Sickle Red Blood Cells Directly Activate Neutrophils

In sickle cell disease (SCD), abnormal sickle RBCs (SSRBCs) circulate and interact with both the endothelium and circulating blood cells. Among hematopoietic cells, neutrophils have emerged as a critical contributor to disease pathophysiology and vaso-occlusion.¹

Sickle RBCs promote inflammation and cellular activation. SSRBCs have increased expression and activity of proteins in the β2-adrenergic–Gαi protein–adenylyl cyclase–protein kinase A and MAPK pathways.^2,3 Both pathways activate the erythroid adhesion molecules Lutheran/basal cell adhesion molecule (Lu/BCAM),³ intercellular adhesion molecule 4 (ICAM4),² and CD44⁴, as well as NADPH oxidase pathways.⁵ Adhesion of SSRBCs promotes expression of pro-adhesive and pro-coagulant proteins by endothelial cells, increasing endothelial interactions with circulating blood cells.⁶ SSRBC hemolysis, with release of hemoglobin and microparticles, may also lead to neutrophil activation.¹ However, whether SSRBCs can directly stimulate neutrophils is unknown. In this study, we used SSRBCs derived from SCD patients at steady-state to determine if SSRBCs can directly activate healthy donor (HD) neutrophils.

With institutional approval and informed consent, blood samples were obtained from HDs and SCD patients (HbSS) at steady-state to characterize neutrophil adhesion and degranulation and to examine the ability of SSRBCs to activate normal neutrophils. To measure adhesion, a flow chamber system with human umbilical vein endothelial cells (HUVECs) grown to confluence on glass slides was used.² Neutrophils from HDs or SCD patients were incubated with or without ABO-compatible RBCs from HDs or with SSRBCs for 30 minutes. RBCs were then lysed in a 30-second process, and cells were washed in a 5-minute step. Subsequently, neutrophils were infused into the flow chamber and then washed out at a constant flow rate producing shear stresses of 1–5 dynes/cm²; adherent neutrophils were then enumerated. To examine levels of endogenous neutrophil degranulation, plasma matrix metalloprotease-9 (MMP9), a marker of tertiary granule release, was measured by ELISA (R&D Systems, Minneapolis, MN). To determine if SSRBCs could directly induce neutrophil degranulation, neutrophils from HDs were incubated for 30 minutes with buffer or ABO-compatible SSRBCs (1:4 to recapitulate physiologic conditions), and released MMP9 was measured.

As shown in Figure 1A and consistent with prior studies,⁷ neutrophils from SCD patients were significantly more adherent to HUVECs compared to HD neutrophils (HD 19.97 ± 9.26% vs. SCD 64.49 ± 15.04%, p<0.0001). Also consistent with prior reports,⁸ SCD patients have evidence of activated and primed neutrophils. In Figure 1B, plasma levels of MMP9 were higher in steady-state SCD patients compared to age and sex-matched HDs, indicating that neutrophils from SCD patients were activated at baseline (HD 47.93 ± 45.14 ng/mL vs. SCD 351.7 ± 355 ng/mL, p=0.0007). When incubated with the bacterial tri-peptide N-Formylmethionine-leucyl-phenylalanine (fMLF 1μM), a non-maximal stimulant of degranulation, neutrophils from SCD patients released markedly more MMP9, indicating that neutrophils from SCD subjects are primed (Figure 1C, HD 605.2 ± 422.7 ng/mL vs. SCD 2231 ± 892.8 ng/mL, p<0.0001).

Figure 1. Evidence of neutrophil activation in normal blood donors and blood of patients with SCD.

A. Adherence of neutrophils to HUVECs from healthy (n=19) or SCD subjects (n=7) under flow at 1 dyne/cm². B/C. MMP9 release after whole blood from healthy or SCD subjects was incubated with buffer (Panel B) or fMLF (Panel C, 1 μM, n=19) for 30 minutes. ***p<0.001, **** p<0.0001

To determine whether SSRBCs could directly activate neutrophils independent of exogenous cytokine stimulation or free hemoglobin, we first measured the ability of HD neutrophils to adhere after incubation with washed ABO-matched RBCs from either HDs (HbAA) or SCD donors. In addition, we examined neutrophil adhesion after incubation with RBCs that had or had not been exposed to epinephrine (20 μM, 1 min),^2,3 which activates RBC adhesion receptors BCAM/Lu and ICAM-4, both of which mediate binding to leukocyte integrins. As shown in Figure 2A, AARBCs failed to stimulate neutrophil adhesion, regardless of pre-treatment with epinephrine (neutrophils alone vs. AARBCs, p=ns; neutrophils alone vs. epi-treated AARBCs p=ns). However, SSRBCs did induce neutrophil adhesion to HUVECs (neutrophils alone vs. SSRBCs 27.04 ± 10.89%, p<0.01). Epinephrine-treated SSRBCs induced even greater neutrophil adhesion (neutrophils alone vs. epi-treated SSRBCs 41.32 ± 14.01%, p<0.001; SSRBCs vs. epi-treated SSRBCs p<0.001), regardless of individual baseline levels (Supplemental Figure 1A). RBC lysate prepared from both healthy RBCs and SSRBCs did not stimulate neutrophil adhesion (data not shown).

Figure 2. Intact SSRBCs but not AARBCs stimulate neutrophil adhesion and degranulation.

A. Neutrophils from healthy donors were incubated with AA normal (n=3) or SSRBCs (n=22) after treatment with sham or epinephrine (20 μM) for 30 minutes. After red cells were lysed and remaining cells were washed, neutrophils were infused into the flow chamber and adherence to HUVECs was quantified. B. Normal neutrophils were incubated with buffer or SSRBCs (n=7) for 30 minutes, and MMP9 release was measured. *p<0.05, **p<0.01, ***p<0.001, **** p<0.0001

Next, we determined if SSRBCs could directly induce HD neutrophil degranulation. Indeed, SSRBCs caused significant degranulation of HD neutrophils (buffer 132.7 ± 131.1 ng/mL vs. SSRBCs 229 ± 175.7 ng/mL, p=0.016, Figure 2B). Despite variable levels of degranulation at baseline, the addition of SSRBCs consistently caused an increase in MMP9 release (Supplemental Figure 1B, p=0.016).

From these studies, we corroborate prior work demonstrating that SCD patients have activated neutrophils that are more adherent, are prone to degranulation, and have heightened degranulation responses upon stimulation (Figure 1). These results are in keeping with prior studies that showed that neutrophils from SCD patients appear to be primed, possibly due to free heme, inflammatory cytokines, circulating iron, and cross-activation from stimulated platelets.^9–12 We now demonstrate that SSRBCs obtained from SCD patients at steady-state can directly activate HD neutrophils, inducing both adhesion and degranulation (Figure 2). Our findings suggest that SSRBCs themselves are likely responsible, to a significant degree, for the long-observed activation of circulating neutrophils in SCD.

It is presently unclear what specific biologic characteristics of SSRBCs enable them to activate neutrophils. Although SSRBC microparticles may contribute to cellular activation, we did not employ sufficient centrifugal force to pellet microparticles during RBC isolation, suggesting that microparticles are not responsible for our observations. Similarly, RBC hemolysis is unlikely to account for the differences we observed in neutrophil adhesion/degranulation, as measurements of free hemoglobin did not differ when samples were incubated with SSRBCs compared to healthy AARBCs (Supplemental Figure 2). Furthermore, lysate prepared from both AARBCs and SSRBCs did not enhance adhesion of normal neutrophils to HUVECs. Instead, as we have previously shown with monocytoid leukocytes,¹³ it is possible that known properties of SSRBCs, including activated adhesion receptors, membrane phosphatidylserine exposure, or altered sialic acid expression¹⁴, contribute to neutrophil activation. Ultimately, it is important to more thoroughly investigate and understand the mechanism by which SSRBCs activate neutrophils and to delineate the full extent of SSRBCs effects on neutrophil activation, as well as the contribution of specific SSRBC subpopulations.¹⁵ Our findings suggest that SSRBC-neutrophil interactions are major contributors to SCD, implying also that the impact of new therapeutic agents on the ability of SSRBCs to active neutrophils may be critical to therapeutic efficacy.