Causing leukocyte activation and upregulation of adhesion molecules on endothelial cells. CD16+ monocytes, also known as endothelial patrolling monocytes, normally scavenge the damaged cells and debris from the vasculature. As compared to other monocyte subsets or immune cell types, the CD16+monocyte subset expresses higher levels of the anti-inflammatory heme oxygenase 1 (HO-1), a heme degrading enzyme. Given the role of CD16+ monocytes as scavengers of debris on endothelial cells, we tested the hypothesis that this subset may protect SCD vasculature from the ongoing hemolytic insult through expression of high levels of HO-1. We found roughly 35% of circulating CD16+ monocytes from SCD patients expressed very high levels of HO-1 as compared to 5% in healthy controls. The HO-1hi SCD monocytes expressed significantly (30%) less TNF-a compared to HO-1lo monocytes following stimulation, consistent with anti-inflammatory effects of HO-1. We hypothesized that uptake of free hemoglobin/heme was responsible for high HO-1 expression levels in SCD CD16+ monocytes. To test this, healthy donors (HDs) or SCD patient monocytes were treated with different doses of free heme or hemolysed RBCs. We found dose-dependent HO-1 induction (five-fold at 20mM heme) in purified CD16- monocytes, but surprisingly none in CD16+ subset. However, upon co-culture with human umbilical vein endothelial cells (HUVEC), continuous or prior exposure to heme induced HO-1hi expression exclusively in CD16+ monocytes (5 fold in HD and further two fold in SCD compared to non-heme treated cocultures, p<0.001). Using imagining flow cytometric analysis, we found marked increase in uptake of heme-exposed endothelial cell-derived material by CD16+ monocytes (HD: 2% to 13% ± 3%; in SCD: 20% ± 3% to 30% ± 4%, p< 0.001) but none by CD16- monocytes. Our transwell studies demonstrated that cell-cell contact between CD16+ monocytes and heme-exposed HUVEC was required for HO-1hi expression. We found roughly 4-fold increase in expression of phosphatidylserine (PS, annexin V+), ICAM-1 and vCAM-1 on heme-treated HUVEC cells. Antibody blocking studies identified PS moieties as well as ICAM-1 as key molecules involved in monocyte-HUVEC interactions that mediated HO-1hi induction, suggesting that high levels of HO-1 expression in SCD CD16+ monocytes is in part the result of attachment to and engulfment of apoptotic, activated endothelial cells damaged by heme. SCD patients suffer from vaso-occlusive crisis (VOC), resulting from increased attachment of SCD RBCs to damaged and activated endothelium. We hypothesized that inadequate numbers or lower HO-1hi levels in CD16+ monocyte will predispose SCD patients to episodes of VOC due to decreased removal by CD16+ monocytes of damaged endothelial and sickle RBCs. Amongst SCD patients receiving chronic transfusions, we found a two-fold lower frequency of circulating CD16+ monocytes and half the numbers of CD16+HO-1hi monocytes in patients with a recent history of VOC episode as compared to those without VOC (p< 0.01); the former group also expressed higher levels of circulating sVCAM-1 (997 ± 210 vs 765 ± 236 ng/m, p=0.02), a marker of endothelial activation. To formally test the role of patrolling monocytes in endothelial damage induced by SCD RBCs and heme, we injected RBCs from Townes SCD mice alone or after 24hrs with heme into Nr4a1-knockout mice which have a selective loss of patrolling monocytes. Immunofluorescence analysis of liver vasculature showed a 3 fold increase in the activated endothelial marker, ICAM-1 within 24hrs following injection of sickle RBCs and two-fold increase in circulating sVCAM-1 levels in mice treated with sickle RBC plus heme (p <0.001). Transfer of HO-1+ patrolling monocytes (LY6Clo), but not a classical monocyte subset (LY6C+) reversed activated endothelial phenotype, indicating that patrolling monocytes can inhibit SCD-induced endothelial activation. Altogether, these data suggest that SCD patrolling monocytes remove hemolysis-damaged endothelial cells, resulting in HO-1 upregulation and dampening of vascular inflammation. Perturbations in CD16+ monocyte numbers resulting in lower local HO-1 levels can predispose SCD patients to VOC, thus identifying HO-1+ patrolling monocytes as key players in VOC pathophysiology and as therapeutic targets.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution