Isolation of plasma and nuclear membranes of thymocytes. I. Enzymatic composition and ultrastructure

The purpose of this work was to isolate thymocyte plasma membranes at high yield and purity to study specific surface molecules in their structural context. A procedure was developed in which 92-95% of the cells were disrupted by homogenization in a dense viscous medium, while nuclei remained intact. Differential centrifugation of the homogenate was avoided; instead, only a brief (2 h) centrifugation at equilibrium-density of membrane components was used. Five fractions were obtained, three by flotation. Membrane-bound enzymatic activities indicated a 60-80% yield of plasma membranes in the three floated membrane fractions, which comprised 1.6% of the homogenate protein. Enrichment factors for three ectoenzymes, alkaline phosphatase, gamma-glutamyltransferase, and ouabain-sensitive adenosine triphosphatase were respectively, 70-74, and 40-50 in the two lightest fractions. Nuclear membranes were then isolated from the remaining whole nuclei and were found to be enriched in esterase and NADH-cytochrome c reductase. Plasma membranes and light nuclear membranes appeared as pure unit-membrane vesicles in thin sections and freeze-etching electron microscopy. Some aggregation of intramembranous particles occurred in plasma membrane vesicles.     

Studies on plasma membranes: XIX. Isolation and characterization of a plasma membrane fraction from calf thymocytes

A plasma membrane fraction was isolated from calf thymocytes by a modification of the method of Wallach and Kamat (Wallach, D. F. H. and Kamat, V. B. (1966) in Methods in Enzymology) (Colowick, S. P. and Kaplan, N. O., eds), Vol. 8, pp. 164–172, Academic Press, New York). Fractions were examined electron microscopically and subjected to chemical and enzymic assays.With respect to the cell homogenate and the final microsomal fraction, respectively, the plasma membrane fraction was enriched by a factor 23 and 5.1 in cholesterol, 11 and 2.4 in phospholipid, 5.1 and 4.2 in sialic acid, 20 and 5.2 in Mg²⁺-ATPase (EC 3.6.1.3), and 8 and 2.6 in 5′-nucleotidase (EC 3.1.3.5). Succinate:cytochrome $\text{c}$ oxidoreductase (EC 1.3.99.1) was lacking, and DNA was hardly if at all present in the plasma membrane fraction. The major part of the RNA found in this fraction (30 μg · mg^?1 protein) was concluded to be an authentic component of the plasma membrane.The concept of membrane “markers” was briefly considered and the conclusion was reached that by current criteria and electron microscopic evidence the plasma membrane fraction obtained from calf thymocytes consisted of reasonably clean plasma membranes.     

Lipid composition and microviscosity of subcellular fractions from rabbit thymocytes. Differences in the microviscosity of plasma membranes from subclasses of thymocytes.

There are indications from freeze-fracture experiments that subclasses of rabbit thymocytes show different mobilities of plasma membrane components. Consequently, one would expect differences in the fluidity of the plasma membrane. For this reason, rabbit thymocytes were separated on a Ficoll/Metrizoate gradient yielding three subclasses representing various levels of cell differentiation. These thymocyte subclasses did not show any significant differences in the degree of fluorescence polarization using the probe 1,6-diphenyl-1,3,5-hexatriene. The fluorescence polarization of the plasma membrane may be overshadowed by the contribution of all cellular lipids due to penetration of the fluorescent probe into the cell. Therefore, plasma membranes were isolated from rabbit thymocytes using a cell-disrupting pump, differential centrifugation, and sucrose density gradient centrifugation. As shown by biochemical and electron microscopical analyses, plasma membranes with a high degree of purity were obtained. As expected the plasma membrane fractions showed a higher microviscosity than the other subcellular fractions. This was attributed to a higher cholesterol to phospholipid molar ratio and a higher degree of saturation of phospholipid fatty acid chains. Subsequently, the microviscosity was measured of plasma membrane preparations obtained from two main subclasses of thymocytes representing mature and immature lymphocytes. The immature thymocytes yielded two plasma membrane fractions with higher microviscosity than the mature cells. These finding is in line with earlier observed differences in the glycerol-induced clustering of intramembranous particles. Furthermore, the results of this study support the view that the fluorescence polarization technique applied to whole cells does not exclusively monitor the plasma membrane.     

Lipid composition and microviscosity of subcellular fractions from rabbit thymocytes. Differences in the microviscosity of plasma membranes from subclasses of thymocytes

There are indications from freeze-fracture experiments that subclasses of rabbit thymocytes show different mobilities of plasma membrane components. Consequently, one would expect differences in the fluidity of the plasma membrane. For this reason, rabbit thymocytes were separated on a Ficoll/Metrizoate gradient yielding three subclasses representing various levels of cell differentiation. These thymocyte subclasses did not show any significant differences in the degree of fluorescence polarization using the probe 1,6-diphenyl-1,3,5-hexatriene. The fluorescence polarization of the plasma membrane may be overshadowed by the contribution of all cellular lipids due to penetration of the fluorescent probe into the cell. Therefore, plasma membranes were isolated from rabbit thymocytes using a cell-disrupting pump, differential centrifugation, and sucrose density gradient centrifugation. As shown by biochemical and electron microscopical analyses, plasma membranes with a high degree of purity were obtained. As expected the plasma membrane fractions showed a higher microviscosity than the other subcellular fractions. This was attributed to a higher cholesterol to phospholipid molar ratio and a higher degree of saturation of phospholipid fatty acid chains.Subsequently, the microviscosity was measured of plasma membrane preparations obtained from two main subclasses of thymocytes representing mature and immature lymphocytes. The immature thymocytes yielded two plasma membrane fractions with higher microviscosity than the mature cells.     

The isolation and characterization of plasma membranes from calf thymocytes.

A simple method is described for the isolation and characterization of plasma membranes from calf thymocytes. The procedure involves extraction of thymocytes in a hypotonic medium containing borate and EDTA. Membrane ghosts, obtained by centrifugation of the cell lysate, are purified by passage through a column containing glass beads. The purity of plasma membranes was checked by chemical analysis, by assay of marker enzymes and also by electron microscopy. Polyacrylamide gel electrophoresis of the calf thymocyte plasma membrane produced a number of protein bands as well as a major band which stained for carbohydrate. The method is rapid and could be applied to isolate plasma membranes from nucleated cells of various types in large quantities.     

Isolation of rat hepatocyte plasma membranes. II. Identification of membrane-associated cytoskeletal proteins

Rat liver plasma membranes were isolated as presented in the preceding paper (Hubbard, A. L., D. A. Wall, and A. Ma., 1983, J. Cell Biol., 96: 217-229) and found to contain many filaments associated both with desmosomes along the lateral surface and with the cytoplasmic aspects of membranes comprising each of the three domains (lateral [LS], bile canalicular [BC] and sinusoidal [SF] ). Exposure of the plasma membranes to alkaline media (up to pH 11) resulted in loss of recognizable filaments without loss of domain morphology or membrane enzyme activities. Electrophoretic analysis of solubilized components from control and alkaline-extracted plasma membranes revealed that three major polypeptides present at 43, 52, and 56 kdaltons in the control had been released by alkaline treatment (pH 11) and could be quantitatively recovered in the supernate. The 43-kdalton component was identified as cytoplasmic actin by comparison of its tryptic 125I- peptide map to those of muscle (alpha) and brush border (beta, gamma) actins. The 52- and 56-kdalton polypeptides were identified as tonofilament components by their solubility properties and their ability to reassemble into 9.5-nm filaments from monomers present in an alkaline extract.     

Density of newly synthesized plasma membrane proteins in intracellular membranes II. Biochemical studies

Using two independent methods, incorporation of radioactive amino-acid and quantitative immunoblotting, we have determined that the rate of synthesis of each of the Semliki Forest virus (SFV) proteins in infected baby hamster kidney (BHK) cells is 1.2 X 10(5) copies/cell/min. Given the absolute surface areas of the endoplasmic reticulum and Golgi complex presented in the companion paper (Griffiths, G., G. Warren, P. Quinn , O. Mathieu - Costello , and A. Hoppeler , 1984, J. Cell Biol. 98:2133-2141), and the approximate time spent in these organelles during their passage to the plasma membrane (Green J., G. Griffiths, D. Louvard , P. Quinn , and G. Warren 1981, J. Mol. Biol. 152:663-698), the mean density of each viral protein in these organelles can be calculated to be 90 and 750 molecules/micron 2 membrane, respectively. In contrast, we have determined that the density of total endogenous integral membrane proteins in these organelles is approximately 30,000 molecules/micron 2 so that the spike proteins constitute only 0.28 and 2.3% of total membrane protein in the endoplasmic reticulum and Golgi, respectively. Quantitative immunoblotting was used to give direct estimates of the concentrations of one of the viral membrane protein precursors (E1) in subcellular fractions; these agreed closely with the calculated values. The data are discussed with respect to the sorting of transported proteins from those endogenous to the intracellular membranes.     

Isolation of Raja erinacea basolateral liver plasma membranes: characterization of lipid composition and fluidity.

Developing a method for isolating skate (Raja erinacea) basolateral liver plasma membranes, as well as characterizing the lipid composition and fluidity of these membranes, was the primary purpose of this study. Membranes were isolated using self-generating Percoll gradients. Marker enzyme studies indicate that this preparation is highly enriched in the basolateral domain of the liver plasma membrane and largely free of contamination by intracellular organelles or canalicular membranes. Further, these membranes contain the agency responsible for Na(+)-dependent alanine transport. This finding indicates that this membrane preparation will be useful for the study of skate liver plasma membrane transport processes. The lipid composition and fluidity (as assessed by the fluorescence anisotropy of 1,6-diphenyl-1,3,5-hexatriene) of the skate basolateral liver plasma membrane shows little variation among preparations. Further, DPH anisotropy plotted as a function of temperature yields a straight line (r = 0.99) which indicates that there is no lipid phase change in these membranes from 4 degrees to 37 degrees C. The membrane preparation does contain substantial phospholipase A2 activity. The function of this enzyme is, in part, to modify membrane lipid composition and fluidity in response to temperature variations; therefore, this finding suggests that in situ lipid metabolizing enzymes may play a central role in the adaptation of skate basolateral liver plasma membranes to changes in the ambient temperature.     

Isolation and lipid composition of nuclear membranes from macronuclei of Tetrahymena pyriformis

A procedure was developed for isolation of macronuclei and nuclear membranes from the ciliated protozoan Tetrahymena pyriformis E, and the lipid composition of the isolated nuclear membranes was determined.This method involves cell lysis with octanol, separation of the nuclear membrane with 0.2 M phosphate–1M NaCl and purification on a discontinuous sucrose gradient. By phase-contrast and electron microscopic examinaton, our preparations were pure and preserved the typical nuclear membrane morphology: inner and outer nuclear membranes, and nuclear pore complexes. As for lipid distribution, the three major phospholipids in the membranes were phosphatidylcholine (31.0%), phosphatidylethanolamine (26.1%) and 2-aminoethylphosphonolipids (23.3%) and the molar ratio of a sterol-like lipid, tetrahymanol to phospholipid phosphorus was 0.036. These results were compared to other membrane fractions of Tetrahymena.     

Isolation of plasma membrane glycoprotein from bovine thymocytes

Glycoprotein was isolated from a purified thymocyte membrane preparation by two methods: lithium diiodosalicylate-phenol extraction and hot 75% ethanol extraction. A higher yield of membrane sialic acid was obtained by the latter method. The preparations had similar apparent molecular weights on sodium dodecyl sulfate gel electrophoresis. Both had similar receptor activities against a panel of hemagglutinins, although the 75% ethanol extract was more active on a weight basis. However, there were significant differences in carbohydrate and amino acid compositions of the two thymocyte extracts. The lithium diiodosalicylate-extracted material had much more glucose, ribose, and glycine than the ethanol extract. The glycoprotein preparations from thymocytes were quite distinct from the glycoprotein of bovine erythrocytes in both composition and receptor properties.     

Isolation and characterization of Neurospora crassa plasma membranes.

The isolation and characterization of plasma membranes from a cell wall-less mutant of Neurospora crassa are described. The plasma membranes are stabilized against fragmentation and vesiculation by treatment of intact cells with concanavalin A just prior to lysis. After lysis, the concanavalin A-stabilized plasma membrane ghosts are isolated by low speed centrifugation techniques and the purified ghosts subsequently converted to vesicles by removal of the bulk of the concanavalin A. The yield of ghosts is about 50% whereas the yield of vesicles is about 20%. The isolated plasma membrane vesicles have a characteristically high sterol to phospholipid ratio, Mg2+-dependent ATPase activity and (Na+ plus K+)-stimulated Mg2+ATPase activity. Only traces of succinate dehydrogenase and 5'-nucleotidase are present in the plasma membrane preparations.     

Isolation and protein composition of membranes of neurosecretory vesicles and plasma membranes from the neural lobe of the bovine pituitary gland.

1. Homogenates of neural lobes of bovine pituitary glands were fractionated by differential ultracentrifugation. 2. Neurosecretory vesicles were isolated by sucrose-gradient ultracentrifugation and membranes were obtained after hypo-osmotic lysis of the particles. 3. A method is described for the isolation of a preparation of purified neuronal plasma membranes by using a fraction enriched in nerve endings as a starting material. 4. The purity of the subcellular fractions was estimated by enzyme assays and by examination with the electron microscope. 5. On the basis of the results it was estimated that neuronal plasma membranes constitute more than 30% of the protein of the nerve endings and neurosecretory vesicles more than 45% of the total amount of protein in the homogenate. 6. The proteins of membranes of neurosecretory vesicles and of plasma membranes were solubilized by means of sodium dodecyl sulphate. Polyacrylamide-gel electrophoresis of such preparations showed that both membranes contained a large number of proteins, including three glycoproteins.     

Analysis of the pinocytic process in rat kidney II. Biochemical composition of pinocytic vesicles compared to brush border microvilli,lysosomes and basolateral plasma membranes

Pinocytic vesicles, brush border microvilli, lysosomes and basolateral plasma membranes were isolated from rat kidney cortex and their biochemical composition and membrane turnover compared. Pinocytic vesicles are devoid of marker enzymes of brush border microvilli, such as alkaline phosphatase and 5′-nucleotidase, and of lysosomes, such as acid phosphatase and β-glucuronidase. The protein pattern as revealed by polyacrylamide gel electrophoresis differs for all four membranes. Analysis of the phospholipid composition shows that pinocytic vesicles are rich in the negatively charged phospholipid phosphatidylserine and have a low content of sphingomyelin and phosphatidylethanolamine.[¹⁴C]guanido-arginine, [³H]fucose and $\text{myo-}\text{[}{}^3\text{H]inositol}$ were preferentially incorporated into the pinocytic vesicles. Using a double label technique with leucine also, evidence of a more rapid turnover of the pinocytic vesicle membrane proteins was obtained.The results suggest that pinocytic vesicles are not derived from the brush border microvillous membrane but are independent entities that are newly synthesized during the pinocytic process.     

Isolation and characterization of plasma membranes from krebs II ascite cells using Percoll gradient

1. Plasma membranes were isolated from Krebs II ascite cells grown in the mouse. Cells were disrupted by nitrogen cavitation in an isotonic alkaline buffer containing magnesium and ATP. Isolation was performed in an alkaline-buffered self-generating gradient of Percoll with an angular rotor. At each step of the preparation, the pH appeared as the critical aspect of our procedure. 2. External membrane markers were concanavalin A and 5'-nucleotidase (EC 3.1.3.5). They reached a relative specific activity of 10, whereas this value was only of 0.7 for the endoplasmic reticulum marker, NADH dehydrogenase (EC 1.6.99.3). 3. Plasma membrane from 4 ml packed cells were isolated within 1 h after homogenization with good yield: 50% and 67% of total [3H]concanavalin A and 5'-nucleotidase, respectively, were recovered in the two plasma membrane fractions. 4. Electron microscopy examination showed the presence of vesicles of different sizes devoid of other structural contaminants. 5. Using the specific binding of concanavalin A to the external cell membrane, it was calculated that about 50% of the total cell phospholipid and 10% protein are located in the plasma membrane. Their sphingomyelin content is much higher than in the whole cell, in contrast to phosphatidylinositol, known as a more specific endoplasmic reticulum phospholipid.