Transfusion-Induced Impairment of Macrophage Responses Is Partially Restored By the Iron Chelator Deferasirox

Iron overload is common in MDS, as a consequence of chronic red blood cell (RBC) transfusions and to a lesser extent, increased intestinal iron absorption to support erythropoiesis. Clinical evidence suggests that patients with transfusional iron overload show an enhanced susceptibility to infections, which represent one of the most common causes of morbidity and mortality in low risk MDS patients. Iron overload may increase the risk of infections by supporting bacterial growth and altering the immune response.

We asked whether transfusions impair macrophage effector functions. Therefore we investigated the impact of transfusions on the phenotypic plasticity of monocytes and macrophages and their functional responses to infectious cues. We applied a mouse model of repeated transfusions of allogeneic RBCs in wild-type and NUP98-HOXD13 MDS mice. Monocytes as well as hepatic and splenic macrophages were analyzed in transfused and non-transfused mice, subjected or not to LPS challenge.

Under steady-state condition, MDS mice displayed a mild but significant iron phenotype, hallmarked by elevated systemic and hepatic iron levels, due to inappropriately low hepcidin levels. Repeated transfusions increased serum iron levels and transferrin saturation and promoted iron deposition and oxidative damage in tissues.

Transfusions caused macrophage iron loading, monocyte recruitment to the liver and spleen and massive splenic macrophage cell death. Macrophages from transfused mice developed an M2-like anti-inflammatory phenotype, hallmarked by elevated expression of M2 markers (CD206, Arg-1, Ym1) and reduced expression of M1 markers (MHCII, CD86). Accordingly, circulating pro-inflammatory cytokines (IL-6, TNFα, INFγ, IL-1β) were reduced, and anti-inflammatory cytokines (IL-10) increased. Consistently, after LPS challenge macrophages show a markedly reduced expression of M1 markers and inflammatory cytokines, and increased expression of M2 markers in transfused compared to non-transfused mice. Similar results were obtained both in wild-type and MDS mice. All together these observations indicate that transfusion-induced erythrophagocytosis dampens the inflammatory response to infectious cues.

In an attempt to relieve macrophages from iron overload, transfused mice have been administered the iron chelator deferasirox during the transfusion period, up to 1 week after the last transfusion. Deferasirox significantly reduced iron levels in liver and spleen. Interestingly, chelation therapy partially restored the phenotype of macrophages by rescuing the expression of M1 markers, including MHCII and CD86, and M2 markers, such as CD206. Similar results have been obtained in the setting of transfusion with and without LPS stimulation.

Our results show that transfusions lead to iron-overload mediated toxicity in macrophages blunting their inflammatory response to infectious stimuli by affecting cell plasticity. Our data support a role for transfusions in triggering a phenotypic switching of macrophages towards an anti-inflammatory phenotype, with reduced ability to counteract infections. By restoring cell polarization, iron chelation likely improves the functional response of macrophages to invading microorganisms. Combined with marked cytopenias, the weak pro-inflammatory activation of macrophages might explain the increased propensity of transfused MDS patients to develop infections. Transfusion practice might therefore increase the risk of infections not solely through increased iron availability to microorganisms, but also through the impairment of macrophage functions. Our results suggest that transfused MDS patients might benefit from chelation therapy by both restricting iron from pathogens and improving the innate immune response.