Hematologists often battle life-threatening infections in patients with hemolytic anemia. For example, in children with sickle cell disease, encapsulated bacteria such as pneumococcus are a constant threat to the host, due in part to hyposplenism. An unusual predilection for salmonella osteomyelitis has been cited in numerous texts on sickle cell disease, purportedly due to gastrointestinal or gallbladder leak of these bacteria into the blood with subsequent seeding of skeletal infarcts. In sub-Saharan Africa, invasive nontyphoid Salmonella (NTS) infection is a common and often fatal complication of Plasmodium falciparum infection. Malaria infection lyses RBCs, releasing hemoglobin, which in turn bathes the vasculature with cytotoxic, pro-oxidative heme. Heme oxygenase-1 (HO-1) detoxifies heme by metabolizing it to biliverdin and bilirubin, releasing carbon monoxide and iron, with the latter constituent being safely sequestered in ferritin. Heme induction of HO-1 affords protection against non-cerebral forms of severe malaria, in part through generation of carbon monoxide.1, 2, 3 However, HO-1 induction may be a two-edged sword with respect to tolerance of malaria, as Cunnington et al. from the London School of Hygiene and Tropical Medicine now show that induction of HO-1 during malarial hemolysis impairs resistance to NTS by limiting production of bactericidal reactive oxygen species.
The investigators demonstrate that co-infection of mice with Plasmodium yoelii 17XNL (Py17XNL) and Salmonella typhimurium causes acute, fatal bacteremia with a high bacterial load. Similarly, if hemolysis is induced by phenylhydrazine (PHZ) or mimicked by hemin administration, none of the mice survived 16 hours after Salmonella infection, with all such treated mice having high bacterial loads compared with controls (saline-treated mice). S. typhimurium localizes in granulocytes following hemolysis or hemin infusion, but the bacteria are not killed. Infection with Py17XNL inhibits the granulocyte oxidative burst, and hemolysis induces dysfunctional granulocyte mobilization. Treatment with hemin or PHZ followed by infection with Py17XNL caused marked depletion of Gr-1hi (a marker of mature granulocytes) cells from bone marrow. For PHZ and hemin treatment, loss of Gr-1hi cells from bone marrow was accompanied by an increase in granulocytes in peripheral blood, confirming the effect of free heme in mobilization of neutrophils from the bone marrow into the periphery. Although the proportion of circulating granulocytes did not increase during Py17XNL infection, granulocyte mobilization may have been obscured by an overall increase in leukocyte count or by granulocyte redistribution. HO-1 is induced in immature bone marrow myeloid cells; however, PHZ or hemin treatment accompanied by Py17XNL infection led to induction of HO-1 in peripheral blood monocytes, and HO inhibition using tinprotoporphyrin restored resistance to S. typhimurium.
In Brief
Together, these data indicate that intravascular heme (released during hemolysis) mobilizes granulocytes from bone marrow and simultaneously impairs their capacity to generate an oxidative burst. Thus, in the setting of concurrent intravascular, granulocytes entering the circulation in response to infection are able to phagocytose S. typhimurium but, owing to their reduced oxidative burst capacity, fail to kill them, providing instead a safe house for bacterial replication and dissemination.
This elegant study provides another example of why systems that regulate heme clearance and toxicity are numerous and redundant. In patients with malarial infection, HO-1 provides tolerance and cytoprotection but may also exacerbate bacterial infections. The potential role of heme-derived iron in promoting bacterial infections and blockade of Toll-like receptor-4 also merits further investigation.4
References
Competing Interests
Dr. Vercellotti indicated no relevant conflicts of interest.
