Background: Transfusions of red blood cells (RBCT) are necessary and life-saving in premature and critically ill infants, who experience severe anemia due to physiologic and iatrogenic factors. Recently, we showed that severe anemia was associated with increased intestinal permeability, demonstrated by identifying endotoxin in the bloodstream. The risks of experiencing severe anemia during critical developmental periods must be balanced with the risks of transfusions, which can lead to the Systemic Inflammatory Response Syndrome (SIRS). In this study, we investigated the role of severe anemia in endotoxin sensitization of the liver, and of RBCT-triggered cytokine storm, in murine neonates.

Methods: C57BL/6 (WT) and TLR4-KO mouse pups were studied in 4 groups (n=6 each): (1) naïve controls; (2) non-anemic RBC transfused; (3) anemic (hematocrit 20-24%); and (4) anemic RBC transfused. Pups in anemic groups underwent facial vein phlebotomy on days P2, 4, 6, 8, and 10. Leukoreduced and 7-day refrigerator-stored packed RBCs from allogeneic (FVB) adult donors were transfused intravenously into the retro-orbital plexus (20 mL/kg) of P10 pups. Blood was collected 2, 4, and 6 hours post-transfusion, and plasma was separated for quantifying inflammatory cytokines (IL-1β, TNF-α, CXCL-1, IL-6, IFN-γ) using a Luminex assay. Plasma non-transferrin bound iron (NTBI) levels were measured at 2 hours post-transfusion. To measure leaky gut associated liver endotoxin sensitization, genetically-modified E. coli, expressing green fluorescent protein (GFP; 108 colony-forming units), were administered by gavage; these bacteria were measured in the liver of anemic pups by immunofluorescence using anti-GFP antibody. To investigate the endotoxin-associated monocyte-macrophage system in the liver, single cell suspensions were prepared and flow cytometry was performed by staining with antibodies against CD45, CD11b, F4/80, CD11c, Ly6G and Ly6C.

Results: RBC transfusions led to the acute release of inflammatory mediators (IL-1β, TNF-α, CXCL-1, IL-6, IFN-γ) within 2 hours, in pups with a normal hematocrit ("transfusion controls"), as well as in pups rendered anemic. However, inflammatory cytokine levels were significantly higher in RBC-transfused anemic pups as compared to transfusion controls, and remained elevated for several hours. We excluded the inflammatory response at 6 hours post-transfusion, because intestinal inflammation was observed after that time by the release of intestinal fatty acid binding protein (iFABP). No changes were found in plasma NTBI levels between transfusion control and anemic-transfused pups, indicating that the acute inflammatory response in the anemic-transfused groups is independent of liver iron accumulation due to stored RBC clearance. However, severely anemic mouse pups had elevated amounts of orally administered E. coli-GFP in the liver relative to control mice, indicating that anemia allows endotoxin delivery and sensitizes the liver by bacterial translocation across the anemic intestine. The TLR4-KO mice study showed that overall plasma levels of inflammatory cytokines (TNF-α, IL-6, IFN-γ) were reduced, as compared to anemic-transfused WT pups, but remained elevated in the plasma of anemic-transfused TLR4-KO pups, as compared to TLR4-KO transfused-control pups, indicating that this acute inflammatory response does not depend on TLR4. By flow cytometry, anemic (P10) pups showed an increased number of CD11bhiF4/80loLy6C+ hepatic monocytes as compared to control, but no changes in CD11bhiF4/80hi hepatic macrophages (i.e., Kupffer cells); qRT-PCR studies confirm that these anemic hepatic monocytes contained mRNA transcripts of IL-1β, TNF-α, CXCL-1, IL-6, and IFN-γ. In addition, the anemic mouse pups that were treated with recombinant haptoglobin (rHp, 100µg/kg) showed reduced plasma levels of IL-1β, TNF-α, CXCL-1, IL-6, and IFN-γ at 2-6 hours post-transfusion of packed-stored RBCs.

Conclusions: In this mouse model, the severe anemia-associated low-grade inflammatory state in the liver leads to increased numbers of hepatic monocytes with inflammatory cytokines mRNA; in this condition, RBC transfusions activate these monocytes and cause systemic inflammation. Pre-treatment of rHp was reduce the RBCT-associated systemic inflammatory response.

No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution