Myeloperoxidase Attenuates Pathogen Clearance during Plasmodium yoelii Nonlethal Infection

ABSTRACT

Myeloperoxidase (MPO), a leukocyte-derived enzyme mainly secreted by activated neutrophils, is known to be involved in the immune response during bacterial and fungal infection and inflammatory diseases. Nevertheless, the role of MPO in a parasitic disease like malaria is unknown. We hypothesized that MPO contributes to parasite clearance. To address this hypothesis, we used Plasmodium yoelii nonlethal infection in wild-type and MPO-deficient mice as a murine malaria model. We detected high MPO plasma levels in wild-type mice with Plasmodium yoelii infection. Unexpectedly, infected MPO-deficient mice did not show increased parasite loads but were able to clear the infection more rapidly than wild-type mice. Additionally, the presence of neutrophils at the onset of infection seemed not to be essential for the control of the parasitemia. The effect of decreased parasite levels in MPO-deficient mice was absent from animals lacking mature T and B cells, indicating that this effect is most likely dependent on adaptive immune response mechanisms. Indeed, we observed increased gamma interferon and tumor necrosis factor alpha production by T cells in infected MPO-deficient mice. Together, these results suggest that MPO modulates the adaptive immune response during malaria infection, leading to an attenuated parasite clearance.

INTRODUCTION

Plasmodium falciparum infection is still one of the major tropical infectious diseases. The control of the symptomatic blood-stage infection by the host requires the innate and the adaptive immune systems (1, 2). Using mouse models, it was shown that macrophages (3), CD4⁺ T cells (4–6), and B cells (7, 8) are essential for effective parasite clearance. However, an effective immune activation and regulation is necessary for rapid parasite control and for the prevention of host pathology (5, 9). Myeloperoxidase (MPO), a heme protein secreted mainly by neutrophils, could be an important mediator in both antiparasitic defense and host pathology. Being 5% of the total protein, MPO is the most abundant protein in neutrophils (10). During infections and inflammatory diseases, MPO is increasingly released from azurophilic granules of activated neutrophils into the phagocytic vacuoles and extracellular space (11). MPO has also been found in monocytes, where it represents about 1% of the total protein (12). Monocytes later lose their ability to secret MPO during their differentiation toward macrophages. Macrophages, however, are able to take up MPO by neutrophils or MPO alone (13). The most prominent feature of MPO is its ability to catalyze the oxidation of halogenides and various substrates together with its cofactor, hydrogen peroxide (14, 15). Besides this ability to generate reactive oxygen species (ROS), MPO is involved in neutrophil extracellular trap (NET) formation (16). NETs are extracellular structures of chromatin and antimicrobial proteins that are released from activated neutrophils and can bind to the pathogen (17). These antimicrobial features enable MPO to participate directly in the defense of fungal (16, 18) and various bacterial infections (17, 19, 20). In addition, MPO further activates neutrophils (21) and promotes their recruitment (22), leading to an enhanced proinflammatory immune response. Whether MPO also promotes parasite clearance in malaria infection is unknown. So far, it has been shown that neutrophils are able to phagocytize merozoites (23) and parasitized red blood cells (pRBC) (24). Additionally, leukocyte-derived ROS (25) may contribute to parasite killing. Moreover, patients with P. falciparum malaria have increased MPO plasma levels (26–28), which are higher than levels observed in patients with cardiovascular disease or sepsis (28, 29), for which MPO has been intensively studied in recent years. Based on these findings, we hypothesized that increased MPO production would lead to increased ROS levels, which would contribute to accelerated Plasmodium parasite clearance. In this study, we used infection with the nonlethal Plasmodium yoelii strain 17NL as a murine model for malaria infection. With this model, we analyzed the role of MPO in immune activation and parasite clearance using C57BL/6J wild-type (WT) and MPO-deficient (MPO^−/−) mice.

RESULTS

MPO plasma levels are highly elevated in P. yoelii 17NL-infected wild-type mice.

MPO plasma levels increased significantly on days 6 and 12 postinfection (p.i.) compared to levels for naive mice (Fig. 1). The peak MPO plasma level was reached on day 6 p.i., followed by a constant decrease until the end of infection.

FIG 1.

MPO plasma levels of WT mice are increased during the course of P. yoelii 17NL (PyNL) infection. Mice were infected i.p. with P. yoelii 17NL. On selected days, mice were sacrificed and the plasma concentration of MPO was determined (n_naive = 6; n_{day 6 p.i.} = 5; n_{day 12 p.i. =} 5; n_{day 18 p.i. =} 5). Data are representative of 2 to 3 independent experiments and are shown as means ± standard errors of the means (SEM). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

MPO deficiency leads to more rapid parasite clearance.

To analyze the effect of MPO on parasite clearance, we compared the parasitemia between infected WT and MPO^−/− mice. Infected MPO^−/− mice showed parasite levels similar to those of WT mice at the beginning of the infection but were able to clear the infection significantly earlier (Fig. 2A). The peripheral neutrophil counts increased similarly in WT and MPO^−/− mice at the beginning of the infection (see Fig. S2 in the supplemental material). In addition to the parasitemia, we measured spleen enlargement, normalized by tibia size, as an indicator for parasite clearance. The spleen/tibia ratio of both MPO^−/− and WT mice increased during P. yoelii 17NL infection, with no significant differences in their organ ratios during early infection (Fig. 2B). However, the spleen/tibia ratio of MPO^−/− mice decreased significantly on day 18 p.i., when parasite loads were markedly lower than those of WT mice.

FIG 2.

MPO^−/− mice clear parasite infection earlier than WT mice. Mice were infected i.p. with P. yoelii 17NL and monitored for parasitemia (A) and spleen/tibia ratio (B). (A) Each group consisted of 7 mice and was monitored for 21 days. (B) Graph shows combined results of individual experiments per time point. The following numbers of mice per group were used: n_naive = 4; n_{day 6 p.i.} = 5; n_{day 12 p.i.} = 7; n_{day 18 p.i.} = 6. Data are representative of 3 independent experiments and are shown as means ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

We next analyzed whether a decrease in parasitemia was also seen if the neutrophils, the main producers of MPO, were depleted with an anti-Ly6-G antibody before infection (Fig. 3A and B). This depletion was transient, and neutrophils repopulated after 5 to 8 days (data not shown). A significantly decreased parasitemia was not observed in the group of mice receiving the depletory antibody compared with the group of mice receiving a control antibody (Fig. 3C). Furthermore, if mice received an additional shot of antibody at day 3 p.i., the parasitemia of the mice which experienced antibody-mediated neutrophil depletion was elevated on day 9 p.i. (Fig. 3D). However, this might be influenced by the fact that MPO plasma levels did not significantly decrease after anti-Ly6-G-mediated neutrophil depletion. Mice that received anti-Ly6-G antibody before infection showed MPO plasma levels similar to those of control mice on day 5 p.i. (Fig. 4B). On the same day, assessed parasitemia did not differ between the groups (Fig. 4A), although mice that received anti-Ly6-G antibody showed a slightly diminished parasitemia.

FIG 3.

Influence of antibody-mediated depletion of neutrophils on parasitemia in WT mice. Representative dot plot (A) and frequency (B) of GR-1⁺ Ly-6C^intermediate CD11b⁺ peripheral leukocytes show that neutrophils are depleted on the day of infection. (C) Mice received i.p. anti-Ly-6G antibody or control IgG 48 and 24 h before infection with P. yoelii 17NL. Parasitemia was assessed during the course of infection (n_{control IgG} = 7; n_anti-Ly-6G = 6). (D) Mice received an additional shot of antibody at day 3 p.i. Parasitemia was assessed during the course of infection (n_{control IgG} = 6; n_anti-Ly-6G = 6). Data are representative of 2 independent experiments and are shown as means ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

FIG 4.

Antibody-mediated depletion of neutrophils in WT mice does not lead to decreased plasma levels of MPO. Mice received i.p. anti-Ly-6G antibody or control IgG 72 and 48 h before infection with P. yoelii 17NL (n_{control IgG} = 8; n_anti-Ly-6G = 8). On day 5 p.i., parasitemia was assessed (A), mice were sacrificed, and the plasma concentration of MPO (B) was determined. Data are representative of 2 independent experiments and are shown as means ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

MPO deficiency does not decrease parasitemia in the absence of the adaptive immune system.

To investigate if there is an influence of MPO on parasite clearance when only the innate immune system is present, we infected RAG1^−/− × MPO^−/− mice and compared them to infected RAG1^−/− mice. Because we did not observe a difference in parasite load between the groups (Fig. 5), we hypothesized that MPO has an influence on the adaptive immune system. Based on this assumption, the T cell response of the MPO^−/− mice was assessed in the next step.

FIG 5.

MPO deficiency does not increase pathogen clearance in the absence of the adaptive immune system. RAG^−/− × MPO^−/− mice (n = 5) and RAG^−/− mice (n = 6) were infected i.p. with P. yoelii 17NL and monitored for parasitemia. Data are representative of 2 independent experiments and are shown as means ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

MPO deficiency leads to enhanced proinflammatory cytokine production of T cells during P. yoelii 17NL infection.

To investigate the influence of MPO on the proinflammatory T cell response, we performed a flow cytometry analysis of the T cells. An expanded population of CD4⁺ and CD8⁺ T cells was detected on day 18 p.i. in the MPO^−/− mice (Fig. 6A). We further analyzed the frequencies of gamma interferon (IFN-γ)- and tumor necrosis factor alpha (TNF-α)-producing CD4⁺ and CD8⁺ T cells during the course of infection. A representative dot plot of the intracellular cytokine staining of CD4⁺ and CD8⁺ T cells is shown (Fig. 6B and C, respectively). Gating strategy, the corresponding parasitemia, and spleen/tibia ratios of the analyzed mice used in this experiment are displayed in Fig. S3 in the supplemental material. Analysis of the intracellular cytokine staining showed increased frequencies of IFN-γ- and TNF-α-producing CD4⁺ and CD8⁺ T cells in infected MPO^−/− mice (Fig. 6D and E, respectively). T cells from naive MPO^−/− mice either did not differ in their frequencies of cytokine-producing T cells from naive WT mice or showed decreased frequencies.

FIG 6.

Increased expansion of splenic IFN-γ- and TNF-α-producing T cells in MPO^−/− mice during P. yoelii 17NL infection. On selected days splenocytes were isolated and stimulated with phorbol myristate acetate and ionomycin. The cells were stained and subsequently analyzed using flow cytometry. Parasitemia of analyzed WT and MPO^−/− mice and the gating strategy are shown in Fig. S3 in the supplemental material. (A) Displayed are the absolute counts and percentages of CD4⁺ and CD8⁺ T cells. Representative dot plots show IFN-γ and TNF-α staining of CD4⁺ (B) and CD8⁺ (C) T cells on day 6 p.i. Frequencies of IFN-γ and TNF-α production in the CD4⁺ (D) and CD8⁺ (E) T cell compartment are shown. The graph shows combined results of individual experiments per time point. The following numbers of mice per group were used: n_naive = 4; n_{day 6 p.i.} = 5; n_{day 12 p.i.} = 7; n_{day 18 p.i.} = 6. Data are representative of 2 to 3 independent experiments and are shown as means ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

DISCUSSION

Parasite clearance of P. yoelii 17NL infection occurred more rapidly in MPO-deficient mice than in WT control animals. To our knowledge, this has not been described before. This finding was surprising, because we anticipated from existing evidence in bacterial (20) and fungal (16) infections that high MPO levels would contribute to parasite elimination. As mentioned before, high MPO plasma levels had been shown in patients with P. falciparum infection (26–28). Correspondingly, in the P. yoelii 17NL model applied here, MPO plasma levels were also significantly increased during infection. The high plasma levels of MPO, which result mainly from MPO secretion by neutrophils (10), may play an important role in immune regulation. In human malaria, increased neutrophil counts (30–32) as well as activation (27) during the acute phase of infection have been observed and, arguably, contribute to increased MPO plasma levels. Neutrophils are the first line of defense of the innate immune system against microbes (33). In malaria, neutrophil-mediated phagocytosis of parasites (23, 24), leukocyte-derived ROS, and nitric oxide have been postulated to be involved in parasite killing (25). In addition, an adhesion of NET structures to pRBC and RBC of children with P. falciparum infection was observed (34) and could also play a role in host defense. However, our findings for whether neutrophils are essential for the host defense against P. yoelii 17NL infection are inconclusive. Although the peripheral neutrophil counts rapidly increased after infection, an antibody-mediated depletion of neutrophils in WT mice before infection did not lead to increased parasite levels. This could be explained by the fact that the neutrophil depletion was only transient. However, the parasitemia did increase on day 9 p.i. if neutrophil depletion was repeated 3 days p.i., which suggests a role of neutrophils in host defense after all. On the other hand, MPO^−/− mice did not show an increase in parasitemia but were able to clear the infection even more rapidly. Interestingly, and in contrast to our initial assumption that the antibody-mediated depletion would lead to decreased MPO plasma levels, the depletion did not have a significantly decreasing effect on the MPO levels detected in plasma. In fact, the underlying mechanism of an anti-Ly6-G-mediated depletion is still not clear. Neutrophil death via complement-dependent and -independent mechanisms has been proposed (35), but it is not known how these mechanisms would affect MPO release by the apoptotic neutrophils. Odobasic et al. recently reported a correlation between neutrophil apoptosis and MPO release of aged bone marrow neutrophils in vitro (36). Correspondingly, an anti-Ly6-G-mediated depletion could lead to increased MPO release in vivo.

We speculate that the presence of neutrophils at the beginning of the infection is not required to control the infection. In any case, the absence of MPO throughout the course of infection is associated with a more rapid parasite clearance. In contrast to individuals with chronic granulomatous disease (37, 38), individuals with MPO deficiency do not develop an increased susceptibility to infections except for an enhanced incidence of candida infection (39–41). Likewise, murine studies reported impaired fungicidal activity in MPO^−/− mice (42, 43). We, like others, observed that MPO^−/− mice developed normally (42, 43) and without signs of spontaneous infection under specific-pathogen-free conditions (44). It has been discussed that an increased production of superoxide anion or peroxynitrite compensates for the MPO deficiency (45). Therefore, these molecules might be more effective in parasite killing, which may explain the observed rapid parasite clearance in MPO^−/− mice. However, it was shown that in vitro treatment of pRBC with peroxynitrite did not influence parasite viability (46). Furthermore, a deficiency of the enzyme NADPH oxidase, which is involved in superoxide anion production, does not have an effect on the parasitemia of Plasmodium-infected mice (47). The fact that the RAG1^−/− × MPO^−/− mice did not show enhanced parasite clearance compared with RAG^−/− mice suggests that MPO deficiency does not modulate the innate immune response against P. yoelii 17NL infections, as RAG deficiency goes along with the lack of the adaptive immune response (48). Thus, we hypothesized that the parasite-decreasing effect seen in MPO^−/− mice depends on the adaptive immune system. We therefore investigated the influence of MPO on the adaptive immune response in greater detail, observing the proinflammatory cytokine production in T cells.

In malaria infection, CD4⁺ T cells are necessary to clear the infection effectively (4–6), whereas the primary wave of malaria infection is controlled by macrophages (3). From day 5 on, CD4⁺ T cells are the main producer of IFN-γ in P. yoelii 17NL infection. IFN-γ is needed to efficiently clear parasites during the course of infection (49). We observed that MPO^−/− mice showed increased frequencies of IFN-γ- and TNF-α-producing CD4⁺ and CD8⁺ splenic T cells. We speculate that the increase in IFN-γ and TNF-α production contributes to the enhanced parasite clearance in MPO^−/− mice. This raises the question of why the proinflammatory cytokine production is enhanced in MPO^−/− mice. Indeed, increased proinflammatory cytokine levels have been reported from MPO^−/− mice with experimentally induced glomerulonephritis (50) or in lung injury after bone marrow transplantation (51). Evidence indicates that ROS generated by the myeloperoxidase-hydrogen peroxide system could lead to a suppression of T cells. Suppression (52–54) or induced apoptosis (55) of T cells by an oxidative environment have been described recently. Additionally, studies showed that MPO, or its products, could modulate the initiation of the T cell response via suppression of dendritic cells (36, 56). Taken together, these facts support our hypothesis that MPO suppresses T cells in malaria infection, leading to reduced proinflammatory cytokine production. We further hypothesized that an increased proinflammatory cytokine production is responsible for the more rapid parasite clearance in MPO^−/− mice.

In summary, the absence of MPO leads to accelerated parasite clearance in P. yoelii 17NL malaria, possibly mediated by enhanced IFN-γ and TNF-α production by T cells, which could arise due to a lack of an inhibitory effect of MPO-derived ROS. Taking into account that MPO production is also strongly induced in P. falciparum infection, MPO could influence the outcome of this disease by modulation of the immune response.

MATERIALS AND METHODS

Mice and parasites.

WT mice were purchased from Charles River Laboratories. MPO^−/− mice, which were originally obtained from Jackson Laboratory, were bred in the animal facility of the University Medical Center Hamburg-Eppendorf and, later, in-house at the Bernhard-Nocht Institute for Tropical Medicine (Hamburg, Germany). WT mice used for the control experiment (see Fig. S1 in the supplemental material) were also bred in-house. Recombination activation gene 1 (RAG1)-deficient mice, which do not have any mature T and B cells (48), and RAG1 and MPO double-deficient (RAG1^−/− × MPO^−/−) mice were bred on a C57BL/6J background in the animal facility of the University Medical Center Hamburg-Eppendorf. Mice were kept under specific-pathogen-free conditions. Female mice aged 7 to 10 weeks were used for experiments.

For infection with blood-stage parasites, mice received either 2 × 10⁴ pRBC from a previously infected mouse or 2 × 10⁶ pRBC from a cryopreserved stock (parasitized blood frozen in a solution of 0.9% NaCl, 4.6% sorbitol, and 35% glycerol). All infections were performed via intraperitoneal (i.p.) injection of a 200-μl dilution of pRBC in phosphate-buffered saline (PBS). Parasitemia was determined with optical microscopy of blood smears from tail vein blood, which were stained with Giemsa or Wright's stain. All experiments were in accordance with local Animal Ethics Committee regulations.

In vivo antibody treatment.

For depletion of neutrophils, mice received i.p. 500 μg of anti-Ly-6G antibody (clone 1A8; BioXCell) 48 and 24 h before infection. Control mice received i.p. 500 μg of rat IgG isotype control antibody (Dianova). In another experiment, mice received an additional shot of depletory or control antibody at day 3 p.i. Additionally, the depletory effect of anti-Ly-6G antibody was verified with flow cytometry of peripheral leukocytes.

Isolation of splenocytes and determination of spleen/tibia ratio.

Spleen weight and tibia length of sacrificed mice were measured. We used the spleen/tibia ratio in order to normalize for age-dependent differences in organ size (57). Spleens were homogenized through a 70-μm cell strainer (BD Biosciences). RBC of homogenized cell suspensions were lysed by the addition of a solution of Tris-HCl (0.01 M) in ammonium chloride (0.14 M). After washing with culture medium (500 ml RPMI 1640, 25 ml fetal calf serum, 5 ml l-glutamine [200 mM], 2.5 ml gentamicin [10 mg/ml]), approximately 1 × 10⁶ cells were stimulated with phorbol 12-myristate 13-acetate (50 ng/ml) and ionomycin (400 ng/ml) at 37°C with 5% CO₂ for 4 h. After the first hour of stimulation, 1 μl/ml monensin (2 mM; BioLegend) was added.

Flow cytometry.

For control of neutrophil depletion, 20 μl of tail vein blood was extracted from every mouse, and neutrophil frequency was measured. RBC were lysed and leukocytes were stained with anti-CD11b (M1/70; BD Biosciences), anti-Ly-6C (HK1.4; BioLegend), and anti-GR-1 (RB6BC5; BD Biosciences). To determine the frequencies of proinflammatory cytokine-producing T cells, splenocytes were stained with anti-CD3 (145-2C11; BD Biosciences), anti-CD4 (L3T4; BD Biosciences), anti-CD8 (53-6.7; eBiosciences), anti-IFN-γ (XMG 1.2; Biolegend), and anti-TNF-α (MP6-XT22, BioLegend). Intracellular cytokine staining was performed by permeabilizing and fixating cells with a FoxP3 staining buffer set (eBiosciences). Data were acquired using an LSR II or Accuri C6 (both from BD Biosciences) setup and analyzed with FlowJo (TreeStar) software. To determine the peripheral neutrophil counts, blood was collected from the submandibular vein, and peripheral neutrophils were quantified using a ProCyte Dx hematology analyzer (IDEEX, Germany).

MPO ELISA.

MPO plasma levels of naive and P. yoelii 17NL-infected mice were determined using a standard capture enzyme-linked immunosorbent assay (ELISA) kit for murine MPO (Hycult Biotech).

Statistical analysis.

Figures were constructed and statistical analysis was performed using Graph Pad Prism software. Statistical significance was calculated using unpaired Student's t test or one-way analysis of variance with Dunnett's posttest.