Relations between plasma membrane and lysosomal membrane. 1. Fate of covalently labelled plasma membrane protein

To quantify the kinetics of the plasma membrane flow into lysosomes, we covalently labelled at 4 degrees C the pericellular membrane of rat fibroblasts and followed label redistribution to the lysosomal membrane using purified lysosomal preparations. The polypeptides were, either labelled with 125I by the lactoperoxidase procedure, or conjugated to [3H]peroxidase using bisdiazobenzidine as a bifunctional reagent. Both labels were initially bound to plasma membrane, as indicated by their equilibrium density in sucrose or Percoll gradients and their displacement by digitonin, as well as by electron microscopy. Upon cell incubation at 37 degrees C, both covalent labels were lost from cells with diphasic kinetics: a minor component (35% of cell-associated labels) was rapidly released (half-life less than 1 h), and most label (65%) was released slowly (half-life was 20 h for incorporated 125I and 27 h for 3H). Immediately after labelling up to 30 h after incubation at 37 degrees C, the patterns of 125I-polypeptides quantified by autoradiography after SDS-PAGE were indistinguishable, indicating no preferential turnover for the major plasma membrane polypeptides. The redistribution of both labels to lysosomes was next quantified by cell fractionation. At equilibrium (between 6 and 25 h of cell incubation) 2-4% of cell-associated 125I label was recovered with the purified lysosomal membranes. By contrast, when 3H-labelled cells were incubated for 16 h, most of the label codistributed with lysosomes. However, only 6% of cell-associated 3H was bound to lysosomal membrane. These results indicate that in cultured rat fibroblasts, a minor fraction of plasma membrane polypeptides becomes associated with the lysosomal membrane and is constantly equilibrated by membrane traffic.     

Characteristics of a membrane reservoir buffering membrane tension.

When membrane-attached beads are pulled vertically by a laser tweezers, a membrane tube of constant diameter (tether) is formed. We found that the force on the bead (tether force) did not depend on tether length over a wide range of tether lengths, which indicates that a previously unidentified reservoir of membrane and not stretch of the plasma membrane provides the tether membrane. Plots of tether force vs. tether length have an initial phase, an elongation phase, and an exponential phase. During the major elongation phase, tether force is constant, buffered by the "membrane reservoir." Finally, there is an abrupt exponential rise in force that brings the tether out of the trap, indicating depletion of the membrane reservoir. In chick embryo fibroblasts and 3T3 fibroblasts, the maximum tether lengths that can be pulled at a velocity of 4 microm/s are 5.1 +/- 0. 3 and 5.0 +/- 0.2 microm, respectively. To examine the importance of the actin cytoskeleton, we treated cells with cytochalasin B or D and found that the tether lengths increased dramatically to 13.8 +/- 0.8 and 12.0 +/- 0.7 microm, respectively. Similarly, treatment of the cells with colchicine and nocodazole results in more than a twofold increase in tether length. We found that elevation of membrane tension (through osmotic pressure, a long-term elevation of tether force, or a number of transitory increases) increased reservoir size over the whole cell. Using a tracking system to hold tether force on the bead constant near its maximal length in the exponential phase, the rate of elongation of the tethers was measured as a function of tether force (membrane tension). The rate of elongation of tethers was linearly dependent on the tether force and reflected an increase in size of the reservoir. Increases in the reservoir caused by tension increases on one side of the cell caused increases in reservoir size on the other side of the cell. Thus, we suggest that cells maintain a plasma membrane reservoir to buffer against changes in membrane tension and that the reservoir is increased with membrane tension or disruption of the cytoskeleton.     

Electrostatics of membrane adhesion.

We consider electrical double layer interaction under the conditions typically encountered during membrane fusion. Within the physiological concentration range of monovalent electrolytes the interaction is repulsive and the Poisson-Boltzmann calculation may be used to evaluate the force. When divalent counterions are added, strong ion-ion correlations make the Poisson-Boltzmann approximation inapplicable. We use the anisotropic hypernetted chain method to show that in the presence of small amounts of divalent counterions in adsorption equilibrium with the surfaces, the double layer interaction turns into attraction. This attractive electrostatic force may be the balancing contribution controlling membrane adhesion.     

Spiroplasma membrane lipids.

Membranes of six spiroplasma strains belonging to different Spiroplasma species and subgroups were isolated by a combination of osmotic lysis and sonication in the presence of EDTA to block endogenous phospholipase activity. Analysis of membrane lipids showed that in addition to free and esterified cholesterol the spiroplasmas incorporated exogenous phospholipids from the growth medium. Sphingomyelin was preferentially incorporated from phosphatidylcholine-sphingomyelin vesicles or from the serum used to supplement the growth medium. Palmitate was incorporated better than oleate into membrane lipids synthesized by the organisms during growth. The major phospholipid synthesized by the spiroplasmas was phosphatidylglycerol. The positional distribution of the fatty acids in phosphatidylglycerol of Spiroplasma floricola resembled that found in Mycoplasma species, in which the saturated fatty acids prefer position 2 in the glycerol backbone and not position 1 as found in Acholeplasma species and elsewhere in nature. Electron paramagnetic resonance analysis of spin-labeled fatty acids incorporated into S. floricola membranes exhibited homogeneous single-component spectra without immobilized regions. The S. floricola membranes were more rigid than those of Acholeplasma laidlawii and less rigid than those of Mycoplasma gallisepticum.     

Self-assembly of membrane junctions.

We present a mechanism for the aggregation of mobile intermembrane junctions, such as the connexon dyad of gap junctions. The model demonstrates that intermembrane repulsion provides a powerful self-assembly pressure. If the membrane repulsion is strong enough to prevent membrane adhesion, then the self-assembly pressure is of effective infinite range.     

Localization of membrane proteins in membrane vesicles of Bacillus subtilis.

Electrons can be transferred to the respiratory chain in whole cells and in membrane vesicles of Bacillus subtilis W 23 by the membrane impermeable electron donor reduced 5-N-methyl-phenazonium-3-sulfonate as efficiently as by the membrane permeable electron donor reduced 5-N-methyl-phenazonium methyl-sulfate, indicating that the respiratory chain is accessible from the outside of the membrane.Succinate is oxidized by whole cells and membrane vesicles at a low rate and does not energize transport of l-glutamate. In the presence of 5-N-methyl-phenazonium-3-sulfonate or 5-N-methyl-phenazonium methyl-sulfate, the oxidation rate and the rate of l-glutamate transport are increased considerably. The electrons are transferred directly from succinic dehydrogenase to these acceptors. Succinic dehydrogenase must therefore be exposed to the outside surface of the membrane in both membrane vesicles and whole cells. The exposure of succinic dehydrogenase to the outside is also indicated by the observations that only a 5% increase in the oxidation rates of succinate-5-N-methyl-phenazonium methylsulfate and succinate-5-N-methyl-phenazonium-3-sulfonate is observed upon solubilization of the membrane with the nonionic detergent Brij-58. Furthermore, treatment of membrane vesicles with trypsin decreases by more than 95% these oxidation rates.NADH is oxidized at a high rate and energizes transport of l-glutamate in whole cells and membrane vesicles effectively. The NADH-oxidation is not effected by trypsin treatment of the vesicles indicating that the oxidation occurs at the inside-surface of the membrane. Trypsin treatment of the vesicles, however, significantly decreases the rate of l-glutamate transport driven by NADH. Therefore component(s) of the transport system for l-glutamate must be effected by trypsin treatment. No apparent differences could be observed in the localization of membrane-bound functions between membrane vesicles and whole cells. This strongly supports the contention that the vesicle membrane of B. subtilis has the same orientation as the cytoplasmic membrane of whole cells.     

Plant membrane transport.

Recent developments in plant membrane transport, particularly concerning the vacuolar and plasma membranes, have increased our understanding of molecular aspects of primary pumps, carrier systems and ion channels.     

Outer membrane proteins of Escherichia coli. II. Heterogeneity of major outer membrane polypeptides

When E. coli outer membrane protein is dissolved in sodium dodecyl sulfate (SDS) solution and boiled briefly, a single major peak (peak B) with a molecular weight of 42,000 daltons is observed on SDS-containing polyacrylamide gels. If the protein is dissolved in SDS solution at 37 °C and applied to gels without further treatment, peak B disappears and two other major peaks appear: Peak A, which is composed of aggregates and migrates more slowly than peak B, and peak C which is composed of monomeric protein not fully reacted with SDS and which migrates faster than peak B. When cyanogen bromide peptides of protein from peak A and peak C were compared, it was evident that peak A and peak C contained entirely different polypeptides. This was further confirmed by differential labeling studies with methionine and leucine. The cyanogen bromide peptide profiles of protein from peak A suggested that this peak was composed of two polypeptides, and this was confirmed by electrophoresis in an alkaline gel system which resolves peak B into three subcomponents. Two of these were derived from peak A and the third was derived from peak C. These results indicate that the outer membrane of E. coli contains at least three nonidentical major polypeptides, each of which has a nearly identical molecular weight of about 42,000 daltons. These polypeptides are present in identical proportions in the soluble and insoluble fractions obtained when the outer membrane is treated with Triton X-100 plus EDTA.