David W. Allmann

Formation of membranes by repeating units

Abstract Electron microscopic evidence is presented for membrane formation by the repeating units of (1) the individual complexes of the electron transfer chain; (2) the outer membrane of beef heart mitochondria; (3) the mitochondrial membranes of rat liver; (4) the chloroplast membrane of spinach; (5) the outer segments of bovine photoreceptors; (6) a bovine liver microsomal membrane; and (7) the plasma membrane of bovine erythrocytes. Several hitherto unexplained phenomena, such as the reconstitution of the electron transfer chain and the requirement of lipid for certain enzymic activities, are rationalized in terms of this capability of the repeating units to form membranes. On the basis of the properties of lipid-containing and lipiddepleted membranes, a theory of how membranes are formed has been developed. This theory depends primarily on the restrictions, imposed by phospholipids, on interactions between repeating units. Lipid-containing repeating units can associate into a continuum, one unit or macromolecule thick that can be identified with a membrane. By contrast, lipid-depleted repeating units tend to form a bulk phase, which can be identified with a random three-dimensional polymeric aggregate. Enzymic activities cannot be fully demonstrated when the repeating units exist in the state of a bulk phase.

The membrane systems of the mitochondrion: II. The K fraction of the outer membrane of beef heart mitochondria

Beef heart mitochondria contain at least two membranes: an inner membrane with infoldings (cristae) of a tubular nature, and with distinctive tripartite subunits; and an outer membrane with repeating subunits of a different nature and size. The inner membrane, which can be isolated in a relatively pure form, contains the four complexes of the electron transfer chain as well as enzymes such as ATPase. The outer membrane, which has also been isolated in a relatively pure fraction, contains the ancillary mitochondrial enzyme systems that implement citric acid cycle oxidation, fatty acid oxidation, fatty acid elongation, and substrate level phosphorylation. The outer membrane can be isolated either as a composite fraction (F) or in two distinct and separate fractions, one a particulate membranous fraction (K), and the other a soluble nonmenbranous fraction (S). In the present series of three communications we have described the preparation and properties of the soluble membrane fraction (S), the particulate membrane fraction (K), and the composite outer membrane fraction (F). The S fraction contains the citric acid cycle enzymes of low molecular weight, e.g., isocitric dehydrogenase, malic dehydrogenase, condensing enzyme, fumarase, and aconitase. This fraction also contains significant amounts of the enzymes concerned in oxidation and elongation of fatty acids as well as of enzymes with transfer function such as myokinase. The S fraction has been designated as the detachable sector of the outer membrane.

Membrane systems of mitochondria: VI. Membranes of liver mitochondria

Abstract Beef and rat liver mitochondria were isolated by procedures designed to eliminate contaminating microsomal activities (e.g., glucose 6-phosphatase and the rotenone-insensitive DPNH-cytochrome c reductase). Mitochondria thus prepared were completely competent in respect to electron transport and citric acid cycle activity. The matrix space of these mitochondria was ruled out as the site of the citric cycle activities on the basis of two lines of evidence: the inability of carboxydextran to penetrate the inner boundary membrane during the time of the release of the S fraction; and the apparent morphological integrity of the inner boundary membrane of mitochondria exposed to the action of oleate or phospholipase during the period of release of the S fraction. Liver mitochondria were fractionated into the soluble S fraction, the membranous K fraction, and the R2 fraction. Each of these fractions accounted for about 33% of the total mitochondrial protein. The S fraction, which contained all the readily solubilizable citric cycle activities, was shown to be localized between the two boundary membranes. The enzymes of this fraction serve as spacer units which link together the two boundary membranes. The K fraction, which contained the activities of particulate complexes such as the α-ketoglutarate dehydrogenase complex, was identified with the outer membranes. Finally, the R2 fraction, which contained all the activities of the electron transfer chain and ATPase and monoamine oxidase activity, was identified with the cristae. The repeating units of the outer membrane are devoid of projecting elements, and show up in electron micrographs as spheres, about 100 A in diameter. Cardiolipin accounts for about 15% of the total phospholipid phosphorus of the outer membrane fraction. The cholesterol content of the outer membrane is relatively low. The analytical data for the lipid composition of the outer membrane are compatible with the thesis that the cristae and outer membranes have the same set of phospholipids and the same ratio of phospholipid to cholesterol.

The membrane systems of the mitochondrion: III. The isolation and properties of the outer membrane of beef heart mitochondria

The outer membrane of beef heart mitochondria is composed of a detachable sector (S) and an invariant sector (K). These separate fractions (K and S) have been isolated and examined in detail, and the results were described in the preceding two communications. This communication describes the preparation and properties of the composite outer membrane fraction. This membrane fraction contains all the enzymes of the citric acid cycle and other ancillary enzyme systems present in the S and K fractions. As isolated, it is essentially devoid of any electron transfer components, accounts for the estimated mass of the outer membrane, contains about 25–30% phospholipid by dry weight, and resembles the intact outer membrane.

Resolution of the repeating unit of the inner mitochondrial membrane

Abstract The tripartite repeating unit of the inner mitochondrial membrane has been resolved into the following components: basepiece, headpiece, and headpiece-stalk. The headpiece contains rutamycin-insensitive ATPase activity; the basepiece contains activities of the electron transfer chain. Ultrastructural and chemical evidence is presented in support of two theses: ( 1 ) the headpiece-stalk sector accounts for the rutamycin-sensitive ATPase function; and ( 2 ) the stalk determines rutamycin sensitivity of the ATPase in the headpiece. An attempt has also been made to bring our observations on the rutamycin-sensitive ATPase and those of E. Racker into accord.

Studies on the transition of the cristal membrane from the orthodox to the aggregated configuration. I: Topology of bovine adrenal cortex mitochondria in the orthodox configuration

Bovine adrenal cortex mitochondria examined by electron microscopyin situ orin vitro in 0·25 M sucrose have an unusual cristal membrane structure. The cristae usually appear as unconnected vesicles within a double membrane system. A few of the vesicles appear to be attached to the inner boundary membrane or to one or more other vesicles. The configuration of such mitochondria will be defined as the orthodox configuration. In this communication we will provide evidence that the inner membrane is not composed of multiple vesicles, but is one continuous membrane with tubular invaginations, and that these invaginations alternately are ballooned out and squeezed down. A mechanism has been proposed to account for the differentiated structure of the cristae of adrenal cortex mitochondria.

The membrane systems of the mitochondrion. IV. The localization of the fatty acid oxidizing system.

Abstract The localization of the enzymes concerned in the various steps of fatty acid oxidation has been determined by the enzymic analysis of the isolated inner and outer membrane fractions of beef heart mitochondria. The enzymes that activate fatty acids, the carnitine-long chain acyl transferase, and the enzymes of the β-oxidation, have been found to be localized exclusively in the outer membrane. None of the activities associated with these enzymes could be detected in the preparation of the inner membrane. The enzymes for the complete oxidation of palmitate are localized in the outer membrane. The requirement for the carnitine-mediated transfer of the acyl group across the outer membrane from exterior to interior was also demonstrated. This finding establishes the outer membrane as the carnitine “barrier.” Atractyloside (an inhibitor of several mitochondrial reactions) is a potent inhibitor of fatty acid oxidation. This inhibition is exerted on enzymes associated with the membrane-forming sector of the outer membrane, and also on enzymes associated with the detachable sector of the outer membrane. The atractyloside “barrier” for fatty acid oxidation has been established as the outer membrane.

Mitochondrial structural protein I: Methods of preparation and purification: Characterization by gel electrophoresis

Abstract Structural protein prepared from heavy beef heart mitochondria by a variety of methods, was shown to be heterogeneous by polyacrylamide gel electrophoresis. Procedures are described which have been found to be effective in purifying structural protein to a stage at which it shows one band when electrophoresed on polyacrylamide gel. When structural protein so purified was oxidized with performic acid, and subjected to electrophoretic analysis, four major protein bands were visualized. A possible explanation of this observation is discussed. The solubility properties of structural protein varied according to the method of preparation.

Studies on the transition of the cristal membrane from the orthodox to the aggregated configuration. II: Determinants of the orthodox-aggregated transition in adrenal cortex mitochondria

Bovine adrenal cortex mitochondria when isolated in a medium 0·25 M in sucrose contaminated (with calcium) have tubular cristae which are periodically expanded to spherical vesicles and contracted to flattened connecting sections. This scalloped tubular form of the cristae corresponds to the nonenergized orthodox configuration—a configuration in which the matrix space is maximally expanded. When adrenal cortex mitochondria are isolated in media in which the free calcium content is relatively low, e.g., a medium 0·25 M in sucrose and 0·1 mM in EDTA, the cristae assume the aggregated configuration—a nonenergized configuration in which the matrix space is maximally contracted. The composition of the isolation medium determined the configuration. Procedures have been described for isolating bovine adrenal cortex mitochondria (a) predominantly in the orthodox configuration, (b) predominantly in the aggregated configuration, or (c) in a mixed (1∶1) population of configurations. The concentration of Ca2+ bound to the mitochondrion was found to be a determinant of the nonenergized configuration. When the level of bound Ca++ was 20–25 μmoles/mg protein, the cristae of the mitochondria were entirely in the orthodox configuration. Addition of Ca2+ could induce the transition of cristae from the aggregated to the orthodox configuration whereas addition of Mg2+ could induce the transition of cristae from the orthodox to the aggregated configuration. The configurational transition could be followed by any of several methods—a change in 90° light scattering, a change in O.D.520 mμ, a change in pH, or examination by electron microscopy. The orthodox to aggregated transition is energy-independent since it proceeds even in presence of inhibitors both of electron transfer and of ATP hydrolysis. The binding of Ca2+ is independent of the binding of Mg2+; this independent binding is consistent with the opposite effects induced by Ca2+ and Mg2+, respectively. Whereas Ca2+ induces a proton release, a decrease in 90° light scattering and a decrease in O.D.520 mμ (when the cristae are initially in the aggregated configuration), Mg2+ induces equal and opposite changes (when the cristae are initially in the orthodox configuration).

Site of action of atractyloside in mitochondria. II. Inhibition of oxidative phosphorylation.

Abstract The site of inhibition by atractyloside of oxidative phosphorylation and related processes in mitochondria prepared from beef heart was localized in the outer membrane. Support for this conclusion came from several lines of evidence: (a) the outer membrane is impenetrable to charged molecules of the size of acyl-CoA esters, adenine nucleotides, and pyridine nucleotides; (b) up to 60% of the total adenine nucleotides is localized in the outer membrane; (c) the release of AMP bound to the outer membrane is mediated by atractyloside; (d) exogenous adenine nucleotides can be bound by the outer membrane; (e) the loss of atractyloside inhibition parallels the loss (or alteration) of the outer membrane; (f) oxidative phosphorylation in submitochondrial particles such as ETPH is insensitive to atractyloside; and (g) the outer membrane contains a phosphoryl transferase (nucleoside monophosphokinase) which is atractyloside-sensitive.