Functional Role of the Chemokine Binding Duffy Antigen Receptor Complex (DARC) in Human Inflammation In Vivo.

Background: The Duffy receptor is a promiscuous receptor for chemokines that binds selected members of both the C-X-C and C-C families with high affinity. Duffy antigen may influence plasma levels of proinflammatory cytokines by acting as a “chemokine sink”. Additionally, Duffy knockout mice experienced an exaggerated response to endotoxin in comparison to wild-type mice. The current trial was designed to elucidate the functional role of the Duffy blood group antigen in human inflammation. We hypothesized that “Duffy negative“ volunteers might show an increased inflammatory response to endotoxin (LPS) infusion in terms of cytokine response. The human endotoxemia model is a well established model of systemic inflammation, where a well defined, self-limited inflammatory stimulus permits the elucidation of key players involved in the inflammatory response. We therefore used this model to investigate the functional role of the Duffy Antigen Receptor Complex (DARC) in the inflammatory response after endotoxin challenge.

Methods: Thirty-two healthy male volunteers received an intravenous infusion of 2ng/kg endotoxin, 16 Caucasians (Duffy antigen positive on erythrocytes) and 16 subjects of African descent (“Duffy negatives”). Cytokines, chemokines, as well as their receptors were quantified by ELISA, RT-PCR and FACS.

Results: Plasma levels of TNF, IL-6, IL-8 and IL-8 mRNA in whole blood increased to a similar extent in both groups after LPS infusion. In contrast, peak MCP-1 levels at 3 hours were roughly 2-fold higher in Duffy positive subjects 16ng/mL as compared to Duffy negative subjects (7ng/mL p<0.001). Similarly, GRO-alpha levels were 2.5-fold higher in Duffy positive subjects at 2 hours after LPS infusion (210pg/mL vs 85pg/mL; p= 0.0001), whereas baseline values showed no difference between both groups. There was no difference in the LPS induced release of MIP-1 beta (which was used as a negative control). Baseline levels of MCP-1 and GRO-alpha in erythrocyte lysates were significantly higher in Duffy positive individuals, whereas IL-8 and MIP-1beta levels were comparable between both groups. Values of MCP-1 and GRO-α, both known to bind to the Duffy antigen, increased extensively in erythrocyte lysates of Duffy positive subjects over time (P=<0.001 vs. time). In contrast, MCP-1 only moderately changed over time in Duffy negative subjects (P<0.03 vs. time), whereas GRO-alpha remained comparable to baseline at all time points (P=0.2 vs. time). As a consequence, MCP-1 and GRO-alpha were several hundred-fold higher in Duffy positive subjects at all time points and reached peak levels at 2–4 hours. Additionally, IL-8, which also binds to erythrocytes via the Duffy antigen, was roughly 7-fold higher in erythrocyte lysates of Duffy positive subjects 2 hours after LPS infusion. As expected, MIP-1beta remained similar at all time points and was comparable between both groups. The expected increase in CD11b, and decrease in CCR-2 and CXCR-1, CXCR-2 occurred without differences between groups.

Conclusion: This study characterized for the first time the in vivo function of the Duffy Antigen in humans. The blood group antigen has a profound effect on erythrocyte-bound chemokines and free chemokines in a model of human endotoxemia.