Reduction of dapsone hydroxylamine to dapsone during methaemoglobin formation in human erythrocytes in vitro--II. Movement of dapsone across a semipermeable membrane into erythrocytes and plasma.

We have used an in vitro two-compartment model, to investigate the ability of dapsone, formed by erythrocyte-mediated detoxification of its hydroxylamine metabolite, to escape the cells and cross a semi-permeable membrane into both plasma and other erythrocytes. Both diethyl dithiocarbamate (DDC) treated and untreated erythrocytes were incubated with dapsone hydroxylamine and dialysed against either fresh cells or plasma. Methaemoglobin was predominantly detectable in compartment A although the presence of low levels of methaemoglobin in compartment B indicated that the hydroxylamine itself had crossed the membrane. In contrast to methaemoglobin disposition, recovery of dapsone was higher (P < 0.05) in compartment B compared with A for all three treatment groups at 30 and 60 min, but not at the remaining time points. Regression analysis of the cumulative recovery of dapsone over 150 min in all three treatment groups for both compartments A and B showed correlation coefficients close to unity. In compartment A, analysis of the mean slopes of the regression lines indicated that, overall, significantly more dapsone was recovered from group 1 (erythrocytes, hydroxylamine and DDC dialysed against untreated red cells) compared with group 3 (erythrocytes and hydroxylamine dialysed against plasma) (0.22 +/- 0.05 vs 0.09 +/- 0.005; P < 0.025). Also in compartment A, significantly more dapsone was recovered from group 2 (erythrocytes and hydroxylamine dialysed against untreated red cells) compared with group 3 (erythrocytes and hydroxylamine dialysed against plasma: 0.16 +/- 0.02 vs 0.09 +/- 0.005). In compartment B, dapsone recovery was significantly greater in group 1 (erythrocytes, hydroxylamine and DDC dialysed against untreated red cells; slope of regression line: 0.59 +/- 0.05) compared with group 2 (erythrocytes and hydroxylamine dialysed against untreated red cells; slope of line: 0.28 +/- 0.02, P < 0.005). In addition, dapsone recovery was significantly greater in group 1 (0.59 +/- 0.05) compared with group 3 (erythrocytes and hydroxylamine dialysed against plasma; 0.21 +/- 0.02, P < 0.005). Dialysis of erythrocytes with dapsone itself over 120 min caused no detectable methaemoglobin formation. The process of erythrocyte-mediated dapsone formation from its hydroxylamine may feasibly occur in vivo and contribute to the systemic persistence and therapeutic effect of dapsone.