystal image.

Cytochrome c oxidase is a complex integral membrane protein consisting of 13 different polypeptide chains and four metal centers having a total molecular weight of approximately 200,000 daltons. It can be isolated in two 2-dimensional crystalline forms differing in aggregation state of the enzyme. One crystal form consists of cytochrome oxidase dimers (approximately 400,000 daltons) embedded unidirectionally in the lipid bilayer of a collapsed vesicle while the other form consists of crystalline sheets of cytochrome oxidase monomers. Both crystal forms have been studied by electron microscopy during the past two decades, and this paper summarizes the results of early structural studies as well as more recent results applying techniques of cryoelectron microscopy and digital image processing. The structure of frozen-hydrated cytochrome oxidase dimers at 20 A resolution is discussed as well as the packing of monomers within dimers and the site of cytochrome c binding.

The polarized X-ray absorption spectra of the copper, iron and zinc sites of mitochondrial cytochrome oxidase in oriented membrane multilayers have been examined. The copper X-ray absorption edge spectra indicate the presence of a tetragonal copper, which we assign as CuB, oriented with the long axis approximately orthogonal to the membrane normal. We have also detected the presence of a relatively long (2.6 A) Cu-S or Cu-Cl interaction, which we assign to a copper-thioether (probably Met210) coordination at the CuA site, with the bond oriented along the membrane normal. The coordination of the zinc, the iron and the CuB heme a3 binuclear site are discussed.

The structure of cytochrome oxidase from beef heart mitochondria has been analysed by cryo-electron microscopy of vesicle crystals of the space group p22(1)2(1), with cell dimensions a = 102 A, b = 123 A, gamma = 90 degrees. Several methods of specimen preparation were applied to the vesicular two-dimensional crystals in the electron microscope, to ensure that the structure was preserved to the maximum resolution. The two most informative density maps were from specimens embedded in ice and from negative staining in a 1:1 mixture of glucose and uranyl acetate. The three-dimensional structure of the ice-embedded molecule shows a single, well resolved, but convoluted density, which represents in size and shape one cytochrome oxidase dimer. At the bottom of the molecule, a substantial part of the protein is embedded in the lipid bilayer of the vesicle. The molecule then extends upwards, out of the bilayer, into the internal space within the vesicle. Here, the structure first passes through a region within the molecule containing a hollow cavity that lies roughly at the centre of mass of the dimer, and then branches into two well-resolved halves at some distance from the membrane. The negatively stained structure, in contrast, shows a stain-excluding region in the centre of the vesicle at the level of the cavity in the ice-embedded structure, but otherwise has a similar overall external shape. In addition, there is a small rotation of the whole molecule by approximately 25 degrees relative to the orientation of ice-embedded specimens. We interpret these differences to mean that the central cavity seen in the ice-embedded structure is too small to allow the stain to penetrate during the drying process and that the drying process causes the rotation. The structures described here are consistent with one another and allow an interpretation at higher resolution than from previous work.

We have studied the spatial relationship between cytochromes a and a3 by the enhancement of the spin relaxation of cytochrome a3-NO EPR signals by the paramagnetic a heme at 15 K. An Fe-Fe distance of 12-19A is estimated from the absence of dipolar broadening and from the observation of spin relaxation enhancement in the a3-NO complex. When this result is combined with resonance x-ray diffraction data reported by Blasie et al. (Blasie, J. K., Pachence, J. M., Tavormina, A., Dutton, P. L., Stamatoff, J., Eisenberger, P., and Brown, G. (1982) Biochim. Biophys. Acta 679, 188-197) and the contribution from the exchange interaction is considered, we can limit the iron-iron distance to 12-16 A and estimate the angle between the Fe-Fe vector and mitochondrial membrane normal as 30-60 degrees. We also consider the possible effects of CuA on cytochrome a3-NO.

When cobaltous ion is bound to glutamine synthetase [L-glutamate:ammonia ligase (ADP-forming), EC 6.3.1.2], the two-layered hexagonal molecules polymerize face-to-face, to form long strands. The strands then wind round each other to form three- and seven-stranded cables. The structures of these cables are not immediately evident from electron micrographs because of the confusing superposition of front and back portions of the cables. But optical diffraction and filtering by the procedure of Klug and DeRosier leads to interpretable images of the cables. Because a micrograph of the seven-stranded cable contains 24 views of the glutamine synthetase molecule, it is possible to reconstruct the three-dimensional electron density of a cable and its constituent molecules at a resolution of 30--50 A. This reconstruction confirms that the symmetry of a glutamine synthetase molecule is D6. It suggests that the single subunit is an oblate ellipsoid with its minor axis (about 48 A) roughly parallel to the 6-fold axis of the molecule and its major axis (about 63 A) perpendicular to the 6-fold axis of the molecule. The subunits of the two hexagonal layers of a molecule are eclipsed. Neighboring molecules along a strand also have their hexagonal faces together, but they are rotated about the strand axis by about 7 degrees with respect to one another, rather than being eclipsed. Six outer strands are coiled about a straight central strand, and each forms identical contacts with the central strand. Moreover, these contacts between central and outer strands are apparently similar to the contacts between neighboring outer strands.

By use of mixed infections with conditional lethal mutations in the head genes and an osmotic shock-resistant mutant we have demonstrated that osmotic shock resistance is controlled by gene 24. Using acrylamide gel electrophoresis combined with the "immune replicate" technique, we confirmed the positions of gene products 24 and 24* (P24 and P24*). In this paper we have still used the notation "P24," etc., for designating the product of gene 23, etc., although we prefer and use in general the designation "gp23" as introduced by Casjens and King (Annu. Rev. Biochem. 44:585, 1975). The reason for using the old notation is because the illustrations were prepared several years ago.) P24 ts showed a significantly slower mobility. Both osmotic shock-resistant and -sensitive mature phages contain 24*. Giants constructed with the Osr phage showed the same surface lattice as normal phage. Through temperature-shift experiments with 24(tsL90) alone and in combinations, we studied the phages which are matured after the shift to permissive temperature in the absence of new protein synthesis. Our results strongly suggest that only a fraction of the total phage complement of gene 24-controlled proteins is involved in determining the phenotype of shock resistance, and the remainder is necessary to mature the head.

The structure and the orientation of cytochrome c oxidase molecules in crystalline cytochrome c oxidase membranes (Vanderkooi, G., Senior, A.E., Capaldi, R.A., and Hayashi, H. (1972) Biochim. Biophys. Acta 274, 38-48) were studied by image analysis of electron micrographs and by reacting the crystalline preparations with immune gamma-globulins against individual cytochrome c oxidase subunits. Binding of gamma-globulins to the membranes was detected by the following two methods: (a) electrophoretic identification of gamma-globulin polypeptides in the washed membranes; (b) electron microscopic examination of the negatively stained membranes. The membranes bound immune gamma-globulins against subunit IV (which faces the matrix side in intact mitochondria) but failed to bind immune gamma-globulins against subunits II + III (which face the outer side of the inner membrane in intact mitochondria). In contrast, solubilized cytochrome c oxidase bound either of the two immune gamma-globulins. All cytochrome c oxidase molecules in the crystalline membranes are thus asymmetrically arranged so that subunit IV faces outward and subunits II + III face toward the interior. This orientation is opposite to that found with intact mitochondria. The data also suggest that the crystalline membranes form closed vesicles which are impermeable to externally added gamma-globulins.

Electron microscopic images of selectively contrasted cytochrome oxidase dimer crystals are interpreted in a manner consistent with the structure of monomers determined by Fuller et al. (J. Molec. Biol. 134, 305-327). The arms of the y-shaped monomers lie within and perpendicular to the lipid bilayer protruding approximately 25 A on the matrix side of the membrane. The cytoplasmic-side tails of two monomers spread apart in a dimer forming a large cleft. Decoration of the exposed matrix side of vesicle crystals with antisubunit IV antibody fragments indicates that subunit IV lies along the a-crystal axis roughly 20 A from the center of the dimer. A membrane propensity algorithm applied to the sequences of cytochrome oxidase subunits predicts a total of 19 transmembrane alpha-helices per monomer.

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.

We have used various techniques for preparation of specimens for electron microscopy in order to selectively contrast different regions of vesicle crystals of cytochrome c oxidase dimers. The results are consistent with a dimer composed of two y-shaped monomers [Fuller et al., J. Mol. Biol. 134 (1979) 305] aligned along one pair of arms with the other pair of arms approximately 70 A apart. The four arms of the monomers lie within and perpendicular to the lipid bilayer in which the dimer is embedded, and the arms protrude approximately 25 A from the lipid bilayer on the matrix side of the membrane. The cytoplasmic side domains of the two monomers split away from one another forming a large cleft in the dimer. Monovalent antibodies (Fab fragments) to subunit IV appear to bind to the two monomer arms which are closely apposed across the two-fold axis of the dimer.

A simple and rapid method for the isolation of a large quantity of cytochrome c oxidase from bovine heart mitochondria was developed, based on selective solubilization of mitochondrial protein with first Triton and then lauryl maltoside. Gel filtration shows that the lauryl maltoside-solubilized oxidase preparation is in a hydrodynamically homogeneous state with a Stokes radius of 7.5 +/- 0.2 nm. It contains 8.0 mumol of haem (with an a/a3 ratio of 1)/g of protein. The catalytic constant (maximum turnover number) with respect to cytochrome c approaches 600 S-1. After further purification of the solubilized enzyme on a sucrose-gradient centrifugation, the purified enzyme has a haem content of 10.3 mumol/g of protein and eight major polypeptide bands shown on SDS/polyacrylamide-gel electrophoresis.

Most contrast agents used in NMR imaging studies to date have been paramagnetic. However, it is also possible to obtain selective contrast with a ferromagnetic agent, and these agents are potentially more sensitive than paramagnetic compounds because of their large magnetic moments. The water relaxation ability of ferromagnetic, albumin-coated magnetite (Fe3O4) particles has been investigated. These particles are quite effective at reducing both T1 and T2 at relatively low particle concentrations. The potential applications of these particles include improved visualization of the liver, gastrointestinal tract, and genitourinary tract, as well as specific targeting and detection of small tumors or other cells with unique surface receptors.