Sickling times of individual erythrocytes at zero Po2.

A rapid-reaction parallel-plate flow channel was used to study the kinetics of erythrocyte sickling upon sudden deoxygenation with sodium dithionite. The erythrocytes were recorded on 16-mm film or video tape and visually tracked in time. Sickling was identified by morphologic criteria. At the flow rate used in these studies, the rate of sickling was a reaction-limited process. There was no loss of cellular deformability or membrane flicker before the onset of sickling. Typical sickling times for sickle (SS) cells and trait (AS) cells at room temperature in isotonic buffer were 2.0 and 70 s, respectively. Increasing the buffer osmolality resulted in shorter sickling times and under hypotonic conditions the time required for sickling was prolonged. Between pH 6.4 and 7.0 there was little change in the time required for 50% of the originally discoidal cells to sickle (t50); whereas a marked increase in t50 occurred between pH 7.4 and 7.6. Whole populations of AS and SS erythrocytes were separated into three fractions after centrifugation. The t50 of the fractions progressively decreased from top to bottom, which paralleled an increase in mean corpuscular hemoglobin concentration (MCHC). The t50 decreased as the temperature was increased from 13 degrees to 34 degrees C. This temperature effect was more pronounced for cells that had osmotically induced reductions in MCHC. A two-step process for erythrocyte sickling is proposed: an initial lag phase, during which there is little or no change in internal viscosity, followed by a rapid phase of cellular deformation. The lag phase is altered by changes in MCHC, pH, and temperature.

Effect of heat on the circular dichroism of spectrin in hereditary pyropoikilocytosis.

Hereditary pyropikilocytosis is a hemolytic anemia in which the erythrocytes show increased sensitivity to heat-induced fragmentation. Circular dichroism measurements were employed to study the effect of heat on the secondary structure of pyropoikilocyte membrane proteins. The magnitude of the ellipticity at 222 nm over the temperature range from 25 degrees to 70 degrees C was determined for erythrocyte ghosts, spectrin, and ghost residue after extraction. In pyropoikilocyte ghosts, protein denaturation began at a lower temperature and the midpoint of the structural transition was displaced from 49 degrees C (the value for normal ghosts) to 44 degrees C. This thermal transition was present in spectrin, but not in the ghost residue after extraction. We conclude that an abnormality in the spectrin molecule alters the physical and morphologic properties of the erythrocyte membrane in pyropoikilocytosis.

Influence of hemoglobin precipitation on erythrocyte metabolism in alpha and beta thalassemia.

Certain aspects of the metabolism of centrifuged young and old erythrocytes in hemoglobin H disease have been examined and compared with similar studies of beta thalassemia and normal cells. Glycolysis, hexose monophosphate shunt activity (HMPS), potassium flux, and glutathione (GSH) content were measured. The distributions of hemoglobins H and F, as well as the activities of erythrocyte glucose-6-phosphate dehydrogenase (G6PD) and glutamic oxalacetic transaminase (GOT), were utilized for estimations of the relative ages of the cell samples. The young erythrocytes in hemoglobin H disease differed in several respects from older hemoglobin H cells. They contained more soluble hemoglobin H and GSH and, after splenectomy, fewer inclusions. HMPS activity was subnormal in hemoglobin H young cells and rose to normal activity in old cells. Potassium flux tended to increase in old cells when inclusions were present.Beta thalassemia young cells contained less hemoglobin F and, after splenectomy, more inclusions than old cells. In addition, they had markedly increased glycolysis and HMPS activity. GSH was randomly distributed. Potassium flux was increased in younger cells and particularly increased when inclusions appeared in younger cells after splenectomy. The results are interpreted to indicate that inclusion formation is associated with increased erythrocyte cation permeability in the thalassemia syndromes. This is not related to the level of intracellular GSH. The decreased HMPS activity in young hemoglobin H cells may be due to the presence of the extra thiols of soluble hemoglobin H which can act as a reducing agent. The substitution of hemoglobin H for glutathione in this capacity would then spare the NADPH-requiring glutathione reductase system. As a consequence, HMPS activity would decline. However, in older cells the oxidized hemoglobin H precipitates; these must rely upon GSH and glutathione reductase activity for thiol reduction capacity. Accordingly, HMPS activity increases to normal in the old cell population.

A common type of the spectrin alpha I 46-50a-kD peptide abnormality in hereditary elliptocytosis and pyropoikilocytosis is associated with a mutation distant from the proteolytic cleavage site. Evidence for the functional importance of the triple helical model of spectrin.

We studied nine individuals from five unrelated families with alpha I/46-50a hereditary elliptocytosis (HE) or hereditary pyropoikilocytosis (HPP), including one of the original HHP probands first reported by Zarkowsky and colleagues (1975. Br. J. Haematol. 29:537-543). Biochemical analysis of erythrocyte membrane proteins from these patients revealed, as a common abnormality, the presence of the alpha I/46-50a peptide after limited tryptic digestion of spectrin. The polymerase chain reaction was utilized to study the structure of the DNA encoding the alpha I domain of spectrin in the affected individuals. The DNA sequence of the alpha-spectrin gene encoding the region of the alpha-spectrin chain surrounding the abnormal proteolytic cleavage site was normal. We identified a point mutation causing the replacement of a highly conserved leucine residue by proline at position 207 in the alpha-spectrin chain, a site 51 residues to the amino-terminal side of the abnormal proteolytic cleavage site. Analysis of the proposed triple helical model of spectrin repeats reveals that the mutation occurs in helix 2 at a position directly opposite the abnormal proteolytic cleavage site in helix 3, making this the first report of a mutation occurring in helix 2 of a repeat in the alpha I domain of spectrin. These results add to the molecular heterogeneity of mutations associated with HE/HPP and provide further support for the proposed triple helical model of spectrin. Disruption of this proposed alpha-helical structure by helix-breaking proline substitutions may result in a functionally defective spectrin chain.

Molecular and functional changes in spectrin from patients with hereditary pyropoikilocytosis.

The structural and functional properties of spectrin from normal and hereditary pyropoikilocytosis (HPP) donors from the two unrelated families were studied. The structural domains of the spectrin molecule were generated by mild tryptic digestion and analyzed by two-dimensional electrophoresis (isoelectric focusing; sodium dodecyl sulfate-polyacrylamide gel electrophoresis). The alpha I-T80 peptide (Mr 80,000) is not detectable in two related HPP donors; instead, two new peptides (Mr 50,000 and 21,000) are generated and have been identified as fragments of the normal alpha I-T80. A third sibling has reduced levels of both the normal alpha I-T80 and the two new peptides. A similar analysis of spectrin from another HPP family indicates that their spectrins contain reduced amounts of the alpha I-T80 and the 50,000 and 21,000 fragments of the alpha I domain. The HPP donor also has other structural variations in the alpha I, alpha II, and alpha III domains. The alpha I-T80 domain of normal spectrin has been shown to be an important site for spectrin oligomerization (J. Morrow and V.T. Marchesi. 1981. J. Cell Biol. 88: 463-468), and in vitro assays indicate that HPP spectrin has an impaired ability to oligomerize. Ghost membranes from HPP donors are also more fragile than membranes from normal erythrocytes when measured by ektacytometry. In both the oligomerization and fragility assays, the degree of impairment is correlated with the amount of normal alpha I-T80 present in the spectrin molecule. We believe that a structural alteration in the alpha I-T80 domain perturbs normal in vivo oligomerization of spectrin, producing a marked decrease in erythrocyte stability.

Combined sickle cell disease and autoimmune hemolytic anemia.

Four patients experienced combined sickle cell disease and autoimmune hemolytic anemia within the past ten years. A fifth patient had positive direct antiglobulin test results without verified autoimmune hemolysis. Severely accelerated hemolysis was observed in four patients; anemia was severe, and the reticulocyte count rose into the 60% to 88% range. During the period of active autoimmune hemolysis, decline of the reticulocyte count rose into the 60% to 88% range. During the period of active autoimmune hemolysis, decline of the reticulocyte count into the 6% to 16% range was associated with rapid decrease in the hemoglobin level, requiring transfusion. All five patients were already alloimmunized by transfusions administered before onset of the autoimmune hemolytic anemia; two or more allospecificities were identified in four of five patients. The presence of autoantibody notably compromised compatibility testing; three patients experienced posttransfusion hemoglobinuria, and in vivo cross matching with 51Cr-labeled donor RBCs was employed on three occasions. All patients responded to corticosteroids; mercaptopurine was also administered to one patient. The direct antiglobulin test result reverted to negative in all patients after hospital discharge.

Rheoscopic investigation of red cell deformability in sickle cell anemia.

Adequate erythrocyte deformability is crucial to microvascular function. In sickle cell anemia, a significant fraction of the circulating red cells lose deformability and assume highly abnormal shapes when exposed to low plasma oxygen tension (PO2). The loss of deformability is believed to induce blockage of flow in capillaries with consequent painful crisis or organ infarcts. The deformability of sickle erythrocytes at graded levels of PO2 were investigated in the rheoscope, a viscometric device consisting of transparent counter-rotating cone and plate. Quantitative indices of deformability obtained from still photographs and videotape recordings of cells subjected to shear flow were: fraction of all suspended cells capable of deformation, steady-state elongation, and time course of transient shape recovery following abrupt flow cessation. Suspensions of unfractionated cells were first equilibrated against gas mixtures (O2, N2, CO2) with PO2 = 160, 40 or 20 mm Hg at room temperature and then sheared under the same atmosphere. Results obtained with blood samples from ten pediatric patients being treated at St. Louis Children's Hospital show strong donor-to-donor variations and significant impairment of deformability in the unsickled members of the cell populations relative to normal controls.

Membrane-active agents and heat-induced erythrocyte fragmentation.

Morphological observations of heated erythrocytes were made after preincubation with lysolecithin, diamide, p-chloromercuribenzoate, or N-ethyl maleimide and after ATP-depletion. Diamide (1.0 mM) reduced the critical temperature of fragmentation from 49 degrees C to 47 degrees C, and at higher concentrations the critical temperature was further reduced. The other sulphydryl reacting agents had little or no effect. Drug-induced spherocytes and ATP-depleted cells did not fragment. Membrane-active agents can effect the response of membranes to heat and the effect is dose-dependent.

Heat-induced erythrocyte fragmentation in neonatal elliptocytosis.

Erythrocytes from neonates with elliptocytosis were studied for their pattern of heat-induced fragmentation. Membrane alterations began at 44 degrees C. There was a gradual progression in shape changes as the temperature was increased to 47 degrees C, at which point frank fragmentation occurred. Normal erythrocytes show no morphologic changes until the critical temperature of fragmentation, 49 degrees C is reached. Heat studies were repeated a few months later, at a time when the patients' erythrocyte morphology had become typical of elliptocytosis. Morphologic changes occurred abruptly at 48 degrees C with complete fragmentation. Increased thermal sensitivity of the red cell membrane has previously been demonstrated for pyropoikilocytes, and these studies suggest that some cases of elliptocytosis may be mild expressions of a similar membrane defect.

Erythrocyte porphyrin analysis in the detection of lead poisoning in children: evaluation of four micromethods.

We evaluated four procedures for determination of erythrocyte porphyrin: double extraction with ethyl acetate/acetic acid-HCl, single extraction with ethanol, single extraction with acetone, and direct solubilization with detergent-buffer. The ethyl acetate procedure, when used with two portions of HCl, apparently gives complete recovery of porphyrin and is suitable for reference as a comparison method. The ethanol procedure gives a high and consistent recovery and is technically simpler. The acetone procedure gives low and variable recovery of porphyrin, and the detergent-buffer method is subject to serious hemoglobin interference; neither of these two procedures offers any technical advantage. Stability of samples and methods for standardization were explored. A procedure for expressing results in terms of erythrocyte Zn-protoporphyrin content is given. Because of its stability, coproporphyrin is useful as a daily working standard. The ethyl acetate and ethanol methods are about equally efficient for detecting lead intoxication. Because of its simplicity, the ethanol method seems to be the best for use in screening.

A congenital haemolytic anaemia with thermal sensitivity of the erythrocyte membrane.

Microspherocytes, measuring 2-3 mum in diameter, and cells with blunted projections or triangular in shape characterized the erythrocoyte morphology in three children with congenital haemolytic anaemia. Since the erythrocyte morphology resembled that associated with thermal injury, heat-induced changes in erythrocyte morphology and membrane composition were studied. Erythrocytes developed filaments and spheroid bodies which fragmented, resulting in microspherocyte transformation. Normal cells required exposure to 49 degrees C, whereas the patients' cells fragmented at 45 degrees C. Fragmentation was also observed during incubation of patients' cells at 37 degrees C for 17h. The heat-induced transformation of the patients' cells was associated with an increase in the membrane cholesterol:phospholipid and cholesterol:protein ratios. The phospholipid:protein ratio was unchanged. This suggests that fragmentation produces a selective loss of membrane components. Splenectomy ameliorated the haemolytic process. We propose that the patients' red-cell morphology is the result of in vivo fragmentation, and that the spleen is the major site of microspherocyte and poikilocyte destruction.

Accurate determination of mean cell volume by isotope dilution in erythrocyte populations with variable deformability.

Variations in erythrocyte deformability and morphology lead to artifacts in electronic determinations of mean cellular volume (MCV) by the aperture-impedance method. The micropipette-aspiration technique loses accuracy when applied to severely aberrant cells such as dense sickle cells. A new light-scattering technique requires that the cells be capable of undergoing isovolumetric sphering. In contrast, the isotope-dilution (ID) method measures absolute mean volume and is free of artifacts associated with abnormal deformability or morphology. It does not depend on any algorithms or correction factors and does not subject the cells to any stringent processing, not even centrifugation. The ID method can be used to determine the mean volume of red cells in hypo- or hypertonic media or in the presence of pharmacologic agents. It requires no more than a 1-ml aliquot of suspended cells at a hematocrit of at least 30%. The cells can be readily recovered, washed, and reused. Using EDTA labeled with 57Co as an extracellular space marker we have used ID to determine the MCV of fractionated normal human red blood cells (RBC), unfractionated RBC containing SS hemoglobin, and RBC from four other mammalian species. In the case of human RBC obtained from eight normal donors, we obtained mean MCV values (+/- SD) of 83.6 +/- 3.0, 87.5 +/- 3.9, and 76.5 +/- 5.3 fl for unfractionated and top and bottom 10% density fractions, respectively. The value 83.6 is significantly lower than the generally accepted range of 89-91 indicated by electronic analyzers calibrated against spun microhematocrits. The discrepancy of about 7% can account for the difference between mean cell hemoglobin concentration (MCHC) data determined by a calibrated Coulter Counter and corresponding data obtained with paired samples using a cyanmethemoglobin procedure specified in NCCLS Standard H15-A and corrected for trapped plasma.

Bacteremia in sickle hemoglobinopathies.

We analyzed 178 episodes of bacteremia that occurred during 13,771 patient-years of follow-up of 3451 patients with sickle hemoglobinopathies. Age-specific incidence rates of bacteremia were calculated for patients with sickle cell anemia (SS) and sickle cell-hemoglobin C (SC) disease. The incidence rate was highest among children with SS and SC younger than age 2 years. Children with SC showed an abrupt decrease after age 2 years, whereas children with SS had a gradual decline in rate from 2 to 6 years of age. The predominant pathogen in patients younger than 6 years was Streptococcus pneumoniae (66%); gram-negative organisms were responsible for 50% of bacteremias in patients 6 years and older. Urinary tract infection was present during 73% of Escherichia coli bacteremias, and 77% of Salmonella bacteremias were associated with osteomyelitis. In contrast, no focus of infection was present in 52% of pneumococcal bacteremias. The incidence of pneumococcal bacteremia in children with SS younger than age 3 years was 6.1 events/100 patient-years; the case fatality rate for pneumococcal sepsis in this age group was 24%. No hematologic or demographic variables were associated with occurrence of pneumococcal bacteremia in young children. Retrospective analysis of pneumococcal bacteremia suggests that the prophylactic use of penicillin may decrease the incidence in children younger than 3 years of age.