Abstract
Protein disulfide isomerase (PDI), a multifunctional oxidoreductase that regulates thrombus formation, leukocyte adherence to the endothelium and nitric oxide delivery, is present at high levels and regulates KCNN4 (Gardos Channel) activity in erythrocytes from humans with Sickle Cell Disease (SCD) and sickle mouse models. We reported evidence of elevated erythrocyte-surface associated and circulating PDI activity in humans and mouse models of SCD when compared to either wild-type mice, transgenic mice expressing normal human globins or otherwise healthy individuals; results that suggest a novel role for PDI in the pathophysiology of SCD through unclear mechanisms. We studied the effects of deoxygenation (5% O2) on PDI activity in endothelial cells and sickle erythrocytes from humans with SCD and two sickle transgenic knockout mouse models expressing human sickle hemoglobin, BERK and βSAntilles. Our data shows that deoxygenation of human and mouse sickle erythrocytes increased surface-associated reductive capacity that was sensitive to monoclonal antibodies against PDI (mAb PDI [RL-90]). We then studied sickle human erythrocytes and showed that PDI inhibitors (quercentin-3-rutinoside [Q3R] and mAb PDI) significantly reduced deoxy-stimulated sickle cell dehydration and Gardos channel activity (n=5; P <0.03). We characterized erythrocyte density with a phthalate density-distribution assay, generated density distribution curves and calculated the D50. Both mAb PDI and Q3R significantly reduced D50 when compared to vehicle (1.113±0.002 to 1.102±0.001 [P <0.0001]; and 1.101±0.002 [P <0.0001]; respectively). In contrast, incubation with exogenous PDI (3 nM) increased cellular dehydration (from D50= 1.110 ± 0.001 to D50= 1.115 ±0.001, P <0.01). We also measured the effect of hypoxia and endothelin-1 (ET-1; 10nM) in the human vascular endothelial cell line, EA.hy926. We observed that hypoxia induced PDI secretion that was further enhanced by co-incubation with ET-1 (10nM; n=3; P <0.05). The selective inhibitor of ET-1 subtype B receptor antagonist, BQ788 blocked ET-1 stimulated PDI increases in these cells as was previously reported in red blood cells. Consistent with these results hypoxia was associated with increased mRNA expression of MCP-1, VEGF-a but not ICAM by qRT-PCR and Taqman probes (n=3; P <0.05). Of importance we observed that early cultures of mouse aortic endothelial cells from BERK mice showed similar results. We then evaluated the effects of PDI on erythrocyte hemolysis by exposing cells from patients with SCD to 20 mM 2-2'-azo-bis- (2-amidinopropane) dihydrochloride (AAPH) with or without Q3R or mAb PDI. AAPH-induced hemolysis was dose-dependently blocked by Q3R (IC50: 4.1nM; n=6, P <0.0001 compared to vehicle) or mAb PDI (IC50: 11nM; n=3). Irrelevant IgG did not affect hemolysis under these conditions. Experiments performed in blood from BERK or βSAntilles mice showed similar results. We then studied sickle mice that express varying levels of HbF; BERK (<1% HbF), BERKγM (25% HbF), and BERKγH (45% HbF). BERKγH had the lowest circulating and cell associated PDI activity among the three mouse types that was associated with lower circulating ET-1 levels (n=2; P <0.05). Consistent with these results we observed an inverse correlation between levels of HbF and PDI activity in cells from humans with SCD (n=4). Thus we posit that in SCD elevated erythrocyte PDI activity is important for cell survival and stability and that its inhibition may represent a novel therapeutic approach for improving both the hematological and vascular complications of SCD as it may not only increase sickle erythrocyte survival but may likewise interfere with cellular adhesion leading to reduced vaso-occlusive episodes. [Supported by NIH: HL090632 (AR) and HL096518 (JRR)]
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.