Phosphorylation in membranes of intact human erythrocytes.

Studies of phosphorylation in membranes of intact human erythrocytes were performed by incubating erythrocytes in inorganic [32P]phosphate. Analysis of membrane proteins by polyacrylamide gel electrophoresis showed a pattern of phosphorylation similar to that observed when ghost membranes were incubated with [gamma-32P]ATP. Membrane lipid phosphorylation was also similar in intact cells and ghosts. The most heavily phosphorylated lipid, polyphosphoinositide, was closely associated with glycophorin A, the major erythrocyte membrane sialoglycoprotein obtained when the sialoglycoprotein fraction was isolated by the lithium diiodosalicylate-phenol partition procedure. Only 1 molecule of glycophorin A out of every 100 was found to be phosphorylated, and the phosphate exchange occurred specifically in the COOH-terminal intracellular portion of glycophorin A. These studies show that the human erythrocyte can be used as a model for membrane phosphorylation in an intact cell system.     

Requirements of drug-induced endocytosis by intact human erythrocytes.

Study of drug-induced endocytosis in intact human erythrocytes continues to provide an opportunity for correlating membrane functions such as invagination and fusion with erythrocytic energetics and other determinants of plasma membrane function like Ca++. The studies reported indicate that high concentrations of vinblastine and chlorpromazine can produce endocytic vacuoles, albeit in reduced amounts, even in severely ATP depleted erythrocytes. In contrast, primaquine-induced endocytosis seems definitely dependent upon persistence of erythrocytic ATP stores. The ionophore mediated entry of Ca++ into erythrocytes potentiates primaquine endocytosis, inhibits vinblastine endocytosis, and has no regular effect on chlorpromazine endocytosis. Sodium lactate enhances primaquine endocytosis, probably by causing an increase in the entry of primaquine into erythrocytes. Cytochalasin B neither enhances nor inhibits erythrocytic endocytosis, thereby suggesting that microfibrils or analogues of microfibrils in erythrocytes are not involved in endocytosis. Cyclic nucleotide inhibition of endocytosis is confined to a very high concentration range of nucleotides in the medium. Primaquine and chlorpromazine endocytosis are inhibited by cyclic nucleotides as is vinblastine endocytosis.     

Fluorescent photochemical surface labeling of intact human erythrocytes.

A photolabile nitrene precursor, 3-azido-(2,7)-naphthalene disulfonate (ANDS), has been synthesized and used as a membrane-impermeable probe. The aryl azide was nonfluorescent. When activated by light, a highly reactive nitrene was generated which was capable of nonspecific covalent modifications of hydrophilic regions of cell surfaces. The products of the photolysis were highly fluorescent and modified proteins could be identified by their characteristic fluorescence after electrophoresis on sodium dodecyl sulfate polyacrylamide gels. When intact human erythrocytes were labeled with ANDS, Protein 3, the major membrane protein, and the sialoglycoproteins were modified. No proteins of apparent molecular weight greater than Protein 3 were labeled by ANDS, suggesting that none of these membrane components was exposed to the hydrophilic external surface of the red blood cell. When open erythrocyte stroma were labeled with ANDS, virtually all protein bands detectable by Coomassie blue staining could be shown to contain some fluorescence label. The significance of these findings are discussed with relation to the use of various aryl azides as surface labels of membranes.     

Hypoxanthine-guanine exchange by intact human erythrocytes

The uptake and release of [14C]hypoxanthine by human erythrocytes, suspended in a tris(hydroxymethyl)aminomethane (Tris)-glucose-NaCl isotonic medium (pH 7.4), have been studied at 37 degrees C. The uptake of hypoxanthine, mediated by its incorporation into inosine 5'-monophosphate (IMP), was markedly stimulated by preincubating the cells in phosphate-buffered saline. After a lag time, [14C]IMP-enriched erythrocytes released [14C]hypoxanthine in the medium. Formycin B, at concentrations known to inhibit purine nucleoside phosphorylase in intact erythrocytes, affected hypoxanthine uptake and release and led to an increase in the intracellular concentration of inosine, suggesting that the main catabolic path of IMP is the sequential degradation of the nucleotide to inosine and hypoxanthine. The addition of guanine to a suspension of [14C]IMP-enriched erythrocytes led to an increase in the rate of [14C]hypoxanthine release, which was unaffected by the presence of formycin B. During the guanine-induced hypoxanthine release, guanine was taken up by the cells as GMP. These results suggest that the presence of guanine in the incubation medium activates a catabolic path in human erythrocytes leading to IMP degradation without formation of inosine.     

Metabolism of S-adenosyl-L-methionine in intact human erythrocytes

Freshly isolated human erythrocytes contain S-adenosyl-L-methionine (AdoMet) at a concentration of about 3.5 mumol/l cells. When such cells are incubated in a medium containing 30 microM L-methionine, 18 mM D-glucose and 118 mM sodium phosphate (pH 7.4), intracellular AdoMet levels continuously decrease to a value of about 0.1 microM after 24 h. This occurs in spite of the fact that the cellular concentrations of the substrates for the AdoMet synthetase reaction, ATP and L-methionine, remain relatively constant. In a search for incubation conditions that lead to stable levels of AdoMet in incubated cells, we have developed a sodium-Hepes-buffered medium which includes 1 mM adenine and a stoichiometric excess of MgCl2 over its ligand, phosphate. The inclusion of magnesium ion (and a reduction in phosphate) appears to increase intracellular free Mg2+, which is required for full activity of the erythrocyte AdoMet synthetase. Even in the presence of MgCl2, however, the AdoMet pool level can drop 4-6-fold within the first 2 h of incubation. We present evidence that suggests that this initial fall in the cellular AdoMet level may be due to the activation of AdoMet-dependent protein carboxyl methyltransferase, an enzyme which accounts for a large fraction of the total cellular AdoMet utilization. Adenine, or related compounds in the medium may prevent this activation, although the mechanism of this action is not clear at present.     

Effect of colchicine on virus-induced fusion of human erythrocytes.

Colchicine was found to stimulate the virus-induced fusion of human erythrocytes. Colchicine also stimulated the rate of hemolysis, but had no effect on its final extent, suggesting that the enhanced rate of envelope fusion, $\text{i}$. $\text{e}$. virus to cell, caused by colchicine resulted in the stimulation of cell to cell fusion. The fact that effective doses of colchicine were at millimolar concentrations, together with the absence of microtubules in human erythrocytes, indicates that the target of colchicine action is not this subcellular tubular system. Instead, the peripheral membrane protein, spectrin, may be a likely candidate for the site of colchicine action.     

Carbonic anhydrase activity of intact carbonic anhydrase II-deficient human erythrocytes

Intact erythrocytes from subjects with deficiency of blood carbonic anhydrase (CA) II and from normal subjects were assayed for enzyme activity by use of an 18O exchange technique in a solution containing 25 mM (CO2 + NaHCO3) plus 125 mM NaCl. At 25 degrees C and pH 7.4, the catalyzed reaction velocity was 0.32 +/- 0.04 M/s for the CA II-deficient and 1.60 +/- 0.12 M/s for the normal cells, a ratio of 1:5. Under the same conditions at 37 degrees C the relative difference between the CA II-deficient and normal cells was much less: the velocity for the CA II-deficient cells was 0.84 +/- 0.07 M/s and for the normal cells 1.60 +/- 0.32 M/s, a ratio of 1:1.9. Results were comparable for the hemolysates with the NaHCO3 reduced to 85 mM (the corresponding intracellular concentration): at 25 degrees C CA II-deficient cells had a velocity of 0.36 +/- 0.01 M/s compared with 1.12 +/- 0.04 M/s for the normal cells, a ratio of 1:3.1. At 37 degrees C again the relative difference between hemolysates from CA II normal and deficient cells was much less: the CA II-deficient cells had a reaction velocity of 1.17 +/- 0.22 M/s vs. 2.60 +/- 0.36 M/s for the normal cells, a ratio of 1:2.2. The greater fractional reduction of enzyme velocity of CA II-deficient cells at 25 degrees C compared with 37 degrees C appears to be explained by a greater chloride inhibition of the presumed CA I at the lower temperature.(ABSTRACT TRUNCATED AT 250 WORDS)     

Phosphorylation of glycophorin A in membranes of intact human erythrocytes

The qualitative and quantitative contribution of glycophorin A phosphorylation to the general and specific pattern of membrane protein phosphorylation in intact erythrocytes pre-incubated with 32Pi was examined. Intense 32P-labeled bands at 88,000 and 38,000 Mr were identified as phosphorylated glycophorin A dimer and monomer respectively on the basis of several criteria. Quantitatively, phosphorylated glycophorin A dimer accounted for about 70% of 32P in the band 3 region. This value is at least three times that previously reported. The results of ancillary experiments involving selective extraction of ghosts in acidified chloroform/methanol solvents and electrophoresis in the presence of detergents make it unlikely that the 32P associated with glycophorin A was due to bound polyphosphoinositides.     

Photochemical labeling of the cytoplasmic surface of the membranes of intact human erythrocytes.

N-(4-azido-2-nitrophenyl)-2-aminoethyl sulfonate (NAP-taurine), a photolabile nitrene precursor, has been shown to permeate the human erythrocyte membrane at 37degrees but not at 0 degrees. Utilizing this differential permeability, we have loaded intact erythrocytes with NAP-[35S]taurine in the dark at 37degrees, cooled them to 0 degrees, washed them free of external NAP-[35S]taurine in the dark and cold, and photolyzed them, resulting in labeling of hemoglobin and of proteins on the cytoplasmic surface of the membrane. These experiments complement those previously reported on the labeling of the external surface of the membranes with this reagent (Staros, J. V., and Richards, F. M. (1974) Biochemistry 13, 2720-2726).     

Effect of phosphatidylcholine chlorohydrins on human erythrocytes

Hypochlorite generated in vivo under pathological conditions is a known oxidant and chlorinating agent, able to react with proteins and lipids, which affects the stability of biological membranes. Reaction with unsaturated fatty acyl chains in glycerophospholipids such as phosphatidylcholine results in the formation of chlorohydrins. The aim of this study was to determine the effects of chlorohydrins formed by the reaction of hypochlorite with 1-stearoyl-2-oleoyl-, 1-stearoyl-2-linoleoyl-, and 1-stearoyl-2-arachidonylphosphatidylcholine on biophysical properties of bilayers and their effects on human erythrocytes. Using electrospray mass spectrometry we observed complete conversion of the lipids into chlorohydrins, which resulted in a decrease in the rotational correlation time and an increase in the order parameter of liposomes. Unilamellar chlorohydrin liposomes had a lower permeation coefficient for calcein than liposomes made of parent lipids. Flow cytometry demonstrated fast incorporation of uni and multilamellar chlorohydrin liposomes labeled with NBD-phosphatidylethanolamine into erythrocytes. This effect was accompanied by changes in erythrocyte shape (echinocyte formation) and aggregation. Similar but less pronounced effects were noticed for parent lipids only after longer incubation. Chlorohydrins showed also a stronger hemolytic action, proportional to the lipid:erythrocyte ratio. These results are important for understanding the effects of HOCl on mammalian cells, such as might occur in inflammatory pathology.     

Effect of hydration on the water content of human erythrocytes.

An ideal, hydrated, nondilute pseudobinary salt-protein-water solution model of the RBC intracellular solution has been developed to describe the osmotic behavior of human erythrocytes during freezing and thawing. Because of the hydration of intracellular solutes (mostly cell proteins), our analytical results predict that at least 16.65% of the isotonic cell water content will be retained within RBCs placed in hypertonic solutions. These findings are consistent not only with the experimental measurements of the amount of isotonic cell water retained within RBCs subjected to nonisotonic extracellular solutions (20-32%) but also with the experimental evidence that all of the water within RBCs is solvent water. By modeling the RBC intracellular solution as a hydrated salt-protein-water solution, no anomalous osmotic behavior is apparent.     

Triton shells of intact erythrocytes

About 40% of human erythrocyte membrane protein is resistant to solubilization in 0.5% Triton X-114. These components comprise a structure called a Triton shell roughly similar in size and shape to the original erythrocyte and thus constitute a cytoskeleton. With increasing concentrations of Triton the lipid content of the Triton shell decreases dramatically, whereas the majority of the protein components remain constant. Exceptions to this rule include proteins contained in band 3, the presumed anion channel, and in band 4 which decrease with increasing Triton concentration. The Triton-insoluble complex includes spectrin (bands 1 and 2), actin (band 5), and bands 3′ and 7. Component 3′ has an apparent molecular weight of 88,000 daltons as does 3; but unlike 3, it is insensitive to protease treatment of the intact cell, has a low extinction coefficient at 280 nm, and is solubilized from the shells in alkaline water solutions. Component 7 also has a low extinction coefficient at 280 nm. Spectrin alone is solubilized from the Triton shells in isotonic media. The solubilized spectrin contains no bound Triton and coelectrophoreses with spectrin eluted in hypotonic solutions from ghosts. Electron micrographs of fixed Triton shells stained with uranyl acetate show the presence of numerous filaments which appear beaded and are 80–120 Å in diameter. The filaments cannot be composed mainly of actin, but enough spectrin is present to form the filaments. Triton shells may provide an excellent source of material useful in the investigation of the erythrocyte cytoskeleton.     

Phosphorus-31 relaxation times of 2,3-diphosphoglycerate in intact human erythrocytes

The reciprocals of the spin-lattice relaxation times (T₁s) of the 2-P and 3-P nuclei of 2,3-diphosphoglycerate (DPG) increased linearly as percent DPG bound was raised in model hemoglobin solutions. The 2-P T₁ was slightly greater in intact erythrocytes than in model solutions under similar experimental conditions. The change in the 3-P T₁ with cellular deoxygenation was anomalous indicating that this nucleus should not be used to estimate DPG binding inside intact erythrocytes.