Search

A method utilizing sodium dodecyl sulfate--polyacrylamide gel electrophoresis for the sepration of the alpha and beta chains of hemoglobin has been adapted to quantify hemoglobin chains by simultaneous electrophoresis of globin standards with unknowns, staining with Coomassie blue, densitomeric scanning, and planimetry. This method has been used to quantify the globin chains bound to the human red cell membrane during sterile incubation and ATP depletion in vitro. In thoroughly washed (6-step) ghosts of incubated cells; (1) more beta than alpha chains were bound (beta/alpha ratio 1.28); (2) only about one third of the bound chains contained heme; and (3) globin accounted for less than 20% of the "excess" protein present in the incubated cells compared to fresh cells. Four-step ghosts contained more hemoglobin, a smaller proportion of heme-free alpha and beta chains, and approximately equal numbers of alpha and beta chains (beta/alpha ratio 1.09).

An 18-yr-old female with chronic active hepatitis developed a severe anemia due to a lack of red cell production. Her bone marrow showed many large proerythroblasts but an almost complete lack of more mature erythroblasts. Incubation of the marrow cells in a normal medium with erythropoietin concentrate led to increased erythropoiesis as indicated by the development of mature erythroblasts as well as a ninefold increase in hemoglobin synthesis. The patient's plasma was cytotoxic for erythroblasts. Following splenectomy, a remission of the disease occurred. This study indicates that in some cases the anemia associated with abundant marrow proerythroblasts and the absence of mature erythroblasts has the same pathogenesis as pure red cell aplasia and that splenectomy may be beneficial when there is a lack of response to immunosuppressive drugs.

Increased adhesive forces between sickle erythrocytes and endothelial cells (EC) have been hypothesized to play a role in the initiation of vasoocclusion in sickle cell anemia. Erythrocyte/human umbilical vein EC interactions were studied under controlled flow conditions for normal (AA), homozygous sickle cell (SS), sickle cell trait (AS), mechanically injured normal, and "high-reticulocyte control" RBC by using video microscopy and digital image processing. The number of adherent RBC was determined at ten-minute intervals during a washout period. Results indicate that SS RBC were more adherent than AA RBC. Mechanically injured (sheared) AA RBC were also more adherent than control normal cells but less adherent than SS RBC. AS RBC did not differ significantly in their adhesive properties from normal RBC. Less-dense RBC were more adherent to EC than dense cells for normal, SS, and high-reticulocyte control RBC. The number of cells adherent at a given time during washout was a very strong function of wall shear rate. In addition, at all shear rates studied, the average velocity of individual SS RBC in the region near the EC surface was approximately half that of AA RBC at the same bulk volumetric flow rate through the flow chamber. These findings suggest that the increased adhesion of sickle RBC is at least partially related to the increased numbers of less-dense RBC present. Increased adherence of the less-dense cells to the EC lining vessel walls could contribute to microvascular occlusion by lengthening vascular transit times of other sickle cells.

During sterile incubation of normal human erythrocytes at 37 degrees C., intracellular nonhemoglobin protein is bound to the membrane prior to hemolysis. These studies have characterized this phenomenon further. Protein binding to the membrane began after 12 hours incubation when cellular ATP was depleted and increased to 36 hours incubation. The binding was prevented by adding adenosine or glucose at the start of incubation and was arrested by adding adenosine to regenerate ATP during the course of incubation. However, protein, once bound, was not released by regeneration of ATP. The amount of protein bound was not altered by: (1) addition of Ca++ or EDTA to the medium, (2) blockade of sulfhydryl groups with N-ethylmaleimide, or (3) stabilization of heme-globin bonds by conversion of hemoglobin to cyanmethemoglobin. Conversion of hemoglobin to carboxyhemoglobin by incubations under carbon monoxide inhibited protein binding, but this appeared to be an effect of exclusion of oxygen rather than stabilization of heme-globin bonds since incubation under nitrogen had a similar effect. The morphological counterpart of this chemically-measured membrane-bound protein was visible in red cell ghosts stained with crystal violet as small membrane-associated particles resembling Heinz bodies. Sodium dodecyl sulfate acrylamide gel electrophoresis of membranes of incubated cells revealed a new protein band that was identical to globin monomers. This membrane binding of globin during incubation provides a model for the study of Heinz body formation in clinical disorders.

The abnormal adherence of sickle erythrocytes to endothelial cells (EC) has been hypothesized to play a role in the initiation of vaso-occlusion in sickle cell anemia. We studied erythrocyte/endothelial cell interactions under controlled flow conditions for normal (AA), homozygous sickle cell (SS), sickle cell trait (AS), mechanically injured normal, and "high reticulocyte control" red blood cells (RBC). Human umbilical vein endothelial cells grown to confluence on glass slides formed the base of a parallel plate flow chamber into which RBC suspensions were perfused at a constant flow rate, producing a wall shear stress of 1 dyne/cm2. Adhesion was monitored using video microscopy, and the number of adherent RBC was determined at ten-minute intervals during a wash out period. Results indicate that SS RBC were more adherent than AA RBC. Mechanically injured (sheared) RBC were also more adherent than control normal cells, but less adherent than SS RBC. AS RBC did not differ significantly in their adhesive properties from normal RBC. Less dense (younger) RBC were more adherent to EC than dense (older) cells for normal, SS and "high reticulocyte control" RBC. These findings suggest that the increased adhesion of sickle RBC is at least partially related to the increased numbers of young RBC present. Increased adherence of young cells to the EC lining vessel walls could contribute to microvascular occlusion by lengthening vascular transit times of other sickle cells. Microvascular occlusion is a major clinical problem in sickle cell anemia. This obstruction to blood flow could be due to decreased deformability of the cell and its inability to pass through small vessels. If this were the case it would be reasonable to expect that the most severely deformed sickle cells, the irreversibly sickled RBC (ISC), would play an important role in the initiation of vaso-occlusion. However, the number of circulating ISC is not well correlated with the frequency of painful crises and other microvascular occlusive phenomena. Recent evidence suggests that microvascular occlusion may be associated with increased adhesion of sickle cells to vascular endothelial cells. A strong correlation between erythrocyte adherence and clinical vaso-occlusive severity has been reported by Hebbel et al. Hoover et al. and Hebbel et al. demonstrated increased adhesion in static tests.(ABSTRACT TRUNCATED AT 400 WORDS)

The interactions of normal erythrocytes and erythrocytes from patients having hemoglobin S hemoglobinopathies with normal human endothelial cells (EC) were investigated under flow conditions. When EC supernatant, containing 2.8-11.0 U/dl of von Willebrand factor (vWF) antigen and vWF multimeric forms larger than those present in normal plasma, was the red blood cell (RBC)-suspending medium instead of serum-free medium (SFM), the adhesion of sickle RBC, but not normal RBC, to endothelial cells was greatly increased (range of enhancement of sickle RBC adhesion, 2- to 27-fold). Adhesion of sickle RBC to endothelial cells was reduced to near serum-free levels when EC supernatant was immunologically depleted of vWF forms. Sickle RBC suspended in SFM containing 200 U/dl of purified vWF multimers of the type found in normal human plasma or 300 micrograms/ml human fibronectin were only slightly more adhesive to endothelial cells than sickle RBC suspended in SFM alone. These data indicate that unusually large vWF multimers produced by endothelial cells are potent mediators of the adhesion of sickle erythrocytes to endothelial cells. Vaso-occlusive crises in sickle cell anemia may be caused, at least in part, by adhesive interactions between the abnormal surfaces of sickle RBC and the endothelium after the release of unusually large vWF multimeric forms from stimulated or damaged endothelial cells.

The topography of the external surface of the human red cell membrane has been studied using an impermeant radioactive probe, [125I]diazodiiodosulfanilic acid, which binds covalently to protein groups of the membrane following reaction with intact cells. The pattern of labeling was assessed by sodium dodecyl sulfate-polyacrylamide disc gel electrophoresis followed by sequential analysis of single gels for carbohydrates (by staining with the periodic acid-Schiff (PAS) reagent), for proteins (by staining with Coomassie blue), and for radioactivity (by counting gels sliced in 2-mm segments). The radioactive probe bound to membrane polypeptides with apparent molecular weights of 94,200, 58,100, and 46,500 (Peaks A, B, and C, respectively). Peak A co-migrated with a small periodic acid-Schiff-positive band and protein Band 3 (nomenclature of Steck) (Steck, T.L. (1974)J. Cell Biol. 62: 1-19). Peak B migrated with protein Band(s) 4.5 slightly ahead of the major membrane glycoprotein (PAS-1). Peak C migrated like glycoprotein PAS-2 and protein Band 5, the actin-like, water-soluble membrane protein. In contrast to lactoperoxidase iodination and a number of other probes, [125I]diazodiiodosulfanilic acid reacted minimally with the major membrane glycoprotein, glycophorin. When it was reacted with isolated ghosts, all molecular weight classes of polypeptides were labeled. Treatment of labeled cells with neuraminidase or trypsin altered the glycoprotein staining pattern, but not the radioactive peaks. On the other hand, Pronase eliminated the Mr=94,200 radioactive peak, diminished the other two radioactive peaks, and profoundly changed the glycoprotein and protein staining patterns. Treatment of the membranes of labeled cells in a low ionic strength alkaline medium did not alter radioactive peaks and demonstrated that Peak C differed from the actin-like membrane protein. A nonionic detergent, Triton X-100, solubilized all radioactive components. The studies have defined the binding of [125I]diazodiiodosulfanilic acid to external proteins of the human red cell membrane. Its pattern of reaction differs quantitatively and qualitatively from other commonly used reagents, and it provides a useful additional vectorial probe for the study of membrane topography. Its reactions provide further evidence of the organizational complexity of the red cell membrane and emphasize the fact that interpretation of information derived from the use of membrane probes must take into account the differences resulting from the properties of the probing reagents themselves.

The conformation of the outer surface of the human red cell membrane has been studied under various conditions using the impermeant probe [125I]diazodiiodosulfanilic acid. At least seven polypeptides were labeled by the reagent, including the three extractable glycoproteins separable by the electrophoretic method employed. The Mr = 43,000 protein band was shown to contain two labeled species, one a glycoprotein, in addition to its major constituent, red cell actin. The extent and pattern of labeling were very sensitive to changes in pH and temperature. Total labeling increased with increasing pH and was greater at 4 degrees C than 37 degrees C. Binding of the probe to the Mr = 90,000 polypeptide and the major glycoprotein were relatively increased with increasing pH and temperature while opposite effects were observed for the Mr = 43,000 peptide(s). The pH effects on external membrane labeling were rapidly reversible. Results were similar in cells of different densities, suggesting that the pH and temperature effects were not related to cell age. The data presented emphasize the lability of membrane conformation and reactivity and thus the necessity to consider carefully the conditions of labeling in interpretation of studies using impermeant probes.