Abstract
Red blood cell alloimmunization can be a life-threatening complication for patients with sickle cell disease (SCD) receiving therapeutic transfusions. However, it remains unknown why only some (20-60%) SCD patients develop alloantibodies whereas others do not. Because of ongoing hemolysis in SCD, we have investigated the effects of toxic hemin on T cell responses of patients with SCD on chronic transfusion therapy. We found impaired monocyte control of proinflammatory T cells in response to hemin in alloimmunized SCD patients, partly due to defective levels of monocyte heme-oxygenase-1 (HO-1), an anti-inflammatory heme degrading enzyme that catabolizes heme into iron, biliverdin and carbon monoxide (CO). Monocytes are precursors of dendritic cells (DCs), which represent the antigen presenting cells most able to initiate and regulate immune responses. However, little is known about the functional properties of DCs in SCD patients. We therefore hypothesized that DCs of alloimmunized SCD patients would also have impaired response to hemin. Monocyte-derived DCs (moDCs) from healthy donor controls (n=11) and a cohort of SCD patients (aged 15-30 on chronic transfusion therapy every 3-4 weeks for at least two years using C,E,K phenotyped-matched, leukodepleted units) grouped either as "non-alloimmunized" (no history of antibody production, n=6), versus "alloimmunized" (with a history of having produced at least one alloantibody, n=7) were matured in the absence or presence of hemin (5uM and 20uM) with TLR agonists: LPS/IFNg (TLR4 agonist+STAT1 activation) or R848 (TLR7/8 agonist). Hemin-exposed mature and immature (no TLR agonist) moDCs were then assayed for HO-1 expression, cytokine production, co-stimulation, and T cell priming. Interestingly, upon hemin exposure, immature moDCs from healthy donors and non-alloimmunized patients upregulated more HO-1 (1056±206 fold; mean fluorescent intensity (MFI): 12283±1818 at 20uM hemin) than alloimmunized patients (fold increase 494±49; MFI: 7422±959, p<0.03). In addition, there were no differences among healthy donors and either patient groups in the following mature moDC associated cytokine/surface marker expression after hemin exposure: IL-12p70, IL-6, and TNFa production and surface expression of activation (CD80 and CD86) and antigen presentation (HLA-DR) markers, suggesting that these mature moDC-associated molecules are unlikely to be associated with alloimmunization status. As with healthy donors, hemin inhibited (by 35%) the upregulation of DC-specific maturation marker CD83 expression on mature moDCs of non-alloimmunized patients (MFI before hemin 5199±689 after 20uM hemin 3471±401, p=0.003). Furthermore, HO-1 breakdown product CO, but not biliverdin, inhibited CD83 upregulation on mature moDCs to a similar degree as hemin, suggesting that hemin-mediated CD83 inhibition in mature MoDCs involves HO-1. Remarkably, however, hemin did not alter CD83 expression on mature moDC in alloimmunized patients (before hemin 4382±949 after hemin 4825±1051, p=0.5). To determine whether hemin modulation of CD83 expression affects mature moDC driven differentiation of naïve CD4+ T cells (ie. priming), we performed mixed lymphocyte reactions: whereas hemin inhibited (roughly 30%) moDC-mediated priming of pro-inflammatory IFNg secreting TH1 cells in healthy donors and non-alloimmunized patients (%CD4+ IFNg+ without hemin 58%±9% with 20uM hemin 41%±8%, p=0.005), TH1 cell responses were not suppressed in the alloimmunized group (without hemin: 52%±10% with 20uM hemin: 54%±10%, p=0.4). In summary, hemin-exposed moDCs from healthy donors and non-alloimmunized SCD patients suppress CD83 upregulation and inhibit ensuing TH1 cell responses, possibly through HO-1 induction and subsequent production of CO. In contrast, CD83 upregulation and proinflammatory T cell priming are not inhibited by hemin in alloimmunized patients, probably representing a mechanism by which heme-associated inflammation is maintained in alloimmunized SCD, thereby increasing their risk of alloimmunization. Deciphering this mechanism not only offers a potential biomarker for alloimmunization, but also opens the way to designing innovative treatments for these patients.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.