In a comparative study of erythrocyte metabolism of vertebrates, the specific activity of glucose-6-phosphate dehydrogenase (G6PD) of the Brazilian opossum Didelphis marsupialis in a hemolysate was shown to be high, 207 ± 38 IU g-1 Hb-1 min-1 at 37ºC, compared to the human erythrocyte activity of 12 ± 2 IU g-1 Hb-1 min-1 at 37ºC. The apparent high specific activity of the mixture led us to investigate the physicochemical properties of the opossum enzyme. We report that reduced glutathione (GSH) in the erythrocytes was only 50% higher than in human erythrocytes, a value lower than expected from the high G6PD activity since GSH is maintained in a reduced state by G6PD activity. The molecular mass, determined by G-200 Sephadex column chromatography at pH 8.0, was 265 kDa, which is essentially the same as that of human G6PD (260 kDa). The Michaelis-Menten constants (Km: 55 µM) for glucose-6-phosphate and nicotinamide adenine dinucleotide phosphate (Km: 3.3 µM) were similar to those of the human enzyme (Km: 50-70 and Km: 2.9-4.4, respectively). A 450-fold purification of the opossum enzyme was achieved and the specific activity of the purified enzyme, 90 IU/mg protein, was actually lower than the 150 IU/mg protein observed for human G6PD. We conclude that G6PD after purification from the hemolysate of D. marsupialis does not have a high specific activity. Thus, it is quite probable that the red cell hyperactivity reported may be explained by increased synthesis of G6PD molecules per unit of hemoglobin or to reduced inactivation in the RBC hemolysate.

Eighty micrograms red blood cell (RBC) ghosts from patients who had previously exhibited the cutaneous form of loxoscelism (presenting localized dermonecrosis) and the viscerocutaneous form of loxoscelism (presenting dermonecrosis, hemoglobinuria, hematuria, and jaundice) and from controls were incubated with 2.5 µg crude Loxosceles gaucho venom in 5 mM phosphate buffer, pH 7.4, at 37ºC. Among all membrane proteins, quantitative proteolysis of the important integral transmembrane protein 3 increased with venom dose and with incubation time from 30 to 120 min, as demonstrated by gel densitometry. Similar quantitative data were obtained for RBC ghosts from patients and from control subjects, a fact that argues against the possibility of genetic factors favoring the hemolytic viscerocutaneous form. These data suggest that the clinical forms may be different types of the same disease, with the viscerocutaneous form being the result of large amounts of intravascularly injected venom and the superficial form being the result of in situ venom action. Since protein 3 is a housekeeping integral membrane protein, whose genetic deficiency leads to hemolytic anemia, it is reasonable to relate it to the hemolysis which occurs in the viscerocutaneous form of loxoscelism. The venom protease responsible for the process was not inhibited after 120-min incubation by 0.2 mM paramethylsulfonyl fluoride or by 0.2 mM N-ethylmaleimide but was inhibited by 25 mM ethylenediaminetetraacetic acid (a calcium-chelating agent) in 5 mM phosphate buffer at pH 7.4, which suggests that the enzyme is a calcium-dependent metalloprotease.

Fetal hemoglobin was measured in HIV1/2 patients under treatment with combined therapy (zidovudine and a protease inhibitor). A total of 143 patients and 103 normal individuals were investigated by the quantitative method of Betke and the semi-quantitative acid elution method of Kleihauer. In the normal person, hemoglobin F makes up less than 1% and an increase higher than 1.5% was observed in 21.4% of HIV patients by the method of Betke and in 24.8% of HIV-infected patients by the method of Kleihauer. The quantitative biochemical method of Betke showed that the populations were significantly different (two-tailed Mann-Whitney test). The reason for this hemoglobin F increase might be ascribed to the effect of zidovudine or to direct viral action on gamma chain expression. The finding of a higher F cell frequency indicated by the method of Kleihauer rather suggests that there is an increased F cell clone proliferation rather than an increase in hemoglobin F level in every cell.

Red blood cells (RBC) are viable if kept in an adequate preservative solution, although gradual changes in morphology and metabolism may occur. There is a gradual decrease in adenosine-5'-triphosphate (ATP) concentration, pH, glucose consumption, and enzyme activity during preservation. The normal discocyte shapes are initially replaced by echinocytes and stomatocytes and, at final stages, by spherocytes, the last step before splenic sequestration. Post-transfusional survival has been correlated with the ATP concentration. RBC preserved in ADSOL, a solution containing adenine, dextrose, sodium chloride, and mannitol, are viable for transfusion for up to 6 weeks. Erythrocytes from 10 blood units taken from healthy adult donors were preserved for 12 weeks in ADSOL at 4oC. We now report a significant correlation (r2 = 0.98) between the percentage of discocytes (89 to 7%) and ATP (100 to 10%) concentration in ADSOL-preserved RBC. The results suggest that the percent of discocyte shapes used as an indicator of ATP concentration may be a useful indicator for quality control of RBC viability in centers which have limited assay facilities.