The effect of complement activation on oxidative stress in PNH-like RBCs, in the presence of C5 blockade, depends on their G6PD status. Levels of ROS, expressed as mean fluorescence intensity (MFI), were quantitated by flow cytometry with dichlorodihydrofluorescein diacetate (400 µM),¹⁰  after 24 hours at 37°C since complement activation. (A) Bar diagram of ROS levels in non-PNH (green bars) and in PNH-like RBCs (red bars) from 6 healthy control subjects (Healthy control RBC; solid bars) and from 6 G6PD-deficient male subjects (G6PD-deficient RBC; striped bars) after exposure to sera in which complement (C) had been heat-inactivated (C-inactivated sera). The gray bars show ROS levels when non-PNH RBCs were exposed to complement activated by acidification (C activated); as expected, PNH-like RBCs were lysed when exposed to activated complement (data not shown). Average and standard errors are shown. In these experiments, sera were acidified even when complement was due to be heat inactivated to make sure that acidification as such had no effect. (B) Box and whisker plot of levels of ROS after complement activation in the presence of C5 blockade by eculizumab in non-PNH (green boxes) and PNH-like RBCs (red boxes) from 11 healthy control subjects (Healthy control RBC; solid boxes) and from 8 G6PD-deficient male subjects (G6PD-deficient RBC; striped boxes). ROS levels are expressed as MFI, the bottom and top of the box show the 25th and 75th percentile, the horizontal line within the box shows the median, and the ends of the whiskers represent the minimum and the maximum value. Nonparametric tests for paired (Wilcoxon) and unpaired (Mann-Whitney) samples were performed, as appropriate. Statistical significance was accepted for P ≤ .05. NS, not statistically significant.