Invasive non-typhoid salmonella is the leading cause of bacteremia in many parts of Sub-Saharan Africa and contributes significantly to the high regional malaria mortality rate. So far the common assumption has been that hemolysis-induced macrophage dysfunction is responsible for the increased systemic bacterial (salmonella) load following salmonella and malaria co-infection, however no direct evidence has documented macrophages acting as the primary refuge of NTS in the context of hemolysis.
(SLIDE)
The hemolytic effects of Malarial infection results in the liberation of heme into the bloodstream, inducing neutrophil migration and activating the neutrophil oxidative burst by catalyzing the production of ROS.|||| Concurrently, free heme leads to the expression of HO-1 (an enzyme responsible for degrading heme to biliverdin, CO and Fe), limiting the formation of ROS and sparing the body of their cytotoxic effects.
(SLIDE)
Thus the production of HO-1 presents an interesting dichotomy, while ROS are critical in resistance to certain pathogens (such as Salmonella) their inherent cytotoxicity means they are targeted by cytoprotective enzymes. This raises the possibility that tolerance to one pathogen may sometimes come at the price of loss of resistance to another.
(SLIDE)
We hypothesized that liberation of heme by intravascular hemolysis may lead to HO-1 induction and impairment of resistance to NTS with increased bacterial replication
(SLIDE)
(1) In order to determine whether heme liberated by hemolysis impairs resistance to NTS infection we compared survival and bacterial loads of Salmonella typhimurium infected mice with or without preceding Py17XNL co-infection, PHZ or hemin treatment
(SLIDE)
(1) Py17XNL resulted in a self-resolving infection which was accompanied by a chronic/progressive hemolytic anemia, in contrast; PHZ and hemin treatments caused acute hemolysis. In all 3 cases of preceding infection an increased mortality rate was accompanied by higher bacterial loads in whole blood, spleen, liver and bone marrow (FIGURE 1d+e). It is important to note that in the absence of Salmonella infection, Py17XNL, PHZ and hemin did not cause any mortality indicating that hemolysis and heme impair resistance to NTS bacteremia
(SLIDE)
(2+3) Following Py17XNL infection, PHZ or hemin treatment; the proportion of Salmonella localized in granulocytes was substantially increased when compared to control mice. As there was no obvious defect in the uptake of Salmonella by macrophages and monocytes, accumulation of Salmonella in blood granulocytes likely resulted from impaired bacterial killing and/or a more permissive intracellular environment for bacterial replication. To investigate these possibilities we compared the phagocytic ability of CD11b+ cells from Py17XNL-infected mice and un-infected mice.
(SLIDE)
(2+3)(Figure 3 a+b, Supp 3 image). As assumed, the rate of phagocytosis between infected and un-infected mice was relatively unchanged indicating no defect in the uptake of Salmonella. The presence of intracellular GFP+ bacteria was confirmed by confocal microscopy. Following lysis, the live bacterial recovery from Py17XNL infected mice was significantly higher from control mice indicating considerable impairment of intracellular killing of Salmonella.
(SLIDE)
(2+3) As HO-1 reduces the production of ROS, and since NADPH oxidase is essential for resistance to Salmonella early in infection we investigated whether Py17XNL infection impairs the granulocyte oxidative burst. Using the oxidation of dihydrorhodamine to its fluorescent derivative rhodamine as an indicator of oxidative burst, we observed progressive suppression of oxidative burst capacity within blood granulocytes of Py17XNL infected mice (Figure 3c).In contrast, blood granulocytes isolated from PHZ or hemin treatment did not differ from control mice in their oxidative burst capabilities. Considering PHZ-mediated hemolysis and hemin treatments elicit a rapid increase in free heme, these results suggested that it was instead the chronic/progressive nature of malarial anemia that was responsible for the observed decrease in oxidative burst capacity.
(SLIDE)
We wondered whether the chronic hemolysis associated with malaria induced, HO-1 expression in immature bone marrow myeloid cells, suppressing their oxidative burst capacity as they matured and introducing a gradual accumulation of dysfunctional cells into the circulation. We confirmed the effect of free heme in mobilizing granulocytes from bone marrow into the periphery as treatment with hemin, PHZ and Py17XNL caused marked depletion of mature cells from bone marrow which was accompanied by an increase in granulocytes in peripheral blood. Salmonella infection caused granulocyte mobilization in control mice and, when paired with Py17XNL, PHZ or hemin treatment markedly exacerbated granulocyte mobilization therefore accounting for the presence of immature granulocytes within the bloodstream of co-infected mice. This data indicates that intravascular heme mobilizes granulocytes from bone marrow and simultaneously impairs development of their oxidative burst and subsequent functional maturation. As a result, granulocytes entering the circulation in response to infection are able to phagocytose Salmonella but are unable to kill them and instead (due to the increased iron concentration as a result of heme breakdown) provide a niche for bacterial replication.
(SLIDE)
Since heme-oxygenase is responsible for the increased susceptibility to Salmonella, logic would dictate that inhibition of heme oxygenase would then restore resistance to Salmonella. Py17XNL-Salmonella co-infected mice were pretreated with the competitive heme-oxygnease inhibitor tin protoporphyrin 9(SnPP). Following treatment, bacterial loads in blood, spleen and liver were not significantly different from control mice (FIGURE 5a NIGGA) and SnPP administered in control mice had no effect on bacterial load. Meaning that inhibition of HO-1 alone was responsible for the observed decrease in bacterial load. Interestingly cobalt protoporphoyrin (CoPP), which induces HO-1 in the absence of hemolysis or free heme did not impair resistance to NTS. Therefore both heme and heme-oxygenase are necessary for impaired Salmonella resistance. It is important to note that inhibition of HO by SnPP did not restore oxidative burst capabilities of already mobilized granulocytes but instead reversed the gradual accumulation of impaired and immature granulocytes. Indicating that inhibition of heme oxygenase restores normal development and maturing of bone marrow granulocytes.

(SLIDE)
Our results have shown that loss of resistance to Salmonella requires hemolytic release of cell-free heme and subsequent induction of HO-1 and that inhibition of HO-1 reverses this susceptibility. So although HO-1 is essential for tolerance to the cytotoxic effects of free heme resulting from malarial infection it simultaneously impairs resistance to Salmonella.
This diagram portrays our proposed mechanism. In (a) we see Salmonella infection alone causes emergency granulopoiesis and mobilization of granulocytes from bone marrow to the blood stream. These granulocytes phagocytose the salmonella bacterium and generate an oxidative burst to terminate the pathogen. (b) shows the series of events following malaria or PHZ treatment. Liberation of heme induces HO-1 expression in immature myeloid cells which degrades heme into Biliverdin, CO and Fe. In the case of PHZ treatment where there is acute hemolysis the oxidative burst in granulocytes is activated normally. While in the case of malaria when there is chronic hemolysis, functionally immature granulocytes have enough time to accumulate in the blood, ultimately resulting in reduced oxidative burst capacity. (c) is the combination of malaria and Salmonella. The accumulation of low oxidative burst capacity granulocytes as a result of malarial infection allows Salmonella to survive within the granulocyte following phagocytosis. The increased iron concentration resulting from the breakdown of heme promotes bacterial reproduction.
(SLIDE)
In conclusion these findings offer a plausible explanation for the particular susceptibility to Non typhoid salmonella bacteremia in individuals with hemolysis. Since salmonella has evolved to survive and replicate within mononuclear phagocytes and chronic hemolysis provides a non-threatening environment for bacterial replication. We have identified a possible adjunct therapy that might enhance resistance to NTS in patients with hemolytic disease. However care must be taken in order to avoid impairing tolerance to heme and exacerbating its toxic effects.
On a grand scale we have demonstrated that tolerance and resistance mechanisms identified from studies of single pathogens may not easily translate to the real world where people are simultaneously exposed to multiple pathogens.