A Novel Selectin Antagonist, GMI-1070, Prevents Vaso-Occlusion in Sickle Cell Mice by Inhibiting Leukocyte Adhesion and Activation.

Abstract

Previous studies using intravital microscopy in a sickle cell disease (SCD) mouse model (Berkeley) suggest that adherent leukocytes (WBCs) play a key role in vaso-occlusion by capturing circulating erythrocytes (RBCs) in venules. In addition, mice deficient in both P-and E-selectins are protected from vaso-occlusion (VOC) induced by surgical trauma and TNF-α stimulation, suggesting that targeting selectins or their ligands represents a potentially useful strategy. Selectins bind to specific sialylated and fucosylated carbohydrate structures presented by glycoprotein or glycolipid ligands. Here, we tested the effect of novel small glycomimetic selectin inhibitors, GMI-1070 and GMI-1077, on leukocyte behavior and sickle cell VOC. Berkeley SCD mouse bone marrow was transplantated into lethally irradiated C57BL/6 animals to generate age- and gender-matched genetically identical cohorts of SCD mice. Fully engrafted male SCD mice were treated with TNF-α and prepared for intravital microscopy examination of the cremaster muscle 90 min later. GMI-1070, GMI-1077 (both 20 mg/kg) or vehicle (PBS) were administered immediately prior to cytokine stimulation (t=0 min), and an additional dose was given at t=70min. Another group of mice was injected with antibodies against P-and E-selectins (PES, 1 mg/kg) as positive control. Several post-capillary and collecting venules were examined between t= 90min and t= 150min. Antibody blockade of endothelial selectins completely ablated leukocyte rolling, whereas GMI-1070 and GMI-1077 significantly increased the rolling flux fractions (PBS: 5.0±1.2 GMI-1070: 10.6±1.3%%; GMI-1077: 9.9±1.0%; p< 0.001). Furthermore GMI-1070 and GMI-1077 significantly reduced the recruitment of adherent leukocytes (914±172 and 1433±119 cells/mm2, respectively) compared to sickle mice injected with PBS control (2400±392 cells/mm2, p< 0.001). Although the reduction in leukocyte adhesion was not as marked as with anti-P and E-selectins (61±25 cells/mm2, p< 0.001), GMI-1070, in particular, dramatically inhibited the capture of sickle RBCs by adherent leukocytes (PBS: 0.9±0.4, GMI-1077: 0.6±0.2, GMI-1070: 0.07±0.05 and PES: 0.01±0.01 RBC interactions/WBC/min, p< 0.05) and markedly improved the blood flow in venules (PBS: 312±24, GMI-1077: 398±41, GMI-1070: 710±68 and PES: 683±75 nL/s, p< 0.001), to levels observed in non-sickle mice. The increased leukocyte rolling fluxes by these glycomimetics suggest that they inhibit E-selectin > P-selectin. Since the hallmark of E-selectin-mediated adhesion is the slow leukocyte rolling, we analyzed leukocyte rolling velocities in the various group and indeed found a 2-fold increase in rolling velocities in sickle mice treated with GMI-1070 compared to PBS control (PBS: 21±1 μm/s, GMI-1070: 38±1 μm/s, p<0.001). Consistent with these results, other studies using a parallel plate flow chamber (0.9 dynes/cm2) revealed that GMI-1070 was much more potent (1000-fold difference) in inhibiting the binding of human PMNs to TNF-α-stimulated (to induce E-selectin) endothelial cells (HUVEC) than with IL-4 and histamine stimulated HUVECs (to induce P-selectin). Further, competitive inhibition assays revealed that the IC50 of GMI-1070, relative to the standard glycyrrhizin, was much lower for E-selectin than P-selectin. These studies suggest that E-selectin-mediated adhesion/signaling may play a more important role than previously appreciated in the pathophysiology of SCD, and suggest that GMI-1070 may be beneficial for the treatment of sickle cell vaso-occlusion.