Density separation of human red blood cells on self forming Percoll gradients: correlation with cell age

Human red blood cells were density separated on self-forming Percoll gradients. Redistribution of density fractionated red blood cells was studied by recentrifugation on self-forming Percoll gradients. A protocol that avoids centrifugation of red cells prior to removal of white cells and introduces EDTA before red cell pelleting completely avoided redistribution. Dense red cells separated according to this method were senescent on the basis of a biochemical and a physical criterion: the increase in the band 4.1a:4.1b ratio (Mueller, T., Jackson, C.W., Dockter, M.E. and Morrison, M. (1987) J. Clin. Invest. 79, 492-499) and the loss of maximum deformability. Characterization also included the relative content of two surface proteins (complement receptor 1, CR1 (Ripoche, J. and Sim, R.B. (1986) Biochem. J. 235, 815-821); decay accelerating factor, DAF) on density fractionated red cells. Unlike cytoplasmic proteins, these proteins face similar conditions, whether located on circulating reticulocytes or aging red cells. Both components were lost linearly within experimental errors with cell density and were lower by 60 and 40% in dense than light cells, respectively.

Peripheral catabolism of CR1 (the C3b receptor, CD35) on erythrocytes from healthy individuals and patients with systemic lupus erythematosus (SLE)

The present study investigated the rate of catabolism of CR1 (the C3b receptor, CD35) on erythrocytes (E) in vivo, in relationship with the expressed number of CR1/E, the CR1.1 HindIII quantitative CR1 polymorphism, and cell age. The relationship between the number of CR1/E and cell age was analysed by measuring G6PDH activity in E that had been sorted according to high or low expression of CR1 (CD35), by assessing the expression of CR1 (CD35) on E separated according to cell density, and by comparing the number of CR1 (CD35) antigenic sites on reticulocytes and on E. A physiological catabolism of CR1 (CD35) manifested by a reduction in the number of CR1 (CD35) antigenic sites/E with cell ageing was consistently observed in healthy individuals. The number of CR1/E decreased with ageing of E according to a complex pattern that associated an exponential decay and an offset. Calculated half-lives of CR1 (CD35) ranged between 11 and 32 days in healthy individuals. A more rapid loss of CR1 (CD35) with cell ageing occurred on cells from individuals expressing high numbers of CR1/E. In patients with systemic lupus erythematosus (SLE), half-lives of CR1 (CD35) on E were in the same range as those of healthy individuals with a similar quantitative CR1 genotype; the number of CR1 (CD35) on reticulocytes was reduced and linearly related to the number of CR1/E, independently of the patients' quantitative CR1 genotype. Transfusion experiments with E bearing high or low amounts of CR1/E indicated the lack of preferential removal of E bearing high numbers of CR1 (CD35) in patients with SLE. These results indicate that the rate of loss of CR1 (CD35) from E with cell ageing is directly related to the quantitative CR1 phenotype and suggest that enhanced peripheral catabolism is not the sole mechanism of the acquired loss of CR1 (CD35) on E in patients with SLE.

Covalent binding of detergent-solubilized membrane glycoproteins to 'Chemobond' plates for ELISA

An ELISA method is presented which is based on covalent binding of detergent-solubilized membrane proteins to surface-modified polystyrene plates (Chemobond plates). These plates carried 0.52-0.65 nmol of aldehyde groups per well (150 microliters) and allowed coupling of protein by Schiff base formation either at high pH and subsequent reduction with NaBH4 or by trapping reduced imines at pH 6-6.8 with cyanoborohydride. They bound 15 times the amount of normal plates. Sodium chloride (0.5 M) increased binding 2-3-fold. Binding was essentially resistant to elution by 1% sodium dodecyl sulfate. Reduction of uncoated plates with NaBH4 eliminated the high extent of binding. ELISA tests on Chemobond plates with a rabbit anti-band 3 antibody gave a ten-fold higher signal than plates to which band 3 protein was merely adsorbed. The use of an antigen-enzyme conjugate to detect bound antibody allowed to perform antibody binding and detection of bound antibody simultaneously in the presence of 0.05% Triton X-100. A competitive, one step ELISA system allowed determination of rabbit anti-band 3 antibodies in diluted serum with a sensitivity range of 0.02-0.4 microgram/ml.

Fatty acid acylation of membrane skeletal proteins in human erythrocytes

Fatty acid acylation of membrane proteins was studied on human erythrocytes by measuring incorporation of [3H]palmitate at different specific radioactivities. A 55 kDa polypeptide within the band 4.5 region was the main acceptor protein for acylation by fatty acids (palmitate, stearate, oleate), while other polypeptides (80, 65, 48, 30 kDa) incorporated [3H]palmitate slowly, in substoichiometric amounts. Integral membrane proteins were preferentially fatty acid acylated. Skeletal membrane proteins were, however, poorly labeled. Neither purified ankyrin nor band 4.1 protein were fatty acid acylated in human erythrocytes. On the other hand, label associated with high molecular weight skeletal proteins resisted low and high ionic strength extractions, and was extracted selectively by urea [corrected] along with a small subpopulation of spectrin which was also tightly associated with the membrane.

Haemolytic anaemia in analpha-lipoproteinaemia (Tangier disease): morphological, biochemical, and biophysical properties of the red blood cell

A patient with familial analpha-lipoproteinaemia (Tangier disease) was found to have stomatocytosis and haemolytic anaemia. The analysis of the red cell membrane constituents revealed a low cholesterol content (90 nmol/ml red cells, control 130 nmol/ml red cells), a decreased cholesterol/phospholipid ratio (0.54, control 0.78), high phosphatidylcholine (41.5%, control 30.6%) and low sphingomyelin (18.8%, control 27.6%). The electrophoretic membrane protein pattern was normal. Osmotic gradient ektacytometry and osmotic resistance showed a decreased surface/volume ratio, which caused an increased filtration resistance in 3 microns pores. The elasticity of the membrane was unchanged. Functional membrane properties were altered: the anion exchange rate was increased, whereas alkali cation fluxes were normal. The capacity to release vesicles was reduced. This case represented a new type of stomatocytosis. It contributes to the understanding of the role of cholesterol and phospholipids in the red cell membrane and biomembranes in general.

Electrophoretic properties of human IgG and its subclasses on sodium dodecyl-sulfate-polyacrylamide gel electrophoresis and immunoblots

Unreduced human immunoglobulin G (IgG) which was not aggregated showed anomalous apparent molecular masses on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). It migrated mainly as three distinct bands with apparent molecular masses from 190 to 240 kDa on gels containing 8% polyacrylamide, when denatured at 37 degrees C. Generation of this banding pattern has two reasons: (a) the pattern is a superposition of bands originating from the four IgG subclasses that differ in molecular masses and structures; and (b) the complexity of the band pattern is further increased, because IgG myeloma proteins of the IgG1 and IgG2 subclass migrated as doublets, while IgG3 and IgG4 formed primarily one band with slightly different apparent molecular masses. These properties were independent of the type of light chain in all myeloma proteins studied. Generation of doublets suggests heterogeneities of monoclonal proteins. The two separable protein populations from IgG1 differ in their susceptibility to reduction. Reduction at 37 degrees C cleaved the larger into heavy and light chain, while it generated heavy chain dimer and light chain from the smaller species. Hence, it is possible that monoclonal IgG1 are comprised of at least two subpopulations of molecules with different S-S bonds. Doublet formation of IgG2 remains unexplained, since both species were equally sensitive to reduction. Knowledge on the anomalous properties of IgG on SDS-PAGE is a prerequisite to run immunoblots from unreduced cellular antigens without confounding cell-associated IgG with cellular antigens.

Spectrin-like proteins in the paraflagellar rod structure of Trypanosoma brucei

A polyclonal, monospecific rabbit antibody to human erythrocyte spectrins cross-reacted with two sets of proteins (a doublet of 180/200K and a triplet of 67–66-65K; K = 10(3) Mr) in the parasitic protozoon Trypanosoma brucei brucei. Except for the 66K protein, the cross-reacting proteins are localized in the flagellum, on the basis of evidence from cell fractionation and immunofluorescence microscopy. Immunogold labelling and electron micrographs further revealed that the spectrin-like proteins are confined to the paraflagellar rod structure. The spectrin-like proteins with apparent molecular weights of 180 and 200 share homology with spectrin band 1, since V8-protease from Staphylococcus aureus generated similarly sized, antigenic peptides from these proteins. The results indicate homology between the cross-reacting proteins and human red cell spectrin.

[Alternative complement pathway activation by anti-band 3 antibodies]

Naturally occurring antibodies against the anion exchange protein of red cells (band 3 protein) can elicit in whole serum a strong C3b deposition to red cells under conditions which favor alternative complement pathway activation. Such a mode of opsonization calls for generation of an alternative C3 convertase nucleated by C3b covalently bound to anti-band 3. Senescent, but not young, red cells should also carry "C3b-anti-band 3" complexes, if clearance of in vivo aged red cells occurred by the same mechanism. We succeeded in isolating covalently linked complexes of C3b and IgG primarily from membranes of senescent red cells.

Naturally occurring anti-band-3 antibodies and complement together mediate phagocytosis of oxidatively stressed human erythrocytes

Treatment of erythrocytes with the thiol-specific oxidant azodicarboxylic acid bis(dimethylamide) (diamide) enhances their phagocytosis by adherent monocytes. Phagocytosis of diamide-treated erythrocytes required that the cells were opsonized with whole serum, since complement inactivation abolished phagocytosis. Opsonization with whole serum containing 20-100 times the physiological concentration of naturally occurring anti-band-3 antibodies enhanced phagocytosis of diamide-treated erythrocytes. High inputs of anti-band-3 also restored phagocytosis of erythrocytes that had been incubated with complement-inactivated serum. Elevated concentrations of anti-spectrin antibodies were ineffective in whole and complement-inactivated serum. Specific recognition of diamide-treated erythrocytes by anti-band-3 antibodies may be due to generation of anti-band-3 reactive protein oligomers on intact diamide-treated erythrocytes. Generation of such oligomers was dose-dependent with respect to diamide. Bound anti-band-3 alone was not sufficient to mediate phagocytosis. It resulted in deposition of complement component C3b on the cells through activation of the alternative complement pathway in amounts exceeding that of bound antibodies by two orders of magnitude. Thus, anti-band-3 and complement together mediate phagocytosis of oxidatively stressed erythrocytes, which stimulate senescent erythrocytes with respect to bound antibody and complement.

Red cell associated, naturally occurring anti-spectrin antibodies

Images

Re-evaluation of the structural integrity of red-cell glycoproteins during aging in vivo and nutrient deprivation

Results presented in this paper show that removal of white-cell contaminations from human red blood cells by filtration through cellulose [Beutler, West & Blume (1976) J. Lab. Clin. Med. 88, 328-333] is a necessity whenever red cells are incubated at elevated temperatures or haemolysed after density separation. Omission of this precaution results in proteolysis of sialoglycoproteins in membranes from less-dense (young), but not dense (old), subpopulations. This proteolytic damage occurs during haemolysis of the cytoplasmic domain of glycophorin. A different type of proteolysis occurs if white-cell-contaminated red cells are incubated in the absence of glucose at elevated temperatures. Red cells release sialoglycopeptides. This process is stimulated by Ca2+ ions and is accompanied by the release of vesicles that differ from spectrin-free vesicles [Lutz, Liu & Palek (1977) J. Cell Biol. 73, 548-560]. This sialoglycopeptide release is dependent on white-cell contamination and is not required for the release of spectrin-free vesicles.

Calmodulin inhibits the phosphorylation of spectrin in vitro

In vitro phosphorylation of purified spectrin dimer was studied in the presence of Ca2+-calmodulin (CaM). CaM inhibited autophosphorylation of the beta subunit of spectrin. The inhibitory effect (65% at a 32-fold molar excess) appeared to be due to a weak interaction of CaM with spectrin. CaM was similarly effective in a phosphatase-stimulated autothiophosphorylation of the beta subunit with [gamma-35S]ATP. Hence, its inhibitory effect was not due to stimulation of a spectrin-associated phosphatase activity. Phosphorylation of spectrin by the catalytic subunit of a cAMP-dependent protein kinase occurred in both subunits (1984, FEBS Lett. 169, 323). CaM selectively inhibited a cAMP-dependent phosphorylation of the alpha subunit of spectrin to 30% at two CaM per spectrin. It was ineffective on the cAMP-dependent phosphorylation of the beta subunit up to a 32-fold molar excess. These results yield functional evidence for a CaM-spectrin interaction. They further suggest that CaM can regulate the extent of a cAMP-dependent phosphorylation of the alpha subunit of spectrin.

Red cell spectrin phosphorylation and cytoskeletal anchorage

A cAMP-dependent phosphorylation of spectrin occurred in intact human red blood cells supplemented with cAMP. A cAMP-dependent phosphorylation of spectrin altered its binding properties. Spectrin was resistant to low ionic strength extraction and remained associated with inside-out vesicles (IOV). A cAMP-dependent phosphoform of spectrin contained label in both subunits, when generated in vitro. In vivo, the labeling of spectrin band 1 increased with red cell age. The low extent of spectrin band 1 phosphorylation in young cells could be due to a Ca2+-calmodulin-spectrin interaction, because Ca2+-calmodulin selectively inhibited a cAMP-dependent labeling of spectrin band 1, when tested on purified spectrin dimer.

Naturally occurring autoantibodies to exoplasmic and cryptic regions of band 3 protein, the major integral membrane protein of human red blood cells

Immunoglobulin G (IgG) of healthy human blood donors and IgG from pooled sera (Sandoglobulin) contain natural (auto)antibodies to band 3 protein, the major integral membrane protein of human red blood cells. Affinity-purified and 125I-iodinated anti-band 3 antibodies bound specifically to band 3 protein on immunoblots from membrane proteins in the presence of unlabeled, absorbed IgG. Purified (auto)antibodies also bound nonspecifically to band 4.2 and weakly to band 5 and 6, when assayed with second antibody and 125I-iodinated protein A. The antibodies were directed to regions of band 3 protein that were cryptic and in part exoplasmic but with a low accessibility to surface modifications. The antigenic sites were located within the 65K, but not the 38K-dalton chymotryptic fragment of band 3 protein. Antigenic band 3 protein was equally present in membranes of young and senescent red cells. Hence, if these antibodies were involved in recognizing a few exoplasmic sites of band 3 protein on senescent red cells, antigen exposure would require alterations in band 3 accessibility (conformation, topology) rather than an enzymatic generation of antigenic sites.

Naturally occurring autoantibodies to exoplasmic and cryptic regions of band 3 protein, the major integral membrane protein of human red blood cells

Immunoglobulin G (IgG) of healthy human blood donors and IgG from pooled sera (Sandoglobulin) contain natural (auto)antibodies to band 3 protein, the major integral membrane protein of human red blood cells. Affinity-purified and 125I-iodinated anti-band 3 antibodies bound specifically to band 3 protein on immunoblots from membrane proteins in the presence of unlabeled, absorbed IgG. Purified (auto)antibodies also bound nonspecifically to band 4.2 and weakly to band 5 and 6, when assayed with second antibody and 125I-iodinated protein A. The antibodies were directed to regions of band 3 protein that were cryptic and in part exoplasmic but with a low accessibility to surface modifications. The antigenic sites were located within the 65K, but not the 38K-dalton chymotryptic fragment of band 3 protein. Antigenic band 3 protein was equally present in membranes of young and senescent red cells. Hence, if these antibodies were involved in recognizing a few exoplasmic sites of band 3 protein on senescent red cells, antigen exposure would require alterations in band 3 accessibility (conformation, topology) rather than an enzymatic generation of antigenic sites.

A cyclic AMP-dependent phosphorylation of spectrin dimer

In contrast to the properties of spectrin obtained from [32P]phosphate-labeled red cells, purified spectrin dimer could be phosphorylated by a cAMP-dependent protein kinase from bovine heart. Both spectrin bands were phosphorylated. Spectrin band 2 contained in addition to autophosphorylated peptides several phosphopeptides that were distinct from autophosphorylated ones. The cAMP-dependent phosphorylation of spectrin band I was modulated by reducing agent and the concentration of spectrin. At high concentrations spectrin band 2 was predominantly labeled. The cAMP-dependent phosphoform of spectrin band 2 had a pI slightly higher than that of autophosphorylated spectrin band 2, but lower than that of ankyrin.

Binding of autologous IgG to human red blood cells before and after ATP-depletion. Selective exposure of binding sites (autoantigens) on spectrin-free vesicles

Binding of autologous IgG to fresh, ATP-depleted red blood cells as well as to spectrin-free vesicles was studied by a non-equilibrium binding assay using 125I-iodinated protein A from Staphylococcus aureus. IgG binding was 14-times higher to spectrin-free vesicles than to ATP-maintaining red blood cells and 4-times higher than to ATP-depleted erythrocytes from which these vesicles were released. Protein A binding to vesicles that were released from washed and nutrient-deprived erythrocytes, was dependent on added autologous IgG. However, spectrin-free vesicles that were spontaneously released from erythrocytes conserved in whole blood, bound similar amounts of protein A with or without added autologous IgG (0.45-0.55 ng/micrograms band 3 protein). These findings demonstrate that opsonization of spectrin-free vesicles by autologous IgG occurs not only in the test tube, but also under blood blank conditions. The binding characteristics of IgG to spectrin-free vesicles are indicative of a natural autoantibody rather than an unspecific binding of autologous IgG. The preferential binding of IgG to spectrin-free vesicles implies a selective exposure of corresponding autoantigens in membrane regions that have lost cytoskeletal anchorage and bud off.

Glycoprotein topology on intact human red blood cells reevaluated by cross-linking following amino group supplementation

Protein-protein interactions were studied at 0-4 degrees C with amino group specific cross-linkers on intact human erythrocytes after introducing free alkylamino groups into glycoproteins to overcome the scarcity of accessible amino groups. Amino group supplementation is based on the formation of aldehydes by either enzymatic or chemical oxidation. Subsequently, an imine is formed between aldehyde groups and 2-(4-aminophenyl) [1-14C]ethylamine (arylalkyldiamine) in the presence of sodium cyanoborohydride (NaCNBH3). The arylamino group of arylalkyldiamine forms imines at least 280 times more rapidly than the alkylamino group. This property leaves the majority of alkylamino groups free for subsequent cross-linking with bifunctional reagents. Amino group supplementation enhances the cross-linking probability of glycophorins in glycophorin-containing vesicles. When applied to intact human erythrocytes of any age, in conjunction with the cross-linker disuccinimidyl 3,3'-dithiobis(propionate), the glycoproteins, band 3 and glycophorins, did not undergo substantial cross-linking (less than 2% of the total label cross-linked). However, substantial cross-linking (20% of the total label cross-linked) of either glycoprotein was detected on spectrin-free vesicles [Lutz, H. U., Liu, S. C., & Palek, J. (1977) J. Cell Biol. 73, 548-560] that are devoid of cytoskeletal restraints. The inability to cross-link these proteins on intact cells is not due to a lack of accessible amino groups on the surface of these glycoproteins, because amino group supplementation enhanced their ability to form cross-links on intact cells and on spectrin-free vesicles by the same factor (2-3-fold). This and various controls suggest a monomeric arrangement of the exoplasmic portions of band 3 on intact erythrocytes.

Naturally occurring autoantibodies to skeletal proteins from human red blood cells

The IgG fraction of sera of healthy human subjects contains natural antibodies to cytoskeletal elements of the donors own red blood cell membranes. Autoantibodies to spectrin are characterized in more detail: their Fab portion binds to the antigen. Autoantibodies, affinity-purified on immobilized spectrin band 1, precipitate 0.4 microgram of spectrin dimer per 1 microgram of autoantibody. They bind to band 1 and cross-react with band 2 of spectrin as well as with breakdown products of spectrin on blots from separated membrane polypeptides. Autoantibodies purified on spectrin band 2 after absorption on band 1 do not cross-react with band 1. The evidence strongly suggests the existence of such autoantibodies in healthy human subjects. This finding indicates that autoantibody production to normally unexposed antigens is not suppressed in ontogeny. These anti-cytoskeleton autoantibodies may have a physiologic role in clearance of debris from lysed cells. Their existence may open a new understanding of elevated anti-spectrin autoantibody concentrations in diseases with different etiologies.

Naturally occurring autoantibodies to skeletal proteins from human red blood cells

The IgG fraction of sera of healthy human subjects contains natural antibodies to cytoskeletal elements of the donors own red blood cell membranes. Autoantibodies to spectrin are characterized in more detail: their Fab portion binds to the antigen. Autoantibodies, affinity-purified on immobilized spectrin band 1, precipitate 0.4 microgram of spectrin dimer per 1 microgram of autoantibody. They bind to band 1 and cross-react with band 2 of spectrin as well as with breakdown products of spectrin on blots from separated membrane polypeptides. Autoantibodies purified on spectrin band 2 after absorption on band 1 do not cross-react with band 1. The evidence strongly suggests the existence of such autoantibodies in healthy human subjects. This finding indicates that autoantibody production to normally unexposed antigens is not suppressed in ontogeny. These anti-cytoskeleton autoantibodies may have a physiologic role in clearance of debris from lysed cells. Their existence may open a new understanding of elevated anti-spectrin autoantibody concentrations in diseases with different etiologies.

On the mechanism of vesicle release from ATP-depleted human red blood cells

The release of spectrin-free vesicles from ATP-depleted human red blood cells (Lutz et al. (1977) J. Cell. Biol. 73, 548) can be considered the final step of a shape change from discocytes to echinocytes. The study of physical and chemical properties of released vesicles suggests that vesicle release is not merely a consequence of charge alterations within either monolayer of the budding membrane. Fresh membranes and released vesicles have within experimental error the same sialic acid content per surface area and the same electrophoretic mobilities. Vesicle release cannot be stimulated by doubling the charge density on the outer monolayer by means of a phospholipase D-treatment, but correlates with a breakdown of polyphosphoinositides to diacylglycerol on the inner monolayer. This breakdown does not lead to a significant change in the negative charge density on the inner monolayer, because an increased phosphatidate content compensates for this alteration. Furthermore, polyphosphoinositide breakdown and diacylglycerol production are not the rate-limiting step in vesicle release from ATP-depleting red blood cells. This is evident from the fact that 10 mM EDTA inhibits vesicle release to 75% without affecting polyphosphoinositide breakdown and diacylglycerol production. Hence, diacylglycerol formation may be sufficient for membrane budding as suggested earlier (Allan et al. (1976) Nature 261, 58), but vesicle release requires a second, as yet unidentified process.

[Elimination of old erythrocytes from the circulation: exposure of a cell-age specific antigen on aging erythrocytes]

Selective phagocytosis of senescent human red blood cells (RBC) requires a molecular alteration on the surface of aging RBC. This cell-age determining signal represents neither a surface-charge density change nor extensive desialylation, as assumed earlier. Since IgG autoantibodies have been detected which specifically bind to senescent RBC, the required surface alteration represents the exposure of a cell-age-specific antigen (CAS antigen). The CAS antigen has been identified as a membrane protein with apparent molecular weights of approximately 100 K or approximately 200 K. Immunoprecipitation of Triton extracts from chymotrypsin-treated RBC as well as preliminary peptide analyses of the surface 125I-iodinated part of a CAS-antigen enriched fraction suggest that the CAS antigen has the protein structure of protein band 3. The CAS-antigenic sites are not formed during senescence but are already present in young RBC. Exposure of the CAS antigen in senescent RBC appears to be due to an increased probability of CAS-antigen dimers. This apparently minor change is sufficient for the following reasons: IgG autoantibody binding to monomeric CAS antigen appears to be monovalent and transient. Thus, CAS antigens may even be exposed rather than cryptic on the surface of young RBC. In contrast, IgG autoantibodies form thermodynamically considerably more stable complexes with dimeric CAS antigen because of a bivalent binding. The dimerization probability of CAS antigens in the plane of the membrane can be increased experimentally by destroying the anchorage between integral membrane proteins and the cytoskeleton. This loss of anchorage is followed by drastically enhanced IgG-autoantibody binding.