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.

Erythrocyte membrane proteins from 44 representative mammals were studied. Protein 4.2 was not detected in guinea pigs (Cavia porcellus) (N = 14), Southern Brazilian swamp large rats (Myocastor coypus) (N = 2), cutias (Dasyprocta sp) (N = 4), and horses (Equus caballus) (N = 13). These animals also presented high ankyrin concentrations except for the horse which did not exhibit a sharp band, although minor components located between proteins 2 and 3 could account for the ankyrin family. The rodents studied did present band 6, which was not detectable in other common rodents such as white rats (Rattus norvegicus) (N = 9) and mice (Mus musculus) (N = 12). Since the absence of protein 4.2 does not disrupt the cytoskeleton membrane, we suggest that it is not an essential protein. Its absence may be compensated physiologically by the higher ankyrin concentration observed.

Glucose-6-phosphate dehydrogenase (G6PD) activity and the affinity for its substrate glucose-6-phosphate were investigated under conditions similar to the physiological environment in terms of ionic strength (I: 0.188), cation concentration, pH 7.34, and temperature (37oC). A 12.4, 10.4 and 21.4% decrease was observed in G6PD B, G6PD A+ and G6PD A- activities, respectively. A Km increase of 95.1, 94.4 and 95.4% was observed in G6PD B, G6PD A+ and G6PD A-, respectively, leading to a marked decrease in affinity. In conclusion, the observation of the reduced activity and affinity for its natural substrate reflects the actual pentose pathway rate. It also suggests a much lower NADPH generation, which is crucial mostly in G6PD-deficient individuals, whose NADPH availability is poor.