Fig. 4: Viability and superoxide production of neutrophils isolated from patients with long COVID syndrome.

The viability of neutrophilic granulocytes was tested with flow cytometry after Annexin A5-FITC (A5) and Propidium Iodide (PI) staining (a). The percentage (mean ± SEM of 27 [LCS], and 17 [control] independent experiments) of viable (double negative), early apoptotic (A5 + PI−), late apoptotic (double positive), or necrotic (A5-PI+) cells is shown. LCS: long COVID syndrome, n.s.: no significant difference. Mann–Whitney U test was performed. Basal superoxide production of PMN isolated from patients with long COVID syndrome (LCS) or healthy volunteers (control) (b). Unstimulated superoxide production was determined with lucigenin, and the relative luminescence unit (RLU) was measured in 90 minutes at 37 °C with a luminometer. The mean ± SEM of 17 (LCS) and 11 (control) independent experiments is plotted. Data were analyzed using a two-way repeated measures ANOVA. ****p < 0.0001 represents the significant interaction (time x disease). The maximal superoxide production of unstimulated PMN cells (c), the slope of the curves (d), and the area under the curve (e) were also determined. The mean ± SEM of 17 (LCS) and 11 (control) independent experiments is shown. Data were analyzed with an unpaired t test. **p < 0.01; ***p < 0.0001, compared to control. Superoxide production of neutrophilic granulocytes of children with long COVID syndrome (LCS) and healthy, COVID-19 convalescent (CG+) was measured after stimulation with PMA (f) or pooled human serum-opsonized Zymosan (g). Mean ± SEM of 19 (LCS) and 12 (control) independent experiments are plotted. Data were analyzed with two-way repeated measures ANOVA. *p < 0.05 represents the significant interaction (time x disease). n.s.: no significant interaction. The maximal superoxide production (h) the slope of the superoxide curves (i), and the area under the curves (j) are calculated. The mean ± SEM of 19 (LCS) and 12 (control) independent experiments (except h and j–control group Zymosan: 11) is plotted. Data were analyzed using a Mann–Whitney U test. n.s.: no significant difference. k PMA-, or pooled human serum-opsonized zymosan-induced superoxide production (l) normalized to the unstimulated samples. Mean + SEM of 17 (LCS) and 11 (control) independent experiments are plotted. Data were analyzed using two-way repeated measures ANOVA. **p < 0.01 represents the significant interaction (time x disease). n.s.: no significant interaction. Characteristics of the stimulated superoxide production—maximal superoxide production (m), the slope of the superoxide curves (n), and the area under the curve (o)–normalized to the unstimulated values, were calculated. Mean ± SEM of 17 (LCS) and 11 (control) individual experiments (except m—LCS group Zymosan: 15, PMA: 16, and o—LCS group Zymosan: 16) are shown. Statistical analysis was performed with the Mann-Whitney U test. *p < 0.05; **p < 0.01, compared to control.