Defining Mechanistic Responses to Different Vaso-Occlusive Triggers in Sickle Cell Disease: A Step Towards Personalized VOC Reversal Strategies

Introduction: Vaso-occlusive crises (VOC) are the most common complication of sickle cell disease (SCD) and can be elicited by different physiological events, including infections, dehydration, cold temperature and psychological stress. Currently, there exists no specific treatment to resolve VOC, or limit its intensity, other than pain management and rehydration. Agents targeting platelet (plt) aggregation, hemoglobin polymerization and selectin molecule activity can prevent vaso-occlusion (VO) in standardized models of VOC in SCD mice. However, all have, so far, failed to bring relief to patients in real-life conditions, where VOC triggers are polymorphic. Hypothesizing that VOCs may differ in their molecular mechanisms, depending on their triggers, we compared the effects of physiological triggering events and inflammatory signaling molecules on in vivo and in vitro models of SCD VO.

Methods: Townes HbSS mice (20 wks) were submitted to molecular triggers, consisting of tumor necrosis factor (TNF)α, hemin, lipopolysaccharide (LPS), or interferon (IFN)α. Other mice were submitted to hypoxia/reoxygenation, cold temperature or dehydration. Cutaneous VO was then evaluated by laser Doppler perfusion monitoring (LDPM), cremaster microvascular leukocyte recruitment was observed by intravital microscopy, and neutrophil-plt aggregate (NPA) formation in peripheral blood was quantified by flow cytometry. For in vitro protocols, peripheral blood from sickle cell anemia (SCA) patients was incubated with TNFα, LPS, IFNα or RBC lysis products. We then investigated blood cell adhesion to endothelial cells using microfluidic adhesion protocols, the architecture and composition of the VO aggregate using confocal imaging, and the phenotype of neutrophils and their aggregation with RBC/ plts using flow cytometry.

Results: In the SCD mouse model, LDPM indicated that all of the molecular and physiological triggers promoted VO, significantly decreasing both blood flow velocity and perfusion in the cutaneous microcirculation. The physiological triggers all augmented microvascular leukocyte rolling and moderately increased adhesion, while hypoxia/ reperfusion and dehydration, but not cold, induced circulating NPA formation. Notably, the molecular triggers differed more distinctly in their mechanistic responses. In SCD mice, the bacterial component, LPS, caused extensive microvascular leukocyte adhesion without NPA formation, while TNFα cytokine amplified leukocyte rolling, adhesion and caused significant NPA formation. Consistently, stimulation of blood from SCA patients with LPS led to global activation of neutrophils, with increased expressions of the MAC-1 integrin (CD18/CD11b), CD66b, CD64 and CD33, and RBC recruitment at the heart of the VO aggregate, indicating enhanced RBC adhesion to other blood cells, rather than to endothelial cells. TNFα stimulation of SCA patient blood, however, caused a specific increase in neutrophil MAC-1 expression, but not of other adhesion molecules, and an overrepresentation of the plts inside the VO cluster. In contrast, the major effect of hemin in SCD mice was an acceleration of leukocyte rolling, but without significant leukocyte adhesion. Once again, this was consistent with the lack of overexpression of adhesion or activation molecules on neutrophils after the addition of RBC lysis content to SCA whole blood. Finally, in SCD mice, the viral response cytokine, IFNα, induced leukocyte rolling, in association with moderate adhesion, but with a very significant increase in NPA formation. Significantly, IFNα also drastically increased the adhesion of plts from SCA blood to endothelial cell layers during microfluidic flow assays, together with increased RBC adhesion both to endothelial cells and to other cell types, but in the absence neutrophil phenotype alterations.

Conclusions: Our study is the first to demonstrate that VO is not a uniform phenomenon, and probably depends on the specific trigger of the crisis. Molecular triggers amplified inflammatory responses more than physiological events, revealing specific VOC mechanistics. For instance, phenotypic leukocyte changes were crucial to bacterial component-induced VO, while plt activity was central to the IFNα response. Findings represent an important step towards defining targeted strategies for personalized VOC reversal in patients, according to the triggering event.