Fig. 2. Mannose expression correlates with clinical anemia in sickle cell disease.
a Normalized geometric mean fluorescence (gMFI) of Galanthus nivalis Agglutinin (GNA) lectin staining of RBCs from peripheral blood samples, comparing RBC from patients with sub-types of sickle cell disease (milder phenotypes: HbSC (compound heterozygotes for hemoglobin S with C) (n = 34), HbS/beta-thal (n = 8)(compound heterozygotes for hemoglobin S with β-thalassemia), HbSS microcytic (n = 12))(compound heterozygotes for hemoglobin S homozygosity with α-thalassemia) and HbSS HPFH (n = 3)) (compound heterozygotes for hemoglobin S homozygosity with hereditary persistence of fetal hemoglobin) versus healthy donors (n = 45), sickle cell trait (n = 57)(HbAS) and β-thalassemia (n = 6). The dotted line indicates the 90th centile of GNA lectin binding within healthy samples. Data are shown as median +/− IQR, 2 tailed Mann–Whitney, p values shown, numerous (>20) experiments. 2 tailed p values relative to HbSS: HbAA, p = 3.9 × 10−14, HbAS, p = 3.9 × 10−16, HbSC, p = 1.8 × 10−6. Each data point derived from an independent RBC donor. No adjustments made for multiple comparisons. b Plot of RBC count against normalized GNA gMFI for sickle cell disease: HbS/B + and HbSC indicates compound heterozygosity for HbS with β-thalassemia and HbC respectively; Spearman’s rank correlation, p = 7.0 × 10−23, numerous (>20) experiments. Plots of: (c) RBC count versus serum lactate dehydrogenase (LDH), Spearman’s rank, d) RBC count vs GNA binding, Spearman’s rank, (p = 5.2 × 10−9) e LDH vs. GNA lectin binding; HbSS RBCs for which corresponding serum LDH values were available; Spearman’s rank correlation, numerous (>20) experiments. f Percentage phagocytosis of HbAA and HbSS RBCs by human monocyte-derived macrophages analyzed by microscopy. Mannan inhibition as shown. n = 4–7 independent RBC donors over two independent experiments. Data are shown as median +/− IQR, 2 tailed Mann–Whitney.