Epinephrine-induced activation of LW-mediated sickle cell adhesion and vaso-occlusion in vivo

Abstract

Sickle red cell (SS RBC) adhesion is believed to contribute to the process of vaso-occlusion in sickle cell disease (SCD). We previously found that the LW RBC adhesion receptor can be activated by epinephrine to mediate SS RBC adhesion to endothelial alphavbeta3 integrin. To determine the contribution of LW activation to vaso-occlusive events in vivo, we investigated whether in vitro treatment of SS RBCs by epinephrine resulted in vaso-occlusion in intact microvasculature after RBC infusion into nude mice. Epinephrine enhanced human SS but not normal RBC adhesion to murine endothelial cells in vitro and to endothelium in vivo, promoting vaso-occlusion and RBC organ sequestration. Murine sickle RBCs also responded to epinephrine with increased adhesion to postcapillary endothelium in nude mice. Epinephrine-induced SS RBC adhesion, vaso-occlusion, and RBC organ trapping could be prevented by the beta-adrenergic receptor (beta-AR) antagonist, propranolol. Infusion of soluble recombinant LW also significantly reduced adhesion and vaso-occlusion. In addition, epinephrine-treated SS RBCs induced activation of murine leukocyte adhesion to endothelium as well. We conclude that LW activation by epinephrine via beta-AR stimulation can promote both SS RBC and leukocyte adhesion as well as vaso-occlusion, suggesting that both epinephrine and LW play potentially pathophysiological roles in SCD.

Figures

Figure 1

Epinephrine induces increased human and murine SS RBC adhesion to murine EOMA cells in vitro. Results are presented as percent adherent RBCs at a shear stress of 2 dynes/cm2; error bars show SEM of 3 different experiments. RBCs were sham treated or stimulated with 20 nM epinephrine (epi) for 1 minute. (A) Human RBCs: Inhibition of adhesion with antibody was performed as described in “Materials and methods.” *P < .001 compared with sham-treated normal RBCs; **P < .001 compared with epi-treated SS RBCs. (B) Murine normal and SS RBC adhesion: *P < .01 compared with sham-treated normal RBCs; **P < .001 compared with sham-treated SS RBCs.

Figure 2

Immune interactions and survival of human RBCs in nude mice. (A) Anesthetized nude mice were infused with Dil (rhodamine)–labeled sham-treated or epi-treated normal RBCs. Blood samples collected 40 minutes after infusion were analyzed for murine immunoglobulin, which was detected using FITC-labeled antimurine immunoglobulin, bound to human RBCs. Insignificant murine immunoglobulin was bound to circulating sham- or epi-treated human normal RBCs compared with noninfused cells. One representative experiment is shown (n = 3). (B,C) Anesthetized mice were injected with a 1:1 mixture of Dil-labeled sham- (■) and DiO-labeled epi-treated (□) RBCs ([B] normal, [C] SS RBCs; n = 3 for each). Error bars show SEM of 3 different experiments. The percentages of sham- and epi-treated normal RBCs circulating in the bloodstream of animals at all times were not significantly different. A significantly greater percentage of sham-treated than epi-treated SS RBCs was retained in the circulation at all times (P < .001). (D,E) Normal RBCs were sham treated or epi treated. Sham-treated normal RBCs were detected to some degree in the spleen. Epi treatment did not lead to a significant increase in splenic trapping of normal RBCs. (F,G) Epi treatment had a significant effect on trapping of SS RBCs in the spleen compared with sham-treated cells. *P < .05 compared with sham-treated. Scale bar for panels D and F = 100 μm.

Figure 3

Epinephrine stimulates activation of SS, but not normal, RBC adhesion to vessel walls in nude mice. (A-D) Microscopic observations of postcapillary venules were conducted through implanted window chambers after infusion of RBCs into the tail vein of nude mice using × 5 and × 20 magnification to observe human RBCs and × 20 magnification to observe murine RBCs. Vessels without adherent cells appear gray due to rapidly moving fluorescent RBCs. (A) Infusion of sham-treated (n = 8), forskolin (fsk) + IBMX–treated (n = 3) or epi-treated (n = 8) human SS RBCs. Sham-treated human SS RBCs showed little adhesion to vessel walls, whereas fsk + IBMX–treated and, to a greater extent, epi-treated human SS RBCs showed marked adhesion to postcapillary venules, with intermittent vaso-occlusion, as indicated by arrows. (B) Infusion of sham-treated (n = 3) or epi-treated (n = 3) normal human RBCs. Sham-treated human normal RBCs showed no adhesion to venule walls. Epi had little to no effect on human normal RBC adhesion to vessel walls. Scale bar = 200 μm. (C) Infusion of sham-treated (n = 4) or epi-treated (n = 4) murine sickle cells. Sham-treated murine SS RBCs showed very weak adhesion to vessel walls. Epi increased adhesion of murine SS RBCs to postcapillary venules. (D) Infusion of sham-treated (n = 3) or epi-treated (n = 3) normal murine RBCs. Sham- or epi-treated normal murine RBCs showed no adhesion to venule walls. (E) Fsk + IBMX or epi enhances human SS RBC occupation of venular length. SS RBCs were sham treated, IBMX + fsk treated, or epi treated (n = 3 for each treatment) prior to infusion into the tail vein of nude mice. The values of 30 segments of vessels analyzed were averaged among groups of animals to represent the mean percent venular length occupied by SS RBCs. Error bars show SEM of 3 different experiments for each treatment. *P < .05 compared with sham-treated regardless of the vessel diameter within the ranges specified. (F) Percentage venular length occupied by human SS RBCs related to venular diameter. Animals were injected with sham- (■) or epi-treated (▴) SS RBCs. Error bars show SEM for 6 different experiments for each treatment. *P < .001 compared with sham-treated. Data were compared using 1-way ANOVA analysis followed by Bonferroni corrections for multiple comparisons.

Figure 4

Epinephrine increases SS RBC organ sequestration. (A,B) SS RBCs were sham treated, fsk + IBMX treated, or epi treated. Sham-treated SS RBCs were detected to some degree in the lung but only minimally in the kidney. Treatment with fsk + IBMX did not significantly increase SS RBC trapping in the lung or kidney, and epi treatment led to significant increases in RBCs trapping only in the lung. Error bars show SEM of 3 different experiments. *P < .05 compared with sham-treated. The 2 parts in panel A represent 2 different experiments with similar results. (C,D) Normal RBCs were sham treated or epi treated. Epi treatment only had a significant effect on trapping of normal RBCs in the lung compared with sham-treated cells. Error bars show SEM of 3 different experiments. *P < .05 compared with sham-treated. Scale bar = 100 μm.

Figure 5

Effect of propranolol on epinephrine-induced SS RBC adhesion and organ sequestration. (A) β- and β2-AR antagonists abolish epi-induced SS RBC adhesion to HUVECs in vitro at a shear stress of 2 dynes/cm2. SS RBCs were sham treated, epi treated, or pretreated with phenoxybenzamine, propranolol, atenolol, or butoxamine followed by treatment with epi (n = 3 each). Error bars show SEM of 3 different experiments. *P < .01 compared with sham-treated; **P < .001 compared with epi-treated. (B) In vitro propranolol treatment had no effect on SS RBC adhesion (n = 5) in vivo, whereas epi dramatically increased SS RBC adhesion and vaso-occlusion to venule walls (n = 5). Propranolol treatment of SS RBCs significantly reduced subsequent epi-stimulated adhesion and stasis (n = 5). (C) Effect of propranolol on percent venular length occupied by SS RBCs. Video frames showing more than 30 vessel segments were used to quantify the length of venules occupied by SS RBCs in animals infused with SS RBCs treated as described for panel A (n = 5 for each treatment). The values were averaged among groups of animals to represent the mean percent venular length occupied by SS RBCs. Error bars show SEM. *P < .01 compared with propranolol-treated RBCs for vessels up to 25 μm in diameter, and P < .001 for vessels more than 25 μm in diameter. **P < .01 compared with epi-treated RBCs regardless of the vessel diameter. (D-E) SS RBCs were epi treated or propranolol + epi treated. Epi-treated SS RBCs were extensively trapped in the lung and spleen. Propranolol significantly reduced epi-treated SS RBC trapping in the lung and spleen. SS RBC trapping in the kidney was similar for all cell preparations. Error bars show SEM of 5 different experiments. *P < .05 compared with sham-treated (not pictured); **P < .05 compared with epi-treated. Scale bar = 100 μm.

Figure 6

Epinephrine-activated SS RBCs adhere through LW and induce activation and adhesion of murine leukocytes. (A-B) Epinephrine-induced SS RBC adhesion and vaso-occlusion are mediated via LW. (A) Inhibition of SS RBC adhesion with soluble proteins was performed as described in “Material and methods.” Sham-treated SS RBCs showed little adhesion to vessel walls, whereas epi-treated SS RBCs showed marked adhesion to postcapillary venules, with intermittent vaso-occlusion. Infusion of sCD44 (as a negative control) 30 minutes prior to intravenous administration of epi-treated SS RBCs (n = 5) had no effect on epi-treated SS RBC adhesion to postcapillary venules, whereas infusion of sLW 30 minutes prior to injection of epi-treated SS RBCs (n = 5) markedly inhibited adhesion to postcapillary vessels. (B) Involvement of LW in SS RBC adhesion to endothelium and vaso-occlusion. Animals were infused with sham- or epi-treated SS RBCs 30 minutes after intravenous administration of sCD44 or sLW (n = 5 for each treatment). Thirty segments of venules were analyzed to quantify the length of venules occupied by SS RBCs. The values were averaged among groups of animals to represent the mean percent venular length occupied by SS RBCs. Error bars show SEM of 5 different experiments. *P < .001 compared with sCD44/epi-treated (animals/cells) for vessels up to 25 μm in diameter, and P < .05 for vessels more than 25 μm in diameter. (C) Epi-activated human SS RBCs induce adhesion of murine leukocytes to endothelium. Fluorescently labeled (red) sham- (n = 3) and epi-treated SS RBCs (n = 3) were infused 30 minutes after injection of rat FITC-labeled antimouse CD45 or LFA-1 antibodies. In the presence of epi-treated human SS RBCs, murine leukocytes adhered to postcapillary endothelium, and this adhesion was blocked by antibody against mouse LFA-1. Arrows indicate identical areas visualized with filters for red and green fluorescence to show where human SS RBCs and murine leukocytes are adherent to endothelium. In some areas, both murine leukocytes and human SS RBCs were adherent, whereas in other areas only one adherent cell type was present. Top panels show adhesion of human SS RBCs, bottom panels show adhesion of murine leukocytes.

Figure 7

Model of SS RBC activation by epinephrine. We previously demonstrated that stimulation of ARs by epi affects downstream events via AC through activation of Gαs protein complex. Increased intracellular cAMP as a result of activation of AC leads to activation of PKA, which acts as a downstream effector to up-regulate SS RBC adhesion mediated by activation of LW, which becomes phosphorylated and preferentially recognizes the endothelial αvβ3 integrin. We now propose that epinephrine acts via stimulation of RBC β2-ARs and leads to activation of LW to interact with both endothelial cells as well as leukocytes.